From: IRC B. <wt...@us...> - 2001-10-15 04:27:23
|
******************************************************************* [03:01] omega (om...@om...) joined #gstreamer. [03:01] <omega> bleagh [03:02] <omega> anyway, that's one more reason to do inline asm <g> [03:03] <omega> in fact... [03:03] <omega> it's even better then to use static local vars in a function [03:04] <omega> because then the "got pointer" becomes a single register, and all the references in asm code to: [03:04] <omega> 8048442: 0f 6f 83 dc ff ff ff movq 0xffffffdc(%ebx),%mm0 [03:04] <omega> instead of requiring a full register to be precalculated for each constant [03:04] <omega> more stuff for codecs.org wiki.... [03:04] Action: omega heads there now [03:05] Action: omega fixes the wiki first [03:08] Action: omega finds that all wiki data is GONE [03:09] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [03:10] Shippou (zbl...@cr...) joined #gstreamer. [03:14] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [03:16] Shippou (zbl...@cr...) joined #gstreamer. [03:20] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [03:22] Shippou (zbl...@cr...) joined #gstreamer. [03:24] hds-sleep (ha...@pc...) left irc: mooooh! [03:36] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [03:51] Shippou (zbl...@ot...) joined #gstreamer. [04:22] Zeenix (zeenix@203.128.13.11) joined #gstreamer. [04:23] <Zeenix> hello everyone [04:24] <omega> yo [04:25] <Zeenix> hi Boss [04:25] <wtay> yo [04:25] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [04:26] <Zeenix> wtay: i wanted to ask you of newcaps func attached to a pad [04:26] <Zeenix> wtay: when is it called the first time ? [04:27] <wtay> when a peer element has set caps on his pad [04:28] <Zeenix> you mean the element that connects its pad to this pad [04:28] <wtay> yes [04:29] <Zeenix> so i'll be using it to change the payload type accordingly [04:29] <wtay> yes [04:30] <wtay> the sinkpad of the rtpsrc should have a template that lists all mime-types is knows about [04:30] <Zeenix> its the only work i need to do in rtpsend [04:30] <wtay> in the newcaps function it checks which one it is and sets the payload type accordingly [04:30] <Zeenix> ok [04:31] <wtay> rtprcv should check the payload and set caps on its src pad [04:31] <Zeenix> you are not telling me i'll need to to do that xml stuff you did in the udpsrc [04:31] <wtay> no, not at all [04:31] <wtay> rtp has his own method of sending datatypes [04:32] <Zeenix> yes, the payload type in each rtp packet [04:35] <Zeenix> in rtpsend we are sequencing, timestamping etc ourselves, i wanted to know if we also have to sequence the packets ourselves in rtprecv [04:35] <Zeenix> i'll check the gphone code for this if you dont know about it [04:36] <wtay> I though rtp did that for you [04:36] <Zeenix> librtp must be responsible for that [04:37] <Zeenix> but that sequencing & timestamping code in rtpsend put me into doubt [04:41] Shippou (zbl...@cr...) joined #gstreamer. [04:41] <Zeenix> wtay: what are doing at this hour here, havent slept the whole night ? [04:42] <wtay> yeah, need to sleep soon.. [04:42] <omega> wtay: you need to wake up soon [04:43] <wtay> soon == at 2pm? [04:43] <omega> heh [04:43] Action: Zeenix is looking for the easiest & quickest way to do doc writing [04:44] <wtay> omega: still busy with the cross compiler? [04:44] <omega> getting closer to being done [04:44] <omega> I have tarballs of glib, mad, and gst, and I can use gstreamer in the embedded enviro (chroot) to play mp3's on the netwinder [04:44] <omega> now I need to do some slimming work and get some numbers [04:44] <wtay> cool [04:45] <omega> problem is that they probably want deltas for things like the _FAST stuff and such ;-( [04:46] <omega> so I may have to back those things out or something [04:46] <wtay> deltas? [04:46] <omega> size of libgst.so after each successive attempt to shrink it [04:47] <wtay> ah I see [04:47] <wtay> of course [04:48] <Zeenix> omega: tell your president that stop the attacks, so that our University is On again <g> [04:48] <omega> pfff [04:48] <omega> are you telling me you *want* to go back to school? bleck <g> [04:48] <Zeenix> omega: yes, as its my last semester & i want it to finish soon [04:48] <omega> hrm, true [04:49] <wtay> yay! 59MB leaked in my test case with bytestream [04:49] <omega> oooh [04:50] <wtay> something bother me though... [04:50] <wtay> s/bother/bothers/ [04:50] <omega> ya think [04:50] <omega> ? [04:50] <wtay> real 1m28.441s [04:50] <wtay> user 0m43.920s [04:50] <wtay> sys 0m40.800s [04:50] <wtay> that's an awfull lot of sys overhead [04:50] <omega> yeah [04:50] <wtay> should try the fast memchunk someday [04:51] <omega> yeah [04:51] <Zeenix> wtay: memchunk ? is it a tool to check these things out [04:52] <wtay> Zeenix: nope, a non locking memchunk as a replacement for gmemchunk [04:52] Action: omega wishes for a real playlist... [04:52] <omega> not this stupid linear xmms thing [04:53] Action: Zeenix thought xmms has playlist support [04:53] <omega> yes, it's mostly useless [04:54] <Zeenix> omega: i use my own mp3 player, no playlists & stuff but i'll do that someday [05:22] <wtay> I'm going to sleep, cya [05:22] Nick change: wtay -> wtay-zZz [05:26] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [05:28] Shippou (zbl...@cr...) joined #gstreamer. [05:36] ajmitch (aj...@p5...) left irc: Ping timeout for ajmitch[p54-max10.dun.ihug.co.nz] [05:39] ajmitch (aj...@p5...) joined #gstreamer. [05:57] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [05:58] Shippou (zbl...@cr...) joined #gstreamer. [06:03] Zeenix (zeenix@203.128.13.11) left irc: Ping timeout for Zeenix[203.128.13.11] [06:05] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [06:12] Shippou (zbl...@cr...) joined #gstreamer. [06:18] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [06:19] Shippou (zbl...@cr...) joined #gstreamer. [06:22] <vektor> ss [06:22] <vektor> anu [06:22] <vektor> one alive? [06:22] <vektor> help? [06:22] <ajmitch> eh? [06:23] <vektor> i am extremely rioml [06:23] <vektor> drunk [06:23] <ajmitch> ah [06:23] <vektor> yhis is reeally weiord [06:23] <vektor> they are watcxhing studay nightluve [06:23] <vektor> live [06:23] <vektor> it's really [06:23] <vektor> wore [06:23] <vektor> dd [06:23] <vektor> tjat io am drunk [06:23] <vektor> thius is kinda cool [06:23] <vektor> qctually [06:23] <vektor> causde i can't tyle [06:23] <vektor> type [06:23] Last message repeated 1 time(s). [06:23] <vektor> cause i'm extremey [06:23] <vektor> drunk [06:23] <ajmitch> good thing this is logged for omega & others to laugh at later ;) [06:23] <vektor> kinca [06:24] <vektor> cool [06:24] <vektor> can [06:24] <vektor> and [06:24] <vektor> ;like [06:24] <vektor> i'be nmever been so furunk [06:24] <vektor> drunk [06:24] <vektor> this is [06:24] <vektor> really [06:24] <vektor> cool [06:24] <vektor> actually [06:24] <vektor> i think [06:24] <vektor> anuwaus [06:24] <vektor> anuways [06:24] <vektor> i don't think [06:24] <vektor> anyways [06:24] <vektor> anyone i ,mean [06:24] <vektor> should [06:24] <vektor> evetr [06:24] <vektor> drunj [06:24] <vektor> yhid [06:24] <vektor> much [06:24] <vektor> reaqlly [06:24] <vektor> this is too much [06:24] <vektor> i think [06:24] <vektor> caus [06:24] <vektor> e [06:24] <vektor> i am [06:25] <vektor> too drunk [06:25] <vektor> tp [06:25] <vektor> type [06:25] <vektor> is it/*98 [06:25] <vektor> i'm ok [06:25] <vektor> iu'm ok [06:25] <vektor> i'm [06:26] <vektor> ok [06:26] Last message repeated 1 time(s). [06:26] <vektor> i'm ok [06:26] <ajmitch> yep [06:33] ShrimpX (marius@131.252.244.168) joined #gstreamer. [06:40] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [06:43] Shippou (zbl...@cr...) joined #gstreamer. [06:52] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [06:53] Shippou (zbl...@cr...) joined #gstreamer. [07:09] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [07:12] Shippou (zbl...@ot...) joined #gstreamer. [07:14] Hal69 (Ha...@ca...) joined #gstreamer. [07:15] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [07:22] Shippou (zbl...@cr...) joined #gstreamer. [07:47] Nick change: taaz-away -> taaz [08:20] Nick change: taaz -> taazzzz [08:24] chillywilly (da...@d4...) joined #gstreamer. [08:41] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [08:44] Shippou (zbl...@ot...) joined #gstreamer. [08:57] xav (xav@AMontpellier-201-1-5-223.abo.wanadoo.fr) left irc: Ping timeout for xav[AMontpellier-201-1-5-223.abo.wanadoo.fr] [09:04] Hal69 (Ha...@ca...) left irc: Ping timeout for Hal69[cable-212-10-237-24.anchorageak.net] [09:06] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [09:07] Hal69 (Ha...@ca...) joined #gstreamer. [09:09] Shippou (zbl...@cr...) joined #gstreamer. [09:09] xav (xav@AMontpellier-201-1-5-223.abo.wanadoo.fr) joined #gstreamer. [09:10] Hal69 (Ha...@ca...) left irc: Ping timeout for Hal69[cable-212-10-237-24.anchorageak.net] [09:12] Hal69 (Ha...@ca...) joined #gstreamer. [09:31] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [09:31] Shippou (zbl...@cr...) joined #gstreamer. [09:35] BBB-zZz (BB...@uc...) left irc: Ping timeout for BBB-zZz[ucu-105-116.ucu.uu.nl] [09:35] BBB-zZz (BB...@uc...) joined #gstreamer. [09:36] walken (wa...@zo...) joined #gstreamer. [09:36] <omega> yo [09:36] <walken> boink [09:36] <walken> how r u going ? [09:36] <walken> doing even [09:36] Action: walken reinstalling debian [09:36] <omega> well, sorta have the cross-compile stuff figured out, but still all sorts of gotchas [09:36] <walken> having a telnet on the debian install disks is cute :) [09:36] <omega> doh [09:37] <walken> hard disk crashed :/ [09:37] <omega> oops [09:37] Action: omega needs to get a case for his server, upgrade, and start rsync'ing his laptop to another drive.... [09:37] Action: walken needs to go raid-1 [09:38] <walken> so I dont care about disk crashes in the future [09:38] <walken> and maybe add a few fans to get rid of the increased heat [09:38] <omega> heh [09:40] <walken> hmmm [09:40] <walken> have to quit [09:40] walken (wa...@zo...) left irc: l8r [09:44] <omega> oops, I managed to put m4 into an infinite loop. [10:08] Hal69 (Ha...@ca...) left irc: [10:34] steveb (st...@no...) joined #gstreamer. [10:54] chillywilly (da...@d4...) left irc: [10:55] omega (om...@om...) left irc: sleep [11:47] Uraeus (csc...@c1...) joined #gstreamer. [11:47] <Uraeus> morning all [11:49] <steveb> hi [11:49] <steveb> can you talk about your new job yet? [11:49] <ajmitch> hey Uraeus [11:50] <ajmitch> hi steveb [11:50] <Uraeus> steveb: not in details yet no [11:50] <Uraeus> ajmitch: morning or evening to you [11:50] <ajmitch> steveb: the question is, can we make Uraeus talk? ;) [11:50] <ajmitch> Uraeus: nice hangover? [11:51] <Uraeus> no, not at all, but on the good side I feel that I can have a fresh start on stomach content now [11:52] <Uraeus> guess I learned that there is a limit for me on the number of Vodka Russians I can handle well [11:53] Action: ajmitch is quite lagged, checking out the www module from cvs to fix up spelling mistakes ;) [11:53] <ajmitch> haha, how much of the night do you remember? [11:54] <Uraeus> ajmitch: luckily most of it, so I feel confident that I didn't ruin my frieds wedding :) [11:54] <ajmitch> steveb: http://gstreamer.net/images/dscn0009.jpg <- you are in this shot, right? [11:54] <ajmitch> Uraeus: hehe [11:56] <steveb> yep :p [11:56] Action: Uraeus hits the shower to hopefully feel more fresh, but at least smell less puke [11:56] Nick change: Uraeus -> Ura_shower [12:07] Nick change: Ura_shower -> Uraeus [12:08] <ajmitch> wb Uraeus! [12:08] <ajmitch> that was quick [12:11] <Uraeus> well I don't feel for long showers when I have a headache and feel nauseus [12:11] <ajmitch> hehe [12:13] Action: Uraeus goes to rest on the sofa for a bit, bbl [12:16] Action: ajmitch wonders who wrote the FAQ [12:20] Action: Uraeus thinks ajmitch should not critize the person who wrote the FAQ today [12:22] <ajmitch> Uraeus: can i commit spelling fixes? ;) [12:23] <Uraeus> ajmitch: sure ;) [12:24] <ajmitch> cool thx :) i noticed some in the FAQ, it bugged me ;) [12:24] <Uraeus> guess there where a few in the FAQ ;) [12:25] <ajmitch> yeah, apart from that it's pretty good ;) [12:26] <ajmitch> now i can see what a mess i've made of it (and fix it ;) [12:27] <steveb> GStreamers should be GStreamer's [12:28] <ajmitch> yeah, i thought about that one [12:28] <ajmitch> hmm, what needs to be done to update the pages on the webserver? [12:28] <ajmitch> they update automatically, ro do we need to go to a php page? [12:30] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [12:30] Action: ajmitch sees a cvs-update.php file in the htdocs dir [12:31] Shippou (zbl...@cr...) joined #gstreamer. [12:31] Action: ajmitch wonders whether he should risk running it ;) [12:33] <ajmitch> someone else can dump in apostrophes, i'm going to bed ;) [12:33] <ajmitch> night [12:35] <steveb> see ya [12:37] <Uraeus> night [12:37] Nick change: Uraeus -> Ura_brb [12:38] Zeenix (ze...@ps...) joined #gstreamer. [12:56] <Zeenix> hmm.... wtay still sleepy [13:02] Zeenix (ze...@ps...) left irc: I pay 25 Rs( 0.5$ )/hour [14:03] bstard (Lor...@a2...) joined #gstreamer. [14:33] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [14:35] Shippou (zbl...@cr...) joined #gstreamer. [15:42] xav (xav@AMontpellier-201-1-5-223.abo.wanadoo.fr) left irc: Ping timeout for xav[AMontpellier-201-1-5-223.abo.wanadoo.fr] [15:49] xav (xav@AMontpellier-201-1-5-223.abo.wanadoo.fr) joined #gstreamer. [16:03] Action: BBB-zZz is away: I'm busy [16:03] Action: BBB-zZz is back (gone 00:00:03) [16:03] Nick change: BBB-zZz -> BBB [17:17] thomasvs (th...@ur...) joined #gstreamer. [17:36] BBB (BB...@uc...) left #gstreamer. [17:44] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [17:49] Shippou (zbl...@ot...) joined #gstreamer. [18:05] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [18:06] Shippou (zbl...@cr...) joined #gstreamer. [18:15] <vektor> I will never, never, ever drink again. [18:16] <bstard> lol! [18:16] <bstard> you say that now... [18:16] <bstard> next weekend come this memory test ;) [18:32] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [18:34] Shippou (zbl...@ot...) joined #gstreamer. [18:44] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [18:47] Shippou (zbl...@cr...) joined #gstreamer. [18:48] vishnu (jo...@cx...) joined #gstreamer. [18:51] <steveb> wtay-zZz: can i pick your brains? [19:03] wingo (wi...@rd...) joined #gstreamer. [19:03] <wingo> yo [19:05] <steveb> yo [19:05] <wingo> good job with the apostrophes :) [19:08] <steveb> gee, thanks [19:09] <steveb> so how does publishing work? [19:09] <wingo> cvs update -dP [19:09] <wingo> in the web directory [19:10] <wingo> we tried to write a script to do it for us, but perms held us back [19:10] <wingo> i suppose a cron job would do the trick though... [19:23] <wingo> ok, so in theory it will update now at 2:43 in the morning, PDT i guess [19:24] <wingo> and i ran it manually now [19:27] <steveb> cool [19:28] Ura_brb (csc...@c1...) left irc: Client Exiting [19:46] <vishnu> wingo? [19:46] <wingo> vishnu? [19:47] <vishnu> how do i login to the status update thingy? [19:47] <wingo> one moment... [19:47] <vishnu> lib/db_setup.php [19:47] <vishnu> ?? [19:48] <steveb> we need a status key on that page again [19:48] <wingo> you have shell access, no? less /home/groups/g/gs/gstreamer/htdocs/lib/db_setup.php [19:48] <wingo> steveb: ? [19:48] <steveb> so we know what the colours mean [19:49] <wingo> and the page is gstreamer.net/status/admin.php [19:49] <vishnu> oh .. heh [19:49] <wingo> steveb: agreed... [19:50] <vishnu> oh cool, ok [19:56] <wingo> you've added some screenshots :) [19:56] <vishnu> wingo: yah :-) [19:57] <wingo> bah. /me has to write an essay on the quijote. [19:57] <wingo> i like reading more than writing :) [19:58] <vishnu> quijote ? [19:59] thomasvs (th...@ur...) left irc: [x]chat [20:00] <steveb> quijote == quiote ? [20:00] <wingo> don quijote? [20:00] <wingo> de la mancha? [20:00] <vishnu> oh ok :-) [20:00] <wingo> :) [20:13] <wingo> i would rather learn more about programming an editor than editing [20:13] Action: wingo looks on e-lisp intro site, doh [20:36] Nick change: wtay-zZz -> wtay [20:36] <wtay> hullo [20:37] <wingo> hey. [20:39] <steveb> hi [20:39] <wtay> yo [20:40] <steveb> i'd like to make a couple of changes to the -launch parser [20:40] <wtay> yes? [20:40] <steveb> instead of exiting on error, i'd like to return an error code of 0 [20:41] <wingo> rather than 1? [20:41] <wtay> how do you mean? [20:41] <steveb> wingo: you mean -1? [20:42] <wingo> yeah :) [20:42] <steveb> wtay: s/exit(-1)/return 0/ [20:42] <wtay> oh [20:42] <steveb> then anything that calls gst_parse_launch_cmdline has to check the return code (and maybe quit) [20:42] <wtay> why not return -1? [20:43] <wtay> steveb: yeah [20:43] <wingo> or some gst error code, even [20:43] <steveb> -1 would be ok, that was one thing i was wondering [20:44] <wtay> you could define a couple of GST_PARSE_ERROR_* and use them [20:44] <wingo> you could differentiate between bad syntax and a gstreamer error [20:44] <wtay> yup [20:45] <steveb> either way, if the parser returns an error code, the pipeline may be in an unstable state and the app should act accordingly [20:45] <wtay> yes [20:46] <wtay> we also have a gst_element_error function.. not sure how we are going to use that [20:46] <steveb> ok, the other thing i wanted to do was make the parser pluggable. now that dparams is out of core I had to take out the @freq=440 syntax [20:47] <wtay> well... [20:47] <steveb> so i'd like to be able to extend the parser with a simple callback registration thingy [20:48] <steveb> callbacks are stored in a GSList and are called one at a time until one of them says that it has handled the token [20:48] <wtay> uhm [20:49] <ajmitch> morning [20:49] <wtay> evening.. [20:49] <wingo> lunch? [20:49] <BeNOW> afternoon. [20:50] <steveb> for a start the callback could have 'element' scope only. the arg=val might be the only existing syntax to be made a callback [20:50] <wingo> what other applications might this have, beyond dparams? [20:51] <wtay> we might want to rethink -launch [20:51] <steveb> weeell, a plugin could register its own special syntax if arg=val doesn't cut it. thats about all i can think of [20:53] <steveb> i don't want to rewrite the parser. all i want is to set dparams from -launch [20:53] <wingo> how would we rethink -launch? besides getting -compprep and the like to work ;) [20:54] <wtay> -launch is a bit of a mess now IMO [20:54] <wingo> steveb: is there somthing mutually exclusive between dparams and properties? [20:55] <wtay> I mean, adding a plugin system to it is not going to be easy I think [20:55] <steveb> wingo: what do you mean? [20:55] <wingo> steveb: for each dparam, could you make a property? or would that be a mess? [20:56] <wingo> that would serve for setting constant values, no? [20:57] <steveb> i've thought of automagically exposing dparams as properties but i can't think of a correct behaviour for interaction between setting via dparams and setting via props [20:58] <steveb> if a prop is set in the middle of a running pipeline, does it just set the param to that value until the dparam object resets it to what it wants? [20:59] <wingo> how about, 'undefined interaction' [20:59] <wingo> yeah [20:59] <wingo> if you set it like that, maybe you're a moron anyway? [20:59] <wingo> or, maybe you cause the dparam to stop [20:59] <wingo> which would make more sense [20:59] <steveb> or could props be used to set the default value only, so any prop setting is ignored when pipeline is running [20:59] <wingo> also possible [21:00] <wingo> i think stopping the dparam makes more sense, though [21:00] <wingo> for apps that change things via properties at runtime, it would allow them to work [21:00] <wingo> and be ignorant of dparams [21:01] <steveb> so setting a prop will replace the dparam model with a constant one? i guess that could work [21:04] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [21:07] Shippou (zbl...@cr...) joined #gstreamer. [21:27] Nick change: taazzzz -> taaz [21:28] Action: taaz pokes wingo [21:28] <ajmitch> taaz: don't be a bully [21:28] <taaz> oh? you want a piece of me too? [21:28] Action: taaz pokes ajmitch [21:29] <ajmitch> heh [21:29] <ajmitch> ouch! [21:29] Action: ajmitch pokes taaz [21:30] Action: steveb pokes himself [21:32] <wingo> ok, ok, back from lunch... [21:32] <wingo> what up taaz. yo. [21:33] <taaz> debian archive gone. taaz not like this. [21:33] <wingo> oh oops. let me put that back :-\ [21:33] <taaz> ;) [21:36] <wingo> just the 0.2.1, or do you want 0.2.0 as well [21:36] <taaz> eh? [21:36] <wingo> releases/0.2.1/debian, or releases/0.2.0/debian as well [21:37] <taaz> i forget how i had it setup [21:38] <taaz> i had a symlink to latest from releases/debian i think [21:38] <taaz> actually... i cant remeber at all ;) [21:38] <wingo> you had a symlink :) [21:38] <wingo> so do you just want one debian dir, then? or per-release [21:39] <taaz> is the idea that these huge binary releases go into cvs too? [21:39] <wingo> no [21:39] <wingo> imho [21:39] <wingo> there are only a few source tarballs in cvs [21:40] <wingo> really old ones at that [21:40] <wingo> that dir will be out of cvs control, imo [21:40] <wingo> that ok? [21:40] <taaz> fine with me [21:41] Action: taaz stares out window at the rain [21:41] <wingo> yeah. east coast kids :) [21:43] <taaz> so how is this going to be setup? [21:43] <wingo> ok, debian archive back. do you mind fixing the perms to 664? [21:43] <taaz> all dl info on downloads page? [21:43] <wingo> i think [21:43] <wingo> i just made a /releases/debian dir, no symlink [21:44] <taaz> dude, you own all those files now [21:45] <wingo> ok, /me is a moron :) [21:47] <wingo> i think i've changed all of them to 664, so things should be good now i theory [21:47] <wingo> in theory even [21:54] <vektor> never, ever drink again. [21:54] <wingo> hahahahaha!!!!! [21:55] <wingo> vektor's got a hangover, vektor's got a hangover [21:55] <vektor> i'm as sick as a dog [21:55] <wingo> that sucks. [21:57] <wingo> well, all you can do is wait. [21:57] <wingo> and looking at the computer helps not at all :-\ [22:07] xav (xav@AMontpellier-201-1-5-223.abo.wanadoo.fr) left irc: Ping timeout for xav[AMontpellier-201-1-5-223.abo.wanadoo.fr] [22:09] xav (xav@AMontpellier-201-1-5-223.abo.wanadoo.fr) joined #gstreamer. [22:09] <wtay> vektor: I should mail the IRC log to -devel <grin> [22:09] <ajmitch> wtay: where is the irc log going at the moment? [22:10] <wtay> ajmitch: to /dev/null :) [22:10] <ajmitch> wtay: hehe :) [22:10] <ajmitch> wtay: why not gst-daily? [22:10] <wtay> whee 72 testcases for bytestream and no leaks [22:10] <wtay> ajmitch: 'cause I lost my config for that [22:10] <ajmitch> wtay: bah [22:10] <ajmitch> wtay: i've got irc logs being mailed from here [22:11] <wtay> I should recreate those scripts.. [22:11] <wtay> ajmitch: exim? [22:11] <ajmitch> wtay: yup [22:11] <wtay> ajmitch: mind sending me the script? [22:11] <ajmitch> wtay: sure, i think it was your script anyway ;) [22:11] <wtay> oh [22:11] <wtay> right :) [22:12] <wtay> my backup :) [22:17] thomasvs (th...@ad...) joined #gstreamer. [22:18] <thomasvs> wingo: do I just grab www from cvs to make changes ? [22:18] <wingo> thomasvs: yes [22:19] <thomasvs> what's the module's name ? [22:19] <ajmitch> www ;) [22:19] <thomasvs> heh ok [22:19] <wingo> if you want it to be updated immediately, run /home/groups/g/gs/gstreamer/www-update.sh [22:19] <wingo> otherwise it will get done overnight :) [22:20] <ajmitch> ah, so that's what you do ;) [22:20] <wingo> me? you mean cron :) [22:21] <wingo> actually, we have a contract with paul vixiehimself to update the site every night :) [22:21] <wtay> heh [22:22] <wtay> paul vixie just agreed to run a daily script here too :) [22:22] <wtay> let's try it.. [22:23] ajmitch (aj...@p5...) left irc: Ping timeout for ajmitch[p54-max10.dun.ihug.co.nz] [22:25] <vishnu> "whee 72 testcases for bytestream and no leaks" -- congrats [22:25] <vishnu> wtay: does it handle events too? [22:27] Action: thomasvs needs a new computer [22:27] Action: vishnu gives thomasvs a new computer, gratis [22:27] <wtay> vishnu: nope, not yet [22:28] <thomasvs> vishnu: i wish ;) [22:28] <wtay> doh! [22:28] Action: wtay sent the irc_log to the wrong list [22:28] <thomasvs> heh [22:28] <thomasvs> wtay needs a new brain [22:28] <thomasvs> vishnu gives wtay a new brain, gratis [22:29] <thomasvs> man, www includes all of the tarballs ? [22:29] Action: wtay should have known that there is no list called gst-daily [22:29] <wingo> wtay i do that all the time! [22:29] Action: vishnu gives wtay a new brain and sends the bill to thomasvs [22:29] <thomasvs> damn, and all of erik's slides as well [22:29] <wingo> at least it's not a real project, gst would be fed up with us [22:30] <wingo> thomasvs: not all the tarballs... [22:30] <wingo> just a few of the old ones [22:30] <thomasvs> 20 fucking mb ! [22:30] <thomasvs> sorry [22:30] dschleef (ds...@c7...) left irc: Ping timeout for dschleef[c799921-a.pinol1.sfba.home.com] [22:30] <thomasvs> that's a lot [22:30] Action: wtay waits for the fetchmail timeout.. [22:30] <thomasvs> and i just wanted to correct three little mistakes ;) [22:31] <wingo> doh... i had to trim out a bunch, but being on broadband makes you lose touch... [22:32] <wtay> yay! irc_log back in action [22:32] <thomasvs> wingo: well, it only took about 1 minute ;) [22:32] <thomasvs> ok, fixed [22:32] <thomasvs> wingo: nice work btw ;) [22:32] <thomasvs> where can I fix the plugin status ? [22:33] <thomasvs> directly in mysql ? [22:34] <wtay> thomasvs: that were 4 mistakes :) [22:34] <taaz> eh? why do people wait for fetchmail? if you run it again it wakes the daemon and fetches immediately [22:34] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [22:35] <thomasvs> wtay: yeah, i intended to fix three and caught another one I didn't see before [22:35] <thomasvs> just minor stuff [22:35] Shippou (zbl...@cr...) joined #gstreamer. [22:35] <wingo> thomasvs: thanks. go to gstreamer.net/status/admin.php and log in using passwords in /home/groups/g/gs/gstreamer/htdocs/lib/db_setup.php [22:36] <wingo> the status user/pass for status, etc. gstreamer.net/admin.php is the overview admin page :) [22:37] <wingo> we could remove passwords if they're too much of a pita [22:37] Action: thomasvs loves the smell of a new galeon in the morning [22:37] <thomasvs> wingo: I don't mind, but maybe some secure use of ssh keys would be nice somehow [22:38] <thomasvs> but it's always a pain to get it done right so this is good enough for me ;) [22:38] <wingo> there are two extremes, ssh and the wiki [22:39] <wingo> i'm not sure that status is important enough to somehow merit ssh ;) [22:39] <thomasvs> ok, I'm guessing db_setup is not in the cvs'd www dir ? [22:39] <wingo> right :) [22:41] <thomasvs> ok, that works [22:43] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [22:44] dschleef (ds...@c7...) joined #gstreamer. [22:44] Shippou (zbl...@cr...) joined #gstreamer. [22:47] <wingo> it seems that we need to collapse 'module' and 'type' into one column in the status table [22:48] <wingo> in the audio category it's irrelevant [22:48] <wingo> maybe just call it 'name' or something, and call 'component' 'description' instead [22:49] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [22:50] ajmitch (aj...@p5...) joined #gstreamer. [22:50] evil_monk (trevo@208.141.162.68) joined #gstreamer. [22:51] <ajmitch> damn [22:56] <wingo> ok, i'm going to add a column called 'name' to the status table, to be a dupe of 'type' [22:56] <wingo> we can figure out what to do with type and module later [22:58] <wingo> um, if i can figure out the sql to copy a column... [23:03] <steveb> wingo: could you make the status drop box sticky? [23:03] <wingo> ? [23:04] <steveb> so when you choose 'core plugins', it is still selected when the page loads [23:04] <wingo> oh ok [23:04] Shippou (zbl...@ot...) joined #gstreamer. [23:04] <taaz> what's this </a> commit? [23:05] <taaz> oh never mind... [23:11] <wtay> wingo: I hope the media/ files are not in cvs... [23:11] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [23:12] Shippou (zbl...@ot...) joined #gstreamer. [23:12] <wingo> wtay: they aren't [23:13] <wtay> phew.. :) [23:13] <wingo> steveb: fixed, i think. the status also should look a bit cleaner... [23:13] <wingo> no key yet though [23:13] thomasvs (th...@ad...) left irc: Client Exiting [23:14] <steveb> looks good. you should probably do the same to the drop box at the bottom ;) [23:14] <steveb> or remove it entirely [23:15] <wingo> oh yeah... [23:22] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [23:23] Shippou (zbl...@cr...) joined #gstreamer. [23:27] <wingo> jeez sorry about the email spamming [23:27] <wingo> it still doesn't look right [23:28] <wingo> i'll get a test site set up on my server when i get home, or maybe doing that now would be better... [23:29] <wingo> thing is you can't get to mysql.sourceforge.net outside of sf's shell servers [23:29] Action: wtay wonders how wingo is going to write 5 in french.. :) [23:29] <wingo> did i spell 4 right? [23:29] <wtay> yeah [23:30] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [23:30] <wingo> schweet! :) cinque six i did my time in french class :) [23:30] <wtay> http://www.helio.org/education/french/counting/0-9.html [23:30] <wtay> wingo: wrong :) [23:30] Shippou (zbl...@cr...) joined #gstreamer. [23:31] <wingo> damn :) [23:31] <wtay> hehe [23:31] <wingo> i did do my time though regardless :) [23:31] <wingo> ideas on the key? steveb, anyone? [23:32] <wingo> no labels at all, perhaps? [23:37] <wtay> uhm.. that's a pretty neat french tutorial.. [23:45] ezequiel (eze...@ho...) joined #gstreamer. [23:45] <ezequiel> hi [23:45] <wtay> yo [23:46] <ezequiel> español? [23:46] <wingo> si [23:46] <ezequiel> excelente [23:46] <wingo> aunque no tengo el teclado adecuado [23:46] <ezequiel> ok [23:46] <ezequiel> sabés como hacer andar gstreamer en mandrake? [23:46] <wingo> no, uso debian ;) [23:46] <ezequiel> ok [23:46] <wingo> a lo mejor los rpm te funcionaran [23:47] <ezequiel> probé con los rpm y compilando [23:47] <wingo> pero lo mejor es hacerlo de los codigos fuente [23:47] <wingo> y no te saliste? [23:47] <ezequiel> no [23:47] <wingo> ay. con que error? [23:47] Action: wtay looks for a spanish tutorial [23:48] <wingo> heh :) [23:48] <ezequiel> el core puede compilar ok [23:48] <ezequiel> pero no anda [23:48] <ezequiel> :| [23:48] <wingo> usaste autogen.sh os configure, etc? [23:48] <wingo> s/os/o [23:48] <ezequiel> configure [23:48] <wingo> hm. [23:48] <wingo> y bajaste el 0.2.1, no? [23:48] <ezequiel> sip [23:49] <wingo> has hecho 'gstreamer-register'? [23:49] <ezequiel> sip [23:49] steveb (st...@no...) left irc: advance! [23:49] Zygo (zbl...@24...) joined #gstreamer. [23:49] <wingo> y funcion'o? [23:49] <ezequiel> no [23:49] <wingo> que pas'o? [23:49] <ezequiel> nada [23:50] <ezequiel> dice que hay un caño roto ;) [23:50] Zygo (zbl...@24...) left irc: Read error to Zygo[246.chipworks.com]: Connection reset by peer [23:50] <wingo> que es un ca~no? no es mi lengua nativa :) [23:50] <ezequiel> una tubería, pipe [23:50] <wingo> oh [23:50] <wingo> bueno... [23:51] <ezequiel> preferís hablar inglés? [23:51] <ezequiel> creo que puedo [23:51] <wingo> da igual [23:51] <wingo> como quieras :) [23:51] <ezequiel> ok [23:51] <ezequiel> let's talk in english [23:51] <wingo> pero con ingles a log mejor los demas te pueden ayudar tambien [23:51] <wingo> ok [23:51] <wingo> wtay: there? [23:52] <wtay> si? [23:52] <ezequiel> yup [23:52] <wingo> 'broken pipe' while running gstreamer-register [23:52] <wingo> 0.2.1. [23:52] <ezequiel> yup [23:52] <ezequiel> not in -reg [23:52] <ezequiel> in -launcher [23:52] <wtay> broken pipe... hmm [23:52] <wtay> oh [23:52] <ezequiel> ERROR: no pipeline description found on commandline [23:52] <wingo> what pipeline? [23:52] <wingo> oh... [23:52] <wtay> oh, right [23:52] <wingo> so it might be ok. [23:53] <wingo> try gstreamer-launch disksrc location=foo.mp3 ! mad ! osssink [23:53] Action: wtay was going to ask what he put into the pipe :) [23:53] <ezequiel> I've no idea [23:53] <wingo> see gstreamer.net/wiki/GstLaunchQuickStart [23:53] <wingo> er [23:53] <wingo> see gstreamer.net/wiki/index.php?GstLaunchQuickStart [23:53] <wtay> or gstreamer-launch fakesrc ! fakesink [23:54] <ezequiel> ok, thanks [23:54] <wingo> no prob [23:54] <wingo> (lema) :) [23:54] <vektor> oh mny god i am not recovered [23:55] <wingo> poor vektor [23:55] <wtay> vektor: best cure is to not stop drinking :) [23:56] <wingo> ezequiel: so now you can write the essay i need to write on the quijote, eh? [23:56] <wingo> :) [23:56] Action: wingo puts off his work [00:00] --- Mon Oct 15 2001 [00:04] <ezequiel> ** CRITICAL **: file gstplay.c: line 126 (gst_play_init): assertion `priv->audio_element != NULL' failed. [00:04] <ezequiel> what is that? [00:05] <wingo> what was the pipeline? [00:06] <ezequiel> how can I know it? [00:07] <wingo> what was the command line? [00:08] <ezequiel> gstreamer-launch [00:08] <wingo> gstreamer-launch (pipeline definition goes here) [00:09] <wingo> gstreamer-launch fakesrc ! fakesink [00:09] <ezequiel> ok [00:09] <wingo> does that work? [00:09] <ezequiel> Couldn't create a 'fakesin', no such element or need to run gstreamer-register? [00:09] <ezequiel> I ran gstream-register [00:10] <wingo> perhaps you misspelled it? [00:10] <ezequiel> no [00:10] <wingo> fakesink with a k [00:10] <wingo> oh [00:10] <ezequiel> I copy&paste [00:10] <ezequiel> wait me [00:10] <wingo> but you missed the k, it seems [00:11] <ezequiel> I'll restart somethings [00:11] <ezequiel> ok [00:11] <ezequiel> I'll try again [00:11] <ezequiel> oh [00:11] <ezequiel> wow [00:11] <vektor> the best idea i had was to brush my teeth. nothing like getting the taste of puke out to make you feel better. [00:11] <ezequiel> now my console is full of: fakesink: ******* (fakesink0:sink)< (0 bytes) [00:11] <ezequiel> fakesrc: ******* (fakesrc0:src)> [00:12] <wtay> good [00:12] <ajmitch> vektor: feeling ok yet? ;) [00:13] <vektor> no [00:13] <vektor> almost [00:13] <vektor> :( [00:13] <ezequiel> ok [00:14] <ezequiel> gtg [00:14] <ezequiel> restart [00:14] <ezequiel> thx [00:14] <wingo> ezequiel: now try gstreamer-launch disksrc=tu/mp3/aqui.mp3 ! mad ! osssink [00:14] <wingo> cya [00:14] <wtay> cya [00:14] <wtay> wingo: could fail if libmad was missing... [00:15] ezequiel (eze...@ho...) left irc: Read error to ezequiel[host085155.arnet.net.ar]: EOF from client [00:15] <wingo> true [00:15] <wingo> so much easier than typing the mpg123 mess though :) [00:17] <wtay> nice job on the website btw. [00:19] <wingo> thanks. [00:19] <wtay> so, the procedure is now to make changes in cvs and run some script them to update the site? [00:19] <wtay> s/them/then/ [00:20] <wtay> or does it also happen automagically? [00:21] <wingo> it does happen automagically at some time in the middle of the night [00:21] <wingo> or, you can run /home/groups/g/gs/gstreamer/www-update.sh [00:21] <wingo> or, cvs update -dP in the root web directory [00:21] <wtay> how would we do a new release then? [00:21] <wingo> of what? [00:21] <wingo> status stuff, you mean? [00:21] <wtay> like, we would first work on the new site for a few days and then we would activate the changes [00:22] <wtay> no, a new release, with release notes etc.. [00:22] <wingo> oh. [00:22] ezequiel (ezequiel@200.45.177.99) joined #gstreamer. [00:22] <ezequiel> re [00:22] <wingo> well, you would make a new release dir [00:22] <wingo> and then... [00:23] <wtay> wingo: automatic updates are fine for small updates but not for large ones.. [00:23] <wingo> i dunno. [00:23] <wingo> wtay: good point. [00:23] <wtay> I remember we have directories for different releases though.. [00:23] <wingo> what all does a release entail? i haven't been around for one. [00:23] <wingo> yes [00:23] <wingo> i think there's even something of a web design do [00:23] <wingo> c [00:23] <wtay> we could work on stuff in just one of the dirs.. [00:24] <wingo> gstreamer.net/README.website [00:24] <wtay> and then checkin a symlink or so to activate it.. [00:24] <wtay> ok [00:24] <wingo> it's old, but still relevant [00:24] <taaz> heh... that's a bad sign. wingo not around for any releases yet... [00:25] <wingo> :) [00:25] <wtay> better be here when we do our next release then :) [00:25] <wingo> of course :) [00:25] <wtay> ah :) [00:25] <taaz> how many years did you plan on being here wingo? ;) [00:25] <wingo> heh [00:26] <wtay> wingo: yes, you can answer the question :) [00:26] <wingo> yes, but i prefer not too :) [00:26] <wingo> to even [00:26] JOSF (jo...@li...) joined #gstreamer. [00:27] <JOSF> yo [00:27] <wingo> yo. [00:27] <wtay> yo [00:27] <JOSF> yeah [00:27] <JOSF> yo men [00:27] <JOSF> I'm here [00:27] <wtay> dude [00:27] <JOSF> yeah [00:28] <wtay> like, yeah [00:28] <JOSF> cool [00:28] <wtay> uhl [00:28] <wtay> uhm [00:28] <JOSF> check it out ! [00:29] <JOSF> so? all cool ? [00:29] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [00:30] <JOSF> there he goes... [00:30] <wtay> sorta, yes [00:30] <JOSF> anyway... [00:30] <ezequiel> I'll try building [00:30] Shippou (zbl...@cr...) joined #gstreamer. [00:30] <wtay> I'm going to try something nasty.. [00:30] <JOSF> yo Shippou [00:31] <JOSF> cool: nasty ! [00:32] <JOSF> KDE 40% Gnome 24.5%, not cool :( [00:32] <JOSF> (likings in a poll at LJ) [00:32] <ezequiel> what are that percentiles? [00:32] <wtay> windows 95% other 5% [00:33] <ezequiel> shit [00:33] <JOSF> lol wtay [00:33] <JOSF> Workbench 0.00000000000000000000001 % [00:33] <JOSF> not cool [00:33] <wtay> ADos? [00:33] <JOSF> yo [00:33] <JOSF> wtay: version me ;) [00:34] <JOSF> cool ? [00:34] <wtay> no kidding =) [00:34] <JOSF> yeah :) [00:34] <ezequiel> wow [00:35] <JOSF> It's hooke up to my Linux box. I have the keaboard on my knees, two monitors up n runnin. I am in fuckingheaven :-) [00:35] <ezequiel> does you know what amiga means in spanish? [00:35] <JOSF> girl-friend ? [00:35] <JOSF> or is that amica ? [00:35] <ezequiel> just friend [00:35] <ezequiel> girl-friend is novia [00:35] <JOSF> so, what is amigo ? [00:35] <ezequiel> friend, but male [00:35] <JOSF> i see [00:36] <JOSF> Hola ! :) [00:36] <ezequiel> Hola ! [00:36] <ezequiel> I'm leernin english ;) [00:36] <JOSF> ezequiel: cool, I will learn spanish soon. [00:37] <JOSF> atm I am learning italian [00:37] <wtay> from aye boook! [00:38] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [00:39] <JOSF> wtay: yeah, book and tapes [00:39] <JOSF> there he goes... [00:39] <ezequiel> wow [00:39] <ezequiel> how can I tell to make that I'll compile for a 686? [00:40] <JOSF> In which language ? ;-) [00:40] <ezequiel> er.... [00:40] <JOSF> -arc686 I think [00:40] <JOSF> or so [00:40] <ezequiel> ok [00:40] Shippou (zbl...@cr...) joined #gstreamer. [00:40] <JOSF> man gcc or so ;-) since I dont know it [00:40] <ezequiel> nope [00:40] <JOSF> KMail fav' email client -- berk [00:41] <ezequiel> evolution! [00:41] <JOSF> They never used UMS/Pint, HA ! [00:41] <JOSF> nah, this is a poll [00:41] <ezequiel> JOSF, you got to evolute! [00:41] <ezequiel> use evolution! [00:41] <JOSF> fav' www browser -> konqueror, lol [00:41] <ezequiel> galeon! [00:41] <JOSF> ezequiel: No, I tried evolution. I prefer UMS/Pint [00:41] <JOSF> Opera [00:42] <wingo> um, mutt? :) [00:42] <JOSF> Or Mozilla [00:42] <ezequiel> ok [00:43] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [00:43] <JOSF> UMS is supercool. I use it from my IRC client or from my desktop or my text-editor through IPC scripts. [00:43] <JOSF> there he goes... [00:43] <ShrimpX> wtay: do you know x86 assembly? [00:44] <wtay> ShrimpX: somewhat.. [00:44] <ShrimpX> i'm having hella trouble understading some stuff.... [00:45] <wtay> like? [00:45] <ShrimpX> look here: http://web.pdx.edu/~marius/dl/file.s [00:45] <wtay> k [00:45] <ShrimpX> that's the gcc generated asm for http://web.pdx.edu/~marius/dl/file.c [00:45] <wtay> ok [00:46] <ezequiel> does anyone use mandrake? [00:46] <wtay> what about it? [00:47] <ShrimpX> well, basically, i'm having trouble understanding why the first 6 instructions in main: are doing, before printf is called [00:47] <ShrimpX> i know that $.LC0 is pushed so printf finds it... [00:47] <wtay> ShrimpX: well; you're allocating an array on the stack so it has to grow [00:48] <ShrimpX> ok. but why the numbers $68 and $-12? [00:50] <wtay> ShrimpX: I would say it needs room to push the args for printf and such [00:51] <ShrimpX> cool, i think i'm starting to figure it out. [00:51] Action: ShrimpX diagrams the stack [00:52] <ShrimpX> do you know what the difference is between addl $-12, %esp and subl $12, %esp? [00:52] <wtay> hmm, no.. [00:54] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [00:54] <vektor> yeah, if you're generating code, it's easier to just generate addl instructions rather than both addl and subl ? [00:54] Action: vektor has no clue [00:55] <ShrimpX> only that code uses both addl and subl :) [00:55] Shippou (zbl...@cr...) joined #gstreamer. [00:55] Action: ezequiel hola [00:55] <ezequiel> I was just probing the cmd [00:55] <wtay> ShrimpX: they seem to use addl for accessing local vars and subl to manipulate the stack.. [00:55] <ShrimpX> wtay: thanks for the input [00:55] Action: wtay has no clue [00:56] <ShrimpX> hehe, me neither [00:56] <ShrimpX> but i think i'm getting there [00:56] <ezequiel> I can't compile the lame of gstreamer [00:57] <ezequiel> /bin/sh ../../../libtool --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I../../.. -I/usr/include/glib-1.2 -I/usr/lib/glib/include -D_REENTRANT -I/usr/include/gtk-1.2 -I/usr/include/glib-1.2 -I/usr/lib/glib/include -I/usr/X11R6/include -I/usr/include/gnome-xml -O6 -Wall -I../../.. -I../../../include -c gstlame.c [00:57] <ezequiel> gcc -DHAVE_CONFIG_H -I. -I. -I../../.. -I/usr/include/glib-1.2 -I/usr/lib/glib/include -D_REENTRANT -I/usr/include/gtk-1.2 -I/usr/include/glib-1.2 -I/usr/lib/glib/include -I/usr/X11R6/include -I/usr/include/gnome-xml -O6 -Wall -I../../.. -I../../../include -c gstlame.c -fPIC -DPIC -o gstlame.o [00:57] <ezequiel> In file included from gstlame.c:35: [00:57] <ezequiel> gstlame.h:95: field `lgf' has incomplete type [00:57] <ezequiel> gstlame.c:721: warning: `gst_lame_change_state' defined but not used [00:57] <wtay> ezequiel: what lame version? [00:58] <ezequiel> ?? [00:58] <ezequiel> don't know [00:58] <ezequiel> I think 1.89beta [00:58] <ezequiel> I haven't lame [00:59] <ezequiel> sorry [00:59] <ezequiel> 3.89 [00:59] <wtay> uhm [00:59] <ezequiel> I'll compile lame now [01:00] <wtay> dunno what version we require.. [01:00] <ezequiel> how can I compile lame? [01:01] <ezequiel> the gstreamer lame [01:01] <wtay> you need the right version of lame [01:01] <ezequiel> which one? [01:01] <wtay> I don't know :( [01:02] <wtay> lemme find it.. [01:02] <ezequiel> how can I passby lame? [01:04] <wtay> are you building from CVS? [01:04] <vektor> damn, sox's usage kicks: [01:04] <vektor> sox: Usage: [ gopts ] [ fopts ] ifile [ fopts ] ofile [ effect [ effopts ] ] [01:04] <vektor> wow [01:04] <vektor> that's so informative [01:05] <ezequiel> no [01:05] <ezequiel> I'm building from the 0.2.1 tar.gz [01:05] JOSEF (jo...@li...) joined #gstreamer. [01:05] JOSF (jo...@li...) left irc: Ping timeout for JOSF[line-120-54.dial.matav.net] [01:06] <wtay> ezequiel: you'll have to edit the makefile I'm afraid [01:06] <ezequiel> ah [01:06] <ezequiel> thx [01:06] <ezequiel> ;) [01:06] Nick change: JOSEF -> JOSF [01:08] wingo (wi...@rd...) left irc: My damn controlling terminal disappeared! [01:16] ezequiel (ezequiel@200.45.177.99) left irc: Aplicación Saliendo [01:41] <wtay> bedtime for me, cya [01:41] Nick change: wtay -> wtay-zZz [01:43] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [01:46] Shippou (zbl...@cr...) joined #gstreamer. [02:04] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [02:04] Shippou (zbl...@cr...) joined #gstreamer. [02:04] JOSF (jo...@li...) left #gstreamer. [02:09] wingo (wi...@rd...) joined #gstreamer. [02:28] <wingo> hey kids, we now have a gstreamer test site! [02:28] <wingo> ambient.2y.net/gstreamer/ [02:28] <wingo> i dumped the database and put it in my local mysql server. [02:38] <taaz> we? [02:39] <wingo> those of us that want to play with the web site. i'll have a cvs updater doober done soon. [02:40] <wingo> good for stagin if you don't want to break a live server. [02:40] <taaz> yeah, but 'we' can't use it ;) [02:41] <wingo> what do you mean? you can commit to cvs and then see how the changes work without having to break gstreamer.net. [02:41] <wingo> if they are php changes of the like. [02:41] <wingo> or the like even. [02:41] Action: taaz confused [02:42] <taaz> how are we going to update your site? [02:42] <wingo> accessing a url, that i have to make :) [02:42] <taaz> oh [02:42] <wingo> which is different from gstreamer.net because i don't have to worry about perms on my server [02:42] <wingo> although... [02:42] <wingo> that is a security hole over here... [02:42] <wingo> hum. [02:43] <wingo> yeah, well i guess that nixes that. [02:43] <wingo> we'll file this under 'marginally useful'. [02:43] Action: taaz thinks Monsters, Inc. will be a cool movie [02:43] Action: wingo has not heard of it [02:44] <taaz> new pixar movie [02:44] <wingo> cool [02:45] <taaz> you could password protect it [02:47] <wingo> shell access would be better, but i'd have to make sure the server was reasonable secure first [02:51] <ajmitch> but you trust gst people, right? ;) [02:53] <wingo> except that shifty taaz fellow :) [02:53] <ajmitch> hehe [02:54] <ajmitch> you want insecure? i run a site which runs cvs update as soon as stuff is committed [02:55] <taaz> i dont think that works on sf :( [02:56] <wingo> if you set everything to 666... [02:56] <ajmitch> it doesn't :) [02:56] <wingo> 666! [02:56] mwc (ma...@ly...) joined #gstreamer. [02:56] <ajmitch> it works best when cvs & web server are same machine [02:57] <taaz> yeah but it's not a good idea to leave everything world writable. any of the 100000 users can kill your site then [02:58] <ajmitch> taaz: no issues here, tho [02:58] <ajmitch> cos i got stuff all users :) [02:59] <ajmitch> i think i fixed it by having each cvs user run as one system user (pubcvs) [02:59] <ajmitch> and stuff was checked out by pubcvs |
From: IRC B. <wt...@us...> - 2001-10-16 04:26:35
|
******************************************************************* [03:03] bstard (Lor...@a2...) joined #gstreamer. [03:04] <taaz> eh? i know it's doable. but so what unless it works on sf [03:05] ezequiel (ezequiel@200.45.231.214) joined #gstreamer. [03:05] <ezequiel> hi [03:06] <ezequiel> can anyone help me? [03:06] <ezequiel> I'm trying mandrake&gstreamer [03:06] <ezequiel> who did can? [03:07] wingo (wi...@rd...) left irc: User abort with 5 Ctrl-C's [03:08] Action: ezequiel is thinking he won't can... [03:11] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [03:11] <ezequiel> who can help me? [03:12] <ezequiel> I need to use gstreamer! [03:12] <ezequiel> f**** asf videos! [03:12] <ezequiel> is there any [03:12] <ajmitch> eh? [03:12] <evil_monk> i don't think gstreamer can do asf [03:12] <ezequiel> doc, info, man? [03:12] Shippou (zbl...@cr...) joined #gstreamer. [03:12] <ezequiel> asf, avi, any [03:13] <ajmitch> docs for what? gstreamer? [03:13] <ezequiel> yes [03:13] <ezequiel> I need to use it [03:13] <ajmitch> gst can do avi, mpeg, maybe some asf (depends on the plugins) [03:13] <ajmitch> there are docs in the tarball, iirc, and in cvs, and on the website [03:13] <ajmitch> but they are mainly developer docs [03:13] <ajmitch> you wanting to write an app using gstreamer? [03:14] <ezequiel> no [03:14] <ezequiel> I just need to use gstmediaplay [03:14] <ajmitch> ok [03:15] <ezequiel> If I can do that I'll try mozstreamer ;) [03:15] <ajmitch> what problems are you having? [03:15] <ezequiel> It can't start the oss [03:15] <ajmitch> there aren't very many docs for gstmediaplay, afaik :) [03:15] <ajmitch> ok [03:15] <ajmitch> what is the error message? [03:15] <ezequiel> ** WARNING **: osssink: unable to open the sound device (in use ?) [03:16] <ajmitch> well is something else using the soundcard? like xmms, esd, etc? [03:16] <ezequiel> esd, perhaps [03:16] <ezequiel> how can I know it? [03:16] <ajmitch> run this as root: lsof -n /dev/dsp [03:16] <vektor> Hey, I have a question. [03:16] <ajmitch> vektor: yes? [03:17] <vektor> I have this app I wrote which records audio. I use the OSS API. [03:17] <ezequiel> I haven't lsof [03:17] <vektor> Can you think of a good way of determining when I've dropped or missed a frame of audio? [03:17] <vektor> ezequiel: You should get lsof, it's very useful. [03:17] <vektor> ezequiel: But if you're running mandrake, you probably have aRts running. [03:17] <ajmitch> ezequiel: then run this as root: fuser -v /dev/dsp [03:18] <ajmitch> vektor: dunno how sorry, i haven't done anything related to that stuff yet :) [03:18] <vektor> ok [03:18] <ezequiel> I'm runing mandrake [03:18] <vektor> ezequiel: I know, that's why I mentioned. [03:18] <ezequiel> But I don't use kde [03:18] <ajmitch> ok [03:18] <ajmitch> ezequiel: do you have fuser? [03:19] <ezequiel> I don't have fuser [03:19] <vektor> Sucks to be you then, doesn't it. [03:19] <ajmitch> hmm [03:19] <vektor> So, you have some app using the soundcard and you have no idea at all what it might be. [03:19] <ezequiel> Perhaps I don't have linux ;) [03:19] <ajmitch> ps ax | grep esd [03:20] <ajmitch> ezequiel: heh [03:20] <ezequiel> 5378 ? S 0:00 esd -nobeeps -as 30 [03:20] <vektor> Do you have sound working at all? [03:20] <vektor> Ah, well, there you go. [03:20] <ajmitch> ezequiel: hmm, ok, you could kill esd for now :) [03:20] <ezequiel> I killed it [03:20] <ezequiel> I works now [03:20] <ajmitch> cool [03:21] <ezequiel> thanks [03:21] <vektor> Fucking esd. [03:21] <ezequiel> I'm new in linux [03:21] <ajmitch> vektor: heh, i know it's bad [03:21] <vektor> We need to solve the sound system mess. [03:21] <vektor> This has to stop. [03:21] <ajmitch> i should see if arik can add an output selector in gstmediaplay [03:22] <ajmitch> he's done a bit of work on it lately, so i don't wanna touch it :) [03:22] <ezequiel> how can I select esd for playing? [03:22] <ajmitch> ezequiel: you can't at the moment unless you edit the source ;) [03:23] <vektor> You don't want to though because it will add alot of latency. [03:23] <ezequiel> in which part it could be? [03:23] <vektor> IMHO. [03:23] <ezequiel> IMHO? [03:23] <vektor> IMHO. [03:23] <ezequiel> what it means? [03:23] <vektor> In My Humble Opinion, you don't want to though because it will add alot of latency. [03:24] <ezequiel> ok [03:24] <vektor> And therefore fuck your sync in subtle ways. [03:24] <vektor> But hey, what do I know. [03:24] <ajmitch> if you really *really* want to, change "osssink" to "esdsink" on line 125 of gstplay/gstplay.c [03:24] <ezequiel> ok [03:24] <ajmitch> and recompile [03:25] <ezequiel> thanks [03:25] <ajmitch> i think that's the right line number [03:26] <ezequiel> yes [03:26] <ezequiel> it is [03:28] <ezequiel> thanks [03:29] Last message repeated 1 time(s). [03:29] <ajmitch> no probs [03:30] <ezequiel> :) [03:40] ezequiel (ezequiel@200.45.231.214) left #gstreamer (Aplicación Saliendo). [04:00] ajmitch (aj...@p5...) left irc: Ping timeout for ajmitch[p50-max2.dun.ihug.co.nz] [04:02] ajmitch (aj...@p3...) joined #gstreamer. [04:05] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [04:19] <vektor> hey anyone here ever used libst ? [04:19] <vektor> from sox ? [04:20] Action: vektor feeling brave. [04:22] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [04:22] Shippou (zbl...@cr...) joined #gstreamer. [04:23] vektor (reet@HSE-Kitchener-ppp194670.sympatico.ca) left irc: Ping timeout for vektor[HSE-Kitchener-ppp194670.sympatico.ca] [04:23] vektor (reet@HSE-Kitchener-ppp194670.sympatico.ca) joined #gstreamer. [04:31] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [04:31] Shippou (zbl...@cr...) joined #gstreamer. [05:03] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [05:03] wingo (wi...@rd...) joined #gstreamer. [05:06] Shippou (zbl...@cr...) joined #gstreamer. [05:23] chillywilly (da...@d9...) joined #gstreamer. [06:06] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [06:06] Shippou (zbl...@cr...) joined #gstreamer. [06:13] <wingo> ooh sleepy. [06:13] wingo (wi...@rd...) left irc: peace. [06:31] omega (om...@om...) joined #gstreamer. [06:31] <omega> oooh, logs [06:32] <ajmitch> hehe [06:32] <ajmitch> that's what happens when you remind wtay :) [06:32] Action: omega has been gathering numbers all day [06:32] <omega> yeah <g> [06:33] <omega> you wouldn't believe how huge some of these debugging statements get in asm... [06:33] <omega> gstscheduler.c:1148 [06:33] <omega> GST_DEBUG(GST_CAT_SCHEDULING, "has peer %s:%s\n",GST_DEBUG_PAD_NAME(GST_PAD_PEER(pad))); [06:33] <omega> turns into *110* instructions, with no -O [06:33] <omega> only 77 with -O6 [06:36] <omega> I just wrote a tool to go through the .o's and determine how many instructions are attributable to each line of source [06:36] <omega> very informative, and very encouraging even [06:37] <omega> a whopping 28 lines of perl [06:55] chillywilly (da...@d9...) left irc: [06:56] <taaz> is that useful? [06:57] <omega> what? [06:57] <taaz> instructions per source line [06:57] <omega> yes [06:57] <omega> objdump -S -l gstpad.lo | less [06:58] <omega> search for 'realsrc = GST_PAD_REALIZE' [07:01] <taaz> does alot of that go away with proper non-debug sort of flags? [07:01] <omega> nope [07:01] <omega> though the _FAST stuff is supposed to make it better, and it made it worse [07:01] <omega> so yes, it's useful <g> [07:05] <taaz> ok. [07:06] <taaz> made any improvements with your neat new tools? ;) [07:06] <omega> well, a preview of my results: [07:06] <omega> difference between default build of gstreamer on arm with glib2, nothing turned off: 215272 bytes [07:07] <omega> with all --disable's, no INFO/DEBUG, and G_DISABLE_CHECKS/G_DISABLE_ASSERT defined: 106876 bytes [07:07] <omega> oh, and -Os, but that turns out to be a minor (~3-5%) gain on ARM [07:08] <taaz> do you check out speed too? or just going for min size? [07:08] <omega> just for size atm [07:09] <omega> but on most embedded targets, you should get a pretty linear relationship between the two, because memory bandwidth and caching is typically relatively minimal [07:09] <omega> besides, a *huge* number of conditionals go away [07:11] <taaz> getting close to target size? [07:11] Action: taaz is mr question mark tonight [07:11] <omega> I think so, yes [07:11] <omega> but that's not taking into account glib and gobject, which are huge [07:12] <omega> they total 3/4MB so far [07:12] <omega> problem is there appears to be a compiler bug [07:12] <omega> adding G_DISABLE_CHECKS to gstreamer reduced the total size by 33% in one go [07:12] <omega> adding it to glib on the arm causes assembler errors [07:15] <omega> there are a lot of huge blobs of code in gstpad.lo.... [07:15] <omega> gstpad.lo is the biggest file by 33% (9K to gstscheduler.lo's 6K), and it's because of bulky macros [07:18] Action: vishnu hears the word "macro" [07:19] <omega> specifically, the bulky code is caused by gobject's type checking system [07:21] <vishnu> would this help? http://guidod.4t.com/substruct-c [07:22] <omega> very likely, yes [07:22] <vishnu> heh, we should start a lobby on the gcc list :-) [07:22] <omega> problem is the code in gtype.h is very opaque, and it's unclear how things unravel [07:24] <omega> problem I see with substruct-C and the gobject model is that suddenly you might start getting namespace collisions [07:24] <vishnu> no, i think it's completely backward compatible [07:25] <omega> how? if GstMyFilter has a 'type' member, it will collide with the GTypeInstance's 'type' member [07:25] <omega> or am I not getting the point of substruct-C? [07:26] <vishnu> no it won't, because it should match the current struct before searching up the inheritance list [07:26] <omega> I do like the idea of transparent type casting... [07:26] <omega> ok [07:26] <omega> is there a bypass syntax then, to get the GTypeInstance's type member? [07:27] <omega> or you just revert to the old ptrcasting method? [07:27] <vishnu> same as before, just cast it and then access the field [07:27] <omega> hmm, I'd be happy with nothing but the transparent casting <g. [07:27] <vishnu> yah, it seems like a no-brainer that this feature should be in C [07:27] <omega> which could be implemented without any syntax changes at all, I think [07:28] <omega> compiler just has to be smart enough to see that something of type GstMyFilter* happens to have a GstElement at the front of it, and allow 'casts' to GstElement* without a fight [07:32] <vishnu> hrm, it is similar to the idea of zero length arrays at the end of a struct [07:32] <omega> somewhat, yeah [07:35] <omega> wow. G_OBJECT_TYPE(object) is huge. that's the culprit [07:36] <omega> oh wait, I should be building against glib2 first of all.... [07:37] steveb (st...@no...) joined #gstreamer. [07:37] <omega> yo [07:37] <steveb> yo [07:37] <ajmitch> hey steveb [07:42] <omega> heh, openoffice is 1, kde is 5, and gstreamer is almost 2 [07:43] <omega> (yo) [07:43] <ajmitch> hehe [07:43] <ajmitch> when does gstreamer turn 2? [07:43] <omega> well, the first set of slides are dated Oct 22 [07:43] <ajmitch> ok [07:43] <omega> first time it was made 'public', beyond talking to a few coworkers at OGI [07:44] <omega> of course, the slides were just presented to the OGI people that wanted cookies that day.... [07:44] <omega> the first public release was Oct 31 [07:44] <omega> the GScreamer release [07:44] <ajmitch> fun ;) [07:45] <omega> I think it was waiting for GStreamer to build that prompted me to upgrade from a P5/133 to a PIII/500 a few months later <g> [07:49] Action: omega waits for gstreamer to build [07:51] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [07:52] Shippou (zbl...@cr...) joined #gstreamer. [07:53] <ShrimpX> omega: is there a chance that you could help me figure out some assembly code? [07:53] <omega> I can try [07:53] <ShrimpX> cool! [07:53] <omega> since that's what I've been staring at (uncomprehendingly so far) for a few hours [07:54] <ShrimpX> hehe [07:54] <omega> rather, I haven't tried to understand it yet [07:54] <ShrimpX> this file http://web.pdx.edu/~marius/dl/file.s is the gcc-generated asm for http://web.pdx.edu/~marius/dl/file.c [07:54] <omega> ok [07:54] <ShrimpX> i think i understand _most_ of it at this point [07:55] <omega> warning: calling conventions and stack stuff has always somewhat confused me [07:55] <ShrimpX> hehe [07:55] <ShrimpX> well, "somewhat confused" sounds pretty good to me. :) [07:55] <omega> ok, here's an idea: [07:56] <omega> write a comment next to every line of asm you can decipher [07:56] <ShrimpX> OK [07:56] <ShrimpX> and show it to you again? [07:56] <omega> yeah [07:56] <ShrimpX> ok [08:09] <ShrimpX> omega: is there a 4 byte offset between %ebp and %esp, to make room for the return value of main? [08:12] omega (om...@om...) left irc: Ping timeout for omega[omegacs.net] [08:12] omega (om...@om...) joined #gstreamer. [08:17] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [08:17] Shippou (zbl...@cr...) joined #gstreamer. [08:19] BeNOW (an...@ds...) left irc: Ping timeout for BeNOW[dsl-mr-207-34-113-i106-cgy.nucleus.com] [08:21] BeNOW (an...@ds...) joined #gstreamer. [08:28] <ShrimpX> omega: http://web.pdx.edu/~marius/dl/file.s [08:34] <omega> looks like you've got it figured out [08:34] <ShrimpX> looks ok? [08:34] <omega> dblchecking [08:35] <ShrimpX> so the initial %esp, i assumed it's whatever it already is, when the program begins execution. That's why i'm using all the offsets [08:36] <omega> this looks odd, though afaict your comments are correct [08:36] <ShrimpX> and %ebp is always 32 bits behind the stack pointer, for the program's return value, right? [08:36] <omega> the whole sequence before the first call is very funky looking [08:36] <ShrimpX> yea! [08:36] <omega> no, the C ABI uses eax for return value [08:36] <ShrimpX> gcc does some funky things with offests [08:36] <ShrimpX> oh [08:37] <omega> not sure what the leal is for [08:37] <omega> oh, waitasec.. [08:37] <omega> leal doesn't load, it just does math [08:37] taaz (dlehn@66.37.66.32) left irc: Ping timeout for taaz[66.37.66.32] [08:37] <omega> %esp-56 is str [08:37] <omega> ebx = esp-56 [08:37] <omega> pushl %ebx puts the str pointer on the stack [08:38] <omega> there are actually no memory opcodes in this entire program, except the leal, which is only using the address generator, doesn't actually load/save [08:38] <ShrimpX> so the reading is done entirely within scanf? [08:39] <omega> yes [08:39] <ShrimpX> ok [08:39] <omega> you understand the scanf call now? [08:39] taaz (dlehn@66.37.66.32) joined #gstreamer. [08:39] <ShrimpX> yea, i think so. [08:40] <ShrimpX> so it grabs the 2 values that were last pushed onto the stack [08:40] <omega> if you add something like str[0] = '\0'; to the code, you'll see some memory insns [08:40] <omega> the addl adjusts the stack to clear off the two arguments for the previous printf [08:41] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [08:41] <omega> leal -52(%ebp),%ebx is (in C) ebx = ebp - 52; [08:41] <ShrimpX> aaaah.... [08:41] <omega> it just uses the address generator [08:41] <ShrimpX> i get it [08:41] Shippou (zbl...@cr...) joined #gstreamer. [08:41] <omega> the pushl ebx;pushl $.LC1 puts the args on the stack for the call [08:42] <omega> so you might want to regroup the insns to make things clearer [08:42] <ShrimpX> yeah [08:42] <omega> but the whole stack thing at the front makes no sense to me [08:42] <omega> taaz: any ideas? http://web.pdx.edu/~marius/dl/file.s [08:47] <ShrimpX> could that first %ebx be pushed to collect the return value? [08:47] <ShrimpX> no, n/m [08:47] <omega> the return values are totally ignored in the whole program [08:48] <ShrimpX> i mean the return value of the actual program [08:48] <ShrimpX> that's what I'm still kinda confused about [08:48] <omega> ignored too afaict [08:49] <ShrimpX> cool [08:49] <omega> there is no reference to eax anywhere [08:49] <omega> put a return 1 in the c code and you'll get a movl $1, %eax somewhere [08:49] <ShrimpX> ah, makes sense [08:49] <ShrimpX> this was compiled with -O2, btw [08:49] <omega> oh, don't do that [08:50] <omega> if you want to understand the asm, ever [08:50] <ShrimpX> hehe [08:50] <ShrimpX> so i'm correct to say that %ebp is at %esp-4, since it's 4 bytes in size, right? [08:51] <ShrimpX> (the current %esp) [08:51] <omega> hmm? [08:51] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [08:51] <omega> oh, yeah [08:51] <ShrimpX> ok [08:52] Shippou (zbl...@cr...) joined #gstreamer. [08:53] <ShrimpX> omega: what do leave and ret do? [08:53] <omega> part of the call structure [08:53] <omega> you have the intel asm reference? [08:53] <ShrimpX> i might [08:53] <ShrimpX> i downloaded a bunch of pdf's today [08:54] <omega> lemme check the part number first [08:54] <ShrimpX> k [08:55] <omega> 245471 [08:56] <ShrimpX> k, i'm downloading it [08:56] <omega> why no dcc? [08:56] <omega> it's bz2'd [08:56] <ShrimpX> pdf [08:56] <omega> yeah, bz2'd pdf [08:57] <omega> 2.9MB instead of 6.5 [08:57] <omega> is your irc client sending the rejects? [08:57] <ShrimpX> i dunno [08:57] <ShrimpX> i can't get it [08:57] <omega> what client? [08:58] <ShrimpX> i already downloaded it off intel tho [08:58] <ShrimpX> epic [08:58] <omega> ok [08:58] <omega> hmm, never heard of that [08:58] <omega> I use xchat, it works quite well [08:58] <ShrimpX> epic is a console one. based on ircII [08:58] <omega> ok [09:01] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [09:02] Shippou (zbl...@cr...) joined #gstreamer. [09:03] mwc (ma...@ly...) left irc: Ping timeout for mwc[lychee.ntu.edu.au] [09:03] mwc (ma...@ly...) joined #gstreamer. [09:03] <ShrimpX> omega: thanks a lot for the help. i'll be able to sleep now. :) [09:03] <omega> heh [09:04] Action: ajmitch wonders whether he should file a bug with this build problem ir just try from a clean tree ;) [09:04] <ajmitch> s/ir/or/ [09:04] <omega> what's the problem? [09:05] <ajmitch> trying to link libjpeg [09:06] <ajmitch> it tries to find the lib in the current dir, not in /usr/lib [09:06] <ajmitch> i'll try a clean build [09:06] <omega> ld *always* looks in /usr/lib [09:06] <omega> you sure it's installed there? [09:06] <ajmitch> yeah [09:07] <ajmitch> it is [09:07] <ajmitch> if i use ln -s, the build works [09:07] <omega> very odd [09:07] <omega> how is it refering to libjpeg? [09:07] Nick change: taaz -> taazzzz [09:07] <omega> -ljpeg ? [09:09] <omega> what libjpeg* files exist in that directory? [09:10] <omega> make clean;make;killall -SLEEP ajmitch [09:10] <omega> er, make distclean;./autogen.sh;make;killall -SLEEP ajmitch [09:10] <ajmitch> doh, there's a libjpeg.la there, dated from may. i have already done a make distclean today [09:10] <omega> rm -rf libjpeg* in the plugin dir [09:11] <omega> you might be able to build without reconfiguring even [09:11] <ajmitch> yep, worked [09:11] <omega> cool [09:11] <ajmitch> why didn't make distclean remove it? [09:11] <omega> because it's not a dist file [09:11] <omega> it should have never been there [09:11] <ajmitch> hmmm [09:12] <ajmitch> ok [09:13] Action: ajmitch shoudl go do a full build now [09:14] <ajmitch> how useful is the editor, btw? ;) [09:14] <omega> I was just thinking about that. not very, atm [09:14] <ajmitch> ok [09:14] <omega> it needs a rewrite and a maintainer [09:16] <ajmitch> why, what needs redone? [09:17] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [09:17] <omega> the whole widget structure is rather evolved and hackish, lots of design cruft [09:17] <ajmitch> k [09:17] Shippou (zbl...@cr...) joined #gstreamer. [09:17] <ajmitch> so it'd need someone who knows gst & gtk+ then? [09:18] <omega> I'm trying to get xchat to stop showing me these leave/join messages for those with poor connections, and not having much luck [09:18] <ajmitch> hehe [09:19] <ajmitch> /ignore *@* CHAN [09:19] Action: ajmitch is guessing [09:19] <omega> um, not exactly [09:20] <ajmitch> that wouldn't work ;) [09:20] tnt_ (tnt@24.76.164.254) joined #gstreamer. [09:20] <omega> yo [09:20] <ajmitch> hi [09:20] <tnt_> Hello [09:20] <omega> ajmitch: in xchat something like *!zblaxell@*.* [09:22] <ajmitch> k [09:22] <ajmitch> cool, automake failed for some reason [09:22] <omega> neat [09:22] <ajmitch> it's version 1.5 [09:22] <tnt_> Does anyone here know if gstreamer has video filters (included) to do transitions... like fading, disolves, etc? [09:23] <omega> not yet [09:23] Action: omega is trying to get vektor to convert some of his algos into gstreamer plugins [09:23] <omega> but his stuff is all single-stream stuff [09:24] <ajmitch> omega: you gonna split plugins into a separate cvs module one day, or have the central plugin repository like you wanted? [09:25] <omega> the plugins will all go into a number of cvs modules, and even other projects [09:25] <omega> but I do want to have a web-wide plugin registry [09:25] <ajmitch> k [09:25] <ajmitch> first comes a stable API for plugins [09:25] <omega> including a server that can ID streams for players [09:25] <omega> yes [09:35] Action: vishnu falls asleep on his keyboard [09:35] <vishnu> !@#$%!#$^! [09:35] <vishnu> oops, bye [09:35] <omega> l8r [09:35] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [09:36] <ajmitch> ah, cool, now i get several warnings [09:36] <omega> gstreamer.net/random/omega/insncounts/ [09:37] <omega> gstpad.c is most interesting [09:37] <ajmitch> big [09:37] <omega> yeah, marked up into html, some overhead involved... [09:37] Action: omega wishes more webserves would use http/1.1 compression [09:37] <ajmitch> you been doing this all day? ;) [09:38] <omega> which? [09:38] <omega> gzip -9 gets 10:1 compression on gstpad.c.html [09:38] <ajmitch> hmm [09:38] <omega> down to 13520 bytes. would load very fast if httpd would do it [09:39] <ajmitch> what are the numbers for? asm instructions? [09:39] <omega> yup [09:39] <ajmitch> why are some fractional? [09:40] <omega> because that line was listed several times in the resulting asm, this is the average of how many instructions there are per instance [09:40] <ajmitch> ah, ok [09:40] <omega> I'll update the tables to include more data soon [09:40] <ajmitch> ok, configure runs even after automake 'fails' [09:40] <omega> yeah [09:41] <omega> starting to build an index.html.. [09:42] <ajmitch> you trying to optimise gstreamer core for size quite a bit? [09:43] <omega> yup [09:43] rowenc (rowenc@Hong-231-192.PLU.edu) joined #gstreamer. [09:43] <omega> yo [09:43] <rowenc> hi [09:43] <ajmitch> hi [09:43] <omega> gstpad.lo now has 5420 bytes of text [09:43] <omega> `size gstpad.lo`, how big is yours? <g> [09:44] <ajmitch> i have nothing compiled :) i've cleaned it all up ;) [09:44] <omega> pff <g> [09:46] <rowenc> what's the dvd status at? [09:46] <omega> taaz has been doing some work on dvdsrc to align it with libdvdread [09:46] <omega> not in cvs yet afaik [09:47] <omega> but he does have a basic player functioning on his machine [09:47] <rowenc> cool... does it have event timings yet (pause, etc)? [09:47] <omega> don't think so [09:48] <omega> btw all: I updated the LXR stuff on the website [09:49] <ajmitch> nice [09:49] <ajmitch> so instead of peering thru code here we can look at it online [09:49] <omega> yeah, well, it's still lxr-0.3, not the prettiest code out there [09:49] <omega> the viewcvs stuff looks a lot better [09:50] <ajmitch> where do we access it from? [09:50] <omega> I just wish all these tools (viewcvs, lxr, blame, etc.) could be merged into a single system [09:50] <ajmitch> i wish all multimedia stuff could merge & settle on a single api, like gstreamer for example [09:50] <omega> heh [09:51] <omega> I have a goal in mind: [09:51] <ajmitch> world domination? [09:51] <omega> well, besides that <g> [09:51] <omega> first place in the 2002 Linux Journal readers' choice award in the 'favorite audio tool' category [09:52] <omega> and maybe even spawn a new 'favorite multimedia app' category [09:52] <omega> 2001's winners were 1. xmms, 2. pmg123 (mpg123??), 3. realaudio (bleck) [09:52] <ajmitch> hehe [09:52] <omega> freshmeat has no pmg123, so it must be a typo [09:53] <ajmitch> yeah [09:53] <ajmitch> but to be reader's choice award, it has to be visible to users [09:53] <omega> yup [09:53] <omega> we have a little less than a year <g> [09:53] <ajmitch> at the moment, i'd say it'd only be visible if gstmediaplay gets to be quite good ;) [09:53] <omega> rhythmbox, for instance [09:53] <ajmitch> yeah [09:53] <omega> yeah, or that [09:54] <ajmitch> arik says he has some code for gstmediaplay [09:54] <ajmitch> it's just doing some weird stuff ;) [09:54] <omega> interesting thing though, that gnotices article asking for things that users want in 2.0... has many many many references to gstreamer [09:54] <omega> wtay posted one of those to the list [09:54] <ajmitch> oh? [09:55] Action: ajmitch hardly read any of those comments in the article [09:55] <omega> http://news.gnome.org/1002781202/index_html [09:55] <omega> do a text search for gstreamer just on that page [09:55] <omega> the vast majority of the comments aren't even showing, and I get a lot of hits [09:55] <ajmitch> pretty good :) [09:55] <omega> a dozen just there [09:56] Action: ajmitch waits for the massive page to load [10:00] <ajmitch> curses, i need to setup ssh so i can login without a password [10:00] <tnt_> As I understand it, to make a "video filter", you have to deal with data of type "video/raw" (correct me if I'm wrong)... is there a good reference I could lool at to learn how to do this?... a doc, or some source I could refer to, to help me write a filter. [10:00] <omega> there's the example plugin, but that's a single-input chain-based element [10:01] <omega> I dunno if there are any 'merge' style elements anywhere... [10:01] <omega> I should write one, maybe? <g> [10:02] <ajmitch> yes [10:03] <ajmitch> hmm, seems a few ppl would like gstreamer in gnome 2 ;) [10:03] <ajmitch> you'd better get hacking then omega ;) [10:03] <omega> heh [10:03] <omega> scary thing is I have some writing to do [10:03] <ajmitch> oh? [10:03] <omega> I have to describe gstreamer better than I've ever described it before... [10:04] <ajmitch> whatever for? [10:04] <omega> I'll check with my manager to see when I can say more.... [10:04] <ajmitch> ok then [10:04] <ShrimpX> omega: another quick asm question? [10:04] <omega> let's just say: "you wanted publicity? I'll give you publicity!" [10:04] <omega> ShrimpX: I thought you were going to sleep <g> [10:04] <ShrimpX> apparently not... :) [10:05] <ShrimpX> if you push 2 values onto the stack, and then call scanf, won't the stack pointer be decremented by 8, since the values were taken off the stack by the function? [10:05] <omega> nope, they use the same relative access to %esp [10:06] <ShrimpX> ah [10:06] <omega> because they have to be able to save registers on the stack [10:07] <ShrimpX> oh, i see. but when you call scanf, how does it grab the values that are inteded for it? does it read back 8 bytes? [10:07] <omega> 8(%esp), etc. [10:08] <ShrimpX> cool [10:08] <omega> stack frame is 8 or so bytes, then however man regs it saves * 4 [10:08] <omega> plus its local variables [10:08] <omega> so you get pushl;pushl;call [10:08] <ShrimpX> yea [10:09] <omega> and the function does pushl reg;pushl reg;pushl reg;subl $16,%esp [10:09] <omega> and refers to its args as $36(%esp) and $40(%esp) or something [10:10] <ShrimpX> wow, cool; i think i understand that [10:11] <omega> 36 = 8(stack frame) + 12(3 saved regs) + 16(local variables) [10:12] <tnt_> OK, I looked at the example plugin... so, let me see if I understand it correctly... the video data is transfered in the GstBuffer?... and gst_example_chain is called (potentially) multiple times,... until all the video data gets through? [10:12] <omega> for video, right now, a buffer must contain exactly one frame [10:12] <omega> for single-source filters, that's what you want [10:13] <omega> but for multi-source filters you need to use a different style plugin [10:13] <omega> look at gst/elements/gstidentity.c:gst_identity_loop() [10:14] <tnt_> What is the format of the video data? (And how do you know the other realted data: width, height, etc) [10:14] <omega> the pad GstCaps have that info [10:14] <tnt_> (I'll go look at that.) [10:14] <tnt_> OK [10:15] Action: ajmitch wonders why configure doesn't want to build the xmms plugin [10:16] mwc (ma...@ly...) left irc: [x]chat [10:23] thomasvs (th...@21...) joined #gstreamer. [10:23] <thomasvs> beep [10:23] <omega> yo [10:24] <ajmitch> hey thomasvs [10:24] <thomasvs> omega: planning on doing a release anytime soon ? [10:24] <omega> heh, hopefully [10:24] <thomasvs> I'd like to move our college radio station to gstreamer [10:24] <thomasvs> The hackish thing I wrote three years ago is starting to fall apart ;) [10:24] <omega> heh [10:24] <thomasvs> and gstreamer would be the best thing to transition from automated mix to live dj mode [10:25] <omega> heh [10:25] <thomasvs> and we broadcast at 192 kbit so that would be a good testcase for gstreamer as well [10:25] <omega> yup [10:26] <thomasvs> It'd be so cool : mix from various sound files in the dj system, mix between live shows, and encode on the fly to mpeg and ogg at two bit rates [10:26] <omega> yup [10:26] <omega> that's close to what BeNOW wants to do [10:26] <omega> it might be worth syncing with him [10:26] <thomasvs> Yeah, I'm going to discuss it with him anyway soon [10:27] <omega> cause he's working on backend database for storage, metadata, etc. [10:27] <omega> he could be building the next gen radio-station software <g> [10:29] <thomasvs> BeNOW: not here ? [10:29] <thomasvs> he'll be interested in the automated mix calculation and stuff I can bring in I suppose ;) [10:29] <omega> his stream is having a cow, so probably not here [10:29] <thomasvs> which I want to move to gstreamer as well asap [10:30] Action: ajmitch tries out make distcheck [10:30] <ajmitch> wow, fails straight away with my automake ;) [10:31] <thomasvs> ajmitch:version ? [10:31] <ajmitch> 1.5 [10:31] <tnt_> omega: in gst_identity_loop, is the purpose of << g_signal_emit (G_OBJECT (identity), gst_identity_signals[SIGNAL_HANDOFF], 0, [10:32] <tnt_> buf); [10:32] <tnt_> >> to deal with cothreads? [10:32] <omega> no, it's private to identity [10:32] <thomasvs> Cool, my boss just tells me I have three days to make something that mixes just like our on-air system for two weeks straight without failing ... [10:32] <thomasvs> ... shall I go with gstreamer or something xmms'ish ? [10:32] <omega> it just lets the app keep track of when a buffer gets sent by identity [10:32] <tnt_> OK [10:32] <omega> thomasvs: when did he tell you this? [10:32] <thomasvs> just now, five seconds ago [10:33] <omega> neat. has he ever heard of REALITY??? [10:33] <thomasvs> omega: heh. [10:33] <thomasvs> he trusts in me ;) [10:33] <ajmitch> hehe [10:33] <thomasvs> actually, the problem is that there's too much choice on how to do it [10:33] Action: omega hands thomasvs his Clue-by-Four(tm) to thwack his boss with [10:33] <omega> thomasvs: right [10:34] <ajmitch> 3 days? [10:34] <thomasvs> Well, it's due in 8 days but he's allotted me 3 days [10:34] <thomasvs> but actually it's not that hard, it gives me the chance to do either of two things I should be doing anyway : [10:34] <thomasvs> a) finish my mixer that uses gstreamer (which needs dereferencing and events to work right) [10:34] <thomasvs> b) finish my changes to the xmms crossfade plugin [10:35] <thomasvs> or [10:35] <thomasvs> c) go with my old thesis project which still runs fine and couple that with esd [10:36] <omega> is this a final system, or can it be changed later? [10:36] <thomasvs> omega: this is totally temporary, else I wouldn't consider it [10:36] <omega> ok [10:36] <thomasvs> we have one radio system doing that (commercially) that works and we're going to switch over to another, and during the switchover we're using mine [10:36] <omega> 'that' ? [10:37] <thomasvs> 'that' = automatic mixing [10:37] <omega> with what software? [10:37] <thomasvs> and live assists as well, but we're going full auto at the moment [10:37] <thomasvs> We use Dalet now, which sucks [10:37] <thomasvs> *really* sucks [10:37] <thomasvs> and we switch over to Zenon [10:37] <thomasvs> which is a bit more open-ended [10:37] <omega> what are they? [10:37] <thomasvs> Dalet sucks in the way that you have to set your clock to East-European time to get imports to work right [10:37] <omega> pfff [10:38] <thomasvs> They're both fully automated radio systems, doing live assist and automated mixing [10:38] <thomasvs> as well as scheduling and stuff [10:38] <omega> and very expensive, right? [10:38] <thomasvs> standard software for a lot of radio stations world wide [10:38] <thomasvs> and terribly expensive ;) [10:38] <thomasvs> actually, I consider it a very good field for open source to move into [10:38] <omega> yup. sounds like you and benow need to sync up ASAP [10:38] <omega> mailto:an...@ea... [10:38] <thomasvs> specialty fields like this are where money is being stolen by truckloads [10:39] <tnt_> omega: So basically, if you want to deal with two video sources, you loop until you get your 2 (or more) GstBuffers... maybe checking with "GST_PAD_CAN_PULL", or maybe seeing that "gst_pad_pull" returns something besides NULL... is this correct? [10:39] <thomasvs> Like, my boss was thinking about buying RDS software. [10:39] <thomasvs> It only does one thing right : you can schedule beforehand what message you want to show at what time [10:39] <thomasvs> so nothing dynamic, like traffic info and stuff [10:39] <thomasvs> the price : 10.000 $ [10:39] <thomasvs> !!! [10:39] <thomasvs> I coded something in perl that was dynamic in half a day [10:39] <omega> wow [10:40] <omega> tnt_: yeah, basically [10:40] <ajmitch> thomasvs: where do you work? [10:40] <thomasvs> and they're *still* thinking about buying the software [10:40] <thomasvs> ajmitch: a new commercial radio station in Belgium [10:40] <ajmitch> thomasvs: hmm, interesting [10:40] <thomasvs> well, the commercial is nice because you can buy expensive stuff [10:40] <ajmitch> thomasvs: maybe i should try & sell some of this stuff, my sister works at a nationwide radio staion here :) [10:40] <thomasvs> though I like the attitude more at the college station I work [10:41] <thomasvs> ajmitch: sure, feel free ;) [10:41] <omega> tnt_: for fixed input count plugins, you'll just do while (1) { buf1=gst_pad_pull(pad1);buf2=gst_pad_pull(pad2);dostuff();gst_pad_push(newbuf); } [10:41] <thomasvs> ok, enough chitchat, I should get working on some of this [10:41] <ajmitch> k [10:41] <thomasvs> omega: is event stuff working ok already ? [10:41] <omega> it works as far as we've tried it <g> [10:41] <tnt_> OK. [10:41] <thomasvs> omega: I mean, can I for example do a buffer flush when writing different mp3's ? [10:42] <omega> um, yes and no [10:42] <omega> yes in that that should work, no in that the plugins that matter will say "HUH??" [10:42] <thomasvs> ok, but can I fix those plugins ? [10:42] <omega> yup [10:42] <thomasvs> ok, I'll give it a few hours and see if I can pull it off [10:42] <thomasvs> how do I get a different branch again ? [10:42] <omega> heh, ok [10:42] <omega> cvs update -dP -r BRANCH-EVENTS1 [10:43] <omega> but make sure you have no local changes first [10:43] <thomasvs> I'm going to get a new tree for this [10:43] <omega> I forget offhand how to do a new checkout [10:43] <omega> checking [10:43] <tnt_> Thanks for you help omega... I'll try some coding tomorrow... time for sleep now. See ya. [10:43] <omega> tnt_: cool, l8r [10:43] <omega> thomasvs: cvs -d ... co gstreamer -r BRANCH-EVENTS1 should do it I think [10:44] tnt_ (tnt@24.76.164.254) left irc: Client Exiting [10:45] <thomasvs> close, -r tag before gstreamer [10:45] <omega> ok [10:48] <omega> http://gstreamer.net/random/omega/insncounts/ [10:48] <omega> neat stuff there <g> (IMO) [10:50] <omega> ajmitch: and quite a bit smaller now too ;-) [10:50] <thomasvs> anyone know what these mean btw : [10:50] <thomasvs> NONE:0: /usr/bin/m4: `syncoutput' from frozen file not found in builtin table! [10:50] <thomasvs> NONE:0: /usr/bin/m4: `changesyntax' from frozen file not found in builtin table! [10:51] <omega> tar -xzvf autoconf-2.52.tar.gz;cd autoconf-2.52;./configure;make install [10:51] <omega> ;-) [10:51] <thomasvs> oh ok ;) [10:51] <omega> dunno, I assume your autoconf or automake is screwed up [10:51] <ajmitch> hehe [10:52] <thomasvs> hm, no, I have ac 2.52 [10:52] <thomasvs> and am 1.4p5 [10:52] <ajmitch> hrmm [10:52] <thomasvs> maybe we should have a developer's page where each of us lists their machines and versions of these tools [10:53] <omega> heh [10:53] <ajmitch> yeah [10:53] Action: ajmitch wonders why the slider doesn't work on gstmediaplay [10:53] <ajmitch> shall i file a bugreport? ;) [10:53] Action: omega wonders why ajmitch broke the slider in gstmediaplay <g> [10:53] Action: omega hands ajmitch some superglue to put it back on [10:53] <ajmitch> well, it works, but doesn't move as the song plays ;) [10:53] <omega> any number of reasons for that... [10:54] Action: ajmitch gets the superglue all over his keyboard [10:54] <ajmitch> it's broken? ;) [10:54] <omega> the most likely is that the switch from disksrc to filesrc broke it [10:54] <omega> different param names [10:54] <ajmitch> if i slide the slider, the song changes to match :) [10:54] <omega> that's a start [10:54] <ajmitch> so seeking works [10:55] <omega> well, the pathetic seeking that's always been there [10:55] <thomasvs> hmm, how does that seeking work ? [10:55] <ajmitch> ah well, i'll blame arik ;) [10:55] <thomasvs> I mean if you have a VBR file you'll end up with trouble right ? [10:55] <ajmitch> thomasvs: black magic, i think [10:55] <omega> thomasvs: right now, it just sets the file offset still [10:55] <thomasvs> ok, makes sense [10:55] <omega> thomasvs: we have stuff in the works to help significantly [10:55] <thomasvs> omega: I know ;) [10:55] <omega> I mean the VBR problem <g> [10:55] <omega> see gsttimecache.[ch] [10:55] Action: ajmitch wants to see some of this fancy new stuff omega talks about [10:56] <thomasvs> omega: yeah, I suppose you put black magic in there to make timebased access ? [10:56] <omega> not black, just a little gray <g> [10:56] <ajmitch> hehe [10:56] <thomasvs> good luck for video then [10:56] <omega> the timecache only works as well as the data fed into it [10:56] <thomasvs> well, i'll first compile the damn thing ;) [10:57] <thomasvs> then figure out what to do with it [10:57] Action: ajmitch wonders when the next SF cron update is [11:01] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [11:03] Action: ajmitch thinks some of this editor code is looking dated [11:03] Shippou (zbl...@cr...) joined #gstreamer. [11:03] <omega> ajmitch: ya think [11:03] <omega> ? [11:03] <ajmitch> heh, last commit on a fair bit of this stuff was in 2000 [11:04] <omega> yup [11:04] <ajmitch> january 200 for one file :) [11:04] <ajmitch> 2000, sorry ;) [11:05] <omega> hrm, so we're gonna be celebrating gstreamer's 1802th birthday soon? wow. [11:07] <omega> well, I'm gonna go sleep now [11:07] <ajmitch> k, g'night [11:07] omega (om...@om...) left irc: zzzzzz [11:10] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [11:11] Shippou (zbl...@cr...) joined #gstreamer. [11:30] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [11:32] Shippou (zbl...@cr...) joined #gstreamer. [11:52] ajmitch (aj...@p3...) left irc: Ping timeout for ajmitch[p32-max2.dun.ihug.co.nz] [11:52] ajmitch (aj...@p2...) joined #gstreamer. [12:01] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [12:03] Shippou (zbl...@cr...) joined #gstreamer. [12:16] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [12:17] Shippou (zbl...@cr...) joined #gstreamer. [12:22] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [12:23] Shippou (zbl...@cr...) joined #gstreamer. [12:32] dschleef (ds...@c7...) left irc: Ping timeout for dschleef[c799921-a.pinol1.sfba.home.com] [13:01] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [13:06] Shippou (zbl...@ot...) joined #gstreamer. [13:12] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [13:15] Shippou (zbl...@ot...) joined #gstreamer. [13:23] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [13:25] Shippou (zbl...@ot...) joined #gstreamer. [13:30] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [13:30] Shippou (zbl...@ot...) joined #gstreamer. [13:37] ShrimpX (marius@131.252.244.168) left irc: Ping timeout for ShrimpX[131.252.244.168] [13:39] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [13:39] Shippou (zbl...@ot...) joined #gstreamer. [13:42] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [13:43] Shippou (zbl...@ot...) joined #gstreamer. [13:46] Shippou (zbl...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [13:47] Shippou (zbl...@cr...) joined #gstreamer. [13:51] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [13:51] Shippou (zbl...@cr...) joined #gstreamer. [13:59] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [14:25] Zygo (zbl...@24...) joined #gstreamer. [14:35] Shippou (zbl...@24...) joined #gstreamer. [14:36] glenn (glenn@213.155.150.129) joined #gstreamer. [14:36] asmod (stevec@64.5.222.2) joined #gstreamer. [14:38] glenn (glenn@213.155.150.129) left #gstreamer (Client Exiting). [15:03] bstard (Lor...@a2...) got netsplit. [15:07] bstard (Lor...@a2...) returned to #gstreamer. [15:31] Shippou (zbl...@24...) left #gstreamer (Client Exiting). [15:33] Shippou (zbl...@24...) joined #gstreamer. [15:35] evil_monk (trevo@208.141.162.68) joined #gstreamer. [15:38] Shippou (zbl...@24...) left irc: Ping timeout for Shippou[246.chipworks.com] [15:38] Zygo (zbl...@24...) left irc: Ping timeout for Zygo[246.chipworks.com] [15:39] ShrimpX (marius@131.252.244.168) joined #gstreamer. [15:44] Zygo (zbl...@24...) joined #gstreamer. [15:53] Shippou (zbl...@24...) joined #gstreamer. [15:56] dobey (do...@dr...) joined #gstreamer. [16:21] Shippou (zbl...@24...) left irc: Ping timeout for Shippou[246.chipworks.com] [16:26] thomasvs (th...@21...) left irc: Ping timeout for thomasvs[212-100-172-175.adsl.easynet.be] [16:32] Shippou (zbl...@24...) joined #gstreamer. [16:42] thomasvs (th...@21...) joined #gstreamer. [17:07] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [18:11] evil_monk (trevo@208.141.162.68) joined #gstreamer. [18:31] vekky (wb...@pe...) joined #gstreamer. [18:31] <vekky> hey sup yo? [18:33] <steveb> yo, yo [18:36] <steveb> vekky: feeling better? [18:37] Nick change: taazzzz -> taaz [18:37] vishnu (jo...@cx...) joined #gstreamer. [18:42] arik (ar...@sd...) joined #gstreamer. [18:42] <arik> hi [18:42] <vekky> hi [18:42] <vekky> steveb: kinda. [18:44] <vishnu> yo [18:47] vekky (wb...@pe...) left irc: too bad the scene is dead [19:07] Nick change: dobey -> dobey-eat [19:37] thomasvs (th...@21...) left irc: [x]chat [19:37] arik (ar...@sd...) left irc: hmm... i think i'll go make mix tapes [19:45] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [20:18] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [20:20] Nick change: dobey-eat -> dobey [20:29] Nick change: vishnu -> omega [20:29] Nick change: omega -> vishnu [20:34] chillywilly (da...@d1...) joined #gstreamer. [20:41] evil_monk (trevo@208.141.162.68) joined #gstreamer. [20:52] bstard (Lor...@a2...) joined #gstreamer. [20:57] thomasvs (th...@ad...) joined #gstreamer. [20:57] Nick change: dobey -> dobey-out [21:07] ChiefHighwater (fl...@su...) joined #gstreamer. [21:10] ChiefHighwater (fl...@su...) left irc: Leaving [21:15] Zygo (zbl...@24...) left irc: Ping timeout for Zygo[246.chipworks.com] [21:15] Shippou (zbl...@24...) left irc: Ping timeout for Shippou[246.chipworks.com] [21:17] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [21:19] Zygo (zbl...@24...) joined #gstreamer. [21:30] Shippou (zbl...@24...) joined #gstreamer. [21:36] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [21:37] Nick change: dobey-out -> dobey [21:39] Zeenix (ze...@13...) joined #gstreamer. [21:47] ShrimpX_ (marius@Photon.pdx.netroedge.com) joined #gstreamer. [21:52] <Zeenix> hello everybody [21:52] <Zeenix> #gstreamer so quite today [21:53] <ajmitch> yes [21:53] omega (om...@om...) joined #gstreamer. [21:54] <ajmitch> hey omega [21:54] <omega> yo [21:54] <Zeenix> oh i fogot, could someone remind me whats the typical path to my htdocs dir on zchat.sourceforge.net [21:54] <Zeenix> yo [21:54] <omega> /home/groups/z/zc/zchat/htdocs [21:54] <Zeenix> omega: thanx [21:55] <Zeenix> omega: that librtp doesnt seem to achieve what its meant for at all [21:56] Nick change: wtay-zZz -> wtay [21:56] <wtay> yo [21:56] <ajmitch> hi wtay [21:56] <omega> Zeenix: doesn't surprise me, I've found a large percentage of libs really suck [21:56] <Zeenix> it doesnt seem to abstract you from timing-up the packets [21:56] <wtay> omega: I've been doing some benchmarks of buffer creation/free.. it's horrible [21:56] <omega> oh? [21:56] <Zeenix> wtay: yo [21:56] <wtay> hm [21:56] <wtay> Zeenix: hi [21:57] <wtay> omega: the main culprit being g_mutex_new/free which we do for *every* (sub)buffer and which is *very* expensive [21:57] <Zeenix> & the more horrible thing is that gphone itself also doesnt do that itself [21:58] <omega> wtay: hmm. dunno how to solve that.. [21:58] <wtay> I do :) [21:58] <omega> how? [21:58] <wtay> 1) make a more intelligent buffer gmemchunk [21:58] <wtay> one that keeps the original mutex too [21:58] <Zeenix> here is a view into the code from gphone: [21:59] <Zeenix> switch(rtp_packet_get_payload_type(packet)) { [21:59] <Zeenix> case PAYLOAD_GSM: [21:59] <Zeenix> play_gsm_data(rtp_packet_get_payload(packet), [21:59] <Zeenix> rtp_packet_get_payload_len(packet)); [21:59] <Zeenix> break; [21:59] <wtay> 2) only use the mutex when concurrency is needed [21:59] <wtay> 3) CAS again.. [21:59] <vektor> Zeenix: ah cool you're doing a phone app. [21:59] <vektor> Zeenix: i need to rewrite my SIP stack :( [22:00] <Zeenix> vektor: no an rtprecv plugin [22:00] <omega> great. someone goes and deletes data from sourceforge servers that must be world writable because of things like wiki and so on, and sourceforge now says it's *our* responsibility to do backups of our own stuff [22:00] Zygo (zbl...@24...) left irc: Read error to Zygo[246.chipworks.com]: Connection reset by peer [22:01] Shippou (zbl...@24...) left irc: Ping timeout for Shippou[246.chipworks.com] [22:04] <wtay> omega: so, I'm going to try a few things in test/ [22:04] <omega> ok [22:05] <Zeenix> omega or wtay: i need your comments on this, i dont think implementing this would be easy in plugin's code & neighther would it look nice [22:06] <wtay> Zeenix: implementing the timestamps you mean? [22:07] <Zeenix> wtay: no serializing the packets as they arrive according to the timestamp/seq [22:08] <wtay> Zeenix: what about creating a window? shouldn't be too hard [22:08] <Zeenix> wtay: window ? [22:09] <wtay> Zeenix: create ring buffer and collect the packets in the buckets, assume a packet was dropped if the buffer wraps [22:10] Action: Zeenix needs atleast a day to understand what wtay just said [22:11] <wtay> uhm [22:11] <wtay> like this: [22:11] <wtay> create an array of, say, 16 entries. [22:11] <wtay> packet with timestamp 0 arrives, place it in entry 0 [22:12] <wtay> packet with ts 1, in bucket 1 etc.. [22:12] <Zeenix> ok [22:12] <Zeenix> then [22:12] <wtay> if you have a gap (packet 2 arrived and packet 4 arrived, but not 3), you wait a bit [22:13] <wtay> collect more packets until your buffer is full an then assume packet 3 is lost [22:13] <wtay> s/an/and/ [22:13] <wtay> the window size == worst case latency [22:14] <Zeenix> wtay: whats the window here [22:14] <wtay> the array size [22:15] <Zeenix> wtay: ok you didnt pointed to M$ windows OS <g> [22:16] <wtay> should be only a few lines of code actually [22:16] <Zeenix> wtay: the place for packet 3 remains vaccant till last [22:16] <omega> wtay: http://sourceforge.net/tracker/?func=detail&atid=200001&aid=470948&group_id=1 [22:17] <Zeenix> ? [22:17] <wtay> "wtay: is not a registered protocol" :) [22:19] <wtay> omega: grr [22:20] <wtay> can one use mySQL for the wiki? [22:21] <omega> I want to use cvs, but the question then is how to get write access to cvs.... [22:21] <omega> I've heard there is one wiki that can use cvs as a backend [22:21] <omega> but without pserver write access to gst cvs, that can't be done on sourceforge [22:21] <omega> IMO sf needs to provide a wiki solution itself [22:22] asmod (stevec@64.5.222.2) left irc: [x]chat [22:22] <wtay> or proper permissions on httpd [22:22] <omega> well, yes [22:22] <omega> though that's hard [22:22] <omega> but not impossible [22:22] <omega> you read my latest comments? [22:23] <wtay> apache should be able to do that, if not it's a bug/missing feature IMO [22:23] <wtay> yeah [22:23] <omega> I think it'd be a feature of php, and apache-perl, etc., individually, for stuff run inside the main httpd process [22:24] <wtay> omega: time to reread man chmod :) [22:24] <omega> there is no solution without httpd changes [22:24] <wtay> not really, no [22:24] <omega> either we make the wiki read-only, or we are open to these kinds of attacks [22:24] <wtay> mysql maybe [22:24] <omega> unless it can do decent revision control, I don't want to switch to another wiki [22:24] <wtay> can we run cron jobs to tar it up? [22:25] <omega> we'll have to [22:25] <omega> I'm downloading the restore tarball right now, all 450MB of it [22:25] <omega> only 2:10:38 remaining [22:25] <wtay> wow [22:25] <omega> media/* [22:25] <wtay> was gstreamer.net also removed? [22:25] <omega> yup [22:26] <omega> that's why wingo mentioned that the wiki data had disappeared, good thing he had a backup [22:26] <wtay> yeah [22:26] <Zeenix> wtay: i need to know what exactly you meant by template in your sentence "the sinkpad of the rtpsend should have a *template* that lists all mime-types is knows about" yesterday [22:27] <wtay> Zeenix: GstPadTemplate.. [22:27] <wtay> Zeenix: for example: in osssink, look at the GST_PADTEMPLATE_FACTORY [22:28] <Zeenix> wtay: you mean like GstPadTemplate* [22:28] <Zeenix> sink_factory (void) [22:28] <Zeenix> { [22:28] <Zeenix> return [22:28] <Zeenix> gst_padtemplate_new ( [22:28] <Zeenix> "sink", [22:28] <Zeenix> GST_PAD_SINK, [22:28] <Zeenix> GST_PAD_ALWAYS, [22:28] <Zeenix> NULL); [22:28] <Zeenix> } [22:28] <wtay> Zeenix: yeah, it changed in cvs I think.. [22:29] <Zeenix> wtay: its from gstrtpsend [22:30] <wtay> oh, ok. you need to fill it up some more then.. [22:30] <Zeenix> like ? [22:30] <wtay> like in osssink.c [22:31] dschleef (ds...@c7...) joined #gstreamer. [22:32] Action: Zeenix can access his ext2 FS from windooz [22:32] <Zeenix> s/can/cant [22:32] <omega> read-only, I hope? [22:32] <omega> oh [22:32] <wtay> darn [22:32] <omega> there is a driver somewhere, I think [22:32] <wtay> yeah [22:32] Action: ajmitch can't cos ajmitch doesn't have a windows installation to access it from :) [22:33] <omega> ajmitch: awwww ;-) [22:33] <Zeenix> wtay: the utility you directed me to, could only access the primary partitions, no extended [22:33] <ajmitch> omega: have pity on me, please ;) [22:33] <ajmitch> omega: hmm, i still can't close the one open bug that i submitted ;) [22:33] <wtay> Zeenix: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/gstreamer/gstreamer/plugins/oss/gstosssink.c?rev=1.21&content-type=text/vnd.viewcvs-markup [22:34] <ajmitch> gstmediaplay still dies on ganso_terminator.mpeg, with a massive traceback which bounces around caps nego with colorspace [22:34] <omega> sounds like a wtay bug <g> [22:34] <ajmitch> probably ;) [22:34] <wtay> yeah [22:34] <wtay> it's a feature :) [22:35] <ajmitch> heh [22:35] <wtay> Zeenix: or http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/gstreamer/gstreamer/plugins/avi/gstavitypes.c?rev=1.8&content-type=text/vnd.viewcvs-markup [22:35] <wtay> Zeenix: like the src padtemplate in avitypes [22:35] <omega> wtay: so phpwiki supposedly can use an sql backend [22:35] <omega> http://www.personaltelco.net/index.cgi/WhyWiki [22:35] <omega> and we just moved to phpwiki, so... [22:35] <wtay> yup [22:36] <omega> wtay: you look at the insncounts stuff I did last night? [22:36] <wtay> yes [22:36] <wtay> doesn't mean much to me.. [22:36] <omega> doesn't? it shows major hotspots [22:37] <omega> the code shown is after some minor changes to the GST_PAD macros [22:37] <wtay> on a source code basis, yes. not at runtime [22:37] <ajmitch> sigh, i can't watch heist.mpeg :( [22:37] <Zeenix> omega: do i get a relaxation in the deadline <g> [22:37] <Zeenix> ? [22:37] <omega> wtay: runtime too. every instruction has to be run... or skipped over, but that costs too [22:37] <wtay> ajmitch: get an Xv capable video card/X server [22:37] <ajmitch> wtay: :P [22:37] <omega> Zeenix: <sigh>... I guess.... <g> [22:38] <wtay> omega: depends on how many times it is called [22:38] Shippou (zbl...@cr...) joined #gstreamer. [22:38] <omega> wtay: exactly. most of the hotspots appear around macros [22:38] <omega> these macros do stuff inside that can be very very expensive, and is *not* skipped, but run. [22:38] <omega> if I take out _FAST, those hotspots will multiply like rabbits [22:38] <wtay> omega: combine those figures with gcov and we have a perfect measurement [22:39] <omega> yeah, that's what I want to do [22:39] <omega> what I really want is to integrate this with lxr and viewcvs <G> [22:39] <omega> problem with gcov/gprof is that you have to have a reasonable workload [22:40] <omega> wheee, under 2 hours left to grab gstreamer backup tarball [22:40] <wtay> and I cannot get it to work :( [22:40] <wtay> I tried gprof [22:40] <ajmitch> Program received signal SIGSEGV, Segmentation fault. [22:40] <ajmitch> [Switching to Thread 1024 (LWP 20566)] [22:40] <ajmitch> 0x40d21699 in parse_syshead (mpeg1parse=0x80ff948, buf=0x407f5010 "", size=4080) at mpeg1parse.c:429 [22:40] <ajmitch> 429 if (*outpad == NULL) { [22:40] <wtay> 3 times.. [22:40] Action: ajmitch guesses something is broken ;) [22:41] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [22:41] <wtay> ajmitch: yeah, it needs another check for (outpad == NULL) [22:42] <wtay> ajmitch: what video? [22:43] <ajmitch> wtay: heist.mpeg, you have that in media? [22:43] <ajmitch> the IBM ad [22:43] <wtay> I have in on my disk [22:43] <wtay> s/in/it/ [22:43] <ajmitch> k [22:43] <wtay> crashes here too [22:44] <ajmitch> see, told you it wasn't Xv :P [22:44] Action: ajmitch ducks [22:44] <wtay> EVENTS1? [22:44] <ajmitch> HEAD [22:46] <ajmitch> hmm, what's shihad.mpg in media/? i guess it's a music video [22:46] <wtay> yup [22:46] <wtay> fixed in EVENTS1, you can backport it :) [22:46] <ajmitch> bah [22:46] <wtay> 10 characters, you can manage that :) [22:47] <ajmitch> hehe [22:47] <omega> wtay: we need to just merge events1 back into head [22:48] <wtay> uhm, is it ready? [22:48] <omega> no, but everyone is doing work in it, and I think the remaining changes are non-intrusive [22:48] <wtay> true [22:49] Action: taaz hands out Merge hats [22:49] <wtay> still need to figure out the default handler [22:49] <omega> yeah [22:49] Shippou (no...@ot...) joined #gstreamer. [22:50] Action: thomasvs tries to get a website released [22:50] <thomasvs> php is nice but it sucks as well ;) [22:50] <ajmitch> hehe [22:51] <thomasvs> so does learning fifteen scripting languages which all resemble each other closely but not close enough [22:51] <ajmitch> wtay: wow, your fix works ;) [22:51] <wtay> ajmitch: that's 1 beer, when I meet you.. <g> [22:52] <ajmitch> haha [22:52] <wtay> ajmitch: and 4 for vector :) [22:52] Action: wtay ducks [22:53] <ajmitch> not much point committing the fix to HEAD when your're gonna merge ;) [22:53] rowenc (rowenc@Hong-231-192.PLU.edu) got netsplit. [22:53] dobey (do...@dr...) got netsplit. [22:53] <wtay> bah [22:54] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-26.brain.net.pk] [22:54] steveb (st...@no...) left irc: advance! [22:55] rowenc (rowenc@Hong-231-192.PLU.edu) returned to #gstreamer. [22:55] dobey (do...@dr...) returned to #gstreamer. [23:03] <dschleef> #include <libs/audio/gstaudio.h> [23:03] <dschleef> is correct now? [23:04] <omega> hmm, I dunno how we're gonna deal with those inc... [truncated message content] |
From: IRC B. <wt...@us...> - 2001-10-17 04:27:24
|
******************************************************************* [03:01] <wingo> um, omega? :) [03:02] <wingo> see, omega's busy merging our development branch back into main cvs, right? :) [03:02] <tnt> Oh, one other question... does anyone an alternate to gst_elementfactory_make("mpeg2dec","decode_video"); ? I don't have mpeg2dev.... [03:03] <dschleef> did anyone (wtay?) fix the asm code in mmxidct.S, or should I? [03:05] <tnt> [wtay is sleeping] [03:11] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [03:15] <wingo> ok, i'm headed to skool to see if i can't get something done [03:15] Nick change: wingo -> wingo-out [03:16] Nick change: tnt -> tnt-Ill-be-backk [03:16] Nick change: tnt-Ill-be-backk -> tnt-Ill-be-back [03:19] mwc (ma...@ly...) joined #gstreamer. [03:27] Nick change: tnt-Ill-be-back -> tnt [03:47] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [04:00] <tnt> I hope someone is listening... I've been looking through the gstreamer source trying to find see where factories and elements define their string names... like with << gst_elementfactory_make("mpeg2dec","decode_video"); >> a factory is given the string name "mpeg2dec", and an element under it is given the string name "decode_video". Does anyone know the source file that this is set in? [04:00] <omega> no, it's searching for a factory by the name 'mpeg2dec', this factory must have been registered with gstreamer somewhere in the past [04:01] <omega> it will use that factory to create an element that's an instance, with the given name 'decode_video' [04:01] <tnt> Oh good, you're back. [04:02] <tnt> I had a question before that was more important (to me), what can I use to replace << gst_elementfactory_make("mpeg2dec", "decode_vide"); >>... I don't have mpeg2dec on my system. [04:02] <omega> if you're doing mpeg2 video, there is none [04:03] <omega> if you're doing mpeg1, you have to replace several other element types as well [04:03] <omega> there are examples in tools/README [04:04] <tnt> Do you know where I could get the RPM for the version of mpeg2dec that gstreamer requires (I looked at rpmfind.net but couldn't find it)? [04:04] <omega> hrm, searching. [04:04] <tnt> (Version 0.2.0 of mpeg2dec.) [04:05] <omega> sf isn't responding, seeing if I can find it through the web [04:05] <omega> um, until sf decides to let me log in, I can't find one [04:13] <tnt> Hmmm, I found an RPM for mpeg2dec, but it requires ALSA lib... isn't ALSA an alternate driver system Linux... and not a general library? [04:21] <tnt> omega: SourceForge works for me... what should I search for? (Searching for "mpeg2dec" didn't get me it.) [04:24] chillywilly (da...@d1...) left irc: Read error to chillywilly[d178.as12.nwbl0.wi.voyager.net]: No route to host [04:24] <tnt> I found "libmpeg2", but I don't see any RPMs. [04:28] chillywilly (da...@d1...) joined #gstreamer. [04:45] <vektor> if walken shows up, strangle him until i wake up [04:51] <omega> vektor: what for this time? [04:51] <vektor> No release yet. [04:52] <omega> this is new? <g> [04:53] Shippou (no...@cr...) joined #gstreamer. [04:53] <omega> tnt: looks like the rpms I built went away [04:56] <vektor> omega: no, i'm just getting anxious. [04:56] <tnt> omega: That sucks. Since some of the sample MPEG files are MPEG-1, What should I use as the MPEG-1 'parser' and the `decoder'? [04:57] <omega> yeah [04:57] <omega> see line 67 of tools/README [04:57] <tnt> OK. [05:00] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [05:02] chillywilly (da...@d1...) left irc: [05:16] ajmitch (aj...@p2...) left irc: Ping timeout for ajmitch[p21-max11.dun.ihug.co.nz] [05:19] <dschleef> how do I change the status of plugins on the web pages? [05:20] <omega> there's an admin page, dunno if wingo considers it ready for use yet [05:20] <omega> I'd want some kind of login system there.... send wingo mail for now [05:21] <omega> have you had any specialib ideas recently? [05:28] <vektor> tell walken to msg me if he shows up [05:28] <omega> ok [05:33] <dschleef> omega: yes [05:33] <dschleef> but I'm trying to get gst to compile with the new binutils first [05:33] <omega> heh [05:33] <omega> I'm trying to figure out how to deal with the makefiles... [05:33] <dschleef> cleaning up all the non-PIC code in .S files [05:34] <dschleef> I've been thinking a bit about how/why one might want to approach the problem with something similar to caps [05:34] <omega> which problem? [05:35] <dschleef> kernel implementations [05:35] <omega> hmm [05:36] <dschleef> so a particular kernel implementation would be similar to a gst plugin (sort of), with a caps-like definition of the input and output values [05:36] <omega> hmm [05:36] <dschleef> hopefully lighter weight [05:36] <omega> as opposed to a fixed API for each group of kernels? [05:36] <dschleef> yes [05:36] ajmitch (aj...@p4...) joined #gstreamer. [05:37] <dschleef> I'm trying to assimilate ideas from symbolic math programs like mathematica [05:37] <omega> heh [05:40] <dschleef> I'm afraid that ideas are just that until there's an LDFLAGS+=-lcodec in gst [05:40] <omega> sounds like chicken&egg [05:41] <dschleef> yes [05:41] <dschleef> it's one of the problems i've always had with open source projects -- if I can't hack for a while, run make, then run the program and see results, I lose interest [05:42] <vektor> I agree with that sentiment. [05:42] <vektor> Which is kinda why I like to go start my own project sometimes. [05:42] <omega> yup. that's why I haven't been spending much time on it either ;-( [05:43] <vektor> come on man, I want a -lcodec. [05:43] Action: dschleef needs immediate gratification [05:43] <omega> so do I [05:43] Action: vektor prods omega to deliver [05:48] ChiefHighwater (fl...@su...) joined #gstreamer. [05:48] <omega> yo [05:48] <ChiefHighwater> Ello [05:50] <vektor> omega: you ahven't delivered yet. [05:51] <omega> vektor: um [05:51] <vektor> ;-) [05:51] <ChiefHighwater> omega is now know as UPS [05:51] <vektor> Hey, someone here just got a letter from New Jersey, should we open it? [05:51] <vektor> ;-) [05:52] Action: vektor gets targeted. [05:58] Nick change: ChiefHighwater -> Kelly [06:01] Shippou (no...@cr...) joined #gstreamer. [06:05] Nick change: Kelly -> ChiefHighwater [06:07] <vektor> man it's late. [06:08] <omega> only midnight... [06:09] <ChiefHighwater> hehe..that isn't late, right omega? [06:09] <omega> nope [06:09] <omega> vektor: wimp <g> [06:09] <ChiefHighwater> we set a new record last wekk, right? [06:09] <vishnu> hey, germans are good at big projects, right? [06:09] <omega> yup [06:10] <ChiefHighwater> vishnu:depends on which German I would suppose [06:10] <vishnu> ChiefHighwater: oh, yah. right [06:11] BeNOW_ (an...@ds...) joined #gstreamer. [06:11] <BeNOW_> heya... anyone alive? [06:11] <ChiefHighwater> no, only Zombies here <g> [06:11] Action: vishnu quivers like jello [06:11] <vektor> my parent hasn't waited on me yet :( [06:11] <omega> yup, just us [06:11] <BeNOW_> ooh jello.. kewl. [06:12] Action: ChiefHighwater hands out spoons [06:12] <BeNOW_> I'm at a buddies trying to get gstreamer going and get this error: [06:12] <BeNOW_> libtool: link: error: cannot link shared libraries into libtool libraries [06:12] <omega> upgrade libtool [06:12] <BeNOW_> k, yeah, I remember.. thanks [06:13] <omega> and rerun autogen, or just run `libtoolize --force --copy` [06:35] Nick change: wingo-out -> wingo [06:35] omega (om...@om...) left irc: Ping timeout for omega[omegacs.net] [06:36] <wingo> dschleef: the admin doober is ready for use, gstreamer.net/admin.php [06:42] <vektor> anyone alive? [06:42] <tnt> I'm alive,... just watching TV. [06:42] <wingo> sorta :) [06:43] <vektor> wingo: tell me if you can see a pattern in these numbers. [06:43] <wingo> dcc doesn't work for me :-\ [06:43] <wingo> firewall, etc [06:43] <vektor> www.billybiggs.com/values.txt [06:44] <wingo> checking... [06:46] <wingo> um, general downward trend? :) [06:46] <vektor> you think? [06:47] <vektor> :) [06:47] <wingo> what is it? [06:47] <vektor> um [06:47] <vektor> it's a plot of the difference in how much video i've captured vs how much audio [06:47] <vektor> basically [06:48] <vektor> and i think the slope indicates the error in the sampling rate of my soundcard [06:48] <vektor> i.e., how off it is from 44100. [06:48] <vektor> but i'm having difficulty quantifying it. [06:48] <dschleef> I have that problem with one of my machines. all the mp3s sound about 1.5 tones high [06:49] <vektor> yeah. [06:49] <wingo> what card? [06:49] <vektor> soundcards suck. [06:49] <vektor> I have an es1371. [06:50] <wingo> hum. i have a es1370. [06:52] <vektor> Yep. Basically the same. [06:52] <vektor> Have you attempted to quantify the error? [06:52] <wingo> no [06:52] <vektor> ok. [06:52] <wingo> but i don't do video so it matters less [06:52] <vektor> These numbers should be flipped, btw. [06:52] <vektor> At the beginning, I have less audio than video captured. [06:52] <vektor> By the end, I have more audio than video. [06:53] <vektor> So, it's actually a general upward trend. [06:53] <vektor> The data looks pretty cool using 'gnuplot'. [06:53] <vektor> Do you use gnuplot? [06:53] <vektor> It's cool. [06:53] <vektor> Just run 'gnuplot' and then type: plot 'values.txt' [06:53] <vektor> It's cool. [06:53] <dschleef> isapnp: Card 'ESS ES1869 Plug and Play AudioDrive' [06:53] <wingo> yeah. i just finished my BS in nuclear engineering, i used it all the time [06:54] <vektor> haha, cool. [06:54] <wingo> lots-o-data-analysis [06:54] <vektor> at what school? [06:54] <wingo> north carolina state [06:54] <dschleef> I've done some work on audioscale, but it still sounds like crap without lots of CPU [06:54] <ShrimpX> nuclear engineering. that sounds 3l3373 [06:54] <wingo> :) [06:54] <vektor> neat, my girlfriend's cousin went to nc state :) [06:55] <vektor> just graduated and i think he's starting a masters there. [06:55] <vektor> some engineering something. [06:55] <wingo> nifty. [06:56] <vektor> bah fucking school. [06:56] <vektor> i doin't want to do this assignment. [06:56] <vektor> only, erm, 7 more months of school to go! [06:56] <ajmitch> heh [06:56] <wingo> no kidding! [06:56] <vektor> wingo: so, what are you doing now that you have a BS? [06:56] <ajmitch> only 2 weeks & i have exams! [06:56] <wingo> well, i walk for NE in the fall [06:57] <ajmitch> then i have my 3 month summer break ;) [06:57] <wingo> and then in physics and spanish in the spring :) [06:57] <wingo> after that who knows [06:58] <wingo> what are you studying, vektor? [06:58] <vektor> computer science [06:58] <vektor> but [06:59] <vektor> i'm going to double major in 'combinatorics/optimization' [06:59] <vektor> cause that's just cooler [06:59] <wingo> heh [06:59] <vektor> computer science is so lame. [06:59] <vektor> i took a fuckload of math. [06:59] <dschleef> cs is so lame that I did physics [07:00] <vishnu> cs is so lame that i left school without a degree [07:00] <wingo> right on! [07:00] <dschleef> well, i did that, too [07:00] <wingo> to both! [07:00] <taaz> whoop! got sw:tpm dvd [07:00] <vektor> taaz: you're a lame ass. [07:00] <vektor> :) [07:01] <taaz> vektor: you're an alcoholic lame ass. [07:01] <taaz> ;)) [07:01] <vektor> hrm. [07:02] <taaz> soo.... how do you plan on getting walken to release this calendar year? [07:03] <vektor> i'm waiting for him to show up [07:04] <vektor> so i can POUND HIS ASS HALF WAY AROUND THE GLOBE. [07:04] <vektor> or osmething like that. [07:04] <wingo> 'night. [07:04] <vektor> lates [07:04] Nick change: wingo -> wingo-z [07:04] <wingo-z> , yo :) [07:05] <vektor> yea lates yo [07:18] <taaz> oh the horror! [07:19] <taaz> evil evil lucasfilm [07:19] <taaz> putting ms winders crap on a dvd meant to last a lifetime [07:19] Action: taaz sighs [07:19] ChiefHighwater (fl...@su...) left irc: Leaving [07:22] <vektor> well, i'm going to sleep [07:22] <vektor> lates all [07:22] BeNOW_ (an...@ds...) left irc: [BX] Gary Coleman uses BitchX. Whatchoo talkin bout foo? [07:22] <vektor> kick walken for me [07:22] <vektor> if you see him [07:33] steveb (st...@no...) joined #gstreamer. [07:58] Bala (ry...@sk...) joined #gstreamer. [08:01] Bala (ry...@sk...) left #gstreamer (Client Exiting). [08:07] tnt_ (tn...@h2...) joined #gstreamer. [08:07] mwc (ma...@ly...) left irc: Ping timeout for mwc[lychee.ntu.edu.au] [08:08] tnt (tn...@h2...) left irc: Ping timeout for tnt[h24-83-72-250.vs.shawcable.net] [08:09] mwc (ma...@ly...) joined #gstreamer. [08:20] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [08:42] walken (fo...@c1...) joined #gstreamer. [08:42] <walken> hi [08:43] <taaz> yo [08:43] <walken> whats up ? [08:43] <taaz> hmm... vektor asked us to kick your ass [08:44] <walken> I'm still putting my machine back together [08:44] <taaz> dude... you know what i just recently bought? [08:44] <walken> no ? [08:44] <taaz> an ibm drive [08:44] <walken> eeeek [08:44] <walken> which series ? [08:45] <taaz> like less than a month before i hear from a million people they suck [08:45] <taaz> blah i dunno... 60gxp i think [08:45] <taaz> 40G one [08:45] <walken> I got a DTLA something [08:45] <walken> GXP sucks too [08:45] Action: taaz sighs [08:45] <walken> now I have two western digitals [08:45] Action: walken going raid-1 [08:46] <walken> its not setup yet though [08:46] <taaz> walken: join ogle now [08:46] <taaz> dude talkign idct at me [08:49] <taaz> hmm... so it's not like i can return a drive on speculation it may soon fail [08:49] <taaz> and raid is inconvienient in this particular box... [09:03] thomasvs (th...@ad...) joined #gstreamer. [09:13] <dschleef> my WD Caviar died about a week ago [09:14] <walken> heh [09:14] <walken> all modern disks are crap [09:14] <walken> its really bad [09:15] <taaz> so... where can i find a good raid controller? ;) [09:16] <dschleef> the RIAA must have known I had 12 DVDs backed up on it [09:16] <dschleef> s/RI/MP/ [09:16] <walken> david. do it in software [09:17] <walken> the only drawback is you need a bigger case to put all these disks in [09:17] <walken> well for 2 disks its still fine but you cant put much more in a regular case [09:17] <walken> I got 2*60G now [09:17] <taaz> yeah i could fit another one i guess [09:33] Nick change: tnt_ -> tnt [09:43] <dschleef> can I return a laptop drive that makes too much noise? It makes more noise that the drive it replaced, and I couldn't stand that. [09:47] Nick change: taaz -> taazzzz [09:53] tnt (tn...@h2...) left irc: Client Exiting [10:02] xav_ (xav@AMontpellier-201-1-1-184.abo.wanadoo.fr) left irc: Ping timeout for xav_[AMontpellier-201-1-1-184.abo.wanadoo.fr] [10:08] xav_ (xav@AMontpellier-201-1-1-184.abo.wanadoo.fr) joined #gstreamer. [10:24] walken (fo...@c1...) left irc: l8r [10:31] tnt (tnt@24.83.72.250) joined #gstreamer. [10:31] mwc (ma...@ly...) left irc: [x]chat [10:36] <dschleef> all quiet [10:36] <dschleef> that means nobody will hear me if I start screaming [10:41] <tnt> I heard it :-) [10:46] tnt (tnt@24.83.72.250) left irc: Client Exiting [11:45] xav_ (xav@AMontpellier-201-1-1-184.abo.wanadoo.fr) left irc: Read error to xav_[AMontpellier-201-1-1-184.abo.wanadoo.fr]: EOF from client [12:07] manuel (manuel.gro@213.174.92.50) joined #gstreamer. [12:27] manuel (manuel.gro@213.174.92.50) left irc: Client Exiting [13:06] dschleef (ds...@c7...) left irc: Ping timeout for dschleef[c799921-a.pinol1.sfba.home.com] [13:19] manuel (manuel.gro@213.174.92.50) joined #gstreamer. [13:28] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [13:59] thomasvs (th...@ad...) left irc: Ping timeout for thomasvs[adsl-64938.turboline.skynet.be] [14:00] thomasvs (th...@ad...) joined #gstreamer. [14:02] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [14:03] thomasvs (th...@ad...) joined #gstreamer. [14:04] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [14:06] thomasvs (th...@ad...) joined #gstreamer. [14:07] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [14:09] thomasvs (th...@ad...) joined #gstreamer. [14:10] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [14:11] thomasvs (th...@ad...) joined #gstreamer. [14:12] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [14:14] thomasvs (th...@ad...) joined #gstreamer. [14:15] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [14:17] Zygo (zbl...@24...) joined #gstreamer. [14:32] Shippou (no...@cr...) joined #gstreamer. [14:39] asmod (stevec@64.5.222.2) joined #gstreamer. [15:00] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [15:21] Zygo (zbl...@24...) left irc: Ping timeout for Zygo[246.chipworks.com] [15:31] Zygo (zbl...@24...) joined #gstreamer. [15:35] thomasvs (th...@ad...) joined #gstreamer. [15:50] evil_monk (trevo@208.141.162.68) joined #gstreamer. [15:50] xav_ (xav@AMontpellier-201-1-1-184.abo.wanadoo.fr) joined #gstreamer. [15:54] xav_ (xav@AMontpellier-201-1-1-184.abo.wanadoo.fr) left irc: Read error to xav_[AMontpellier-201-1-1-184.abo.wanadoo.fr]: EOF from client [16:00] Shippou (no...@cr...) joined #gstreamer. [16:01] manuel (manuel.gro@213.174.92.50) left irc: Client Exiting [16:06] thomasvs (th...@ad...) left irc: Ping timeout for thomasvs[adsl-64938.turboline.skynet.be] [16:08] thomasvs (th...@ad...) joined #gstreamer. [16:09] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [16:11] thomasvs (th...@ad...) joined #gstreamer. [16:11] sienap (sy...@s3...) joined #gstreamer. [16:11] <sienap> hej [16:11] <sienap> Can someone help me with a little odd coding problem ? [16:12] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [16:12] <sienap> struct loopstruct {// This is for the FSCKING loops. (experimental). [16:12] <sienap> unsigned int start; // IP start of the loop [16:12] <sienap> unsigned int end; // IP end of the loop [16:12] <sienap> unsigned int level; // Depth (higher is more nested) when 2 loops are at the same level [16:12] <sienap> }; // like in [ [] [] ] then the loop core should keep the [] [] seperated [16:12] <sienap> this is a struct i have declared [16:12] <sienap> void loops(unsigned char *code) { [16:12] <sienap> unsigned int ip = 0; // Fakir IP for loops checker [16:12] <sienap> unsigned int numstarts = 0; [16:12] <sienap> unsigned int numends = 0; [16:12] <sienap> loopstruct loopje; [16:12] <sienap> however this gives an error, being : [16:12] <sienap> [root@Oegaah bf]# gcc -o bfinter bfinter.c [16:12] <sienap> bfinter.c: In function `loops': [16:12] <sienap> bfinter.c:31: `loopstruct' undeclared (first use in this function) [16:12] <sienap> bfinter.c:31: (Each undeclared identifier is reported only once [16:12] <sienap> bfinter.c:31: for each function it appears in.) [16:12] <sienap> bfinter.c:31: parse error before `loopje' [16:12] <sienap> [root@Oegaah bf]# [16:12] <sienap> and that makes me a sad panda [16:12] <sienap> someone a suggestion ? [16:13] thomasvs (th...@ad...) joined #gstreamer. [16:13] <sienap> hej [16:15] sienap (sy...@s3...) left irc: User abort with 5 Ctrl-C's [16:16] bstard (Lor...@a2...) left irc: Ping timeout for bstard[a213-84-40-242.adsl.xs4all.nl] [16:16] bstard (Lor...@a2...) joined #gstreamer. [16:25] xav_ (xav@AMontpellier-201-1-1-184.abo.wanadoo.fr) joined #gstreamer. [17:00] bstard (Lor...@a2...) got netsplit. [17:03] bstard (Lor...@a2...) returned to #gstreamer. [17:23] thomasvs (th...@ad...) left irc: Ping timeout for thomasvs[adsl-64938.turboline.skynet.be] [17:24] thomasvs (th...@ad...) joined #gstreamer. [17:25] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [17:26] xav_ (xav@AMontpellier-201-1-1-184.abo.wanadoo.fr) left irc: Read error to xav_[AMontpellier-201-1-1-184.abo.wanadoo.fr]: EOF from client [17:27] thomasvs (th...@ad...) joined #gstreamer. [17:28] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [17:30] thomasvs (th...@ad...) joined #gstreamer. [17:31] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [17:31] rowenc (rowenc@Hong-231-192.PLU.edu) left irc: I am quitting [17:33] thomasvs (th...@ad...) joined #gstreamer. [17:34] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [17:35] thomasvs (th...@ad...) joined #gstreamer. [17:37] <asmod> Well this is exciting. Certain areas of the building that I am currently in, and the one beside it, have been evacuated because of an Anthrax scare in the US Consulate office that is a few floors above me. [17:39] <thomasvs> I thought they had split up ;) [17:41] <asmod> Well now they are threatening to take over the buildings intercom and play their music on full blast for the rest of the day. [17:51] Nick change: asmod -> asmod_anthrax [17:51] asmod_anthrax (stevec@64.5.222.2) left #gstreamer. [18:00] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [18:00] asmod (stevec@64.5.222.2) joined #gstreamer. [18:17] dobey (dobey@141.154.95.104) joined #gstreamer. [18:25] Nick change: taazzzz -> taaz [18:32] Action: evil_monk is away: I'm busy [19:05] Shippou (no...@ot...) joined #gstreamer. [19:10] tnt (tn...@h2...) joined #gstreamer. [19:10] tnt (tn...@h2...) left irc: Client Exiting [19:10] tnt (tn...@h2...) joined #gstreamer. [19:14] vishnu (jo...@cx...) joined #gstreamer. [19:15] Nick change: vishnu -> walken [19:15] Nick change: walken -> vishnu [19:16] dobey (dobey@141.154.95.104) left irc: Ping timeout for dobey[141.154.95.104] [19:19] Nick change: taaz -> taaz-lunch [19:24] asmod (stevec@64.5.222.2) left irc: Ping timeout for asmod[64.5.222.2] [19:26] asmod (stevec@64.5.222.2) joined #gstreamer. [19:42] <tnt> Is anyone listening? [19:43] <asmod> yup [19:44] <tnt> Has anyone noticed that there's a few functions that are NOT in libgst (in the 0.2.1 release of gstreamer)? [19:45] <tnt> The latest ones I found are gst_caps_get_mime and gst_caps_get_name.. [19:46] <tnt> Would it be better to work from the CVS?... rather than the RPM (like I'm doing now)? [19:46] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [19:48] <tnt> Anyone know how stable the CVS version is right now? [19:51] <vishnu> yes, i only use cvs [19:51] <vishnu> whether it is stable depends on what you are trying to do [19:52] <vishnu> SMP threading is completely broken [19:52] <vishnu> on the other hand, i have a mpeg2 player that is stable enough for my needs [20:03] Action: evil_monk is back (gone 01:30:54) [20:03] <tnt> I just want something that is usable :-) The functions I want aren't in latest RPM version. [20:05] Nick change: taaz-lunch -> taaz [20:19] <vishnu> heh, yah, use cvs [20:20] <tnt> Is there a list of the dependencies for gstreamer, anywhere? [20:20] <vishnu> REQUIREMENTS ? [20:21] <tnt> Like what else it depends on... for example: mpeg2dec, openquicktime, etc.... [20:21] <tnt> I Just realized you are talking about a file in the CvS... sorry. [20:21] ChiefHighwater (pa...@te...) joined #gstreamer. [20:42] arik (ar...@sd...) joined #gstreamer. [20:43] <arik> morning [20:43] <tnt> Good morning. [20:48] wtay-zZz (wi...@ca...) left irc: Read error to wtay-zZz[cable-195-162-215-201.upc.chello.be]: EOF from client [20:56] wtay (wi...@ca...) joined #gstreamer. [20:56] #gstreamer: mode change '+o wtay' by ChanServ!s@ChanServ [20:56] #gstreamer: mode change '-o wtay' by wtay!wi...@ca... [21:11] Nick change: taaz -> taaz-away [21:28] arik (ar...@sd...) left irc: leaving [21:31] BeNOW (an...@ds...) left irc: BitchX-1.0c18 -- just do it. [21:32] omega (om...@om...) joined #gstreamer. [21:37] Zygo (zbl...@24...) got netsplit. [21:37] vektor (reet@HSE-Kitchener-ppp194670.sympatico.ca) got netsplit. [21:37] ajmitch (aj...@p4...) got netsplit. [21:37] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [21:37] vishnu (jo...@cx...) got netsplit. [21:37] asmod (stevec@64.5.222.2) got netsplit. [21:37] evil_monk (trevo@208.141.162.68) got netsplit. [21:37] wtay (wi...@ca...) got netsplit. [21:37] taaz-away (dlehn@66.37.66.32) got netsplit. [21:37] thomasvs (th...@ad...) got netsplit. [21:37] Shippou (no...@ot...) got netsplit. [21:40] vektor (reet@HSE-Kitchener-ppp194670.sympatico.ca) returned to #gstreamer. [21:40] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [21:40] ajmitch (aj...@p4...) returned to #gstreamer. [21:40] Zygo (zbl...@24...) returned to #gstreamer. [21:40] evil_monk (trevo@208.141.162.68) returned to #gstreamer. [21:40] vishnu (jo...@cx...) returned to #gstreamer. [21:40] asmod (stevec@64.5.222.2) returned to #gstreamer. [21:40] arik (ar...@sd...) joined #gstreamer. [21:40] taaz-away (dlehn@66.37.66.32) returned to #gstreamer. [21:40] thomasvs (th...@ad...) returned to #gstreamer. [21:40] Shippou (no...@ot...) returned to #gstreamer. [21:40] wtay (wi...@ca...) returned to #gstreamer. [21:40] #gstreamer: mode change '+o wtay' by ChanServ!s@ChanServ [21:40] <arik> omega: you here? [21:40] <omega> yeah [21:41] <arik> hey [21:41] <arik> finally got everything straightenedout with my school [21:41] <arik> so all is good [21:41] <arik> did you get a chance to look at the code? [21:41] <omega> nope, will do so now [21:41] <arik> alright, thanks ;-) [21:41] <arik> i really want to figure this out cause it makes coding a bit hard ;-) [21:41] <omega> I'm thinking of writing my own gtk2-based player just so I have a better grasp of the issues.. [21:42] <arik> hmm [21:42] <arik> that's sounds like a good idea [21:42] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [21:43] <arik> heh, if you really want (and don't seen anything wrong with gstplay.c) i can send you the whole thing [21:43] <arik> *grin* [21:43] <omega> lemme decipher main.c/gstplay.c first <g> [21:43] <arik> hehe [21:44] <arik> atm i'm still puzzling over the crappiness of the picogui layout system [21:44] <arik> well either the crappiness or my lack of understanding about something that makes it less crappy [21:46] <omega> ok, so what's the basic sequence of calls to gstplay.c? starting with gst_play_new() of course [21:47] <arik> right [21:47] Action: tnt is away: I'm busy [21:47] <arik> then um [21:47] <arik> gst_play_set_uri [21:47] <arik> then gst_play_play [21:47] <arik> then gst_play_get_pipeline [21:47] <arik> then gst_play_stop and on and on [21:48] <omega> ok, you have to iterate the pipeline yourself then [21:48] <arik> yeah at the moment [21:48] <arik> until i add thread [21:48] <arik> s [21:48] <arik> main_play in main.c does the iterating [21:48] <arik> using pgSetIdle [21:48] <omega> ok [21:48] <omega> what do I need to try to run this thing locally? [21:48] <omega> special copy of picogui? [21:48] <arik> heh [21:49] <arik> no, just the standard one from picogui.org (you might need cvs version) [21:49] <arik> but you need a few more files [21:49] <wtay> yo [21:49] <arik> you can of course, just write a basic text frontend to gstplay.c [21:49] <arik> hey [21:49] <arik> but you still need gstplay.h and gstplayprivate.h [21:49] <ajmitch> hey [21:49] <arik> hey aj [21:50] <arik> all i know is that if i add something to any file other then main nothing bad happens [21:50] <arik> but if i add anything to main it tends to core dump [21:50] <arik> until i add something else [21:50] <arik> course it could be in gstplay.c cause i haven't been adding anything to that so i don't know what it would do [21:50] <arik> sorry i can't be more helpful [21:51] <ajmitch> sounds fun [21:51] <arik> heh [21:51] <wtay> heh, from 7.7 sec down to 0.7 seconds with my faster buffer alloc code for 5M buffers [21:52] <omega> wow [21:52] <arik> wow [21:53] Last message repeated 1 time(s). [21:53] <wtay> amazing [21:53] <omega> is this the CAS-based one? [21:53] <wtay> yes.. [21:53] <omega> ok [21:53] <wtay> but the mutex_new happens only once (before putting the buffer in the pool) [21:53] <omega> I think we'll need three impls eventually: mutex, CAS, and atomic inc/dec [21:54] <omega> first two for MP systems, last for known UP systems [21:55] <wtay> atomic inc/dec works for SMP too [21:55] <omega> can it? [21:55] <omega> can you implement spinlocks effectively with them? I've tried... [21:55] <wtay> yes, look at atomic.h, it adds a lock before the inc etc.. [21:56] <wtay> omega: you can't [21:56] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [21:56] <omega> right, that makes them useless for some things [21:56] <omega> but if you can derive a mechanism for the gstmemchunk stuff that uses them, we're set [21:57] <wtay> I renamed it to GstMemPool [21:57] Zygo (zbl...@24...) left irc: Read error to Zygo[246.chipworks.com]: Connection reset by peer [21:58] <arik> omega: so nothing pops out as obivously brain dead? [21:59] <omega> arik: haven't finished reading [21:59] <arik> omega: alright, i'll be patient ;-) [22:00] Action: ChiefHighwater watches arik pace the floor wringing his hands [22:01] <arik> hehe [22:01] <arik> something like that ;-P [22:02] <omega> arik: what pgui modules do I need? [22:02] <arik> omega: cli_c and pgserver should do it [22:03] <omega> 'cli' ? [22:03] <arik> hmm? [22:03] <omega> stands for 'client' I hope? [22:03] <arik> heh yeah [22:03] <arik> c client lib [22:03] <ajmitch> hehe [22:04] <omega> I get an autogen.sh error in automake [22:04] <omega> input/Makefile.am:6: libinput_a_SOURCES includes configure substitution `@INPUT@' [22:04] <omega> vidbase/Makefile.am:7: libvidbase_a_SOURCES includes configure substitution `@VIDBASE@' [22:04] <omega> video/Makefile.am:6: libvideo_a_SOURCES includes configure substitution `@DRIVER@' [22:04] <omega> widget/Makefile.am:24: libwidget_a_SOURCES includes configure substitution `@WIDGET@' [22:04] <arik> um [22:04] <arik> did you do pgserver first? [22:04] <omega> that is pgserver [22:04] <arik> hmm [22:04] <omega> I'm forcing it [22:04] <arik> from cvs or from tar? [22:05] <omega> cvs HEAD [22:05] <arik> hmm [22:05] <arik> should work [22:05] <arik> i think i go those same errors actually [22:05] <arik> and forced it and it worked [22:05] <arik> but i could be remembering wrong [22:11] <omega> ok, I need the rest of the player files now.... [22:11] <ajmitch> arik: this for the picogui player? [22:12] <arik> omega: alright, i'll tar it up for you [22:12] <arik> ajmitch: yep [22:13] steveb (st...@no...) left irc: advance! [22:14] <ajmitch> arik: when do we get to see your wonderful work on gstmediaplay ? ;) [22:14] Action: ajmitch starts backing away... [22:15] <arik> alright i'm sending [22:15] <arik> ajmitch: you better back away! ;-P [22:15] <arik> hehe [22:15] <ajmitch> hehe [22:15] <ajmitch> arik: i'm just curious ;) [22:16] <arik> ajmitch: heh, when i have time to make it work a bit better [22:16] <arik> ajmitch: i would just commit it but i think people might mind that it doesn't really work [22:16] <arik> if you think i'm wrong just let me know ;-P [22:17] <arik> omega: still sending [22:17] asmod (stevec@64.5.222.2) left irc: [x]chat [22:17] <ajmitch> arik: you could commit on a branch & then others can look at it a bit to see what the problems are [22:18] <ajmitch> ask omega first ;) [22:18] <omega> actually, the player should be in a new cvs module.... [22:18] <arik> ajmitch: yeah i should, i just can't remember how to do cvs branches ;-) [22:18] <arik> heh [22:18] <arik> omega: you think so? [22:18] <omega> of course [22:18] <ajmitch> omega: hmm, k [22:18] <omega> this single-module stuff is for the birds [22:18] <ajmitch> hehe [22:19] <arik> omega: can i get it compile without installed gstreamer if it's in a seperate module? [22:19] <ajmitch> plugins shoudl be separate too [22:19] <omega> that's the plan, yes [22:19] <arik> omega: hmm, that could work then [22:19] <omega> need to get some changes into pkg-config, then we can do that [22:19] <arik> damn [22:19] <arik> that's gonna be annoying to update [22:20] <ajmitch> why? [22:20] <arik> omega: ok, file sent [22:20] <arik> why what? [22:21] <ajmitch> what will be annoying to update, and why? [22:22] <arik> gst [22:22] <arik> cause you will have to go to each of the plugin dirs [22:22] <arik> and update seperately [22:24] <ajmitch> hmm, depends on how you split cvs [22:24] <ajmitch> as it is i have a shell script that recurses thru all checked out modules & updates them anyway ;) [22:24] <arik> heh [22:24] <arik> yeah [22:25] <ajmitch> as there is currently gst-bind, www, and gstreamer modules [22:26] <omega> arik: where's picoplay.c ? [22:26] <omega> dead file? [22:26] <arik> yeah [22:26] <arik> dead [22:26] <arik> sorry, shouldn't have included the ~ files [22:26] <omega> I'm setting this up for autoconf [22:26] <arik> oh, thanks [22:26] Action: arik is lazy [22:27] <arik> ;-) [22:28] <ajmitch> heh [22:29] Action: ajmitch is probably far lazier than arik [22:29] <arik> you'd think i would be better with the auto* tools after helix code but i still don't really know what i'm doing with makefiles and such [22:29] <arik> prob aj, prob ;-) [22:30] arik (ar...@sd...) left irc: Changing server [22:31] arik (ar...@sd...) joined #gstreamer. [22:33] <ajmitch> oh python is such fun [22:43] <tnt> Will gstreamer (from the CVS) work with a52dec??? a52dec is what the newest version of ac3dec has been named. Or should I get the older version (... ac3dec-0.6.1)? [22:44] <wtay> tnt: a52dec should work fine [22:44] <wtay> anyone care to guess how many buffers are created in Aliensone.mpg playback? [22:45] <ChiefHighwater> sure, umm...73? [22:45] <wtay> no :) [22:45] Action: ChiefHighwater is sooo clueless [22:45] <ChiefHighwater> higher/lower? [22:45] <wtay> higher :) [22:45] <ChiefHighwater> 412 [22:46] <wtay> higher :) [22:46] <ChiefHighwater> 1768 [22:46] <wtay> higher :) [22:46] <ChiefHighwater> uhoh [22:46] <ChiefHighwater> 12493 [22:46] <wtay> 3519 [22:47] <ChiefHighwater> yeah, that's what I meant..didn't I say that? Laff [22:47] <ajmitch> hehe [22:47] <wtay> almost :) [22:47] <ajmitch> who maintains gst-bind? [22:47] <arik> not me ;-) [22:48] <wtay> I do the java stuff [22:48] <ajmitch> hmm, it's an error in a generated file, Gst_wrap.c in perl5 [22:48] <wtay> I don't maintain the perl bindings :) [22:49] <ajmitch> heh [22:50] <ajmitch> hmm, no python bindings there yet [22:50] <ajmitch> that's not good [22:50] <wtay> 5788 buffers for heist60.mpeg [22:50] <wtay> pretty good I would say [22:50] <omega> these are total allocated buffers (peak)? [22:50] <omega> or cumulative? [22:51] <wtay> cumulative [22:51] <omega> ok, and how many malloc's were done? <g> [22:51] <wtay> 1 [22:52] <wtay> not counting the muxtes allocs (32) [22:52] <omega> what was peak in-flight buffer count? [22:52] <wtay> dunno yet.. [22:56] Action: omega is very tempted to write a new pkgconfig in shell [22:57] <arik> heh [22:57] <arik> do it! [22:57] <arik> but first figure out my problem ;-P [22:57] Action: ajmitch hopes omega isn't serious ;) [22:58] <ajmitch> hehe [22:58] <omega> pkgconfig has some design flaws imo that need to be solved [22:58] <omega> one of which is the fact that it's a binary [22:58] <omega> others have complained about the same thing [22:58] <omega> no other tool requires itself to be installed on a end-user's machine for compilation [22:58] <wtay> max live: 47 [22:59] <omega> oooh [22:59] <omega> libtool copies itself into the distribution [23:00] <wtay> Dobly peeks at 111 live [23:01] <wtay> we should have an event to limit the queue length [23:02] <omega> an event to set the queue depth? [23:02] <omega> or a signal that fires at various points in the queue? [23:03] <wtay> also, yes [23:03] <wtay> there is a suggested buffersize in the mpeg system stream too IIRC [23:03] <omega> yeah [23:06] <wtay> omega: what gives: time ../tools/gstreamer-launch fakesrc num_buffers=500000 silent=true ! fakesink silent=true [23:06] <omega> ? [23:08] <wtay> still running whith HEAD code... [23:08] Nick change: taaz-away -> taaz [23:08] <omega> wtay: what's the problem? [23:08] thomasvs (th...@ad...) left irc: Ping timeout for thomasvs[adsl-64938.turboline.skynet.be] [23:08] <wtay> omega: just trying a benchmark on old vs new code.. [23:09] <wtay> omega: but my events1 code is sufficiently broken to try it.. EOS doesn't work on HEAD [23:09] <omega> oh [23:10] <omega> we need to merge events1 and get that stuff finished [23:10] <arik> yep [23:10] <omega> ASAP [23:10] <arik> EOS doesn't work at all [23:12] <arik> omega: did you get it to build? [23:12] <omega> arik: still fighting with pkgconfig [23:12] <arik> omega: alright [23:13] Shippou (no...@ot...) joined #gstreamer. [23:20] evil_monk (trevo@208.141.162.68) joined #gstreamer. [23:21] Action: tnt is back (gone 01:34:05) [23:23] <omega> arik: why are you linking in gnome, vfs, gal, and ogg?? [23:23] <arik> ugh [23:23] <arik> there is no reason [23:23] <omega> ok [23:23] <arik> cruft [23:23] <omega> do you need any of those? [23:24] <arik> nope [23:24] <omega> ok [23:25] thomasvs (th...@ad...) joined #gstreamer. [23:26] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:27] thomasvs (th...@ad...) joined #gstreamer. [23:28] <arik> alright i'm getting offline for a bit [23:28] <arik> back later [23:28] <omega> l8r [23:28] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:28] arik (ar...@sd...) left #gstreamer. [23:30] thomasvs (th...@ad...) joined #gstreamer. [23:31] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:33] thomasvs (th...@ad...) joined #gstreamer. [23:33] <tnt> I got the CVS version of gstreamer... Where's the ./configure file? [23:33] <omega> there isn't until you run autogen.sh, which runs it for you [23:33] <omega> only use configure directly from a CVS build if you know what options you need to specify [23:33] <tnt> So you run that first, and then do the ./configure, make, make install? [23:34] <omega> somewhere there's supposed to be a documente describing that, but I don't think anyone has written it yet [23:34] <omega> no, autogen runs configure for you [23:34] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:34] <tnt> So then ./autogen.sh, then make, then make install? [23:35] <wtay> no make install is needed [23:35] <wtay> it actually doesn't want to do make install even [23:35] <tnt> So then how does it get installed?... do I do it by hand? [23:35] <wtay> it runs fine from its build dir [23:35] <wtay> you need to rerun ./configure then [23:36] <omega> we must have a doc describing this. didn't someone start one? [23:36] thomasvs (th...@ad...) joined #gstreamer. [23:36] <ajmitch> omega: i gues syou need ppl to write docs :) [23:36] <tnt> Ummm, OK, do I run ./configure (which doesn't exist) or ./autogen.sh? [23:36] <wtay> tnt: ./autogen.sh runs configure with --enable-plugin-builddir [23:36] <wtay> tnt: first run ./autogen.sh [23:37] <wtay> it'll run configure with the wrong options for you.. [23:37] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:37] <wtay> then you run ./configure again with no options and it'll work fine [23:37] Action: omega starts to write a document [23:38] <wtay> I thought there was one.. [23:38] <omega> there's a little bit in the wiki, but it doesn't actually explain anything [23:38] <wtay> omega: http://gstreamer.net/wiki/index.php?GstHowTo [23:38] thomasvs (th...@ad...) joined #gstreamer. [23:38] <wtay> maybe add the extra info there [23:38] <omega> no, writing a new doc [23:39] <omega> needs a different structure, dedicated doc [23:39] <wtay> ok [23:39] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:40] <omega> as soon as we get events done and put out a release, we'll start moving plugins out [23:40] <omega> the sooner the better IMO [23:40] <omega> that will significantly ease the pain of building gstreamer core [23:40] <ajmitch> apparantly make dist needs fixed first [23:40] <omega> no, events1 needs to be merged first <g> [23:40] <tnt> You show add pkg-config to the REQUIREMENTS file. [23:41] <ajmitch> yeah, ok,:) [23:41] thomasvs (th...@ad...) joined #gstreamer. [23:41] <omega> tnt: of 0.2.1, CVS HEAD, or EVENTS1? [23:41] Action: omega adds to HEAD [23:41] <ajmitch> omega: what needs to be done before you can merge? [23:42] <omega> ajmitch: more events stuff [23:42] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:42] <omega> my thought behind starting a simple player of my own was to provide a platform to drive the completion of events [23:42] Action: ajmitch will wait patiently for omega to do this events 'stuff' ;) [23:42] <tnt> omega: What? [23:42] <omega> tnt: nevermind [23:42] bstard (Lor...@a2...) joined #gstreamer. [23:43] <tnt> Oh... REQUIREMENTS... the CVS. [23:43] thomasvs (th...@ad...) joined #gstreamer. [23:44] <omega> um, there appears to be no more pkgconfig home page [23:44] <ajmitch> eh? [23:44] <ajmitch> not good [23:44] <omega> nor a freshmeat entry [23:44] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:45] <tnt> pkg-congif is at http://www.freedesktop.org/software/pkgconfig/ [23:45] <wtay> omega: shall I clutter EVENTS1 some more with a test/bufspeed/ dir (gstmempool.[ch] +test apps +gstbuffer.[ch]) ? [23:45] <omega> oh, duh. I was looking at the wrong domain name [23:46] <taaz> wtay: yea [23:46] #gstreamer: mode change '-o wtay' by wtay!wi...@ca... [23:46] <wtay> uhm thomasvs is kicked? [23:46] thomasvs (th...@ad...) joined #gstreamer. [23:46] <omega> thomasvs: what's going on with your nick? [23:46] <ajmitch> wtay: nah, i think he just doesn't enter his password [23:46] <ajmitch> wtay: and nickserv is unhappy ;) [23:47] <omega> it doesn't autokick [23:47] <wtay> thomasvs: you're gonna get kicked again... [23:47] <ajmitch> omega: depends on the nickserv settings [23:47] <wtay> it doesn't kick me.. [23:47] <omega> oh? I didn't see that feature [23:47] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:48] <ajmitch> /msg nickserv help set kill [23:48] <wtay> taaz: sure? [23:49] thomasvs (th...@ad...) joined #gstreamer. [23:49] Parapraxis (ja...@mc...) joined #gstreamer. [23:49] <omega> yo [23:49] <Parapraxis> yo [23:50] <ajmitch> hi [23:50] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:50] <wtay> yo [23:52] thomasvs (th...@ad...) joined #gstreamer. [23:52] Parapraxis (ja...@mc...) left irc: Read error to Parapraxis[mcut-b-153.resnet.purdue.edu]: EOF from client [23:52] Parapraxis (ja...@mc...) joined #gstreamer. [23:52] <omega> arik's got gnome on the brain, he's using gnome.h just to get glib.h [23:53] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:54] thomasvs (th...@ad...) joined #gstreamer. [23:54] <ajmitch> heh [23:55] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:55] dschleef (ds...@c7...) joined #gstreamer. [23:57] thomasvs (th...@ad...) joined #gstreamer. [23:57] <taaz> anyone motivated to merge EVENTS back to HEAD? [23:57] <omega> I think it's safe to do it now [23:58] <wtay> yeah [23:58] <taaz> yeah, but you two say that every day ;) [23:58] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [23:58] <omega> well, we just need to do it [23:58] <ajmitch> heh [23:58] <taaz> which is why i asked about the motivation ;) [23:59] <taaz> i'm just wondering how the HEAD->EVENTS1 sync will go... [23:59] <omega> hmm [23:59] <taaz> i think some files being edited both places.. [23:59] <omega> wtay: how much longer will you be up? [23:59] <tnt> OK, I ran ./autogen.sh... it made the new automake file which I put in my path,... then I ran ./autogen.sh again,... then I ran ./configure... it had a couple of errors before kicking out: Error #1 << ./configure: line 2755 syntax error near unexpected token `PKG_CHECK_MODULES(GLIB2,' >> and Error #2 << ./configure: line 2755: line 2755: ' PKG_CHECK_MODULES(GLIB2, glib-2.0 gobject-2.0 gthread-2.0 gmodule-2.0,' >>... could this maybe have a p [23:59] <tnt> roblem because I don't have don't have 2.0 for GTK+, etc yet?... or is it something else. [23:59] <wtay> 6 more minutes, I'm afraid [23:59] thomasvs (th...@ad...) joined #gstreamer. [00:00] --- Wed Oct 17 2001 [00:00] <omega> tnt: do not run configure, autogen does that for you [00:00] <tnt> wtay said to. [00:00] <wtay> tnt: you need pkg-config [00:00] <taaz> tnt: autotools didnt find pkgconfig.m4 [00:00] <tnt> I installed it. [00:00] <omega> you need to install it in /usr, not /usr/local [00:00] <wtay> yeah, probably the path issue again [00:00] <omega> autoconf and/or pkgconfig are way too picky about that [00:00] <tnt> I installed it from RPM. [00:00] <taaz> you dont need to [00:00] <omega> rpm -ql pkg-config ? [00:01] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [00:01] <tnt> Hmmm, it says it is not installed... I'll check RedCarpet... I installed it from there. [00:01] <omega> oh, that's waaaaay too old [00:01] <omega> you have to have 0.8.0 [00:02] <tnt> 0.8.0... OK, I'll go get that. [00:02] thomasvs (th...@ad...) joined #gstreamer. [00:02] <omega> rpm -e pkg-config, then make *sure* you configure pkg-config with --prefix=/usr [00:02] <omega> otherwise autoconf won't find it [00:03] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [00:03] <tnt> Well, if I need 0.8.0, I'll have to uninstall the RPM., and get that tarball.... [00:03] <omega> right [00:03] <ajmitch> unless you know the tricks to getting it to work ;) [00:04] <omega> right, which I hope this document I'm writing will eventually cover [00:04] <ajmitch> k [00:04] <tnt> :-) [00:05] thomasvs (th...@ad...) joined #gstreamer. [00:05] <tnt> [thomasvs should try a different nick, until he remembers his password :-) ] [00:06] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [00:07] thomasvs (th...@ad...) joined #gstreamer. [00:08] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [00:10] thomasvs (th...@ad...) joined #gstreamer. [00:11] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [00:12] thomasvs (th...@ad...) joined #gstreamer. [00:13] <wtay> I need to sleep now. cya [00:13] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [00:13] Nick change: wtay -> wtay-zZz [00:14] chillywilly (da...@d3...) joined #gstreamer. [00:15] thomasvs (th...@ad...) joined #gstreamer. [00:16] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [00:17] Action: tnt is away: I'm busy [00:18] thomasvs (th...@ad...) joined #gstreamer. [00:19] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [00:19] Action: omega is tempted to ban thomasvs until he wakes up [00:19] <ajmitch> hehe [00:20] <ChiefHighwater> why is it kicking him though? [00:20] <omega> cause he'll come back [00:20] <omega> which is what's happening now [00:20] <ChiefHighwater> I don't usually sgn in my nick and I don;t get kicked [00:20] <ajmitch> ChiefHighwater: nickserv wants a password [00:20] <ajmitch> ChiefHighwater: he set his to be more secure :) [00:20] <ChiefHighwater> wants one from me too [00:20] <dschleef> don't you have to do something, like talk in a channel, to get kicked by nickserv [00:20] <ajmitch> dschleef: nope [00:20] <omega> doesn't kick by default. IMO that shouldn't be set unless one has a password script [00:21] thomasvs (th...@ad...) joined #gstreamer. [00:21] #gstreamer: mode change '+o omega' by ChanServ!s@ChanServ [00:22] #gstreamer: mode change '+b *!*@*.turboline.skynet.be' by omega!om...@om... [00:22] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [00:22] <omega> there, he won't come back in next time, I'll email him and let him know [00:23] Parapraxis (ja...@mc...) left irc: Client Exiting [00:24] #gstreamer: mode change '+b #gstreamer!*@*' by omega!om...@om... [00:25] #gstreamer: mode change '-b #gstreamer!*@*' by omega!om...@om... [00:25] #gstreamer: mode change '-b *!*@*.turboline.skynet.be' by omega!om...@om... [00:25] #gstreamer: mode change '+b thomasvs!*@*.turboline.skynet.be' by omega!om...@om... [00:25] Action: tnt is back (gone 00:08:24) [00:27] <tnt> OK, I've got pkg-config 0.8.0, and I moved it into /usr/bin (from /usr/local/bin), but I get the same errors when I run ./autogen.sh?... Any ideas? [00:28] <ajmitch> tnt: do you have /usr/share/aclocal/pkg.m4 ? [00:28] <omega> tnt: you must ./configure with --prefix=/usr for pkgconfig [00:29] <omega> all the files have to be configured and installed in the right place from the beginning [00:29] <tnt> ajmitch: Umm.. It's under /usr/local/share/aclocal/pkg.m4... should I move it??? [00:30] <ajmitch> tnt: yeah, would be good [00:30] <omega> tnt: no, uninstall pkgconfig and reconfigure it [00:30] <omega> there's more than just moving files, you have to get pkgconfig compiled with the right path [00:30] <tnt> How do you uninstalled... make uninstall? [00:30] <omega> yup [00:30] <tnt> OK. [00:31] <omega> and manually remove the files that've been moved [00:32] <tnt> (I did.) [00:32] <omega> ok [00:33] <tnt> So, do I run "./configure --prefix=/usr", then "make", then "make install"? [00:35] <wtay-zZz> <subliminal message>omega... starts the merge... today... </subliminal message> [00:35] <omega> yeah, yeah, will do so [00:38] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [00:50] Action: vishnu dreams about the golden merge ... [00:54] <tnt> OK, ./autogen.sh worked (finally :-) ), but after running make I've gotten an error: << libtool: link: error: cannot link shared libraries into libtool libraries >>. I think it was in "gstreamer/types" when it go that error. Any clue? [00:54] <omega> need libtool 1.4 [00:55] <omega> updating REQUIREMENTS [00:55] <tnt> :-) [00:55] <tnt> [There goes all my nice RPMs....] [00:56] <ajmitch> it works best if you use debian sid, imho ;) [00:56] <tnt> I've got Red Hat Linux 7.1 with Ximian GNOME. [00:57] <ajmitch> ugh, poor guy [00:59] evil_monk (trevo@208.141.162.68) joined #gstreamer. [01:03] <tnt> Ummm, I just thougt of something, if I uninstall the old libtool RPMs, the how will I compile the source for the new libtool????? [01:03] <omega> it doesn't need itself [01:04] <omega> libool is just a script anyway [01:04] <tnt> OK. [01:06] <tnt> What about libtool-libs... what's that... will the tarball with libtool 1.4 have that too? [01:07] <omega> yeah [01:07] <omega> remove the rpm [01:08] <tnt> (OK, both removed... along with aspell) [01:08] <omega> uh? [01:08] <tnt> aspell apparently needed something in libtool-libs [01:08] <omega> odd [01:15] arik (ar...@sd...) joined #gstreamer. [01:15] <arik> lo [01:16] <ajmitch> hey arik [01:16] <ChiefHighwater> ello [01:16] <arik> all [01:18] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [01:19] <ajmitch> arik: what's been decided about gstmediaplay? ;) [01:19] <omega> nothing substantial yet <g> [01:19] <arik> heh [01:19] <arik> what he said [01:19] <omega> I'm hoping we can meet in Boise to work through it all.... [01:19] <ajmitch> k [01:19] <arik> did you get picoplay building? [01:19] <ajmitch> we gotta wait that long? ;) [01:20] <omega> yeah, but the GUI doesn't do much [01:20] <arik> yes ;-P [01:20] <arik> omega: what do you mean? [01:20] <omega> I can't select a file, and hitting play doesn't do anything [01:20] <arik> um [01:20] <arik> that's kinda weird [01:20] <arik> it should work fine [01:20] <tnt> For libtool, do I need to give it a special --prefix=/usr,... or is the normal "make install" fine??? [01:20] <arik> it plays files for me anyway ;-) [01:20] <arik> can you try running it with a file as an argument? [01:20] <arik> ./picoplay mp3-file [01:20] <arik> then hit play [01:21] <omega> I killed it [01:21] <arik> what happened? [01:21] <omega> just click the track a few times and it wedges [01:22] <arik> heh yeah i know [01:22] <omega> hitting play doesn't do anything [01:22] <arik> ok [01:22] <arik> that's not right [01:22] <arik> one sec [01:22] <omega> no gst output from it either [01:22] <arik> um [01:22] <arik> it plays files just fine here [01:22] <arik> i just tried it [01:23] <omega> hmm, gotta love unrepeatable stuff <g> [01:23] <arik> hehe ;-) [01:23] <arik> but it compiled and ran in the form i gave you? [01:23] <arik> cause it was core dumping here [01:23] <omega> yeah [01:23] <arik> that's very very weird [01:23] <omega> well, gimme the one that works then! ;-) [01:23] <omega> just the .[ch] [01:24] <arik> which? [01:24] <omega> of all the files you've changed in the last N hours [01:24] <arik> heh [01:24] <arik> none [01:24] <omega> grmbl [01:24] <arik> i haven't changed anything since i gave it to you [01:24] <arik> cause i've been waiting to hear from you [01:24] <arik> so it _should_ work [01:24] <omega> mu. the gui isn't working for me atm, so... [01:25] <omega> you have picogui HEAD from when? [01:25] <arik> what's that about? [01:25] <arik> hmm [01:25] <arik> a few days ago i think [01:25] <omega> hmm [01:25] <arik> last week some time [01:25] <omega> lemme back out picogui [01:25] <arik> ok, the open box doens't work [01:25] <arik> i know that [01:25] <arik> it's in picogui not in my app [01:25] <arik> but if you run it with an mp3 file from the comandline it should work [01:26] <omega> backing out 7 days of pgserver and cli_c changes [01:27] <arik> ok [01:27] <arik> it may be something about my system setup [01:27] <arik> or something [01:30] sienap (sy...@s3...) joined #gstreamer. [01:30] <sienap> hej all.. [01:31] <sienap> a little coding question.. [01:31] <ajmitch> hehe, yeah? [01:31] <sienap> i've got a pointer to some mem which i malloc some mem for [01:31] Action: ajmitch wonders what pr0n viewer sienap is coding ;) [01:31] <sienap> however at some time i'll have to free a PART of it or malloc some more [01:31] <ajmitch> and? [01:31] <sienap> i am coding some kind of VM [01:31] <sienap> verry simple [01:31] <sienap> very simple [01:31] <sienap> however i want to have a dynamic stack [01:32] evil_monk (trevo@208.141.162.68) joined #gstreamer. [01:32] <sienap> so it grows and schrinks [01:32] <sienap> depending on how much there is pushed on and popped off [01:32] <ajmitch> uhuh [01:33] <sienap> you got the idea ? [01:33] <sienap> however i thought it should be correct to do this with a pointer and malloc / free [01:33] <sienap> however when i "remalloc" the old stuff will get lost [01:33] <sienap> which kinda sucks [01:33] <omega> remalloc will copy the data [01:33] <sienap> hmm ? [01:34] <ajmitch> remalloc or realloc? [01:34] <sienap> a little example: [01:34] <sienap> int main () { [01:34] <sienap> char *bla; [01:34] <sienap> bla = malloc(1); [01:34] <sienap> bla[0] = 10; [01:34] <sienap> bla = malloc(2); [01:34] <sienap> bla[1] = 20; [01:34] <ajmitch> http://ajmitch.dhis.org/devel/info/prog-tuts/teachc/ch20/ch20.html <- info on realloc [01:35] <sienap> hmm i should use realloc ? [01:35] <sienap> or something ? [01:35] <omega> realloc, yes [01:35] <sienap> aaah [01:35] <sienap> ic [01:35] <sienap> :) [01:35] <omega> don't think remalloc exists [01:35] <sienap> na with re malloc i meant malloc(1) ; malloc(2) :) [01:35] <sienap> thanks ajmitch [01:35] <sienap> ok that will do it [01:35] <sienap> i am making an VM afaik thing with brainfuck as machine code ;) [01:35] <sienap> just for fun.. [01:36] <vektor> help help help. [01:36] <sienap> however a mate of my just really demotivated me :( [01:36] <vektor> anyone up for a silly problem? [01:36] <vektor> you need gnuplot to play. [01:36] <vektor> come on come on [01:36] <vektor> i know you want to [01:36] <vek... [truncated message content] |
From: IRC B. <wt...@us...> - 2001-10-18 04:27:32
|
******************************************************************* [03:00] <vektor> IRC doesn't lend it self to like, happiness: broadway style, or happiness: teen flavour, or even happiness: madonna flair. [03:00] <vektor> so i have to resort to happiness: warez'in 0-day style. [03:01] <vektor> i need a new rap, yo. [03:01] <dschleef> wingo: the choice is more between known-how-its-broken non-PIC code and unknown-how-it-might-be-broken PIC code [03:01] <wingo> true... [03:01] Shippou (no...@cr...) joined #gstreamer. [03:01] <vektor> all code should be PIC. since there are those of us who might want to use it. [03:01] <wingo> you've tested it for x86? [03:01] <vektor> share the pain! [03:02] <dschleef> wingo: only the idct stuff. And it took a while for me to get it right [03:02] <vektor> dschleef: you PIC'tified an IDCT? or un-PIC'tified? [03:02] <wingo> sounds like you assembler czars will have to convene a pow-wow :) [03:03] <dschleef> that's half the problem. I'm not an i386 asm czar [03:03] <wingo> vektor: what did you write? [03:03] <vektor> I know my way around i386 asm a bit. [03:03] <vektor> wingo: what do you mean? [03:03] <vektor> dschleef: Can I take a look at your hack? [03:03] <wingo> in the r3c0rdn softwarez genre :) [03:04] <dschleef> I know my way around, but, for example, i don't know the difference between lea and leal [03:04] <vektor> wingo: oh, well, i'm working now on ensuring audio sync is perfect even if your soundcard doesn't support true 44100. my app is 'reetpvr' and some of its source is at www.sf.net/projects/reetpvr/ [03:04] <dschleef> vektor: it's on the mailing list [03:04] <vektor> dschleef: oh. [03:04] <vektor> wingo: it's a v4l recorder and will soon also be a dvd recompressor. [03:05] <vektor> wingo: i _never_ drop frames if i can help it. i'm trying to do _everything_ correct for high quality recording. [03:05] <vektor> wingo: for example, i have a near perfect 3:2 pulldown detection/inversion routine. [03:05] <vektor> wingo: and i wrote a huffyuv-style code. [03:05] <vektor> c [03:06] <wingo> i'm not a video fellow. what is 3:2 pulldown detection/inversion? [03:06] <vektor> Oh. [03:06] <wingo> vektor: you need a web page for it :) [03:07] <vektor> 3:2 pulldown is a process where film is transferred to video. So, from a 29.97fps interlaced video stream, I extract the original 24fps progressive film source frames. [03:07] <wingo> why are the fps's different? [03:08] <vektor> I should make a web page for it, but really, I just started a few weeks ago and I'm still hacking a bit. [03:08] <vektor> Um, film standardized on 24fps a million years ago. [03:08] <vektor> And likewise, north american video standardized on 59.94fps. [03:08] <vektor> Most other places in the world standardized on 50fps video. [03:08] <vektor> So, now we all have to just deal with it. [03:08] <vektor> Ugh. [03:10] <vektor> dschleef: have a sec? [03:12] <dschleef> yep [03:12] <vektor> ok [03:12] <vektor> so [03:12] <vektor> question [03:12] <vektor> walken keeps saying that if you compile libmpeg2 -fPIC, then it's like 15% slower. [03:12] <vektor> since now we have to like keep our PIC stuff in those other regs. [03:12] <vektor> do you agree with that sentiment? [03:13] <dschleef> yes, mostly. [03:13] <dschleef> 15% seems a little high [03:13] <vektor> yeah. [03:13] <vektor> That's what he claims though. Ugh. [03:13] <dschleef> that's only on x86, too [03:13] <vektor> Yes. [03:14] <vektor> But he lists that as a main reason he wants to keep distributing libmpeg2 as a static lib on x86. [03:14] <dschleef> on powerpc, with 32 generic registers, using one for a table pointer is irrelevant [03:14] <vektor> Are we going to have problems with this new binutils, if, say, I have a -fPIC lib and I'm using libmpeg2 ? [03:14] <dschleef> yes [03:15] <dschleef> well, maybe [03:15] <wingo> you mean an -fPIC lib like gst? [03:15] <dschleef> if you link it statically with your app, no [03:15] <vektor> I mean if I try and distribute, say, my DVD ripper stuff as a shared lib. [03:15] <dschleef> the problem is a combination of using non-PIC code in a shared object file, i.e., libxxx.so [03:15] <vektor> Yeah. [03:16] <vektor> This is a nasty situation. [03:16] <vektor> I don't like how I can't have a user upgrade libmpeg2 without also upgrading my app either. [03:16] <vektor> Damn walken. [03:16] <vektor> ;-) [03:18] <wingo> that seems to be a common sentiment around these parts :) [03:18] <vektor> haha [03:18] <vektor> nah that's just me :0 [03:18] <vektor> :) [03:19] <dschleef> I suppose there are ways around it. Run-time final linking would be one option [03:19] <dschleef> like how kernel modules work [03:20] <dschleef> kernel modules are .o files, and the symbols are unresolved. insmod has to fix up all the relocations before it copies the module to kernel space [03:20] <wingo> i was about to ask :) [03:20] <vektor> Interesting. [03:21] <vektor> I guess there are no standards for doing that with shared libs. [03:21] <dschleef> the typical way to do this in user space is to use shared libs, but you could use the .o approach [03:21] <vektor> Maybe you can do that with libdl(). [03:21] <wingo> would that be a candidate for a gst util lib? [03:21] <wingo> vektor: hmmm [03:21] <vektor> like, y'know, libdl(), relocate_now or whatever. [03:21] <vektor> RTLD_NOW, [03:21] <vektor> bah [03:21] <vektor> that doesn't do what i want. [03:22] <dschleef> the problem is that the relocations in non-PIC code are in the actual code, not in a separate, read/write space [03:22] <dschleef> of course, the entire purpose of a shared lib is to share it [03:22] <vektor> yeah, i know ;( [03:22] <vektor> makes no sense. [03:22] <vektor> fawk. [03:23] <vektor> so fuck it, i may as well just cp -r mpeg2dec movietime/ [03:23] <dschleef> however, for small segments of code that you really want to run fast, there's a strong reason for not sharing it [03:23] <vektor> and then distribute it as my own source. [03:23] <vektor> well, the system isn't built for that. [03:23] <vektor> is it. [03:23] <dschleef> it would be useful to create the infrastructure. [03:24] <wingo> indeed. [03:24] <vektor> yeah. [03:24] Action: wingo pokes omega_afk [03:24] <wingo> it seems he would be interested as well [03:24] <dschleef> most likely [03:26] <vektor> hey i still have his cell phone number [03:26] <vektor> we could call him ;-) [03:26] <vektor> hahahahahah [03:27] <vektor> or wait [03:27] <vektor> noooooo [03:27] <vektor> it was in my palm pilot [03:27] <vektor> which i toasted. [03:27] <vektor> damn [03:27] <vektor> ;-) [03:34] <wingo> dschleef: should this be a status item? [03:36] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [03:45] <wingo> a non-shared lib loader and translator, i mean [03:45] <wingo> maybe filing a bug would be best? [03:54] <wingo> dschleef: i'm running into strange problems deep in _dl_* with ladspa plugins that once worked for me. could this be due to gst non-pic issues? [04:03] <dschleef> wingo: possibly [04:04] <dschleef> do you have problems with gstreamer-register, too? [04:04] <wingo> no [04:04] <wingo> but with gstreamer-inspect [04:04] <wingo> and certain varieties of ladspa plugins, yes [04:05] <dschleef> run 'objdump -dR' on the offending .so file and output it to a file [04:05] <wingo> you mean the loaded ladspa plugin? [04:05] <dschleef> then look at the first reloc listed [04:06] <wingo> ok [04:06] <dschleef> then look at the last disassembly lines [04:06] <dschleef> the address of the first reloc should be _after_ the last addresss of disasm [04:07] <dschleef> that is, for a non-buggy module [04:07] <wingo> "reloc"? [04:07] <wingo> DYNAMIC RELOCATION RECORDS [04:07] <wingo> OFFSET TYPE VALUE [04:07] <wingo> 00017360 R_386_RELATIVE *ABS* [04:07] <wingo> that sort of thing? [04:07] <wingo> it continues a fair ways [04:08] <dschleef> yes. [04:08] <dschleef> the addr is 17360 [04:09] <wingo> Disassembly of section .fini: [04:09] <wingo> 00013cd4 <.fini>: [04:09] <wingo> 13cd4: 55 push %ebp [04:09] <wingo> so that's at the end [04:09] <wingo> there's more asm [04:09] <wingo> not very much [04:10] <dschleef> that should be ok, then. 17360 > 13cd4 [04:10] <wingo> funny thing is, this .so has many plugins within it (ladspa plugins, not gst) [04:10] <wingo> it works with some of them, not with others [04:10] <dschleef> meaning it loads other stuff? [04:10] <wingo> so, problably not a link problem, no? [04:10] <wingo> yeah [04:10] <wingo> and it loads all of them by name, just can't instantiate some [04:10] <dschleef> well, it's not a problem with libgstladspa.so [04:11] <wingo> that was cmt.so, a ladspa plugin [04:11] <dschleef> but it might be a problem with something that libgstladspa.so depends on [04:11] <wingo> i'll try wih libgstladspa [04:11] <wingo> what about libgst.so? is the non-pic stuff relevant there? [04:12] <tnt> Anyone still listening? [04:12] <wingo> yeah a bit :) [04:12] <dschleef> anything that ends in .so [04:13] <dschleef> but probably only stuff that contains i386 asm [04:13] <wingo> libgst.so looks good just from checking the top and bottom [04:14] <dschleef> i wrote a 3-line program that dlopens argv[1]. [04:14] <dschleef> if it segfaults, there's a problem. [04:14] <vektor> nifty. [04:14] <vektor> that's a good idea. [04:14] <dschleef> then run for each in *.so;do ack $each;done [04:16] <tnt> I got rid of the RPM version I had of gstreamer, and replaced it with the CVS version. After hunting the tarballs from all over the world (and uncompressing, untaring, compiling, etc), I eventually got gstreamer to compile. However, this it is not installed! I tried running some gstmediaplay, and ssome of the examples, but they all dump their core. Anyone know how to get this "style" -- non-installed -- of gstreamer to work??? [04:16] <wingo> tnt: you just ran autogen.sh? [04:16] <wingo> no other parameters? [04:17] <wingo> have you seen the wiki GstHowto node? [04:17] <tnt> I ran autogen a long time ago.. with no parameters... I just got back in... I went out for a while. [04:17] <wingo> i'm plugging it because i wrote it :) [04:17] <tnt> I'll go take a look. [04:17] <wingo> have you run gstreamer-register? [04:17] <tnt> Yes. [04:17] <wingo> it worked? [04:18] <tnt> It seemed to. (I'm not an expert though :-) ) [04:18] mwc (ma...@ly...) joined #gstreamer. [04:18] <wingo> tools/gstreamer-inspect works too? [04:19] <tnt> Ya... the first time I ran it, it told me to run gstreamer-register first... I then did that, then tried running gstreamer-inspect next... it then printed out a whole bunch of stuff. [04:19] <wingo> good. [04:20] <wingo> tools/gstreamer-launch fakesrc ! fakesink ? [04:20] <tnt> Hang on... [04:21] <wingo> we really should make a dichotomous key for this stuff :) [04:21] <tnt> I printed a whole bunch of stuff out... I stopped it with ^C. [04:22] <wingo> good [04:22] <wingo> you know how to use gstreamer-launch? [04:22] <wingo> GstLaunchQuickStart might help [04:22] <tnt> kind of... I'm still learning what to plug into what though. [04:23] <tnt> I try to play an mp3... [04:24] <wingo> yuk! gstmediaplay doesn't work for me either in the uninstalled version. [04:24] <wingo> i think it has to do with libglade stuff [04:25] <wingo> open("gstmediaplay.glade", O_RDONLY) = -1 ENOENT (No such file or directory) [04:25] <wingo> yeah... [04:25] <tnt> I can play mp3s! :-) [04:25] <wingo> good! [04:25] <wingo> ok, gstmediaplay works if you run it from its directory [04:26] <wingo> kind of a hack, but oh well [04:26] <tnt> I'll try again, hang on... [04:26] <wingo> i have to head out for a bit. good luck though. and document your experiences on the wiki :) [04:27] Nick change: wingo -> wingo-out [04:30] Nick change: dschleef -> ds [04:31] <ds> sweet. nickserv lets me use ds now [04:31] <vektor> haha. [04:31] <vektor> anyone here know anything about automata/languages stuff? [04:31] <vektor> formal languages? [04:31] Action: vektor looks for homework help. [04:32] <vektor> Say $L_{1}$ and $L_{2}$ are languages, can anyone here think of the (unique) solution to $X = L_{1} X + L_{2}$ ? [04:32] <vektor> yea yea. [04:32] <vektor> you know you want to. [04:32] <vektor> oh yea, um, assume $L_{1}$ does not contain the empty string. [04:33] <vektor> ahaha i think i got it [04:34] <omega_afk> ya lost me [04:34] Nick change: omega_afk -> omega [04:34] <vektor> you never studied, did you [04:35] <omega> study? [04:35] <vektor> haha [04:35] <vektor> yeah [04:35] <vektor> :) [04:35] <vektor> fuck homework sucks [04:36] Action: ds scores [04:36] <vektor> you scored? [04:36] <ds> I love the modutils developers [04:36] <ds> not in that way [04:36] <vektor> oh [04:36] <vektor> too bad [04:36] <vektor> did the modutils developers solve our problems already? [04:36] <ds> yes [04:37] <omega> which problems? [04:37] <vektor> and package this solution in a neat shared library? [04:37] <vektor> :) [04:37] <ds> I have a program now that loads in a .o file, relocs it, then can give me the address of a function inside [04:37] <vektor> ahahahahahahahahahahaha. [04:37] <omega> hmm [04:37] <ds> like insmod, but for user-space proggys [04:37] <vektor> dude, use the power of dcc [04:37] <omega> what for? uclibc or specialib (or both)? [04:38] <ds> possibly for busybox, perhaps uclibc. but not directly [04:38] <ds> mainly so we can use non-PIC code when it is really important [04:38] <omega> hmmm, interesting [04:38] <omega> any way that could be applied to ld.so in general? [04:39] <ds> well, it sort of defeats the purpose of _sharing_ the libs [04:39] <omega> how? [04:39] <vektor> no, because it lets me upgrade my libs without upgrading my apps. [04:39] <vektor> and i consider that point much more important these daya. [04:39] <vektor> s/daya/days/. [04:39] <vektor> since i have a fucking half gig of ram. [04:40] <ds> because the .text segment is read/write, and gets modified during load [04:40] <omega> oh, right [04:40] Action: ds agrees that the low cost of RAM changes many rules [04:40] <omega> and /me was thinking about ways to do it the other way around recently <g> [04:41] Shippou (no...@cr...) joined #gstreamer. [04:41] <vektor> we should write a paper on this. [04:41] <omega> which one? [04:41] <vektor> 'new thoughts on PIC and performance-oriented code'. [04:41] <omega> ah [04:41] <ds> I think the .so route is better and more sustainable, but for highly optimized code, non-PIC is better [04:42] <vektor> 'a reasonable method of distributing run-time loaded shared code'. [04:42] <ds> actually, thinking about it, it would fit into specialib, too [04:42] <ds> duh [04:42] <vektor> that will be the subtitle. [04:42] <omega> ds: right [04:42] <vektor> we'll make a fortune on merchandising. y'know, selling lunch boxes with our pictures on it and stuff. [04:42] <vektor> it'll be big news. [04:42] <omega> hmm [04:42] Action: ds could patent it [04:43] <omega> that'd go over big [04:43] <vektor> ds: modutils would love you ! :) [04:43] <ds> =) [04:43] <ds> everyone else does it [04:43] <omega> doesn't make it right.. [04:43] <omega> "hey, I know of this cliff, it's really fun to jump off it <g>" [04:43] <vektor> 'dear u.s. patent office. i found this in an open source project and thought it was cool. give me a patent?' patent office: '#87437438734. thank you, come again!' [04:44] <omega> for reference, libmpeg2 *total* is 51KB text [04:44] <evil_monk> gstreamer can't do asf yet can it [04:44] <omega> I think I'm willing to sacrifice 51KB per instance for performance, if it's noticable [04:44] <omega> nope [04:44] <ds> what was the number-of-cycle improvement from an indirect call to a direct call? [04:44] <omega> no one's tried yet [04:44] <omega> ds: don't remember [04:45] <evil_monk> omega: how stable is cvs gstreamer? [04:45] <omega> depends on what you're doing [04:46] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [04:50] <omega> http://www.icrontic.com/dk/Interview/logo2.jpg [04:50] <omega> pretty cool picture <g> [04:50] <ChiefHighwater> hehe [04:50] <omega> ds: wouldn't it be possible to retrofit a .so with this? [04:51] <omega> just find all the GOT calls and replace them? [04:51] <omega> of course, you only get the benefit of a direct call, not the extra register [04:51] <vektor> gee, that would be useful :( [04:51] <omega> well, somewhat [04:52] <omega> might partially offset the cost of the extra reg [04:52] <vektor> i _guess_. [04:52] <ds> omega: no, since the actual instructions are different [04:52] <omega> so what's keeping the math kernels and such from being non-PIC in a PIC environment? if they have to call something, they can pick the GOT ptr out of the stack.... [04:53] Action: ajmitch is back [04:53] <omega> hmm? [04:54] <ds> the application itself is non-PIC, but it is guaranteed to be loaded at a specific location [04:54] <omega> right? [04:54] <ds> but if we have non-PIC .o files, with relocation information, we can load it whereever we want, fix the relocs, and it's just like it was application code [04:55] <omega> right [04:55] <vektor> yeah that's really, really neat. [04:55] <ds> really have to go now. I'm hungry [04:55] <omega> so my question is, why can't we have a non-PIC library used by a PIC library used by a non-PIC app? [04:55] <vektor> locating the .o files sucks though [04:55] <vektor> like [04:56] <omega> like math kernels compiled nonPIC [04:56] <vektor> omega: i'm not totally up on that one either. since 'it worked fine with the old binutils'. [04:56] <vektor> but still [04:56] <vektor> i like the idea of maybe having a PIC library which loads a non-PIC .o file and calls into it. [04:56] <omega> Debian seems to dislike that concept for some reason [04:56] <vektor> that would be neat. [04:56] <vektor> maybe Debian wants everything PIC. [04:56] <vektor> that would be cool. [04:56] <omega> but I don't see the need for PIC to be used for functions that don't call anything [04:56] Action: ajmitch doesn't understand what PIC is :) [04:57] <omega> position independent code [04:57] <vektor> ajmitch: Position Independant Code [04:57] <ajmitch> k [04:57] <omega> all calls go through a jumpgate [04:57] <vektor> ajmitch: so that you can have the same code used by multiple apps at once [04:57] <vektor> ajmitch: and they each relocate on the fly [04:57] <omega> this requires the %ebx (afaik) keep the GOT address, and all calls are relative of that [04:57] <ajmitch> sounds sorta useful [04:57] <vektor> i'm not sure what GOT stands for [04:57] <vektor> ajmitch: it's how shared libs work :) [04:57] <omega> global offset table [04:57] <vektor> yeah thanks omega :) [04:57] <omega> a jumptable [04:57] <vektor> ajmitch: except in linux you don't have to have PIC shared libs. [04:58] <vektor> ajmitch: which i always found kinda funky [04:58] <ajmitch> heh, k [04:58] <vektor> ajmitch: but it's non-portable. [04:58] <vektor> and i think debian wants to stop it, for one. jah? [04:58] <vektor> dunno. [04:58] <omega> so I'm still a little fuzzy on what all is required for PIC code on the side of the caller and callee [04:58] <omega> i.e. why can an app be non-PIC? [04:58] <vektor> because the kernel doesn't put it in a special shared place? :) [04:59] <omega> oh, duh [04:59] <vektor> but you're right. i'm not up on it either. [04:59] <vektor> like [04:59] <vektor> y'know. [04:59] <vektor> but i don't think the kernel can make intelligent decisions. [04:59] <vektor> so i think that fucks with stuff. [04:59] <omega> kernel doesn't care [04:59] <vektor> but the compiler can. [04:59] <vektor> yeah. [04:59] <omega> it's the loader that figures it all out [04:59] <vektor> ah yeah. [04:59] <vektor> ld.so. [04:59] <vektor> my bud and loveer [04:59] <vektor> i should buy ld.so a beer. [04:59] Action: ajmitch sees some gnome people would like gstreamer to be in gnome 2.2 [05:00] <vektor> (lord knows i'll never drink one again) [05:00] <omega> vektor: that would just make your relocations a little shaky [05:00] <vektor> haha. [05:00] <omega> vektor: you learned? [05:00] <vektor> omega: i got school'ed bigtime. [05:00] <vektor> omega: big time. [05:00] <omega> tis a good thing wtay fixed the daily log thing when he did <g> [05:00] <ajmitch> omega: how long do you think it'll be before he drinks again? a week? [05:00] <ajmitch> omega: haha [05:00] <omega> just need to keep a printout of that in vektor's pocket.... [05:00] <vektor> i may have an occasional beer, but man, i'll never drink vodka again for sure. [05:01] <vektor> not that i drink much anyways. [05:01] <vektor> that thing on saturday i guess was the first time i drank since OLS. [05:01] <vektor> dunno. [05:02] Action: ajmitch should get back to python hacking :) [05:02] <omega> http://cube.ign.com/news/39162.html [05:02] <omega> man, that secondary title is perfect... [05:02] <vektor> heheehe [05:03] <vektor> the laugh a new gut thing? :) [05:03] <omega> yup [05:05] evil_monk (trevo@208.141.162.68) joined #gstreamer. [05:09] BeNOW (an...@ds...) joined #gstreamer. [05:09] <omega> yo [05:09] <BeNOW> heya [05:10] #gstreamer: mode change '-b thomasvs!*@*.turboline.skynet.be' by omega!om...@om... [05:10] <BeNOW> so I'm just about to start massacreing icecastsend [05:10] <omega> uh oh <g> [05:10] <ajmitch> hehe [05:11] <BeNOW> I'm putting in mods to allow for sending to shoutcast (icy headers), aswell as to set stream name, desc etc (metadata stuff), and some icecast specific stuff (mountpoint, remote dumpfile, etc) [05:11] <omega> ok, does libshout do this, or do you have to hack on that too? [05:12] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [05:12] <BeNOW> I'm having dropouts here. I believe it's because shoutcasts don't like pulling from icecasts. [05:12] <omega> heh [05:12] <BeNOW> yeah, it's in libshout... the list would be shorter otherwise :) [05:12] <omega> ok [05:13] <BeNOW> and I can't change the public streamers to icecasts... so I'll try to change some of the backend here to use shoutcast. [05:13] <omega> bah [05:13] <BeNOW> is there a good reference for the g_object stuff... is that gtk? [05:14] <BeNOW> glib? [05:14] <BeNOW> yeah, it's a bit blashemous to move away from the nice open-source icecast... [05:15] <BeNOW> I'm sure to come back to it tho, as shoutcast does not support ogg streaming. [05:16] <omega> gtk.org has docs for glib2/gobject [05:16] wingo-out (wi...@rd...) left irc: Read error to wingo-out[rdu25-21-076.nc.rr.com]: EOF from client [05:17] <BeNOW> yeah, just found it... thanks :) [05:17] Action: BeNOW goes back to hacking... [05:19] wingo (wi...@rd...) joined #gstreamer. [05:27] evil_monk (trevo@208.141.162.68) joined #gstreamer. [05:31] robla (ro...@to...) joined #gstreamer. [05:32] <omega> yo [05:32] <omega> no Zeenix atm, I can't keep his timezone straight [05:32] <robla> hey...thanks for the reference, though [05:33] <omega> so you work at real? [05:33] <robla> Yeah [05:33] <omega> developer? [05:33] <robla> Program Manager. I've been a developer here, too. [05:34] <omega> what product? [05:34] <robla> Client Core group. The media engine in the RealPlayer [05:35] <omega> hmm, I've worked with that [05:35] <omega> had my share of frustrating experiences... [05:35] <robla> Haven't we all :) [05:36] <omega> here's a bug that might still exist... lemme see if I can describe it reasonably [05:36] <omega> so, the timestamps on the packets would increment by 33ms (video) [05:36] <robla> Which datatype? [05:37] <omega> custom mpeg1 [05:37] <omega> for whatever reason math-related, we never had a timestamp of 1000 or 2000 , etc. [05:37] <robla> Is this with the DBC plugin, or with the new stuff in RealOne? [05:37] <omega> we found that this caused the buffering system to give us timestamps on the receiving side of 900,933,966,999,1999,1033,etc. [05:38] <omega> would inject a buffer from 1sec (buffer depth) in the future, into the stream... [05:38] <omega> this was with real 7.x iirc [05:38] <omega> a little over a year ago [05:38] <omega> patching up 999 to 1000 etc fixed it ;-) [05:39] <omega> took us, um, a while. [05:39] <robla> k...That was a third party plugin, so I can't help you much with that. However.... [05:39] <omega> the problem was definitely in the real engine [05:39] <robla> We just release our own MPEG renderer in RealOne [05:39] <omega> decoder? better than the old one I hope? [05:39] <robla> I hope [05:40] <robla> It supports RFC 2250 packetization. [05:40] <omega> our hacked up berkeley decoder looked lightyears better than that one <g> [05:40] <omega> that's rtp-based? [05:40] <robla> The old one was done by Digital Bitcasting (now EMC) [05:40] <robla> New one is RTP based. [05:40] <omega> EMC as in the disk guys? [05:40] <omega> or did EMC buy them? [05:41] <robla> Yeah, EMC bought DBC [05:41] <omega> hmm, not for their mpeg codec I hope <g> [05:41] <robla> DBC's core business was video servers [05:41] <omega> ok [05:42] <robla> Have you seen any of Ross Finlayson's toolkit work for RTP/RTSP? http://live.sourceforge.net [05:42] <omega> nope, checking [05:43] <omega> I've been told that librtp is kinda lame, it'd be good to find something better [05:43] <omega> oh, but it's c++. eewww [05:43] <BeNOW> yeah, but mature. [05:43] <BeNOW> and mulicasting, etc. [05:43] Action: BeNOW was looking at that at one point... [05:43] <omega> hmm, I think someone did a C++ plugin somewhere... I did a hybrid at one point... [05:44] <robla> Yeah. I've been playing around with it. It interoperates with our stuff. [05:44] <BeNOW> smae reaction to c++, tho :) [05:44] <omega> robla: here's a question, do you remember Digital Mercury at all? [05:44] <robla> Nope [05:44] <omega> the company where I was doing the real stuff [05:44] <omega> darn. we were talking with Real at one point for quite a while, did demos and stuff [05:45] <omega> killall -9 DigitalMercury about 11mo ago [05:45] <robla> I deal with the guts of the system, so probably wouldn't have been with me. [05:45] <robla> What did DM do? [05:45] <omega> hmm, you'd have been one of the guys we would have worked with if the higherups had ever decided to go anywhere... [05:46] <omega> um... neer did figure out what exactly DMI wanted to do [05:46] <omega> but we were working on scalable videl [05:46] <omega> er, video [05:46] <omega> a team from OGI working on self-scaling video transmission [05:46] <omega> a network protocol to discover bandwidth and send as much data as possible to maximize quality [05:46] <omega> but from a single stream [05:47] <omega> one of these days I'm gonna go implement that in gstreamer.... [05:47] <robla> Layered encoding? [05:47] <omega> yup [05:47] <omega> based on mpeg1, with ~16 levels of quality [05:47] <omega> done in 97 or 98 I think [05:48] <omega> not very advanced, but sufficient for the research [05:48] <BeNOW> aren't wavelets supposed to allow such things? [05:48] <omega> MPEG-4 FGS would be much more appropriate [05:48] <omega> yeah, if you can code them right [05:48] <robla> Layered encodings tend to break down when you start squeezing bits. You get better performance with discrete encodings. [05:49] <omega> robla: in most cases, yes [05:49] <omega> blah, I'm being dragged away to watch a movie... [05:49] <robla> Yeah, I've gotta get out of here (I'm still at work...bleh) [05:49] <omega> but I'm in here all the time, and you want to talk to Zeenix mostly for the rtp/mp3 stuff [05:50] <robla> k....thanks [05:50] <omega> ttyl [05:50] Nick change: omega -> omega_movie [05:50] <BeNOW> robla: is live doing any opensource stuff? [05:50] <BeNOW> sorry, real. [05:50] <robla> See http://www.rtsp.org and http://www.smilgen.org [05:51] <BeNOW> sweet thanks. [05:51] <BeNOW> take it easy :) [05:51] <robla> You too. Bye. [05:51] robla (ro...@to...) left #gstreamer (Client Exiting). [05:57] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [05:59] ShrimpX (marius@131.252.244.168) left irc: Hey! Where'd my controlling terminal go? [06:10] wingo (wi...@rd...) left irc: going home... [06:24] <tnt> Anyone alive out there? [06:24] <tnt> Or has everyone gone to sleep? [06:25] <BeNOW> pong! [06:25] ChiefHighwater (fl...@su...) left irc: Leaving [06:29] <tnt> Does anyone know what's the procedure to compile programs with the CVS (non-installed) version of gstreamer? gstreamer-config does NOT return the correct values (for the libs, includes, etc). (I hope I don't have to had all the flags by hand!) [06:30] <tnt> Are all the gstreamer Guru's away from the computer? :-o [06:30] <ajmitch> use pkg-config? [06:31] <tnt> How should it be used? [06:31] <ajmitch> hmm, i dunno :) [06:31] <tnt> :-D [06:32] <ajmitch> i just know that i needed to export PKG_CONFIG_PATH to build the gst-bind module :) [06:32] <tnt> Some needs to write the tutorial "A Gentle Introduction to GStreamer" [06:33] <tnt> (Like the ML and haskell tutorials by a similar name.) [06:33] <ajmitch> i think omega is writing something [06:34] <ajmitch> anyway, i gotta go, bbl :) [06:34] <BeNOW> I ran ./configure after ./autogen.sh then [06:34] <BeNOW> make install [06:34] <BeNOW> then the paths are correct... but it is installed, of course. [06:34] <BeNOW> I've been told that is not the best way to do it however. [06:35] <BeNOW> but I'm not sure of the better :) [06:35] <tnt> I was told NOT to install it (by omega)... he's the guru, so I just did it. [06:35] <BeNOW> yeah, I was told that after the fact... it works however. [06:36] <BeNOW> if you can, wait for word from omega or someone who knows. [06:36] <tnt> wingo and omega walked me through it... but neither of them are here right now. [06:37] <tnt> I guess I'll wait until tomorrow. [06:37] <tnt> I'm not even sure what time-zone everyone is in, some people seem to just be getting on when it is 2:00am here. [06:39] <BeNOW> yeah, they'll be getting up in europe around then, I believe [06:40] <tnt> I guess we're the only ones from Canada. [06:40] <BeNOW> here (mst) is GMT-7... sweden is GMT+1, I believe. [07:04] ajmitch (aj...@p4...) left irc: Ping timeout for ajmitch[p43-max2.dun.ihug.co.nz] [07:08] ajmitch (aj...@p4...) joined #gstreamer. [07:13] vishnu (jo...@cx...) got netsplit. [07:19] vishnu (jo...@cx...) got lost in the net-split. [07:19] tnt (tn...@h2...) left irc: Ping timeout for tnt[h24-83-72-250.vs.shawcable.net] [07:28] ShrimpX (marius@131.252.244.168) joined #gstreamer. [07:28] arik (ar...@sd...) joined #gstreamer. [07:28] arik (ar...@sd...) left #gstreamer. [07:34] vishnu (jo...@cx...) joined #gstreamer. [07:34] tnt (tn...@h2...) joined #gstreamer. [07:37] <BeNOW> anyone know of a simple function to resolve a host *or* an ip to an ip? [07:38] steveb (st...@no...) joined #gstreamer. [07:38] <BeNOW> (ie it reckognizes the ip as an ip and it doesn't matter what's given to it) [07:38] <BeNOW> heya steveb [07:42] <taaz> eh? [07:42] <taaz> like a c func? [07:44] walken (fo...@c1...) joined #gstreamer. [07:44] <walken> hi [07:44] <steveb> hi [07:51] <BeNOW> sorry, yeah... [07:51] <BeNOW> glib or something is ok too. [07:55] <BeNOW> inet_ntoa(char*), I guess. [07:56] <taaz> gethostbyname usefull too [07:56] <BeNOW> is that a glib func? [07:57] <BeNOW> g_get_host_by_name(char*) [07:57] <BeNOW> or some such beast... [07:57] <taaz> err... no, it's like standard network stuff [07:57] <taaz> look at libgnet if you want glib style interface [07:57] Action: BeNOW is more of a java dude. [07:57] <BeNOW> if it's std, that's ok. [07:58] <taaz> http://www.gnetlibrary.org/ [07:58] <taaz> oh, there's java api for it too [07:59] <taaz> not sure what it is now... haven't done network stuff in java for a few years [08:00] <BeNOW> yeah, no, I'd be all over it if it were java... still finding my C legs, as it were. [08:01] <vishnu> wow, neat [08:03] <taaz> eh? [08:03] <vishnu> gnet is neat. i didn't know about it [08:04] <taaz> yeah... i used like version 0 for something.. convienient at the time... i think it's improved since then though ;) [08:05] <vishnu> heh, maybe someday. tonight i'm extracting the diff algorithm from gnu diff. :-( [08:11] <taaz> why? [08:11] <taaz> so you can create a gstdiff element and compute the difference between streams? ;) [08:12] Nick change: omega_movie -> omega [08:12] <taaz> good mooovie? [08:12] Action: ajmitch will have to throw together some app for a video recorder [08:13] <ajmitch> won't have a VCR next year, might as well use computer ;) [08:13] <vektor> holy shit it is late [08:14] <taaz> help out vektor with reetpvr. he knows his stuff... [08:14] <vektor> ajmitch: just use my video recorder app [08:14] <vektor> it kicks yo azz [08:14] <ajmitch> vektor: yeah, i guess ;) [08:14] <ajmitch> vektor: where is it? [08:14] <ajmitch> what sorta hardware does it need for the compression? [08:14] <vektor> ajmitch: ah. [08:14] <vektor> well [08:14] <vektor> erm [08:15] <vektor> i dunno, what do you have? [08:15] <vektor> you could do some tests. [08:15] <vektor> like, hmm. [08:15] <ajmitch> vektor: don't have the needed hardware yet :) [08:15] <vektor> the source to reetpvr is at www.sf.net/projects/reetpvr/ [08:15] <vektor> but it's _raw_ [08:15] <vektor> like yeah [08:15] <vektor> i'm still _very_ actively working on it. [08:15] <ajmitch> but flatmate will be getting a new computer (and so will i hopefully) [08:15] <vektor> ok [08:15] <vektor> um [08:15] <vektor> let's see. [08:16] <vektor> i assume you want full 720x576, yes? [08:16] <vektor> and lossless quality, yes? [08:16] <ajmitch> i guess [08:16] <ajmitch> PAL here in NZ [08:16] <vektor> hrm. [08:16] <vektor> well, ok. [08:16] <vektor> i can do 480x480 no problem in realtime (lossless, using my huffyuv-style codec) on a p3-733. [08:16] <vektor> so that's like 9GB/half hour or so. [08:17] Action: ajmitch guesses this will take *lots* of diskspace for things like movies :) [08:17] <vektor> i get about 50% compression on my frames. [08:17] <vektor> erm [08:17] <vektor> 50-60%. [08:17] <vektor> lossless. [08:17] <vektor> you can optimize my code though alot. [08:17] <vektor> it's just in C. [08:17] <ajmitch> might go to lossy compression then [08:17] <omega> trick is to go back and use later (night, etc.) cycles on the pvr to recompress [08:17] <vektor> for my compressor. [08:17] <vektor> oh [08:17] <vektor> if you want lossy compression, you can do 720x576 no problem. [08:17] <vektor> and about the same size. [08:17] <vektor> maybe a wee bit less. [08:18] <ajmitch> i'll probably get lotsa diskspace (120GB or so between 4 computers) [08:18] <ajmitch> and if lucky, there'll be 2 computers with 1.4GHz athlon or higher [08:18] <omega> turns out krasic (maintainer of libdv) talked with robla in Ottawa just two weeks ago <g> [08:19] <ajmitch> k, /me has to go, will talk to you later :) [08:19] <vektor> lates. [08:19] <omega> cya [08:22] <tnt> Is any of the video stuff broken in the CVS version? [08:22] <tnt> I tried running the simple example... [08:23] <BeNOW> omega: icecastsend mods just about done... and they work even. [08:23] <tnt> gstreamer-launch disksrc location=video.mpg ! mpeg1parse video_00! queue ! { mp1videoparse ! mpeg_play ! sdlvideosink } [08:23] <tnt> ...but it just hangs with "RUNNING pipeline". [08:23] <omega> probably is ;-( we haven't been keeping up the plugins enough latetly [08:24] <taaz> why not use mpeg2dec instead? [08:24] <omega> no rpm around [08:24] <tnt> Because I just copied and pasted from the README :-) [08:24] <BeNOW> omega: what's the best way to convert "localhost" or "127.0.0.1" to "127.0.0.1"? [08:24] <omega> nslookup <g> [08:25] <omega> dunno, never messed with that at the code level [08:25] <tnt> I've installed gstreamer from CVS, so I have mpeg2dec now. [08:25] <BeNOW> k, will leave it requiring ip then. [08:25] <omega> tnt: mpeg2dec is a wrapper around an external library [08:25] <omega> if you don't have that lib, you don't have mpeg2dec [08:25] <BeNOW> do I ship the code to you? [08:25] <omega> yeah [08:26] <tnt> I downloaded the tarball and installed mpeg2dec... and a whole bunch of other stuff. [08:26] <omega> ok [08:26] <omega> then yeah, use mpeg2dec instead of mp1parse and mpeg_play [08:27] <omega> if you have a video-only mpeg, try to play that with just filesrc ! mpeg2dec ! sdlvideosink [08:27] <omega> grab gstreamer.net/media/thx.mp2 [08:27] <omega> or pond.mpg for mpeg 1 [08:28] <tnt> :-( It gives me error messages (with the mpeg2dec version): "mpeg2parse::parse_packet: not a valid packet time sequence" and a couple others. [08:28] <omega> with thx.mp2 or pond.mpg, don't use mpeg[12]parse, those are just elementary streams, video only [08:28] <tnt> I'll try the pond.mpg (for mpeg 1) [08:29] <omega> just filesrc ! mpeg2dec ! sdlvideosink [08:30] <tnt> :-) YES! FInally I have video output :-) [08:30] <omega> cool [08:30] <BeNOW> omega: modded icecastsend is at: http://benow.ca/misc/icecastsend.tar [08:32] <tnt> Oh, omega, what's the trick to compile programs with the CVS (NON-installed) version of gstreamer... gstreamer-config returns stuff for the installed version. [08:32] <omega> it compiles and works? (I can't test it) [08:32] <omega> tnt: use pkg-config, but you have to play some tricks for now ;-( [08:32] <omega> pkg-config isn't fully mature, so it has annoyances [08:32] <omega> if you run `pkg-config /path/to/gstreamer/gstreamer-uninstalled.pc --cflags --libs`, you'll get what you need [08:32] <tnt> Have any good references on how to use pkg-config. [08:32] <omega> the source ;-( [08:33] <tnt> OK :-) [08:33] <taaz> any good refs on how to use gstreamer? ;) [08:33] <tnt> :-D LOL [08:33] <omega> I'm considering writing a new version of it, in shell (it's in C, therefore can't be shipped with the tarball like libtool and others) [08:33] <omega> taaz: UTSL, um, er, yeah, we have docs, yo [08:34] <omega> BeNOW: committed to EVENTS1 [08:35] <BeNOW> omega: sorry, wanna grab again... extraneous import... is final now for sure. [08:35] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [08:36] <omega> done [08:36] <taaz> omega: huh? we still have an EVENTS1? [08:36] <BeNOW> cheers. [08:36] Action: taaz should stop being a smartalek [08:36] <omega> yeah, I think I'm gonna go merge HEAD in now... [08:36] <omega> um, /me goes to check out a new copy [08:37] <taaz> "If your brain had a butt, this is what would be kicking it." interesting commercial tag line ;) [08:37] <omega> hmm, /me tries to figure that out [08:39] <tnt> omega: that pkg-config command can't find libxml, and idea on how to fix that? (Hmmm, I remember someone mentioning something like that before, before my WM froze up... can't remember exactly what they said though.) [08:39] <omega> what version of libxml ? [08:40] <tnt> The version I have is the latest version with Ximian GNOME... hang on, I'll see what it is. [08:40] <omega> I have 1.8.15-ximian.2, it has /usr/lib/pkgconfig/libxml.pc [08:40] <omega> if your pkg-config wasn't built with --prefix=/usr (not just installed, but built), it won't have /usr/lib/pkgconfig in it [08:41] <omega> bleagh, I have to go look up how to do a merge again :-} [08:41] <tnt> I have that file (libxml.pc) [08:41] <tnt> (And in the same place.) [08:41] <omega> strings `which pkg-config` | grep 'lib/pkgconfig' [08:42] <tnt> Do you want me to run that on the CLI. [08:42] <omega> yes [08:43] <tnt> Well, that returns nothing... but which pkg-config return /usr/bin/pkg-config [08:43] <omega> did you configure pkg-config with /usr as the prefix? [08:43] <tnt> Yes. [08:44] <tnt> (At least I think I did... I did way to make ./configure's, make's, and make install's.) [08:44] <omega> do `pkg-config --debug libxml` and look for the fourth or so line, starting with 'scanning directory' [08:44] <tnt> should be "too many" [08:44] <omega> that's the path(s) being search [08:44] <omega> searched [08:45] <taaz> wuss... real men debug with strace! [08:45] <omega> this is faster.. [08:45] <omega> you want to debug a gstreamer app with s/--gst-mask=-1/strace/ ? [08:45] <taaz> oh take the 'fast' way out [08:46] <taaz> heh [08:46] <tnt> It says, "Cannot open directory '/usr/local/lib/pkgconfig' in packages search path: No such file or directory" [08:46] <omega> taaz: didn't think so [08:46] <omega> you still have a bogus pkgconfig [08:46] <taaz> i think i'm going to debug X with strace tomorrow [08:46] <taaz> seriously [08:47] <omega> need to obliterate all traces of pkgconfig, ./configure --prefix=/usr;make clean;make install [08:47] <tnt> OK. [08:47] <omega> taaz: take your pills [08:47] <omega> you need them, and you've obviously forgotten them for a few days [08:47] <taaz> dude, X hardlocks my machine... what am i supposed to do to fix this? [08:47] <omega> "reinstall!" [08:47] <taaz> haha [08:48] <omega> oh, sorry, that was 1-800-FIX-BSOD..... [08:48] <omega> real cheap number to operate, they just have a 2sec recording [08:48] <taaz> 4.0.x to 4.1.x made using my second video card lock the kernel [08:49] Action: omega closes his eyes and does a grep '^C' UPDATE.OUT [08:49] <taaz> 1280x1024 of screen realestate on my desk is dark for a month... i am clueless how to fix it [08:50] Action: omega opens his eyes [08:50] Action: omega runs for his life [08:50] <omega> 4.0.x [08:50] Action: omega turns around and sees *ZERO* conflicts [08:51] <omega> oh wait, they're hidden [08:51] <omega> 14 of them [08:51] ds (ds...@c7...) joined #gstreamer. [08:51] <omega> yo [08:59] <tnt> OK, pkg-config works now. All add all my wonderful experiences ( :-) ) compiling from the CVS, to the Wiki later... it might help someone. [09:00] <tnt> [Should say "I"] [09:00] <omega> cool ;-) [09:00] <tnt> dammit... [Should say "I'll add" not "All" add] [09:00] <omega> but see, if it were easy, you wouldn't have anything to feel proud about conquering <g> [09:00] <tnt> :-D [09:00] <omega> it'd also be called 1.0.... [09:01] <tnt> LOL [09:01] <tnt> Eventually I'll have an actual application using gstreamer :-) [09:01] <omega> one of these I plan to as well, believe it or not..... [09:02] wingo (wi...@rd...) joined #gstreamer. [09:04] <wingo> i can't sleep :-) [09:04] <wingo> so i am hacking on alsa :))) [09:05] <omega> ooooh [09:05] <omega> you be around tomorrow to mess with the hammerfall finally? [09:05] <tnt> [I should be sleeping... its 12:12AM here... you guys are just on at weird times :-) ] [09:05] Action: ds reads an integer from a .o file [09:05] <wingo> omega: bytestream works beautifully for alsa!!! [09:05] <omega> tnt: I show 12:04 [09:05] <omega> wingo: huh?? [09:05] <wingo> 3:20 am here :) [09:05] <ds> 00:05:49 [09:05] <tnt> I guess my clock is fast [09:06] <wingo> omega: i switched from my hacked sol'n to bytestream [09:06] <omega> no, wingo's clock is fast <g> [09:06] <wingo> and now instead of xruns i get perfect playback [09:06] <omega> wingo: for what part? [09:06] <omega> odd, playback? [09:06] <wingo> not multichannel yet [09:06] <wingo> i'm working on that [09:06] <tnt> Where's wingo?... on the East Coast (of the Americas)? [09:06] <omega> no, I mean is the bytestream stuff for playback or record? [09:06] <wingo> north carolina [09:06] <wingo> omega: within the element, for playback [09:07] <tnt> That's near the Canadian border, isn't it? [09:07] <wingo> yourself, tnt? [09:07] <wingo> tnt, no about in the middle :) [09:07] <tnt> (It's been about 10 years since I lived in the US... Texas.) [09:07] <wingo> on the east coast [09:07] <tnt> I'm in Canada... the west coasr in BC. [09:08] <wingo> 3 128 frame buffers on playback, no xruns on a slow machine with LL patches [09:08] <wingo> without i mean [09:08] <wingo> that's pretty good imho [09:08] <omega> so that's 2.9ms buffers? [09:09] <omega> or 2.666? [09:09] <wingo> 44100 kHz, my brain can't do math [09:09] <omega> 44.1, ok, 2.9ms) [09:09] <wingo> Hz even :) [09:09] <omega> (128/44.1) [09:09] <omega> so try 64 sample frames <g> [09:09] <wingo> trying... [09:09] <omega> I'd like to get to 32 or 48... (1ms or lower) [09:10] <omega> of course, it'd be on a fast machine with LL patches, but it'd be for 32 channels (16+16) [09:10] <wingo> it won't set it right, hrm [09:10] Action: ds calls a function in a .o file [09:10] <wingo> yeah, you're running into pci bus limitations when you get down that low [09:11] <omega> yup. that's why interleaved capture/output is useful [09:11] <omega> because it lets the card burst 32*(32/8)*16 in one go [09:11] <omega> 32 samples, 32-bit aligned, 16 channels [09:11] <omega> or 2KB [09:11] <wingo> ok, i just wasn't setting the right param :) [09:12] <wingo> but it balks on my machine. [09:12] <wingo> many xruns. [09:12] <omega> hmm [09:12] <omega> what card? [09:12] <wingo> ens1370 [09:12] <wingo> i can do it with 6 buffers [09:12] <omega> with LL ? [09:12] <omega> oh, 64x6 [09:12] <omega> hmm [09:12] <wingo> i don't have it enabled atm [09:12] <omega> ok, how about 32? <g> [09:13] <wingo> you need sched_fifo for it to be useful [09:13] <omega> how low can you go? [09:13] <wingo> which gst doesn't use [09:13] <omega> your app can just set it [09:13] <wingo> 64 iirc, checking [09:13] <wingo> giveRTcap from the mustajuuri project could be useful [09:14] <wingo> holy shit, it will do 32 frames/period [09:14] <wingo> going lower... [09:14] <omega> oooooooh [09:14] <omega> everyone's doin the frame-size limbo! [09:14] <wingo> ok, so 16 is an 'invalid argument' on my machine ;) [09:14] <omega> doh [09:14] <omega> but 32 is very cool [09:14] <wingo> indeedy doo. [09:15] <omega> of course, if you could get 32 in, process and 32 out, in <2ms, I'll be happy <g> [09:15] <omega> er ~2ms [09:15] <wingo> me too :) [09:15] <omega> and of course while the 32 in are coming in, 32 are going out, and we're processing another 32 [09:15] <wingo> i didn't try to grok the latency thing in the test dir [09:15] <omega> so it'll get interesting [09:15] <omega> lat.c ? [09:15] <wingo> yeah [09:15] <wingo> but what is gst's latency? [09:16] <omega> that's just a difference between threads and cothreads [09:16] <omega> cothread overhead is very very small [09:16] <omega> chain overhead is even better [09:16] <omega> of course... [09:16] <wingo> this is cool. we just need to get jacksrc/sink, alsa working, and ladspa reliable and things will be very cool. [09:16] <omega> yup [09:17] <omega> I did arecord | lame last sunday, I can't wait to do alsasrc ! lame ! filesink [09:18] <omega> ok, I'm gonna build my merged EVENTS1 before committing <g> [09:18] <wingo> good call :) [09:19] Action: omega lets that chew for a while.... [09:25] <vishnu> cool :-) [09:26] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [09:27] <tnt> :-)... another wonderful experience,... OK, I've got pkg-config working, however, when I try to compile (my own program), I get an error from "ld" (/usr/bin/ld). It says: "cannot find -lgst". The pkg-config command gives "-L/home/tnt/work/gstreamer/gstreamer/gst/" (*** is that end slash OK? ***) and "-lgst"... it should be able to find "libgst.la" (*** is this the correct kind of library file? ***). Any ideas? [09:28] <wingo> do you have a copy of gstreamer installed? [09:28] <wingo> did you just use the snippet out of GstHowTo? [09:28] <tnt> Just the CVS version. [09:28] <omega> link against builddir [09:28] <omega> you need to tack `libtool` onto the front of your gcc commandline [09:29] <wingo> are you using autotools? [09:29] <tnt> Umm, how?... libtool g++ ... [09:29] <tnt> Like that? [09:29] <omega> yeah [09:29] <tnt> OK. [09:30] <omega> merging sucks [09:30] <omega> cvs isn't acting like it's supposed to [09:30] <wingo> then, document your work at http://gstreamer.net/wiki/?DichotomousKey :) [09:31] <tnt> Wohoo!!!, it comiles... now I need to make it do something useful :-) [09:31] <omega> why? <g> [09:32] <wingo> well, i think multichannel is starting to work [09:32] <omega> cool. you think you can test stuff tomorrow? [09:33] <wingo> probably not :-( [09:33] <omega> grrr [09:33] <wingo> you can though :) [09:33] <wingo> i'll get it working perfect :) [09:33] <omega> then again, doing output on that card seems nontrivial anyway [09:33] <omega> there's supposedly some control you have to set to unmute the channels, but there is no such control in the ALSA interface [09:33] <omega> so it's UTSL of the audioengine code [09:34] <omega> if I can find it [09:35] <omega> um, where was ardour's alsa code these days again? [09:35] Nick change: taaz -> taazzzz [09:36] <wingo> -d :pserver:ano...@cv...:/cvsroot/quasimodo co libs/audio_engine [09:36] <wingo> it's in there [09:36] <wingo> -d :pserver:ano...@cv...:/cvsroot/quasimodo co libs/audioengine [09:36] <wingo> without the underscore [09:37] <omega> hrm, nothing relevant in hammerfall.cc [09:37] <wingo> interesting. my card does not support MMAP_NONINTERLEAVED, but it does for INTERLEAVED [09:37] <omega> maybe it'll decide to work [09:38] <omega> not surprising for a consumer card [09:38] <omega> do you know what mode is used in ardour for the hammerfall? [09:38] <wingo> it seems to be more or less generic alsa, once you have asound.conf set up right [09:38] <wingo> non [09:38] <wingo> afaik [09:38] <omega> hmm [09:38] <wingo> interleaved is only done through the plug layer [09:38] <omega> wonder how efficient the PCI transfers are in that case [09:38] <wingo> dunno [09:38] <omega> hmm [09:38] <wingo> does it transfer on write to a mmap'd buffer? [09:39] <wingo> i have to specifically release all mmap'd regions every period [09:39] <wingo> if it does it all at once, you're fine [09:39] <omega> dunno how mmaping works for alsa [09:40] <wingo> boy, i'm pleased as punch. [09:40] <omega> stupid thing is that I think even the hammerfall does bus mastering [09:40] <wingo> so... is the alsa machine alive right now? [09:40] <omega> s/does/doesn't [09:40] <omega> yes [09:40] <wingo> hammerfall i mean [09:40] <omega> but the ADAT interface is off [09:40] <wingo> ssh temple-baptist.com; ssh psi ? [09:40] <omega> yup [09:40] <wingo> doesn't matter, i don't think [09:40] <omega> you won't get sync on the first 8 channels [09:40] <omega> second 8 are looped back [09:41] <wingo> i'll poke around. [09:41] <omega> you have to log in first though <g> [09:41] <omega> the right password helps [09:41] <wingo> i can never quite remember, it takes a couple tries :) [09:41] <omega> actually telnet psi I think [09:42] <wingo> ssh worked [09:42] <omega> ok [09:42] <omega> ADAT1: No Lock [09:42] <omega> ADAT2: Sync [09:42] <omega> ADAT3: No Lock [09:42] <omega> if you can offset to use channels 9-16, you're OK [09:43] <wingo> ok [09:43] <omega> you have your own build to work with? [09:43] <omega> kinda old.... [09:43] <omega> not anymore though <g> [09:43] <wingo> i'm updating it and rebuilding [09:43] <omega> -rw-rw-r-- 1 wingo wingo 41083 Oct 17 00:44 configure.ac [09:43] <omega> so I see <g> [09:43] <wingo> :) [09:44] Action: omega watches wingo turn around abruptly and search in vain for his shadow [09:44] jarl_ (Lor...@a2...) joined #gstreamer. [09:45] bstard (Lor...@a2...) left irc: Ping timeout for bstard[a213-84-40-242.adsl.xs4all.nl] [09:46] Action: omega is getting frustrated with this merge [09:46] <omega> files aren't showing up where they're suppoed to [09:47] Action: wingo gives omega patience. [09:47] <wingo> it's necessary work :) [09:47] <omega> more useful would be the cvs commandline to properly bring in the files that are supposed to be here.... [09:50] <omega> good thing I have gigabytes available now, to handle the N trees I have going right now [09:50] <omega> DSL doesn't hurt either [09:52] <omega> http://ars.userfriendly.org/cartoons/?id=20011017 [09:53] <omega> I've never seen that many mini-punchlines in a single comic before [09:53] <wingo> :) [09:53] <omega> at least 6 [09:54] <wingo> i wish alsa docs were better :-\ [09:54] <omega> I wish alsa docs were. [09:55] <omega> anything at all would be a help... [09:56] <wingo> they aer working on it. really, though. [09:56] <wingo> recent discussions on alsa devel are gathering momentum. [09:56] <omega> hmm, you mean polishing those shiny 0.3.0 docs? [09:56] <omega> <g> [09:57] <wingo> there are some wikis, 0.9 api docs, and talk of docbook. [09:57] <wingo> :) [09:57] <omega> hmmm [09:57] <wingo> automake: both `configure.ac' and `configure.in' present: ignoring `configure.in' [09:57] <wingo> configure.ac:1143: LDFLAGS was set with `+=' and is now set with `=' [09:57] <wingo> make[3]: *** [Makefile.in] Error 1 [09:57] <wingo> that's an error? [09:57] <wingo> outright? [09:57] <omega> uh? [09:58] <wingo> maybe i'll make clean... [09:58] <omega> might have to ;-( [09:58] <omega> I wish that were a faster machine [09:58] <omega> you'll have to build only what you need.... [09:59] <wingo> so... [10:00] <wingo> that copy is not events. [10:00] <wingo> should i wait for your merge, or update -j ? [10:00] <omega> um, it'll be probably an hour before the merge, so do whatever you need to for now [10:01] <wingo> what does that mean wrt my local branch alsa? [10:01] <wingo> events alsa i mean [10:01] <omega> what's it based on? [10:01] <omega> oh, um... [10:01] <wingo> apply a diff later? [10:01] <omega> you'll want to do a cvs diff and re-checkout, apply that diff to the new HEAD [10:01] <omega> or update -dPA and see what happens [10:01] <omega> on a copy, of course <g> [10:01] Hal69 (Ha...@ca...) joined #gstreamer. [10:01] <wingo> i have a couple branches, i think we're ok [10:02] <wingo> does one normally poll a soundcard when you're reading from it? [10:02] <omega> depends on your operating mode [10:02] <omega> you mean poll() ? [10:03] <wingo> yeah [10:03] <wingo> i do it for playback [10:04] <wingo> but it hangs for capture [10:04] <omega> hmm, dunno [10:04] <omega> depends on the API's semantics [10:04] <wingo> i'm reading source... [10:06] <wingo> i think i got it. [10:08] <tnt> What do you use to figure out what elemets are in a factory? (gstreamer-inspect does not seem to list them.) For example, in the factory "disksrc" I have the element "src". How do I find this kind of stuff out? [10:09] <omega> you don't. [10:09] <omega> the factory is just a factory, it doesn't list any instances of itself [10:09] <omega> there's no 'in' a factory [10:09] <tnt> Well, here's my problem, I want to call gst_elementfactory_make("mpeg2dec"," [10:10] <tnt> Well, here's my problem, I want to call gst_elementfactory_make("mpeg2dec","?") but I don't know what the 2nd arg should be. [10:10] <omega> anything you want [10:10] <wingo> just a name [10:10] <omega> it's the name of the element instance you're creating [10:10] <tnt> Oh! OK. [10:10] <tnt> Is that name/label used in any way?... by other functions? [10:10] <omega> it means little unless you want to search for the element later without having a pointer to it, and we don't have those search routines anyway [10:11] <omega> yet [10:11] <omega> debug uses it too [10:11] <tnt> OK. [10:11] <omega> rather extensively [10:11] <omega> you've seen the debug stuff? [10:11] <tnt> Um, not sure. [10:11] <omega> add --gst-mask=-1 to your app's commandline [10:11] <omega> and | less -R to the end <g> [10:11] <tnt> OK... once I finish writing it. [10:12] <omega> heh [10:12] <omega> or add it to -launch or something [10:13] <omega> sit down first though [10:14] Action: omega waits for the response [10:15] <tnt> Ya... that was easy to read. [10:15] <tnt> (I was coding something... that's why it took a while.) [10:16] <omega> you like the debug output? ;-) [10:17] <tnt> Sure... colorful.... now I need to learn how to read a little faster. [10:17] <omega> hehehehe [10:17] <omega> less -R is really useful [10:17] <omega> I've got that aliased. otherwise color disappears [10:17] <omega> alias less='/usr/bin/less -R' [10:18] <tnt> I'll remember that! [10:18] <omega> just add it to your .bash_profile or .aliases files, if that's refered to [10:18] <wingo> it seems alsasrc works, but uses all my cpu :-\ [10:18] <omega> whoops [10:18] <omega> don't spin [10:18] Action: omega compile-tests the merged HEAD->EVENTS1 [10:19] <omega> after re-doing the merge [10:28] <ds> so I have a library now [10:28] <ds> that has an interface similar to libdl [10:29] <ds> but opens and gets symbols from a .o [10:29] <ds> where should I put it? [10:29] <omega> um [10:29] <ds> and what should I call it? libol? [10:30] <omega> um, what's it most useful for? [10:31] <ds> good question. lots of stuff [10:31] <omega> heh [10:33] <omega> if you want to put it on codecs.org, I think you even have write perms already [10:34] <omega> hrm, not. your userid is dschleef? [10:34] <wingo> if i want to sleep while the sound card does its thing, how should I do it? [10:34] <omega> dunno, that's up to the ALSA API [10:34] <wingo> well, i need to stop drinking coffee late at night, but alsasrc :) [10:35] <wingo> n/m, snd_pcm_wait [10:35] <ds> yes, dschleef [10:35] <omega> added [10:36] <omega> bleagh, my right hand is tweaking on me ;-( [10:36] <wingo> my head is tweaking on me :) [10:36] Action: wingo needs sleep [10:37] <omega> if you want, I can manage a new HEAD build for you on psi tomorrow [10:37] <omega> then again, you'll probably wake up about when I do <g> [10:39] <ds> omega: uh, added where? [10:39] <omega> to codecs.org project [10:39] <omega> sf.net/projects/codecs/ [10:41] <ds> must take a while to update [10:43] <wingo> ug ug, i don't understand what i'm doing [10:43] <omega> wingo: welcome to the club [10:46] tnt (tn...@h2...) left #gstreamer (Client Exiting). [10:50] <omega> committing HEAD->EVENTS1 [10:51] <omega> gonna create a lot of mail due to the new commitinfo mailer [10:51] <omega> bleagh [10:51] <omega> brace yourselves [10:51] Action: wingo straps himself in [10:51] <wingo> let me know when it's done so i can commit alsa [10:52] <wingo> alsasrc works fine, it seems, when in alsasrc ! alsasink [10:52] <wingo> timing issues and such [10:52] <omega> for some reason CHW is getting all the commit mail before I do <g> [10:52] <wingo> holy shit that's a lot of mail [10:53] <omega> right. I don't like that 'feature' of this new commitinfo script [10:53] <omega> I think everyone'll agree with me after this <g> [10:53] <omega> 33 messages so far [10:54] <omega> 34 [10:55] steveb_ (ste...@21...) joined #gstreamer. [10:55] <steveb_> merge! [10:55] steveb_ (ste...@21...) left irc: advance! [10:55] <omega> doh [10:55] <wingo> that's funny :) [10:57] mwc (ma...@ly...) left irc: [x]chat [10:58] tnt (tn...@h2...) joined #gstreamer. [10:58] <omega> good thing cvs tag doesn't create mail. [10:59] Vakor (ms...@co...) joined #gstreamer. [10:59] <omega> yo [10:59] <Vakor> hi. [10:59] <ds> omega: that can be configured [10:59] <omega> I hope so [10:59] Action: Vakor is feeling like doing some C programming so as to avoid doing F77 programming for a physics class... [10:59] <omega> doh [11:00] <wingo> i'm actually really well acquainted with f90 :) [11:00] <wingo> but sleepy. [11:00] Action: wingo goes to bed. [11:00] <omega> only 28 people get the -cvs list, so it's not so bad, but it includes poor ajmitch, who'll be downloading for most of tomorrow.... [11:00] Nick change: wingo -> wingo-z [11:00] <omega> ds: you missed the deluge [11:00] <omega> you'll get the next one though, unless I can turn that off first... [11:01] <ds> I do it all the time on RTAI CVS [11:01] <omega> wingo-z: way too early [11:01] <wingo-z> 5 in the morning? :) [11:01] walken (fo...@c1...) left irc: reboot 2.4.12-ac3 [11:01] <omega> wingo-z: no, you committed too soon [11:02] <omega> I have to be careful around those files now when doing the forward merge [11:02] <wingo-z> oh sorry. my bad... er [11:02] <omega> shouldn't be much of a problem [11:02] <Vako... [truncated message content] |
From: IRC B. <wt...@us...> - 2001-10-19 04:27:25
|
******************************************************************* [03:02] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [03:07] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [03:12] ds (ds...@c7...) joined #gstreamer. [03:19] Vakor (ms...@co...) got netsplit. [03:19] wtay-zZz (wi...@ca...) got netsplit. [03:19] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) got netsplit. [03:19] ds (ds...@c7...) got netsplit. [03:19] BeNOW (an...@ds...) got netsplit. [03:20] Vakor (ms...@co...) returned to #gstreamer. [03:20] wtay-zZz (wi...@ca...) returned to #gstreamer. [03:20] BeNOW (an...@ds...) returned to #gstreamer. [03:20] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) returned to #gstreamer. [03:20] ds (ds...@c7...) returned to #gstreamer. [03:21] Nick change: taaz-radio -> taaz [03:37] Action: evil_monk is back (gone 00:50:56) [03:53] omega_afk (om...@te...) joined #gstreamer. [04:15] <omega_afk> wingo: here? [04:15] Nick change: omega_afk -> omega [04:19] Shippou (no...@ot...) joined #gstreamer. [04:29] wtay-zZz (wi...@ca...) got netsplit. [04:29] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) got netsplit. [04:29] ds (ds...@c7...) got netsplit. [04:29] Shippou (no...@ot...) got netsplit. [04:29] BeNOW (an...@ds...) got netsplit. [04:29] Vakor (ms...@co...) got netsplit. [04:29] evil_monk (trevo@208.141.162.68) got netsplit. [04:29] thomasvs (th...@21...) got netsplit. [04:29] wingo (wi...@rd...) got netsplit. [04:29] omega (om...@te...) got netsplit. [04:29] ShrimpX (marius@131.252.244.168) got netsplit. [04:29] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [04:29] tnt (tn...@h2...) got netsplit. [04:29] taaz (dlehn@66.37.66.32) got netsplit. [04:29] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [04:29] ShrimpX (marius@131.252.244.168) returned to #gstreamer. [04:29] thomasvs (th...@21...) returned to #gstreamer. [04:29] tnt (tn...@h2...) returned to #gstreamer. [04:29] taaz (dlehn@66.37.66.32) returned to #gstreamer. [04:29] evil_monk (trevo@208.141.162.68) returned to #gstreamer. [04:29] wingo (wi...@rd...) returned to #gstreamer. [04:29] omega (om...@te...) returned to #gstreamer. [04:29] Vakor (ms...@co...) returned to #gstreamer. [04:30] BeNOW (an...@ds...) returned to #gstreamer. [04:30] ds (ds...@c7...) returned to #gstreamer. [04:30] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) returned to #gstreamer. [04:30] <wingo> omega: yeah, sorta [04:30] wtay-zZz (wi...@ca...) returned to #gstreamer. [04:30] <wingo> i figured out my issues :) [04:30] <wingo> back and forth with pbd on alsa-devel [04:30] <omega> ** WARNING **: error setting period count minimum (8): Invalid argument [04:31] <wingo> alsasrc period_count=2 [04:31] <wingo> hammerfall requires only two periods [04:31] <omega> hmm, ok [04:31] <omega> any way to tell that? [04:31] <wingo> period == fragment in oss-speak [04:31] <omega> right [04:31] <omega> ok, I have audio, checking [04:31] <wingo> i can make 2 the default, but i'm not sure how it would work on most cards [04:31] <omega> levels are low, but nonzero [04:32] <wingo> it's summing all the channels if you are doing src ! sink [04:32] <omega> does that param have to be set, or just left default unless overridden? [04:32] <wingo> see the transform channel [04:32] <omega> xform channel? [04:32] <wingo> omega default unless overridden [04:32] <omega> wingo: the default fails [04:32] <wingo> at the top of the debugging spew [04:32] <wingo> omega i know :) [04:32] <wingo> but only on the hammerfall [04:32] <omega> no, I mean don't set it at all unless param set [04:32] <wingo> see the 0 < 1 * 0.03993 etc [04:33] <wingo> omega: hmm [04:33] <wingo> omega: i'll check into that [04:33] <omega> what's this xform channel? [04:33] <omega> hammerfall has signal right now, if you have time to try it [04:33] <wingo> how it maps the multi channels into one channel, or two [04:33] <wingo> you cna bypass that [04:33] <wingo> if you do: [04:34] <wingo> gstreamer-launch alsasrc 'src3!' alsasink [04:34] <wingo> that gets channel 3 [04:34] <omega> is that mono? [04:34] <wingo> starting from 0 [04:34] <wingo> yes [04:34] <omega> ok, I'll try that [04:34] <wingo> i don't know if it works properly though [04:34] <omega> if I have a hammerfall and a sbawe, what device names do I use? [04:34] <wingo> plughw:0,0 and plughw:1.0 [04:34] <wingo> plughw:0,0 and plughw:1,0 even [04:34] <omega> ok [04:35] Shippou (no...@ot...) got lost in the net-split. [04:35] <wingo> assuming they show up in aplay --list-devices [04:35] <wingo> you can rename them in your asoundrc [04:35] <omega> card 0: card0 [RME Digi9652], device 0: RME Digi9652 [RME Digi9652] [04:35] <omega> card 1: card1 [Sound Blaster 16], device 0: SB16 DSP [DSP v4.16] [04:35] <wingo> yup [04:36] <omega> ok, you're taking off soon again? [04:36] <omega> you have any more code to commit? [04:36] <wingo> well, i have to read a few chapters of don quijote [04:36] <omega> quixote ? [04:36] <wingo> i can clean it up and make sure it works [04:36] <wingo> quijote in modern spanish :) [04:36] <omega> pfff, he's not modern [04:36] <wingo> he's timeless! [04:37] <omega> whatever <G> [04:37] <wingo> :) [04:37] <wingo> so... you doing something soon that you need a commit for? [04:37] <omega> I'm gonna test what you've suggested [04:37] <wingo> did that one-channel soln work for you? [04:37] <wingo> oh [04:37] <omega> checking, I need to grab some hw [04:37] <wingo> good luck :) [04:38] <wingo> i'll be checking irc off and on [04:38] <wingo> i just have to face the other way so i can get some work done :) [04:39] <omega> heh [04:42] Action: taaz writing respones to wingos mail [04:57] Vakor (ms...@co...) got netsplit. [04:57] evil_monk (trevo@208.141.162.68) got netsplit. [04:57] BeNOW (an...@ds...) got netsplit. [04:57] thomasvs (th...@21...) got netsplit. [04:57] wingo (wi...@rd...) got netsplit. [04:57] omega (om...@te...) got netsplit. [04:57] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) got netsplit. [04:57] wtay-zZz (wi...@ca...) got netsplit. [04:57] ShrimpX (marius@131.252.244.168) got netsplit. [04:57] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [04:57] tnt (tn...@h2...) got netsplit. [04:57] taaz (dlehn@66.37.66.32) got netsplit. [04:57] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [04:57] ShrimpX (marius@131.252.244.168) returned to #gstreamer. [04:57] thomasvs (th...@21...) returned to #gstreamer. [04:57] tnt (tn...@h2...) returned to #gstreamer. [04:57] taaz (dlehn@66.37.66.32) returned to #gstreamer. [04:57] wingo (wi...@rd...) returned to #gstreamer. [04:57] omega (om...@te...) returned to #gstreamer. [04:57] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) returned to #gstreamer. [04:57] wtay-zZz (wi...@ca...) returned to #gstreamer. [04:57] evil_monk (trevo@208.141.162.68) returned to #gstreamer. [04:57] BeNOW (an...@ds...) returned to #gstreamer. [04:57] Vakor (ms...@co...) returned to #gstreamer. [05:02] <omega> wingo: ok, I can output mono to src0 of the alsasink on the sbawe. how would I go about outputting to only the left channel? <g> [05:03] <omega> the whole configuration aspect of these elements is gonna be really hairy [05:03] Nick change: omega -> omega_afk [05:10] evil_monk (trevo@208.141.162.68) got netsplit. [05:10] BeNOW (an...@ds...) got netsplit. [05:11] evil_monk (trevo@208.141.162.68) returned to #gstreamer. [05:11] BeNOW (an...@ds...) returned to #gstreamer. [05:13] evil_monk (trevo@208.141.162.68) got netsplit. [05:13] BeNOW (an...@ds...) got netsplit. [05:13] Vakor (ms...@co...) got netsplit. [05:13] thomasvs (th...@21...) got netsplit. [05:13] wingo (wi...@rd...) got netsplit. [05:13] omega_afk (om...@te...) got netsplit. [05:13] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) got netsplit. [05:13] wtay-zZz (wi...@ca...) got netsplit. [05:13] ShrimpX (marius@131.252.244.168) got netsplit. [05:13] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [05:13] tnt (tn...@h2...) got netsplit. [05:13] taaz (dlehn@66.37.66.32) got netsplit. [05:13] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [05:13] ShrimpX (marius@131.252.244.168) returned to #gstreamer. [05:13] thomasvs (th...@21...) returned to #gstreamer. [05:13] tnt (tn...@h2...) returned to #gstreamer. [05:13] taaz (dlehn@66.37.66.32) returned to #gstreamer. [05:13] wingo (wi...@rd...) returned to #gstreamer. [05:13] omega_afk (om...@te...) returned to #gstreamer. [05:13] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) returned to #gstreamer. [05:13] wtay-zZz (wi...@ca...) returned to #gstreamer. [05:13] evil_monk (trevo@208.141.162.68) returned to #gstreamer. [05:13] BeNOW (an...@ds...) returned to #gstreamer. [05:13] Vakor (ms...@co...) returned to #gstreamer. [05:17] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [05:22] <wingo> omega_afk: i'm not quite sure about that. the fact that it gets mixed to stereo is a function of the plug layer. [05:22] <wingo> so maybe if you use hw:0,0 that will work. [05:23] <wingo> otoh, hw:0,0 is very picky about its setup [05:23] <wingo> atm, you have to support noninterleaved access natively on the card for it to work. [05:24] <wingo> i have to write some interleaved memory routines (or better yet, put them in a lib) to get that to work with interleaved access. [05:25] <wingo> omega_afk: i have to clean up the src code, but essentially my cpu-eating error has to do with setting the right poll implements, and me having searched for a kludge instead of poll on sources. i'll fix it later. [05:25] <wingo> s/implements/events [05:29] <wingo> the most robust solution is to play with asoundrc, or do a pan in gstreamer. if you ask for two channels, afaik they should come out on different channels rather than being mixed. [05:29] Shippou (no...@ot...) joined #gstreamer. [05:30] <wingo> omega_afk: check the routing table on gstreamer-launch alsasrc 'src0,src1!sink0,sink1' alsasink [05:39] <wingo> tools/gstreamer-launch alsasrc period_count=2 device=plughw:1,0 'src0,src1!sink0,sink1' alsasink period_count=2 device=plughw:1,0 2>&1 | less -R [05:39] <wingo> that gives me a nice error :) [05:54] BeNOW (an...@ds...) got netsplit. [05:54] BeNOW (an...@ds...) returned to #gstreamer. [05:57] <wingo> hey BeNOW. [05:58] <taaz> so wingo... alsa easy to setup in debian? [06:02] <wingo> it's ok [06:03] <wingo> the only tricky thing is the modutils voodoo [06:03] <wingo> /etc/init.d/alsa puts /etc/alsa/modutils/VERSION in modules.conf for you [06:03] <wingo> even [06:03] <wingo> but without using the normal modutils framework [06:04] <wingo> and irritatingly enough, you need to name VERSION to be 1.0 for it to work (instead of 0.9) [06:04] <wingo> the easiest way to create that file is to use alsaconf, [06:05] <wingo> but that only does 0.5.11, so you copy it to 1.0 and then do insmod, removing args as it tells you, [06:05] <wingo> until it works :) [06:06] <BeNOW> heya wingo [06:07] Action: ds thinks wingo needs to file a bug report or 5 [06:08] <BeNOW> so... gstreamer has issues dealing with mangled mpeg streams (prolly mpg123) and initial sync (again sounds like mpg123) and has buffer problems somewhere after a while and seg faults occasionally. ... I'll try and narrow in on the probs and what causes them [06:08] <BeNOW> 24/7 encoding (256k to 128k) and httpsrc and icecastsend is really testing it. [06:09] <BeNOW> unfortunately the failures occur after 3hrs of streaming, so logging is not all that helpful. [06:29] <wingo> sucks :-\ [06:29] <wingo> use mad if you can [06:40] <BeNOW> hmmm, seems mad doesn't need an mp3parse. [06:41] <BeNOW> httpsrc ! mad ! osssrc works, while, httpsrc ! mp3parse ! mad ! osssrc doesn't [06:43] <BeNOW> sweet... mad now in pipe.. seems fine, sounds a bit better.... I'll see how it does in the long run. [06:49] <wingo> supposedly it's a bit more robust too :) [07:00] omega_afk (om...@te...) left irc: l8r [07:01] <taaz> oh my [07:01] <taaz> head is having serious issues here [07:01] <taaz> smp thread stuff i think :( [07:01] <wingo> really? [07:01] <wingo> ug. [07:01] <wingo> did you just switch to smp? [07:02] <taaz> no [07:02] <taaz> well... just == 22 months ago ;) [07:04] <wingo> :) [07:04] <taaz> hmm... [07:08] <taaz> i can't tell what's not working [07:08] <taaz> i'm just having issue with dvd player... [07:09] <taaz> but that's pretty simple code i think [07:09] <wingo> what's threaded about it? [07:11] <taaz> i dunno... i wrote it months ago ;) [07:13] <taaz> it's just like mpeg2parse3 really [07:14] <ChiefHighwater> bbl [07:14] ChiefHighwater (pa...@te...) left irc: [07:14] chillywilly (da...@d5...) left irc: Read error to chillywilly[d5.as14.nwbl0.wi.voyager.net]: EOF from client [07:17] tnt (tn...@h2...) left irc: Ping timeout for tnt[h24-83-72-250.vs.shawcable.net] [07:35] steveb (st...@no...) joined #gstreamer. [07:41] <wingo> hey steveb [07:41] <wingo> hey, i narrowed down ladspa errors to only ones that have control ports [07:42] <steveb> ok. don't they all? [07:47] <wingo> no, identity_audio, sine_faaa, and all other ones w/o control ports work for me [07:47] <wingo> all of the useless ones :) [07:51] <steveb> can you send me some -launch commands that are failing? [08:06] <BeNOW> there's another live show on now... [08:06] <BeNOW> tables->mac os9->shoutcast... shoutcast->gst->public shoutcast [08:07] <BeNOW> no hiccups since using mad. [08:14] <vektor> BeNOW: Do you write tracks? [08:16] <BeNOW> not really... play around a bit with my little EA1, but mostly listen and dj a bit. [08:16] tnt (tn...@h2...) joined #gstreamer. [08:16] <vektor> Interesting. [08:16] <vektor> I kinda want to get an EA1. [08:16] <BeNOW> ... too busy making it go :) [08:16] <vektor> I have an ER1. [08:16] <vektor> And a shitload of other gear. [08:16] <BeNOW> ahh, the other. [08:16] <BeNOW> sweet. [08:17] <BeNOW> yeah, I know I'd very much enjoy it, but I want to get this done before I do ;) [08:17] <vektor> I have alot of synths ;) [08:17] <vektor> Get what done? [08:17] <BeNOW> the benow stuff. [08:17] <BeNOW> it soaks up the $ and time. [08:17] <vektor> You mean just broadcasting stuff? [08:17] <BeNOW> well, erm, kinda. [08:17] <vektor> So, you just broadcast? [08:17] <ajmitch> BeNOW: one problem with trying to listen to your broadcasted stuff - i don't have the bandwidth :) [08:18] <BeNOW> yeah, there's a 48k stream too... if that's toned down enough. [08:18] <ajmitch> BeNOW: i have a 33.6k modem :) [08:18] <ajmitch> bring on DSL! [08:19] <BeNOW> bummer. yeah, I've a friend that wants a 24k stream, but the quality is pretty low over 24k [08:19] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [08:19] <vektor> It's not worth it. [08:19] <BeNOW> vektor: yeah, I've got schemes and plans tho. [08:19] <ajmitch> i can't get DSL until i move in 3 months [08:19] <BeNOW> initially I want to facilitate rebroadcasting, and finish the playlist stuff. [08:19] Action: ajmitch should see if he can use uni computers :) [08:20] <BeNOW> then get the site acting as a host for others who have stuff they'd like to share... present the albums and tracks through one interface, but the files are located whereever. [08:20] <vektor> BeNOW: I'll be doing a live PA in a few weeks, if you want to rebroadcast that. :) [08:20] <vektor> BeNOW: I could feed it to you live pretty easily. [08:20] mwc (malcolm@138.80.54.113) joined #gstreamer. [08:20] <BeNOW> then tie it in to a p2p metadata network. [08:21] <BeNOW> yeah, it'd be great to broadcast it. [08:21] <vektor> BeNOW: I'll let you know the details soon. [08:21] <BeNOW> the usual way is the dj's set up a shoutcast on their side, then the streamer connects from here and pulls and feeds to the mirrors. [08:21] Action: vektor idle. [08:22] <BeNOW> can have 29@128k and 10@48k... for now. [08:22] <vektor> BeNOW: Yeah, we're having this party on campus, and it's right near an internet feed, so I can get you a live feed right from the party. [08:22] <vektor> There will be many synths present. ;-) Much analog goodness as well. [08:23] <BeNOW> nice... yeah, let me know and I'll put it in the upcoming events. [08:23] <ajmitch> nice [08:23] <vektor> k. [08:23] Action: vektor idle. [08:26] Action: taaz perks up [08:27] <taaz> vektor live! yeah! [08:27] <ajmitch> hehe [08:27] <taaz> i wish i knew someone around here with music gadgets [08:47] Vakor (ms...@co...) left irc: Ping timeout for Vakor[co3016667-a.lowrp1.vic.optushome.com.au] [08:47] <wingo> steveb: even inspect doesn't work [08:48] <wingo> on those plugins [08:50] <wingo> that's off a fresh rebuild too [08:50] <wingo> anyway, i'm going home, it's three in the morning here :) [08:52] wingo (wi...@rd...) left irc: sleep [09:05] steveb (st...@no...) left irc: Ping timeout for steveb[node1ee15.a2000.nl] [09:09] BeNOW (an...@ds...) got netsplit. [09:09] tnt (tn...@h2...) got netsplit. [09:09] mwc (malcolm@138.80.54.113) got netsplit. [09:09] thomasvs (th...@21...) got netsplit. [09:09] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) got netsplit. [09:09] wtay-zZz (wi...@ca...) got netsplit. [09:09] Shippou (no...@ot...) got netsplit. [09:09] ShrimpX (marius@131.252.244.168) got netsplit. [09:09] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [09:09] taaz (dlehn@66.37.66.32) got netsplit. [09:10] ajmitch (aj...@p4...) left irc: Ping timeout for ajmitch[p42-max7.dun.ihug.co.nz] [09:10] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [09:10] ShrimpX (marius@131.252.244.168) returned to #gstreamer. [09:10] thomasvs (th...@21...) returned to #gstreamer. [09:10] taaz (dlehn@66.37.66.32) returned to #gstreamer. [09:10] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) returned to #gstreamer. [09:10] wtay-zZz (wi...@ca...) returned to #gstreamer. [09:10] Shippou (no...@ot...) returned to #gstreamer. [09:10] BeNOW (an...@ds...) returned to #gstreamer. [09:10] tnt (tn...@h2...) returned to #gstreamer. [09:10] BeNOW (an...@ds...) got netsplit. [09:10] tnt (tn...@h2...) got netsplit. [09:10] BeNOW (an...@ds...) returned to #gstreamer. [09:10] tnt (tn...@h2...) returned to #gstreamer. [09:10] steveb (st...@no...) joined #gstreamer. [09:13] mwc (malcolm@138.80.54.113) returned to #gstreamer. [09:35] mwc (malcolm@138.80.54.113) left irc: [x]chat [09:38] tnt (tn...@h2...) left irc: Ping timeout for tnt[h24-83-72-250.vs.shawcable.net] [09:41] Nick change: ShrimpX -> Shrimpsalad [09:44] taaz (dlehn@66.37.66.32) left irc: Ping timeout for taaz[66.37.66.32] [09:44] Nick change: Shrimpsalad -> ShrimpX [09:47] thomasvs (th...@21...) got netsplit. [09:47] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) got netsplit. [09:47] wtay-zZz (wi...@ca...) got netsplit. [09:47] Shippou (no...@ot...) got netsplit. [09:48] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) returned to #gstreamer. [09:48] thomasvs (th...@21...) returned to #gstreamer. [09:48] wtay-zZz (wi...@ca...) returned to #gstreamer. [09:53] Shippou (no...@ot...) got lost in the net-split. [09:57] taaz (dlehn@66.37.66.32) joined #gstreamer. [10:09] ajmitch (aj...@p4...) joined #gstreamer. [10:19] Action: xav_ is back (gone 10:36:47) [10:25] BeNOW (an...@ds...) got netsplit. [10:25] steveb (st...@no...) got netsplit. [10:25] ajmitch (aj...@p4...) got netsplit. [10:25] ShrimpX (marius@131.252.244.168) got netsplit. [10:25] thomasvs (th...@21...) got netsplit. [10:25] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [10:26] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) left irc: Ping timeout for vektor[HSE-Kitchener-ppp195088.sympatico.ca] [10:26] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [10:26] ShrimpX (marius@131.252.244.168) returned to #gstreamer. [10:26] thomasvs (th...@21...) returned to #gstreamer. [10:26] arik (ar...@sd...) joined #gstreamer. [10:26] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) joined #gstreamer. [10:26] ajmitch (aj...@p4...) returned to #gstreamer. [10:26] BeNOW (an...@ds...) returned to #gstreamer. [10:27] steveb (st...@no...) returned to #gstreamer. [10:33] Nick change: taaz -> taazzzz [10:42] <arik> bbl [10:42] arik (ar...@sd...) left #gstreamer. [10:48] Shippou (no...@ot...) joined #gstreamer. [11:02] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) left irc: Ping timeout for vektor[HSE-Kitchener-ppp195088.sympatico.ca] [11:02] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) joined #gstreamer. [11:39] Vakor (ms...@co...) joined #gstreamer. [11:57] <xav_> welcome back to split.openprojects.org [11:58] xav_ (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) left #gstreamer. [11:58] xav_ (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) joined #gstreamer. [12:10] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [12:21] xav_ (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) left irc: Client Exiting [12:23] xav (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) joined #gstreamer. [12:27] Nick change: ShrimpX -> ShrimpZzZ [13:35] Vakor (ms...@co...) got netsplit. [13:35] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) got netsplit. [13:35] xav (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) got netsplit. [13:35] BeNOW (an...@ds...) got netsplit. [13:35] ajmitch (aj...@p4...) got netsplit. [13:35] Shippou (no...@ot...) got netsplit. [13:35] ShrimpZzZ (marius@131.252.244.168) got netsplit. [13:35] thomasvs (th...@21...) got netsplit. [13:35] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [13:35] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [13:35] ShrimpZzZ (marius@131.252.244.168) returned to #gstreamer. [13:35] thomasvs (th...@21...) returned to #gstreamer. [13:35] Shippou (no...@ot...) returned to #gstreamer. [13:35] BeNOW (an...@ds...) returned to #gstreamer. [13:35] ajmitch (aj...@p4...) returned to #gstreamer. [13:35] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) returned to #gstreamer. [13:35] xav (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) returned to #gstreamer. [13:35] Vakor (ms...@co...) returned to #gstreamer. [13:58] bstard (Lor...@a2...) joined #gstreamer. [14:05] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [14:20] Vakor (ms...@co...) left irc: Ping timeout for Vakor[co3016667-a.lowrp1.vic.optushome.com.au] [14:39] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) got netsplit. [14:39] xav (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) got netsplit. [14:39] BeNOW (an...@ds...) got netsplit. [14:39] ajmitch (aj...@p4...) got netsplit. [14:40] BeNOW (an...@ds...) returned to #gstreamer. [14:40] ajmitch (aj...@p4...) returned to #gstreamer. [14:40] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) returned to #gstreamer. [14:40] xav (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) returned to #gstreamer. [14:40] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) got netsplit. [14:40] xav (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) got netsplit. [14:40] BeNOW (an...@ds...) got netsplit. [14:40] ajmitch (aj...@p4...) got netsplit. [14:40] BeNOW (an...@ds...) returned to #gstreamer. [14:40] ajmitch (aj...@p4...) returned to #gstreamer. [14:40] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) returned to #gstreamer. [14:40] xav (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) returned to #gstreamer. [14:40] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) got netsplit. [14:40] xav (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) got netsplit. [14:40] BeNOW (an...@ds...) got netsplit. [14:40] ajmitch (aj...@p4...) got netsplit. [14:40] ShrimpZzZ (marius@131.252.244.168) got netsplit. [14:40] thomasvs (th...@21...) got netsplit. [14:40] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [14:40] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [14:40] ShrimpZzZ (marius@131.252.244.168) returned to #gstreamer. [14:40] thomasvs (th...@21...) returned to #gstreamer. [14:41] BeNOW (an...@ds...) returned to #gstreamer. [14:41] ajmitch (aj...@p4...) returned to #gstreamer. [14:41] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) returned to #gstreamer. [14:41] xav (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) returned to #gstreamer. [14:42] asmod (stevec@64.5.222.2) joined #gstreamer. [14:44] Vakor (ms...@co...) joined #gstreamer. [14:59] ShrimpZzZ (marius@131.252.244.168) got netsplit. [14:59] thomasvs (th...@21...) got netsplit. [14:59] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [14:59] asmod (stevec@64.5.222.2) got netsplit. [14:59] ShrimpZzZ (marius@131.252.244.168) returned to #gstreamer. [14:59] thomasvs (th...@21...) returned to #gstreamer. [14:59] asmod (stevec@64.5.222.2) returned to #gstreamer. [14:59] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [15:00] ShrimpZzZ (marius@131.252.244.168) got netsplit. [15:00] thomasvs (th...@21...) got netsplit. [15:00] asmod (stevec@64.5.222.2) got netsplit. [15:00] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [15:00] thomasvs (th...@21...) returned to #gstreamer. [15:00] asmod (stevec@64.5.222.2) returned to #gstreamer. [15:00] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [15:00] thomasvs (th...@21...) got netsplit. [15:00] asmod (stevec@64.5.222.2) got netsplit. [15:00] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [15:00] thomasvs (th...@21...) returned to #gstreamer. [15:00] asmod (stevec@64.5.222.2) returned to #gstreamer. [15:00] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [15:01] ShrimpZzZ (marius@131.252.244.168) returned to #gstreamer. [15:03] bstard (Lor...@a2...) got netsplit. [15:03] bstard (Lor...@a2...) returned to #gstreamer. [15:04] thomasvs (th...@21...) got netsplit. [15:04] asmod (stevec@64.5.222.2) got netsplit. [15:04] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [15:04] thomasvs (th...@21...) returned to #gstreamer. [15:04] asmod (stevec@64.5.222.2) returned to #gstreamer. [15:04] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [15:06] Shippou (no...@cr...) joined #gstreamer. [15:21] thomasvs (th...@21...) got netsplit. [15:21] asmod (stevec@64.5.222.2) got netsplit. [15:21] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [15:21] thomasvs (th...@21...) returned to #gstreamer. [15:21] asmod (stevec@64.5.222.2) returned to #gstreamer. [15:21] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [15:56] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:15] steveb_ (ste...@21...) joined #gstreamer. [16:27] ajmitch (aj...@p4...) left irc: Ping timeout for ajmitch[p41-max11.dun.ihug.co.nz] [16:29] asmod (stevec@64.5.222.2) left irc: [x]chat [16:36] asmod (stevec@64.5.222.2) joined #gstreamer. [16:41] steveb_ (ste...@21...) left irc: advance! [16:48] bstard (Lor...@a2...) got netsplit. [16:48] Vakor (ms...@co...) got netsplit. [16:48] thomasvs (th...@21...) got netsplit. [16:48] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [16:48] evil_monk (trevo@208.141.162.68) got netsplit. [16:48] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) got netsplit. [16:48] xav (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) got netsplit. [16:48] asmod (stevec@64.5.222.2) got netsplit. [16:48] BeNOW (an...@ds...) got netsplit. [16:48] ShrimpZzZ (marius@131.252.244.168) got netsplit. [16:48] Shippou (no...@cr...) got netsplit. [16:48] Shippou (no...@cr...) returned to #gstreamer. [16:48] bstard (Lor...@a2...) returned to #gstreamer. [16:48] thomasvs (th...@21...) returned to #gstreamer. [16:48] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [16:48] BeNOW (an...@ds...) returned to #gstreamer. [16:48] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) returned to #gstreamer. [16:48] xav (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) returned to #gstreamer. [16:48] asmod (stevec@64.5.222.2) returned to #gstreamer. [16:48] Vakor (ms...@co...) returned to #gstreamer. [16:49] ShrimpZzZ (marius@131.252.244.168) returned to #gstreamer. [16:50] evil_monk (trevo@208.141.162.68) returned to #gstreamer. [16:53] ajmitch (aj...@p4...) joined #gstreamer. [17:16] vishnu (jo...@cx...) joined #gstreamer. [17:25] Vakor (ms...@co...) left irc: Ping timeout for Vakor[co3016667-a.lowrp1.vic.optushome.com.au] [18:25] xav (xav@AMontpellier-201-1-5-32.abo.wanadoo.fr) left irc: Read error to xav[AMontpellier-201-1-5-32.abo.wanadoo.fr]: EOF from client [18:40] <BeNOW> hello... gstreamer made it through the night! [18:57] <thomasvs> hi BeNOW [18:58] <thomasvs> I was wondering if you were willing to cooperate on writing radio automation software ? [19:00] <BeNOW> we [19:00] <BeNOW> well, that's kinda what I'm doing here.... [19:00] uwe (chatzilla@213.174.92.50) joined #gstreamer. [19:00] <BeNOW> a standalone package could be interesting tho. [19:02] <uwe> BeNOW: you're using gstreamer for radio right? [19:02] <BeNOW> yup... full time now. [19:02] <uwe> osssrc or alsasrc? [19:02] <BeNOW> (it made it through the night!) [19:02] <BeNOW> when I go live, it's osssrc. [19:03] <uwe> and lame? [19:03] <BeNOW> haven't played with alsa yet. [19:03] <BeNOW> yeah. [19:03] <uwe> which version? [19:03] <BeNOW> but that's only when live. [19:03] <BeNOW> 3.89beta, I believe [19:03] <uwe> which linux? [19:03] <BeNOW> the other times, gst is used to pull from djs and pull from a local shoutcast, re-encode and send. [19:04] <BeNOW> playlister stuff is still perl for now. [19:04] <BeNOW> Debian unstable. [19:04] <uwe> ! [19:04] <BeNOW> 2.4.6 [19:04] <uwe> osssrc + lame on redhat seq fault [19:04] <uwe> osssrc + lame on suse no seg fault, but no sound [19:05] <uwe> i'll try debian too, thank you [19:05] <BeNOW> humm.. no problems here. [19:05] <uwe> did you compile from cvs or from the tar? [19:06] <BeNOW> I'd suggest trying the pipeline with -launch first... just to eliminate any possibility of one's own code screwing up the works [19:06] Action: BeNOW 's been there. [19:06] <BeNOW> from cvs. [19:06] <uwe> ja, but the pipeline crashes already [19:07] <uwe> uses debian alsa as soundsystem? [19:07] <BeNOW> hmm, this is my pipeline for encoding and broadcasting: [19:08] <BeNOW> nice -n 10 gstreamer-launch osssrc bytes_per_read=16384 format=16 channels=2 frequency=44100 ! { queue maxlevel=1000 ! lame quality=2 bitrate=128 mode=0 ! icecastsend ip=207.34.113.105 port=7128 password=str3am icy=true name="BeNOW Streaming Goodness" genre="techno goa trance ambient jungle dnb hip hop" } [19:08] <uwe> what is the nice for? [19:08] <BeNOW> that's with a modified icecastsend... name, genre, icy won't work for you. [19:09] <BeNOW> just to run it at a bit higher priority... unsure if it helps all that much. [19:09] <uwe> it will partly, we modified icecastsend to ... [19:10] <uwe> because we want to send streaming flash from live input [19:12] <uwe> for what is queue exactly? [19:13] MM (m.m...@39...) joined #gstreamer. [19:13] <BeNOW> buffers input... so things don't get out of sync. [19:14] <BeNOW> with cpu load increase, etc. [19:14] <uwe> ! [19:15] <BeNOW> !? [19:15] <uwe> thanks. will try that for some problems [19:16] <uwe> but your using your program, not a command line pipeline? [19:16] <BeNOW> yep, np. it was a bitch here @ the start too... just starting to come together now after a month and a bit of tinkering. [19:16] <MM> Hay you guys coding keep up the good work :) [19:16] <BeNOW> nope backend is cmd line... or variations. [19:17] <uwe> i try to combine gtk and gstreamer ... but with the new cvs there seems to be problems with threadstuff [19:18] <BeNOW> hmm, yeah, I'm as basic as it gets right now... I'm going to start playing with the gst/java bindings when time becomes available. [19:18] <BeNOW> the playlist stuff will be gstreamer, eventually. [19:18] Action: BeNOW needs his flacs. [19:19] <BeNOW> then I'll basically have a true cd quality remote controlled playlist. [19:20] <uwe> i'll use different pipelines for different tasks, but destroying a pipeline and create a new doesn't work so good :( [19:20] <BeNOW> (which is open sourse of course) [19:20] <BeNOW> yeah, [19:20] <BeNOW> code is much better, but there are more variables to go wrong. [19:21] <BeNOW> I chose to get the basics working, the atleast I know it works... [19:21] <BeNOW> I can now start to build code around it... [19:21] <BeNOW> where needed. [19:21] <uwe> same here, on a cmdline was it easy, but with the gui ... [19:22] <BeNOW> yup, becomes tougher. [19:23] <uwe> we have to change icecast too ... not so easy [19:23] <BeNOW> for code, I'm using gst_parse_launch to parse a pipeline char* [19:23] <BeNOW> works well. [19:24] <BeNOW> you can then manipulate the pipe, if needed. [19:24] <BeNOW> facilitates testing. [19:24] <BeNOW> why change icecast? [19:24] <BeNOW> to do flash? [19:24] <uwe> will try that. change icecast to serve flash [19:25] <BeNOW> why flash? doesn't flash read streaming mp3? [19:25] <BeNOW> I thought shockwave had a streaming mp3 decoder of sorts. [19:26] <uwe> ... [19:28] <uwe> not sure. coworker said no ... [19:28] <BeNOW> may want to verify that. [19:28] wingo (wi...@rd...) joined #gstreamer. [19:28] <BeNOW> it'd save you a bunch of work. [19:28] <wingo> what up, people [19:29] <uwe> oh shit, i hope your'e not right [19:30] <uwe> hi wingo [19:30] <wingo> hi [19:33] <uwe> BeNOW: i think you're right. but i have to check tomorrow, wy we are doing this, not sure anymore ... [19:35] <BeNOW> hehe, great I hope it saves you from having to do the work you haven't already done ;) [19:36] Nick change: wtay-zZz -> wtay [19:36] <wtay> yo [19:36] <uwe> we've done a lot of work yet, but i can remeber there was a reason for using flash [19:37] <uwe> wtay: goede avond [19:37] <uwe> hoe gaat het met jouw? [19:40] <wtay> nogal goed :) [19:41] <wtay> vanwaar ben jij? [19:42] <uwe> duitsland. maar ik werk in rotterdam. [19:42] <wtay> ah [19:43] <uwe> ik was in antwerpen, heel moie stad [19:43] <wtay> antwerpen? hmm.. [19:43] <uwe> and het belgisch bier! [19:44] <uwe> en ... [19:44] <wtay> over het bier zijn we het eens :) [19:44] <uwe> BeNow: did you try different bitrates with gstreamer-launch? [19:45] <uwe> wtay: much better than the dutch! [19:45] <wtay> uwe: of course :) [19:46] <uwe> wtay: do you have a german minority in belgium? i was surprised about german on your bank notes [19:46] <wingo> greets wtay [19:46] <wtay> uwe: yeah a couple of people on the german border. german is still an official language in Belgium [19:47] <uwe> interesting. when my girlfriend is coming to rotterdam, we want to go to belgium again. [19:48] <wtay> uwe: you should visit Ghent [19:48] <wtay> uwe: and party in Leuven :) [19:49] <uwe> will think about that. it's between antwerpern and the sea? leuven is east of brussel? [19:49] <wtay> yeah [19:49] <uwe> mmh. sounds good. [19:51] Nick change: wtay -> wtay-tv [19:54] thomasvs (th...@21...) left irc: [x]chat [19:57] uwe (chatzilla@213.174.92.50) left irc: ChatZilla 0.8.4 [Mozilla rv:0.9.4/20011002] [20:30] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [20:32] Nick change: wtay-tv -> wtay [20:45] Nick change: taazzzz -> taaz [21:00] ChiefHighwater (fl...@su...) joined #gstreamer. [21:09] ChiefHighwater (fl...@su...) left irc: Read error to ChiefHighwater[sub18-226.member.dsl-only.net]: Connection reset by peer [21:13] arik (ar...@sd...) joined #gstreamer. [21:13] <arik> hey all [21:14] <wtay> hi [21:14] <arik> how goes it? [21:14] <wtay> fine [21:15] Action: arik is updating gst to the events1 merge [21:16] <arik> heh [21:16] <arik> is there any way to build a pipeline right now where you can change the source without having to NULL out and make a new pipeline? [21:16] <arik> cause it was broken (afaik) the last time i looked [21:17] <ajmitch> hey arik [21:17] <arik> hey aj [21:17] <arik> how goes school and all? [21:17] <ajmitch> arik: getting there, i got exams soon ;) [21:18] <arik> hehe [21:18] <wtay> arik: I doubt it. nobody did changes to that.. [21:18] <arik> ek [21:19] <arik> shame [21:20] <ajmitch> arik: hehe, get over it ;) [21:20] <arik> damnit no! [21:20] <arik> hehe ;-) [21:20] <wtay> or better: fix it :) [21:20] <arik> hehe [21:20] <ajmitch> arik: you don't want to have to make a new pipeline on changing source? [21:21] <arik> no [21:21] <arik> i don't want to have to null out and create a completely new pipeline [21:21] <arik> if autoplugging worked i think this would work [21:21] MM (m.m...@39...) left irc: [21:22] <ajmitch> arik: so you want to be able to stop a pipeline, change an element, and start it again? [21:23] <arik> i want to be able to change source and reconnect an existing pipeline [21:23] <arik> cause you shouldn't have to make a new pipeline everytime you change songs in an mp3 player [21:23] <arik> it's buggy and it's resource intensive [21:23] <arik> more or less (i don't think i'm doing a very good job of explaining this but i know it was or was supposed to be possible) [21:23] <ajmitch> arik: true [21:23] <arik> but you don [21:23] <arik> ' [21:24] <arik> 't get everything you want all at once ;-) [21:24] <arik> at least EOS is working now supposedly [21:24] <arik> although i'll need to find a way to get it without gtk_signal_connect [21:25] <wtay> arik: EOS only works on some plugins [21:25] <arik> heh [21:25] <wtay> fakesrc and fakesink mainly :) [21:25] <arik> haha [21:25] <arik> wonderful [21:26] <arik> hmm [21:26] <arik> doesn't compile [21:26] <arik> gstwinenc.cc: In function `void gst_winenc_chain (GstPad *, GstBuffer [21:26] <arik> *)': [21:26] <arik> gstwinenc.cc:265: `uint_t' undeclared (first use this function) [21:26] <arik> gstwinenc.cc:265: (Each undeclared identifier is reported only once for [21:26] <arik> each function it appears in.) [21:26] <arik> gstwinenc.cc:265: parse error before `=' [21:26] <arik> gstwinenc.cc:286: `size' undeclared (first use this function) [21:26] <wtay> great.. [21:27] <arik> heh [21:27] <arik> any ideas? ;-) [21:27] <wtay> include the header that provides uint_t or use the latest avifile [21:28] <taaz> so i have a problem [21:28] <arik> get the new avifile [21:28] <arik> alright [21:28] <taaz> my dvd code is halting due to what i think is a thread issue [21:28] <arik> i can wait to try your dvd code [21:28] <arik> er cant [21:29] <taaz> i'll check it in i guess.. [21:29] <arik> does it do anything useful? [21:29] <taaz> i dunno why it has problems and mpeg2parse3 doesn't it's effectively the same code [21:29] <taaz> useful? sure, it plays dvds to some extent [21:29] <arik> heh [21:29] <arik> is there a sample program? [21:30] <taaz> yeah [21:30] <arik> neeto [21:32] omega (om...@om...) joined #gstreamer. [21:32] <arik> hey omega [21:32] <omega> yo [21:32] <wtay> hi [21:33] <omega> anyone know where the gstreamer uml diagrams went? [21:33] <arik> wtay: do i need cvs avifile or is the newest release alright? [21:33] <ajmitch> hey omega [21:34] <wtay> arik: cvs is fine I think [21:34] <arik> wtay: alright [21:34] <wtay> omega: gnostream has one [21:35] <taaz> omega: any idea why dvd playing would have thread issues? it gets locked somewhere [21:35] <omega> threads always have issues [21:35] <arik> heh [21:36] <wtay> http://gnostream.sourceforge.net/images/class_large.png [21:36] <taaz> oh? i was going to blame gst code [21:37] <omega> wtay: that diagram is kinda close, but it appears to be the CORBA layer, not gstreamer [21:37] <vishnu> taaz: don't you know that omega reproduced the bugs we are seeing on my SMP machine? [21:38] <vishnu> taaz: don't use threads [21:38] <wtay> omega: yeah [21:38] <omega> I drew one a while ago that tried to match gstreamer, it didn't get all that close, but... [21:39] <wtay> docs/random/caps.dia is another one... [21:39] <omega> ah, there we go, had to find the right extension [21:40] <omega> but nothing besides plugins and caps [21:40] Action: wtay is trying to use the mad frame API [21:40] <omega> still uses mad_stream* though, right? [21:41] <wtay> yes, it sucks [21:41] <wtay> but maybe I can trick it.. [21:41] <omega> if I can figure out this dct64 thing in mpg123, I can get my layer 1 decoder working and get a demo API built [21:42] <steveb> arik: avifile-0.6.0.20011003 works [21:43] <arik> steveb: k [21:43] <wtay> plugins.dia is outdated.. [21:44] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [21:45] <wtay> omega: please do... [21:47] ChiefHighwater (fl...@su...) joined #gstreamer. [21:48] <ajmitch> hey ChiefHighwater [21:48] <ChiefHighwater> Ello ajmitch 8-] [21:53] Zeenix (ze...@13...) joined #gstreamer. [21:56] <omega> um, anyone else having problems with plugin loading and registry operations? [21:56] <omega> nothing works for me all the sudden [21:56] <Zeenix> omega: me [21:56] <arik> hmm [21:56] <arik> i'll know in a second [21:56] <wtay> omega: works fine here [21:56] ChiefHighwater (fl...@su...) left irc: Leaving [21:56] <omega> wtay: total rebuild? [21:57] <omega> I just ran make clean;autogen;make -C gst,tools and -register has a cow [21:57] <omega> loaded 0 plugins, can't create reg directory '' [21:57] <wtay> nice [21:57] <Zeenix> omega: on my sys gst-register crashes on colorspace plugin [21:57] <omega> oh, you're getting further than me [21:58] <omega> /* #undef GST_CONFIG_DIR */ [21:58] <omega> um, someone commented it out: [21:59] <omega> dnl AC_DEFINE_UNQUOTED(GST_CONFIG_DIR,"$GST_CONFIG_DIR") [22:00] <wtay> i have: AC_SUBST(GST_CONFIG_DIR) [22:00] <omega> yeah, that doesn't define it in config.h [22:00] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: Connection reset by peer [22:01] <wtay> omega: oh, It's not im config either... [22:01] <taaz> vishnu: what do you mean 'dont use threads'? isn't that a core idea in gstreamer? [22:01] <vishnu> taaz: dunno. [22:01] <taaz> vishnu: what's a simple test to reproduce the bugs that could be debugged? [22:01] <vishnu> taaz: cothreads work fine. kernel threads are totally broken [22:02] <vishnu> taaz: see the mailing list archives. you can reproduce it with gstreamer-launch [22:02] <omega> only on SMP though, right? [22:02] <taaz> yo, this is a serious bug imho [22:02] evil_monk (trevo@208.141.162.68) joined #gstreamer. [22:02] <vishnu> omega: yah, only on SMP -- but kernel threads are useless except on SMP so ... [22:02] <taaz> randomly not working is bad [22:03] <omega> no, kernel threads are still useful on UP machines when dealing with any I/O [22:03] <vishnu> taaz: it's not that serious. just don't use kernel threads :-) [22:04] <vishnu> omega: yah, but you don't need kernel threads to get non-blocking I/O. the main point of kernel threads is to be able to use the power of more than one CPU [22:04] <omega> vishnu: no, you don't, but using threads is the best way to do it, because otherwise it would require a full pthreads implementation with cothreads [22:04] <omega> and while I like ngpt and would love to use it, it's got some problems that have to be worked through before it can be utilized effectively in gst [22:05] <vishnu> omega: true ... [22:05] <omega> wtay: turns out the BUILDDIR=$top_builddir didn't work, it really does have to be `pwd`. yuck [22:06] <wtay> I don't understand it.. [22:06] <omega> configure.base:66 [22:06] <Zeenix> wtay: you have mail [22:06] <omega> $builddir was getting set to "", that was trickling through other things to produce empty path strings [22:06] <wtay> and gst/Makefile.am:203 [22:07] <omega> 203 is a dnl [22:07] <omega> oh, makefie [22:07] <omega> yeah [22:07] Nick change: taaz -> taaz-bizy [22:07] <wtay> Zeenix: ok [22:07] <arik> i'll have a second phone line in a few minutes [22:08] <omega> anyone see the thing on /. about the SV24? [22:08] <wtay> omega: that's what makes it work for me.. [22:08] <omega> wtay: yes, but unless you specifically set the config_dir on configure's cmdline, it will be blank as long as the builddir=$top_builddir is in configure.base [22:08] <Zeenix> arik: pray for me that i coud get my first one <g> [22:08] <arik> Zeenix: hehe [22:09] <vishnu> omega: of course we can't use kernel threads to get non-blocking I/O on UP because our whole program will fail if we test it on SMP. not a big deal though, i guess. [22:09] <omega> well, I need to get my hands on an SMP machine and try to debug the problems [22:09] <wtay> omega: it compiles with the -DGST_CONFIG_DIR set correctly.. [22:09] <vishnu> omega: yah, eventually. i'm more interested in events working [22:10] <omega> we've been looking at replacing the procs in my mail server, maybe I'll just go buy them so I have a dual proc test platform finally [22:10] <omega> we really need to either fix configure.base or move the plugins out and then fix configure.base [22:10] <omega> autoconf claims it's a complete disaster [22:11] <Zeenix> wtay: are you reading that or later ? [22:11] <wtay> Zeenix: it's ok [22:12] Action: wtay wishes richardb was here.. [22:12] <Zeenix> wtay: so should i now try to start the other( rtprecv ) [22:12] <wtay> Zeenix: yes [22:13] <Zeenix> wtay: why, cause you dont like writting docs [22:13] <Zeenix> ? [22:13] <wtay> Zeenix: he's the autoconf master.. [22:14] <wtay> Zeenix: how many times did you send that mail? I got it 3 times now :) [22:14] <Zeenix> wtay: linuxfreemail uses MailGate which is very new [22:15] <Zeenix> wtay: they say its stiff infant & they are working on it [22:15] <wtay> Zeenix: they are spamming me! <g> [22:15] <Zeenix> wtay: but i dont see anything new in it since my registration [22:15] <Zeenix> with them [22:16] <wtay> omega: ready to do some more events stuff? [22:16] <Zeenix> wtay: i needed to discus that sequencing process of recieved packets [22:17] <wtay> ok... [22:19] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-49.brain.net.pk] [22:19] <omega> wtay: almost, I have to write something first [22:20] <omega> a 'design spec' for gstreamer [22:20] Zeenix (ze...@13...) joined #gstreamer. [22:20] <Zeenix> dissconnections [22:20] asmod (stevec@64.5.222.2) left irc: [x]chat [22:21] Nick change: ShrimpZzZ -> ShrimpX [22:22] steveb (st...@no...) left irc: advance! [22:22] <wtay> omega: I could redo the UML based on gnostream and the other .dia files if you're intrested [22:22] ajmitch (aj...@p4...) left irc: Ping timeout for ajmitch[p49-max3.dun.ihug.co.nz] [22:22] ajmitch (aj...@p1...) joined #gstreamer. [22:22] <omega> if you want to give it a try, sure [22:23] <wtay> ok [22:23] Action: wtay has nothing better to do [22:23] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-16.brain.net.pk] [22:23] <arik> i'll have to get off soon to rewire the outlet [22:23] <ajmitch> rewire? [22:23] Action: ajmitch wonders what arik is up to [22:27] arik (ar...@sd...) got netsplit. [22:27] evil_monk (trevo@208.141.162.68) got netsplit. [22:27] ajmitch (aj...@p1...) got netsplit. [22:27] ShrimpX_ (marius@Photon.pdx.netroedge.com) got netsplit. [22:27] ShrimpX_ (marius@Photon.pdx.netroedge.com) returned to #gstreamer. [22:27] ajmitch (aj...@p1...) returned to #gstreamer. [22:29] evil_monk (trevo@208.141.162.68) returned to #gstreamer. [22:29] <omega> wtay: why is there a cothread_state in GstRealPad? is it copied from the parent element's state? [22:30] <wtay> omega: no idea.. [22:30] <omega> if so, it's not implemented [22:30] <wtay> it doesn't seem to be used either [22:31] <omega> right [22:32] <omega> wtay: does the pipefilter work properly? [22:32] <omega> I'm not having much luck with it [22:32] <wtay> it should [22:33] arik (ar...@sd...) got lost in the net-split. [22:33] <omega> filesrc ! pipefilter command="madplay -o raw:- -" ! osssink doesn't do anything [22:33] <omega> fakesink doesn't show any buffers [22:34] <wtay> oh [22:34] <omega> mpg123 is no different [22:34] <wtay> will look into it.. [22:34] ChiefHighwater (pa...@te...) joined #gstreamer. [22:34] <omega> I need that to work soon [22:34] <omega> cat doesn't fare any better either <g> [22:35] <omega> it should be rewritten as a DECOUPLED element [22:35] <wtay> it is [22:35] <wtay> right [22:35] <omega> it is? I though it was a chain-based poller [22:35] <wtay> you need a thread [22:35] <wtay> omega: not anymore [22:35] <omega> ok [22:35] <omega> cmdline? [22:36] <wtay> filesrc | pipefilter ! { fakesink } [22:36] <omega> ok, that works [22:37] <omega> cool [22:37] <wtay> nice [22:39] <wtay> not sure it kills it's child process when stopping.. [22:39] <wtay> s/it's/its [22:39] <omega> seems to [22:48] Shippou (no...@ot...) joined #gstreamer. [22:55] <omega> wtay: I'm writing about the GstBuffer, it's gonna be hard to list all the features, but I had a thought related to filesrc [22:56] <omega> once we implement buffer CoW handling, the filesrc has to be modified to allow us to specify a read-only (probably by default) bit on the file [22:56] <omega> separate from filesystem permissions [22:56] <omega> because if someone decides to go rewriting data in a buffer that's a subbuffer of the mmap of a readwrite file, it'll change the file.... [22:56] <wtay> yes.. [22:57] <omega> unless either the file is opened readonly, or the filesrc flags the buffers ad readonly [22:57] <omega> safer to open readonly, in case there's a rogue element [22:58] <wtay> yup [23:03] ajmitch (aj...@p1...) left irc: Ping timeout for ajmitch[p19-max5.dun.ihug.co.nz] [23:07] ajmitch (aj...@p2...) joined #gstreamer. [23:09] <wtay> ooh dvdplay [23:22] Zeenix (ze...@13...) joined #gstreamer. [23:22] <Zeenix> hi again [23:23] <omega> 2MB for $103 [23:23] <omega> doh, wrong window [23:23] <Zeenix> wtay: can we continue before i get disconnected again [23:25] <wtay> sure [23:25] <Zeenix> 1. whenever we need to recieve a buffer( in get function in rtprecv.c) we call rtp_recv( in rtp.c ) [23:25] <Zeenix> 2. rtprecv recieves a packet, get its payload( data ) & payload type & give it to us. [23:25] <Zeenix> 3. if we want to add that sequencing code as you suggested, we would like to add it in rtp_recv ? [23:25] <Zeenix> 4. we would have to buffer the packets for that, in that case what would we be giving to the calling func( get function ) ? [23:26] <Zeenix> omega: did robla contacted you again ? [23:26] <omega> nope. email him [23:26] <wtay> Zeenix: what's the problem? [23:27] <Zeenix> no. 4 is a question [23:27] <Zeenix> omega: i am doing that ATM [23:27] <wtay> Zeenix: what do you mean with buffer the packets? [23:28] <Zeenix> wtay: well we would keep it in an array untill its full [23:29] <wtay> Zeenix: yes, or a linked list [23:30] <Zeenix> wtay: still 4. ? [23:31] <wtay> Zeenix: I don't understand the problem [23:33] <Zeenix> wtay: making it simple, what would be the interface of the rtp_recv() ? [23:33] <wtay> Zeenix: it should be rtp_recv [23:34] <wtay> Zeenix: if it does the sequenceing internally that's fine [23:34] <Zeenix> wtay: if it does, then would it be blocking the get function till its get its llist full [23:35] <wtay> Zeenix: yes [23:35] <Zeenix> wtay: & we would be also having a timeout in the rtp_recv [23:36] <Zeenix> ? [23:36] <Zeenix> wtay: so that it doesnt block forever [23:36] <wtay> yeah [23:37] <Zeenix> wtay: then it would definetelly be return an array or list of buffers [23:37] <Zeenix> recieved [23:37] <wtay> Zeenix: no, just one buffer at a time [23:39] <Zeenix> wtay: then i'll just forget about the sequencing ATM & implement it later [23:39] <wtay> Zeenix: uhm. maybe just try with a simple case first [23:40] vekky (wb...@pe...) joined #gstreamer. [23:40] <vekky> hey yo yo [23:40] <vekky> sup y'all [23:40] <vekky> yea [23:40] Last message repeated 1 time(s). [23:40] <vekky> hey [23:40] <vekky> so [23:40] <vekky> come on [23:40] <vekky> someone here has to be on the Xpert or v4l lists. [23:40] <vekky> come flame me, fukaz! [23:41] <Zeenix> wtay: RTP is only implemented by librtp( gphone ) ? [23:41] <wtay> yo bro [23:41] <wtay> Zeenix: I think so.. maybe our rasta bro vekky has one too.. [23:41] <vekky> RTP is soo easy to implement. [23:41] <vekky> like, 5 minutes [23:41] <vekky> took me like a half an hour cause i'm lame. [23:41] <Zeenix> vekky: sequencing too [23:41] <Zeenix> ? [23:41] <vekky> what do you mean [23:41] <vekky> ? [23:41] <vekky> you're sending audio, right? [23:42] <vekky> or video, even better. [23:42] <vekky> regardless, your input queue is going to be super super small. [23:42] <vekky> you get a packet, you're going to want to play it immediately, if not sooner. [23:42] <vekky> so, it's not like you need to worry too much about sequencing. [23:42] <vekky> drop something if it's ancient, i guess. [23:42] <wtay> it's a user setting.. [23:43] <Zeenix> vektor: that would make rtp into udp [23:43] <vekky> Zeenix: Huh? [23:43] <vekky> Yeah, I put rtp into a UDP packet and send it. ;-) [23:43] <vekky> I have no idea what you mean. [23:44] <Zeenix> vekky: yes but you use rtp to make udp a bit better [23:44] <vekky> If you do anything with an rtp packet besides decode it and put it right into an output buffer, you're fucked. [23:44] <vekky> No, you use RTP to interoperate. [23:44] <vekky> Basically. [23:44] <vekky> Oh and some sanity checking. [23:44] <vekky> Like an SSRC. [23:44] <vekky> but like, whatever. [23:44] <vekky> If you want to talk to others, use RTP. [23:45] chillywilly (da...@d1...) joined #gstreamer. [23:45] <Zeenix> wtay: any other implementation you know abou ? <g> [23:46] <vekky> wtay: ??? You're completely ignoring mine? [23:46] <vekky> erm [23:46] <vekky> s/wtay/Zeenix/ [23:46] <vekky> jeez dude you didn't even ask for my code or anything ;-) [23:46] <vekky> There's the vovida rtp 'stack'. But they're kinda on crack. [23:47] <Zeenix> vekky: no, i am not ignoring you [23:47] <vekky> Zeenix: oh. [23:48] <vekky> so you want my code or what? [23:48] <Zeenix> vekky: i needed a bit popular implementation gst could base its plugin on [23:48] <vekky> so, there's librtp, there's hennings rtp code (is that librtp?), there's the vovida rtp stack, there's my code.... [23:48] <vekky> Um, again, dude, RTP is really trivial to implement, so I don't think you need to be too worried about using some standard lib. [23:49] <Zeenix> vektor: definetelly, if you could provide your code that let me learn how to sequence the recieved packets [23:49] <vekky> ? [23:49] <vekky> You're having problems with sequencing received packets [23:49] <vekky> ? [23:49] <vekky> Again, I don't think that makes sense. [23:49] <vekky> Who's sending you RTP? [23:49] <vekky> This some cisco box or something? [23:50] <Zeenix> vekky: ? Sending me RTP ? what do you mean [23:50] <vekky> Well, you clearly want to implement RTP, and I'm curious about who you're trying to talk to. [23:51] <vekky> I'm assuming you want to make phone calls. [23:51] <vekky> Or ... Hmm. [23:51] <vekky> You can use RTP for video too. [23:51] <omega> vekky: there's a lot more to rtp than just low-latency stuff [23:51] <vekky> Uh... No, it's for low-latency stuff :) [23:51] <vekky> There's no retransmissions :) [23:51] <vekky> So, yeah. [23:51] <omega> yes, it's designed for low latency, but people use it for other stuff [23:52] <omega> Zeenix has to write it to interop with that, so he must have basic reordering in place, if the user decides to turn it on [23:52] <vekky> omega: Like what, sorry? [23:52] <Zeenix> vekky: no i am trying to develope gstrtprecv element ATM [23:52] <omega> like any streamed download, such as Real stuff [23:52] <vekky> Zeenix: Sure, for what? [23:52] <vekky> omega: Which is low latency. [23:52] <omega> howso? [23:52] <omega> only two-way connections are latency-critical [23:52] <vekky> Well, you will... [truncated message content] |
From: IRC B. <wt...@us...> - 2001-10-20 04:27:26
|
******************************************************************* [03:10] Nick change: tnt-away -> tnt [03:39] <vektor> Real Analysis is cool. [03:40] <omega> ? [03:41] <vektor> This cool course I'm attending but not taking. The field is very interesting. [03:41] <vektor> I'm looking at the assignment that's due tomorrow, and being very glad I'm not enrolled. :) [03:48] <omega> bleagh, now people are talking about 'extended' v4l modules. great [03:48] <omega> vektor: I would recommend dropping this root perms thing, because people will only get hung up on it. it's one way to solve it, but there are others, and they're keeping themselves from thinking about those others atm [03:50] <vektor> Well, I require root perms for high accuracy. [03:51] <vektor> There's no other way under Linux. [03:51] <omega> what specifically requires root? [03:51] <vektor> Being SCHED_FIFO ? [03:51] <vektor> Setting /dev/rtc to interrupt at 1024hz? [03:51] <omega> other ways of dealing with that [03:51] <vektor> Like? [03:52] <omega> sorry, other ways of dealing with SCHED_FIFO, like kernel-level permissions systems, user-space launchers, etc. [03:52] <vektor> renice my process and have a fast interrupt? It's still going to chunk if, say, the MPEG2 decoding is slow. [03:52] <omega> what does /dev/rtc give you, and could it be provided other ways? [03:53] <vektor> The /dev/rtc device alows me to ask the kernel to set the x86 timer interrupt to fire (up to) every 8192hz, and sets my process in the wait queue for that event. [03:53] <vektor> This way, since I'm SCHED_FIFO, I'll have an interrupt which will force me to run immediately after it fires. [03:53] <vektor> That way, I can preempt everything else on the system every millisecond. [03:54] <omega> why do you need to do that? isn't vtrace what you want? [03:54] <vektor> If there was a vtrace device with an associated interrupt that would be ideal. [03:54] <omega> exactly. there are other solutions [03:54] <vektor> But I still need the /dev/rtc clock if I'm doing my MIDI sequencer. [03:54] <vektor> Since I can't schedule MIDI events in the kernel. [03:54] <omega> my main point though is that they seem to be stuck in a rathole, so you have to kick them out of it by dropping the subject atm [03:55] <vektor> Well, there's a problem with 'dropping it'. [03:55] <omega> why [03:55] <omega> ? [03:55] <omega> you can bring it up again later [03:55] <taaz> argh! so i just read brandens response to the "why not versioned libs". what is wrong with the libgal thing? that's the whole point of versioning isn't it? or am i missing the point of why we have versioning in the first place? how can they possibly hold back an api -forever- until they decide it's solid as stone. that's a crappy way to develop code imho [03:56] <omega> taaz: libgal1, libgal2, libgal3, libgal4, etc. [03:56] <vektor> Say there was a /dev/vtrace device, what's the point of getting the interrupt if I'm not woken up until 10ms later? [03:56] <taaz> omega: yeah! exactly the way it should be! [03:56] <omega> taaz: HUH??? [03:56] <taaz> how else would you do it? [03:56] <omega> libgal.so.0.1.0, libgal.so.0.2.0, etc. [03:56] <omega> THAT is .so versioning [03:56] <omega> libgal1 and libgal2 are separate packages [03:57] <vektor> While an interface is under development, you either need to be [03:57] <vektor> responsible about bumping the .so version (else you get binaries that [03:57] <vektor> won't load), or you need to ship a static library. [03:57] <vektor> My line, we're advocating responsibility over being lax. [03:57] <omega> gee, I vote for responsiblity... [03:57] <vektor> omega: Then please respond to his post and say that. [03:58] <taaz> omega: you do understand the libgal{1,2,3,4...} thing right? [03:58] <vektor> taaz: Is there something not to get? [03:58] <omega> no, I don't understand why that's necessary [03:59] <taaz> you can only have 1 version of a package installed at a time [03:59] <omega> ok... [03:59] <taaz> so if you want to install two versions of a so you need two differently named pacakges [03:59] <taaz> it's a wart in the system that is hard to fix apparently [04:00] <taaz> would have to have some flag to allow multiple installations of the same pacakge name... debian just not setup to do this yet [04:00] <omega> this indicates to me that libgal development model is screwed up then if they're revving it that fast and causing such nasty dependency problems [04:00] <omega> there are other ways to do that... [04:00] <taaz> so policy says pacakges are named libfooN where N increses for each incompatible API change [04:00] <omega> right [04:02] <taaz> this of course sucks ass for liba52 and libmpeg{1,2,3} because they in theory are liba v52 and libmpeg v1, v2, and v3. yuck. it's kind of a mess in that case. no one has proposed a good fix to policy of naming though [04:02] <vektor> taaz: That's besides the point. :) [04:04] <taaz> not really... branden is upset about naming issues and api revs and how debian deals with it. i see absolutely no reason not to do the same with Xv et al. just rev the package version extension as needed. yeah it sucks ass. [04:04] <vektor> Ok, I'm going to have to form a reply on this 'root' issue. [04:07] <taaz> no one's complained too loud about libc6... that's using the same naming scheme... [04:07] <omega> that one I really did have to be installed in parallel with libc5 afaik [04:07] <omega> still does for some poor souls stuck with old binary software [04:14] wingo_ (wi...@rd...) joined #gstreamer. [04:16] wingo (wi...@rd...) left irc: BitchX-1.0c18 -- just do it. [04:16] Nick change: wingo_ -> wingo [04:16] <wingo> yo. [04:16] <omega> yo [04:18] <wingo> i've got a free evening, so hopefully alsa will reach a more or less stable state tonight [04:19] <omega> cool [04:20] <wingo> what do you think of the recent thread on gst-devel? [04:20] <omega> need to read through it more carefully [04:21] <wingo> should two successive calls to bytestream_peek with the same len return the same data? [04:21] <omega> yes [04:21] <wingo> ok [04:21] <omega> assuming no _flush in the middle, of course [04:22] <wingo> right [04:29] vishnu (jo...@cx...) joined #gstreamer. [04:29] <wingo> hey vishnu [04:33] <vishnu> wingo: yo! [04:34] Action: vishnu gives wingo a high-five [04:34] <vishnu> hey, yah, cool :-) [04:35] Action: omega watches vishnu's hand smash into wingo's face as wingo turns around and says 'huh?' just in time....<g> [04:35] <omega> ouch, that's gotta hurt... [04:35] <vishnu> i can't wait to meet you in person, omega <g> [04:35] <omega> heh [04:35] <omega> aka "damn lag!" [04:35] <wingo> it did, especially considering that i had mouth surgery a week ago :) [04:35] <omega> oops [04:36] <wingo> heh [04:36] <wingo> oh yeah. "huh?" :) [04:37] <wingo> omega: the hammerfall will do 3 periods [04:37] <omega> not 2? [04:37] <wingo> apparently, that's a common case [04:37] <omega> ok [04:37] <wingo> yeah 2 as well [04:37] <wingo> i had to set a default for complicated caps reasons [04:37] <omega> heh [04:37] <wingo> so 3 is the default [04:38] <omega> ok [04:38] <tnt> Does anyone know what library the function "g2g_type_register_static" is from (... I'm guess it is part the the stuff that let's you code with GLIB2.0 but still use GLIB1.x). [04:38] <omega> from gobject2gtk.c [04:38] <tnt> OK. Thanks. [04:38] <omega> g2g_ should never be called directly, but used via the #define's in gobject2gtk.h [04:39] <tnt> I saw it come up when I was linking. [04:39] <omega> ok [04:39] <wingo> my machine skips on mp3 playback when i change x desktops if i don't have 4096 frame periods [04:39] <omega> heh, w w/o LL patches? [04:39] <wingo> w/o [04:39] <omega> ok [04:39] <wingo> actually, sysctl [04:39] <omega> ? [04:40] <wingo> so maybe if i echo 1 > /proc/sys/kernel/lowlatency... [04:40] <wingo> some path like that [04:40] <omega> oh [04:40] <omega> odd that ll patches can be turned off... [04:40] <wingo> it's an option [04:40] <omega> I would expect they'd be rather core reworks of some things [04:41] <wingo> apm and ll sometimes act up, but i'm on a desktop so there's really no reason [04:41] <omega> so what kernel you running otherwise? [04:41] <wingo> atm 2.4. [04:41] <wingo> 7 [04:41] <wingo> so a little old, vm-wise i guess [04:42] <omega> ok, I'm debating whether to use 2.4.12 or 2.4.12-ac3 [04:42] <omega> w/ll [04:42] <wingo> ooh, difficult [04:42] <wingo> 2.2.12 won't compile iirc though [04:42] <omega> I've had problems with 2.4.12, yes <g> [04:42] <wingo> i'd go ac3 for stability [04:42] <wingo> oops :) [04:43] <omega> where does one get the ll patches? [04:43] <wingo> search for "andrew morton low latency" [04:43] <wingo> i can't recall atm [04:44] <wingo> there's a link maybe on ambient.2y.net/wingo/projects/doober/ [04:44] <wingo> lowlatency stuff doesn't help at all for non-root, non-sched_fifo people, does it? [04:44] <omega> dunno [04:44] <wingo> i think that's the deal [04:45] <omega> it should help for everything afaik, because it cuts down on the long kernel paths [04:45] <omega> you may not be able to utilize it without scheduler help though [04:45] <wingo> that's the thing, the scheduler won't help unless you have sched_fifo [04:45] <wingo> afaik [04:45] <wingo> preemptive patch, now that's another thing, i think [04:46] <omega> right, but you *will* see better worst case no matter what [04:46] <omega> yes, they both help [04:46] <omega> I should grab that too [04:46] <wingo> you might not have to be root to use pe [04:46] Action: wingo switches back to the emacs/rxvt window :) [04:47] <wingo> actually, no. so at eos filesrc ! mad 1 alsasink segfaults now. an improvement, i suppose... [04:47] <omega> heh [04:47] <wingo> where should i look for eos handling stuff? [04:47] <omega> um, in the plugins now for the most part [04:47] <wingo> actually, i use bytestream [04:47] <wingo> so [04:47] <omega> oh [04:47] <wingo> i'll have to wait [04:47] <omega> hmm, we haven't worked that out yet [04:48] <omega> I can't get the ll patch [04:48] <wingo> vishnu, where's your redone patch [04:48] <wingo> :) [04:48] <wingo> not too important for me though [04:48] <vishnu> wingo: did someone ask me to redo the patch?? [04:48] <omega> does it handle events? [04:49] <vishnu> omega: yah, i added events to bytestream but wtay reverted it and is slowly adding back the same changes [04:49] <omega> where's the events patch? I don't remember seeing one [04:49] <vishnu> omega: admittedly, it wasn't split out very well [04:50] <wingo> so... alsasrc should set a default rate and format, no? [04:50] <wingo> 44.1/S16? [04:50] <omega> probably, yes [04:52] <wingo> would it be accepted if events were split out? [04:53] <wingo> well, one would have to look of course, but this is apriority [04:53] <omega> define split out? [04:53] <wingo> of vishnu's big big patch [04:53] <omega> well, I want to review it, but yeah, it needs to be dealt with [04:53] <omega> I need to catch up with where bytestream is now, first.... [04:54] Action: taaz notes lack of timestamps again and prepares to get hit on the head for asking again [04:54] Action: vishnu gives taaz a pillow [04:54] <omega> timestamps in bytstream? there's not much to do, it should be relatively easy [04:54] <omega> depends on whether you're using the buffer or byte interface [04:54] Action: taaz falls asleep with his nice new pillow [04:54] <omega> it's already done for the buffer interface [04:55] <taaz> it is? [04:55] <omega> yeah, the buffer returned is given the timestamp of the first represented incoming buffer [04:55] <omega> at least last I looked at it [04:57] <taaz> yeah well this requires new api call afaikt for byte case [04:57] <vishnu> wingo: maybe i'll submit more patches over the weekend [04:57] <taaz> i guess [04:57] <taaz> but that kinda sucks too [04:57] <vishnu> taaz: why do you need the timestamp? [04:57] <omega> I should now have much more time to keep up with things now [04:57] <taaz> vishnu: to pass down the line [04:57] <vishnu> taaz: why? [04:58] <vishnu> taaz: for your dvd player? [04:58] <taaz> vishnu: like a52dec, you want timestamps on output buffers rather than all 0s [04:58] <vishnu> taaz: why does a52dec care? [04:58] <taaz> uh.. it doesnt [04:58] <taaz> it just needs to pass it on [04:58] <taaz> how else will timestamps be preserved? [04:58] <omega> taaz: in the case of a52dec, it almost certainly has to *create* the timestamp [04:59] <omega> filesrc sets timestamp to 0 [04:59] <vishnu> taaz: i'm dubious. isn't there timing information in the mpeg stream? [05:00] <vishnu> taaz: i mean, if you're going to do any work on timing at all then forget about the timestamp. study the mpeg system spec instead [05:00] <omega> well, there is some point, because the mpeg2 parser will set timestamps on the source a52 that are relevant, in many cases [05:01] <omega> there then has to be some cooperation between mpeg2parse, a52dec, and gsta52dec to make sure the timestamps for individual frames are correct [05:01] <vektor> Hello. [05:01] <taaz> what? mpeg2parse -> a52dec -> sink, of course you need to pass the timestamp from mpeg2parse through to sink [05:01] <omega> this requires an API for a52dec that exposes frame-level operations and such [05:02] <vishnu> hrm. ok, maybe [05:02] <taaz> i don't think a52 has timestamps? [05:02] <omega> not itself, no [05:02] <omega> but the PES packets do [05:02] <wingo> what up vektor [05:03] <omega> vektor: you see that .swf I posted? [05:03] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [05:04] BeNOW (an...@ds...) left irc: BitchX-1.0c18 -- just do it. [05:05] <vektor> omega: I haven't looked at it yet. [05:06] <omega> do so <g> [05:06] <omega> it might bring back memories [05:06] <vektor> wingo: I'm postin' up a storm man! Know any mailing lists that have been dull lately? Lemme'at'em! [05:08] <vektor> omega: Thanks! [05:09] <omega> we'll have to make sure drunknap gets that one too [05:09] <taaz> vektor: yes i know one: libmpeg2-devel. start posting a todo list. walken may be more likely to comment on your posts since he apparently cant find the time to post himself [05:10] <wingo> vektor: how about any comments on the plugin build thing, on gst-devel? :) [05:11] <vishnu> wingo: i like the cml2 idea :-) [05:12] <vektor> taaz: What's the post address? [05:12] <vektor> wingo: Huh? [05:12] <vektor> wingo: I'll check that out next ;-) [05:13] <omega> taaz: btw, dvdplay faults after a second on my machine [05:13] <omega> in mpeg2_decode_data [05:14] <taaz> vektor: what? libmpeg2-devel? [05:14] <omega> also, dvdsrc or libdvdread is spewing *huge* amounts of debug [05:14] <taaz> omega: uh.. really? [05:14] <omega> yeah [05:14] <vektor> taaz: Yes, lib...@li... ? [05:14] <wingo> vektor: cml2 does sound really neat [05:14] <taaz> omega: yeah, i forgot i left that debug thing on ;) it's like ifoPrint or something [05:15] <taaz> vektor: wake up, libmpeg2.sf.net [05:15] <vektor> Oh fuck, new mailing list! [05:15] <vektor> Wow, I totally forgot! [05:16] <wingo> OH MY GOD!!! [05:16] <wingo> :) [05:19] <taaz> vektor: you could ask about Xv getting a stride param in its api [05:22] <vektor> taaz: I'm about to. [05:22] <vektor> taaz: As it turns out though, the API kind of supports it. [05:22] <vektor> taaz: I was a bit on crack. [05:25] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [05:27] <wingo> who is evil_monk, and why does he never say anything? [05:34] <wingo> ok, i will sleep for real tonight [05:35] <omega> pfff [05:35] <wingo> 4 hours last night :) [05:35] <omega> you want me to /ban you for the next 9 hours? <g> [05:35] <wingo> lol [05:36] <ajmitch> heh [05:39] <wingo> ajmitch: what time is it there? [05:40] <ajmitch> Fri Oct 19 16:40:17 NZDT 2001 [05:40] <omega> pff, they've even got their own timezone... [05:41] <ajmitch> of course we have our own TZ ;) [05:41] <ajmitch> UTC +1300 [05:41] <wingo> wierdos ;-) [05:43] <ajmitch> hehe [05:43] <ajmitch> Need to get 35.2MB/184MB of archives. After unpacking 52.0MB will be used. [05:43] Action: ajmitch guesses another 3-4 hours.... ;) [05:44] <vishnu> i love debian :-) [05:45] <ShrimpX> unless you prefer installing stuff from source [05:52] Action: omega always figured there was a slackware junky hiding in any decent-sized linux crowd.... [05:53] <vishnu> i tried to install slackware once, didn't succeed [05:53] <omega> I reinstalled^Wupgraded slackware a couple times before moving to redhat [05:58] <wingo> i went redhat -> slackware -> debian [05:59] <wingo> and with debian i stay :) [05:59] Action: ajmitch went mandrake -> debian [05:59] <omega> wingo: you're not still waiting for LinuxOne's distro <g> [06:01] <omega> wow, 2.4.12 has a kernel option right out in the main system config to handle multi-quad NUMA-Q machines [06:02] <omega> even tells you to go get a new lynxer firmware file [06:02] <vishnu> heh, christmas is coming up too :-) [06:02] <omega> flashing the lynxer is, um, entertaining [06:02] <omega> but those machines are sweet [06:03] <omega> too bad I never got to play with the PIII Xeon ones [06:03] <omega> but http://omegacs.net/~omega/STiNG/ has some "pics" and description of the stuff I worked with there [06:05] <wingo> in a 8 level or so back&forth between me and pbd on alsa-devel, i think alsa will work now [06:05] <vishnu> whoa. nice pics [06:05] <omega> heh [06:06] <wingo> it's so nice to have folks take their time to help [06:06] Shippou (no...@ot...) joined #gstreamer. [06:06] <omega> vishnu: without trying very hard at all, I could pull 330MB/sec off those disks [06:09] <vishnu> omega: it's not really fun unless i can do it myself. do you feel like santa claus this year? [06:09] <omega> yeah, but that was sequent hardware, I haven't worked there for ~3.5yrs [06:10] <vishnu> ok, we gotta make you rich on gstreamer stock options first .. no problem [06:10] <omega> pff [06:10] <omega> each quad was ~$100,000 at that time [06:12] <ShrimpX> omega: what did you do there? [06:12] <omega> testing of the volume manager [06:13] <ShrimpX> cool [06:13] <omega> mostly the clustered volume manager [06:13] <ShrimpX> weren't you like 18 or something? [06:13] <omega> I started there when I was 17y2mo or so [06:13] <ShrimpX> impressive [06:13] Action: ShrimpX bows [06:14] <omega> the contracting company figured that out and said "you're too young" [06:14] <omega> so I ended up being a full employee instead <g> [06:14] <ShrimpX> hehe [06:14] <omega> but from those pics you can see how anal I am about clean cabling, where possible [06:15] <ShrimpX> yeah [06:17] <omega> hrm, I don't think I ever posted the OLS pics <g> [06:17] <vishnu> omega: was the pizza good too? [06:17] <omega> vishnu: where? [06:17] <vishnu> omega: at sequent? [06:17] <omega> what pizza? [06:17] <vishnu> hehe [06:17] <ShrimpX> hum [06:18] Action: ShrimpX just had pizza [06:18] Action: ShrimpX grabs another slice [06:18] <vishnu> i mean, what could be better than 330MB/s bandwidth and a fresh hot pizza? [06:19] <omega> earplugs, for when in the lab [06:20] <vishnu> yah, of course earplugs. huge disk bandwidth, fresh hot pizza, and earplugs. it's heaven [06:21] <ShrimpX> hmmmmx0r... [06:22] <omega> 2.4.12 is broken for non-SMP, it won't compile UP [06:23] <omega> oh well, I'll compile SMP for now ;-( [06:23] <vishnu> heh, when is 2.5.1 coming out? [06:23] <omega> depend on how soon we can get linus and alan into the ring for a few rounds [06:23] <ShrimpX> hehe [06:23] Action: ShrimpX always compiles ac kernels [06:24] <omega> this is ac [06:24] <omega> 12-ac3 [06:24] <omega> afaik, dblchecking [06:24] <omega> hrm, nope [06:24] <omega> patching [06:24] <ShrimpX> heh [06:24] <ShrimpX> Linux area51 2.4.9-ac9 #1 Wed Sep 5 19:11:14 PDT 2001 i686 unknown [06:25] Action: ShrimpX needs to upgrade [06:32] <omega> um, where's reiserfs go?? [06:33] <omega> oh, somehow I managed to miss Experimental [06:35] <wingo> ok, going to sleep. at least now i know what i need to do with alsa :) [06:35] <wingo> (for real tonight ;) [06:36] Nick change: wingo -> wingo-zzzzz [06:36] <omega> heh [06:36] Action: omega wonders if wingo will talk in his sleep [06:37] <ShrimpX> omega: what text editor do you use? [06:38] <omega> pico <g> [06:39] <ShrimpX> hehe cool [06:39] <ShrimpX> or nano [06:39] <vishnu> does pico run on unleaded or diesel fuel? [06:39] <omega> um, not sure [06:40] <omega> came from UW, so I would tend to think diesel [06:40] <ShrimpX> pico is wind-powered, man. [06:40] <omega> heh [06:40] <vishnu> oh, yah, right [06:42] <ShrimpX> what's pico licensed under? [06:43] <omega> duno [06:43] <omega> not gpl [06:43] <omega> UW has had their own license for ages [06:43] <vishnu> pico is part of pine, right? [06:43] <omega> yup [06:43] <ShrimpX> they use nano in my cs dept (and no pine), because of "licensing issues" [06:44] <ShrimpX> nano is cooler, but breaks sometimes [06:44] <omega> it has some quirks I don't like [06:44] <ShrimpX> it messes up if you edit large files. [06:48] <ShrimpX> heh, my computer has a little bit of every debian release known to man, some rpm's, and a bunch of stuff built from source... [06:50] <ShrimpX> omega: what other unixes is gst getting ported to? [06:51] <omega> Solaris and Irix have been attempted, dunno the status [06:51] <ShrimpX> any plans for freebsd? [06:51] <omega> that's been tested too, there are problems unless you use linuxthreads [06:51] <ShrimpX> oh [06:58] <tnt> For making you own library, is the info on http://gstreamer.net/wiki/index.php?LibraryWriters still accurate? [07:00] <tnt> [I guess all the gstreamer gurus are hiding :-) ] [07:00] <omega> yes, though that's for libraries that are specifically in gst's libs/ directory, which I'm not so sure I like [07:01] <tnt> Well, I want to make a new Element... where should I put it? [07:01] <omega> what does it do? [07:01] <tnt> It takes two video streams and mixes them (... does a fade from one to the other). [07:02] <omega> that goes somewhere in the plugins/ directory [07:02] <tnt> OK. [07:21] ajmitch_ (aj...@p2...) joined #gstreamer. [07:21] ajmitch (aj...@p5...) left irc: Ping timeout for ajmitch[p59-max11.dun.ihug.co.nz] [07:22] Nick change: ajmitch_ -> ajmitch [07:52] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [07:52] ShrimpX (marius@131.252.244.168) left irc: Hey! Where'd my controlling terminal go? [08:05] Vakor (ms...@co...) joined #gstreamer. [08:27] steveb (st...@no...) joined #gstreamer. [09:01] <tnt> [Hope someone is still listening] Does anyone know what the proper place to use "gst_element_set_loop_function"? Should it be called in the "gst_???_get_type" funtion, "gst_???_class_init", or somewhere else? [09:04] <tnt> I'm gonna guess gst_videomix_init. [09:20] taaz (dlehn@66.37.66.32) left irc: Ping timeout for taaz[66.37.66.32] [09:26] taaz (dlehn@66.37.66.32) joined #gstreamer. [09:54] ShrimpX (marius@131.252.244.168) joined #gstreamer. [09:54] ajmitch (aj...@p2...) left irc: Ping timeout for ajmitch[p22-max7.dun.ihug.co.nz] [09:58] ajmitch (aj...@p2...) joined #gstreamer. [10:03] Nick change: taaz -> taazzzz [10:47] tnt (tn...@h2...) left irc: Client Exiting [10:47] omega (om...@om...) left irc: sleep [11:12] MM (m.m...@39...) joined #gstreamer. [11:19] MM (m.m...@39...) left irc: Ping timeout for MM[39dyn177.com21.casema.net] [11:35] MM (m.m...@39...) joined #gstreamer. [12:17] MM (m.m...@39...) left irc: [12:28] Nick change: ShrimpX -> ShrimpZzZ [12:29] mwc (ma...@ly...) left irc: [x]chat [12:31] Vakor (ms...@co...) left irc: Ping timeout for Vakor[co3016667-a.lowrp1.vic.optushome.com.au] [13:01] Vakor (ms...@co...) joined #gstreamer. [13:29] Uraeus (csc...@c1...) joined #gstreamer. [13:29] <Uraeus> hi [13:32] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [14:30] asmod (stevec@64.5.222.2) joined #gstreamer. [14:33] Shippou (no...@cr...) joined #gstreamer. [15:02] harobed (harobed@ACBA7865.ipt.aol.com) joined #gstreamer. [15:04] <harobed> gstreamer can do coding a video or it can only decode ? [15:05] <harobed> because I search some codec for code and decode some video [15:05] <Uraeus> harobed: it can encode too yes [15:06] <harobed> Uraeus: do you know image magic ? i would code a same program for the video [15:06] <Uraeus> harobed: sound great, well I think gstreamer would be a great library for you to do that with then [15:08] <harobed> exemple : you codec can extracting one part of video into one other format ? [15:09] <harobed> s/you/your [15:09] <Uraeus> harobed: that is possible to do today, and actually if you are interested in a non-linear video editor then wtay-zZz is working on such called gnonlin based on gstreamer [15:09] <harobed> url ? [15:10] <Uraeus> harobed: http://sourceforge.net/projects/gnonlin/ [15:11] <Uraeus> harobed: it is still early stages, but wtay-zZz is a very good hacker so the code should be clean and nice [15:12] <harobed> thank you [15:30] harobed (harobed@ACBA7865.ipt.aol.com) left irc: Read error to harobed[ACBA7865.ipt.aol.com]: Connection reset by peer [15:45] dobey (do...@dr...) joined #gstreamer. [16:07] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:48] wingo-zzzzz (wi...@rd...) left irc: Read error to wingo-zzzzz[rdu25-21-076.nc.rr.com]: EOF from client [17:10] wingo (wi...@rd...) joined #gstreamer. [17:10] <wingo> 'morning. [17:14] wingo (wi...@rd...) left irc: My damn controlling terminal disappeared! [17:27] Vakor (ms...@co...) left irc: Ping timeout for Vakor[co3016667-a.lowrp1.vic.optushome.com.au] [17:29] Nick change: taazzzz -> taaz [17:49] wingo (wi...@rd...) joined #gstreamer. [18:20] <steveb> Uraeus: RMS is a knob [18:23] Action: evil_monk is away: I'm busy [18:27] vishnu (jo...@cx...) joined #gstreamer. [18:30] <wingo> yo [18:31] <dobey> Uraeus: hey [18:41] ChiefHighwater (pa...@te...) joined #gstreamer. [18:43] <dobey> hmm [18:43] Action: dobey waits for yet another email ;-) [18:44] <ChiefHighwater> dobey:who from? [18:45] <dobey> Uraeus: [19:01] dobey (do...@dr...) left irc: Ping timeout for dobey[dreadnought.ximian.com] [19:43] evil_monk (trevo@208.141.162.68) got netsplit. [19:49] evil_monk (trevo@208.141.162.68) got lost in the net-split. [19:53] tnt (tn...@h2...) joined #gstreamer. [19:59] vekky (wb...@pe...) joined #gstreamer. [19:59] ShrimpX_ (marius@Photon.pdx.netroedge.com) left irc: rebooting [19:59] <vekky> Hello all. [20:01] <tnt> Hey [20:18] <wingo> hi [20:29] <tnt> If a (new) plugin has a 'loop function' (and is loop based), then it shouldn't have a 'chain function'... correct? [20:31] thomasvs (th...@21...) joined #gstreamer. [20:31] <thomasvs> yow [20:37] vekky (wb...@pe...) left irc: too bad the scene is dead [20:47] arik (ar...@sd...) joined #gstreamer. [20:50] wingo (wi...@rd...) left irc: Ping timeout for wingo[rdu25-21-076.nc.rr.com] [20:50] arik (ar...@sd...) left #gstreamer. [20:53] wingo (wi...@rd...) joined #gstreamer. [20:54] evil_monk (trevo@208.141.162.68) joined #gstreamer. [21:02] <tnt> Have you guys seen this: http://www.theregister.co.uk/content/4/22354.html ... anyone want to write a source for stuff protected by MS DRM version 2... like .wma files. [21:07] <vishnu> tnt: neat :-) [21:09] <tnt> I'm guessing this is against the DMCA... so a coder living in the US couldn't write it, but are all the coders here burdened by a version of that?... any lawyers here?... is there a similar law to the DMCA in Canada, Sweden, etc etc? [21:10] <tnt> [I especially wonder if Canada has something like that.] [21:10] <vishnu> tnt: international copywrite law is very murky to begin with [21:11] <tnt> This isn't exactly a copyright issue though... it's an issue if you can write a decypter for wma's. [21:12] <vishnu> tnt: some countries have laws but don't enforce them. [21:12] <tnt> Actually doesn't gstreamer have DeCSS in it? So wouldn't that mean someone here could write that kind of stuff. [21:12] <vishnu> tnt: no, all mention of decss was removed recently. we use libdvdread [21:13] <tnt> How does libdvdread handle it?... is it hardware based? [21:14] <vishnu> tnt: libdvdread searches the linker path for another library, libdvdcss, and links to it if available. otherwise it only reads unprotected dvds [21:14] <tnt> OK [21:15] <vishnu> tnt: it's not illegal until you actually use libdvdcss and the binding is done as late as possible [21:16] <tnt> Is libdvdcss illegal though? [21:16] <vishnu> tnt: dunno, maybe [21:28] omega (om...@om...) joined #gstreamer. [21:30] <tnt> If a (new) plugin has a 'loop function' (and is loop based), then it shouldn't have a 'chain function'... correct? [21:30] <omega> right [21:30] <tnt> OK [21:30] <omega> it's possible to have both, but it's tricky, and no one's needed to yet [21:31] <tnt> Like identity?... that's what I'm basing my plugin on, and it came with both. [21:31] <omega> identity is designed to do either/or [21:32] <omega> for testing purposes only [21:32] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [21:37] <ajmitch> hi all [21:48] <wingo> yo [21:49] <wingo> tnt what are you making? [21:49] <tnt> I working on something for matterial <http://matterial.sf.net>... basically an NLE. [21:50] <wingo> a fader? [21:51] <tnt> Oh... I need to make a bunch of video transitions... like those found in common NLE programs (like FinalCutPro, Premiere, etc)... a fader is the first one I'm making. [21:52] <wingo> cool [21:52] <omega> tnt: there are several people interested in NLE stuff, there's a base library called gnonlin[.sf.net] that's been started [21:53] <omega> I'd like to try to make sure that all the NLE projects share as much common code as possible, which is the GStreamer 'way' <g> [21:53] <tnt> I'll go take a look at it. [21:53] <omega> the plugins are a huge part, of course, but there are other parts that can/should be shared too [21:53] Nick change: wtay-zZz -> wtay [21:53] <wtay> yeah [21:54] <tnt> [No webpage for the project.. I'll check the project page/] [21:54] <omega> kinda like the player stuff, which I'm getting stuck in atm... [21:54] <omega> fltk doesn't seem to like [co]threads, even though it's not being used threaded at all [21:57] <tnt> Well, there's isn't a fader (dissolver) yet, so I'm not duplicating anything... I'll continue writing this... and not duplicate anything there. [21:57] <omega> yup [21:57] <omega> there's a lot of work still before we'll hit that point [21:58] <wtay> omega: you should look at _pad_select some time.. [21:58] <tnt> You guys don't use the mailing list for that project much... at all. [21:58] <wtay> the only thing that's holfing us back from doing muxers.. [21:58] <wtay> tnt: nope, it's just some ideas.. [21:59] <omega> wtay: ok, I'll check it out [22:00] <omega> um, glib2 doesn't have g_string_printfa [22:00] <vishnu> yah , it got renamed [22:00] <omega> or ld says it doesn't [22:00] <omega> what's it called? [22:00] <wtay> 53% function docs coverage (341 functions documented, 303 not documented) :( [22:00] Nick change: omega -> PUT_YOUR_NICKNAME_HERE [22:00] <vishnu> try g_string_append_printf or something [22:00] <PUT_YOUR_NICKNAME_HERE> interesting [22:00] Nick change: PUT_YOUR_NICKNAME_HERE -> omega [22:00] <omega> ENOCLUE [22:01] <omega> hmm, ok, it's #defined in gstring.h, but only under DISABLE_DEPRACATED [22:02] <omega> I didn't think I'd defined that in my build, but I guess that's a good thing for development of the gobject2gtk shim.... [22:02] <omega> hrm, I didn't define it. not nice [22:03] <asmod> If my pipeline doesn't seem to be getting enough cpu time when I'm iterating it is there some sort of buffer I can increase the size of to improve the play quality (for mp3's in particular). [22:03] <omega> getting enough CPU time is something that can't be solved by gst [22:03] <omega> but if you're getting *uneven* CPU time, you can add/grow queues in your app to smooth it out [22:04] <asmod> omega: The problem is that I'm running inside of the main mozilla thread and if I give the gstreamer iteration to much priority is screws with mozilla's timeouts/intervals. So I have to give it low priority which results in a funny sounding playback. [22:05] <omega> then you really want to be using a separate thread [22:06] <vishnu> asmod: i use a separate process <g> [22:06] <asmod> omega: Yeah, but all experiments I've tried with that have resulted in seg fault. So I was hoping for a quick fix in the short term. Then later and can look deeper into interating in a separate thread. [22:06] <omega> gstreamer uses what it's given, can't do anything better with that short of trying to optimize the plugins involved [22:07] vektor_ (reet@HSE-Kitchener-ppp195088.sympatico.ca) joined #gstreamer. [22:07] <tnt> Do you have to have Queue's between thread? [22:07] <omega> something DECOUPLED, yes [22:07] <omega> pipefilter is the only other decoupled element atm [22:07] <vishnu> i tried running iterate in a timeout and it sucked. iterate needs to be called like 1000 times per second for some reason [22:07] <asmod> vishnu: I would love to run in a separate process, but I've got to run as a mozilla plugin in order to get video embedded into a mozilla window. [22:08] <omega> asmod: try calling iterate several times per idle timeout [22:08] <tnt> Do it matter which thread you put the Queue into? [22:08] <asmod> omega: I'll try that. [22:08] <omega> tnt: shouldn't [22:08] <tnt> OK. [22:08] <omega> asmod: but if you end up being clocked by the sound card for instance, you'll just end up blocking mozilla [22:08] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) left irc: Ping timeout for vektor[HSE-Kitchener-ppp195088.sympatico.ca] [22:09] <vishnu> tnt: beware, if you use threads you app will fail on SMP machines. the locking isn't working yet [22:09] Nick change: vektor_ -> vektor [22:09] <asmod> omega: I'll test it out. If it's blocking mozilla I'll revert. [22:10] <omega> asmod: maybe increasing OSS buffer depth will give you the breathing room needed to keep that from happening [22:11] <tnt> vishnu : I've just been getting seg faults (with my program that uses threads)... and I thought it might be because I wasn't using Queues. [22:11] <asmod> omega: OSS depth? Is that a property of the osssink plugin? [22:11] chillywilly (da...@d7...) joined #gstreamer. [22:11] <tnt> My workstation isn't a SMP anyways. [22:11] <omega> asmod: yeah, depth is fragment * buffer_size [22:11] <vishnu> tnt: yah, you need queues between threads. the safe thing is to avoid using more than one thread inside gst [22:11] <omega> I think that's in bytes [22:12] <omega> tnt: for now, until I get an SMP machine to debug with [22:12] <vishnu> tnt: if you don't have SMP then you can use threads -- should be fine [22:13] <tnt> vishnu : what do you mean by "avoid using more than one thread inside gst"... in my application I have 3 threads... is this bad? [22:13] <omega> tnt: no, unless you're running SMP, atm [22:13] <vishnu> tnt: no, it's fine. [22:13] <tnt> OK. [22:14] Zeenix (ze...@13...) joined #gstreamer. [22:14] <Zeenix> hello all [22:14] <omega> yo [22:15] <tnt> hey [22:15] <wtay> yo [22:16] <Zeenix> omega: no reply from that robla [22:16] <omega> hmm [22:16] <Zeenix> omega: looks like you scared him with the name of "pakistan" <g> [22:17] <omega> heh [22:18] Action: Zeenix is sending todays rtp to wtay.... [22:20] <Zeenix> whats random32() ? [22:20] <wingo> asmod: oss depth = the fragment setting, see the oss programmer's guide for its syntax [22:21] <wingo> unfortunately, due to a broken api, it has to be complicated [22:21] <asmod> wingo: I see that the default value is 6. What would be a sane value to increase it to? [22:22] <wingo> well, there's number and size. i've always optimized for low latency, i don't know in your case [22:22] <wingo> try tweaking both sides of the equation [22:22] <wingo> osssink fragment=foo [22:23] Nick change: tnt -> tnt-out [22:24] <wingo> asmod: i think the first eight bytes set the size of the fragment, the last eight the number [22:25] <wingo> fragment size defaults to 7fff [22:25] <wingo> which is pretty large [22:26] <asmod> wingo: So what actually do size and number mean? (I admit to knowing nearly nothing about audio programming) [22:26] <wingo> ok, so you have one big audio buffer in the sound card [22:26] <wingo> you can limit its size in software [22:26] omega (om...@om...) left irc: Ping timeout for omega[omegacs.net] [22:26] <wingo> but you have an upper limit. [22:27] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-24.brain.net.pk] [22:27] <wingo> now, periodically, some of the buffer comes available for capture or playback [22:27] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [22:27] <wingo> however long the period is is the "fragment size". [22:27] <wingo> the number of periods is the "fragment count". [22:27] <wingo> decreasing the fragment size will give you better response [22:28] <wingo> increasing it will allow you to buffer more [22:28] <wingo> increasing frag. count also allows you more buffer space, decreasing dropouts [22:28] <wingo> ultimately, you are limited by the internal hw buffer as to how much you can buffer though [22:28] <wingo> it's about 0.5 seconds [22:28] tnt-out (tn...@h2...) got netsplit. [22:28] chillywilly (da...@d7...) got netsplit. [22:28] tnt-out (tn...@h2...) returned to #gstreamer. [22:29] chillywilly (da...@d7...) returned to #gstreamer. [22:29] <asmod> wingo: So I have to determine if I want to buffer more or improve response. [22:29] <wingo> right [22:32] <asmod> wingo: Ok, thanks for the info. I'll experiment with it. [22:33] arik (ar...@sd...) joined #gstreamer. [22:33] <wingo> good luck :) [22:35] <asmod> later all [22:35] asmod (stevec@64.5.222.2) left irc: [x]chat [22:36] Zeenix (ze...@13...) joined #gstreamer. [22:39] omega (om...@om...) joined #gstreamer. [22:39] <Zeenix> 16809 B recieved 187694Bytes sent, my ISP is really fast today [22:39] <arik> heh [22:40] <arik> hmm [22:40] <arik> gstplay crashes if you play to many songs [22:41] <arik> and you can't kill the proces [22:41] <arik> and you can't play another song without rebooting [22:42] <Zeenix> arik: cant kill it even with a cruise missiles, your country have a lot of them <g> [22:43] <arik> hah [22:43] <taaz> Zeenix: you get to see any up close yet? [22:43] <arik> if i had a cruise missle kill program i would kill it [22:44] <arik> anyone know how to kill a process that you can't kill with kill -9? [22:44] <omega> can't be done, though you can try -9, -9, -9, etc. [22:44] <arik> yeah i've tried [22:44] <arik> *sigh* [22:44] <arik> any idea why this is happening? [22:44] <arik> ;-P [22:44] <Zeenix> taaz: all i see is too many Europeans( reporters & stuff ), police & army in the city [22:46] Action: taaz doesn't like this terrorist, biowarefare, bomb dropping mess [22:47] <Zeenix> taaz: me too, but the fall of USSR has led USA to behave as God, UNO is only meant for us not them [22:48] <ajmitch> arik: sounds like you've broke it ;) [22:49] <arik> ajmitch: hah [22:49] <arik> you are _so_ helpful ;-P [22:49] <ajmitch> arik: of course ;) [22:49] <ChiefHighwater> arik:you broke it? now you truly are a GStreamer developer 8-] [22:49] <ajmitch> hha [22:49] <arik> hahaha [22:49] <arik> i guess i am [22:50] Action: arik is proud to be a gstreamer developer [22:51] <Zeenix> wtay: any idea you can give me of the interface of rtp_recv() in rtp.c/h ? [22:53] <Zeenix> wtay: there ? [22:53] <wtay> doing docs ATM [22:56] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-57.brain.net.pk] [22:59] <arik> oh great [22:59] <arik> once gstplay crashes [22:59] <arik> you can't play _any_ audio [22:59] <arik> with _any_ player [22:59] <arik> wonderful [22:59] <omega> cool, sounds like your sound card driver is broken [23:00] <omega> ok, this sucks [23:00] <arik> yeah could be [23:00] <ajmitch> arik: what sound card you got? [23:00] <arik> aj, some yamaha crap thing i need to use alsa to use [23:00] <omega> I can play audio from an fltk app, but only be iterating it manually in a pipeline [23:00] <ajmitch> arik: the driver i use should be pretty stable (its an sb16 card) [23:00] <omega> if I put it in a thread it wedges during cothread_create [23:01] <arik> omega: yeah i can't do thread's with picogui either [23:01] <omega> at all? [23:01] <arik> omega: unless my thread code is wrong, which is _entirely_ possible ;-) [23:01] <omega> where does it fail? [23:02] <arik> hmm [23:02] <arik> one sec [23:02] <arik> i can't remember but i will go find out [23:04] <arik> hmm [23:04] <arik> cothread: trying to switch to same thread, legal but not necessary [23:04] <arik> infinitely [23:04] <omega> odd [23:04] <arik> and picogui gives me an error too [23:04] <arik> but [23:04] <omega> that usually indicates some other pipeline problem... [23:04] <arik> that's not what was happeneing before [23:05] <arik> so i think that's cause the soundcard thing [23:05] <omega> I get stuff at DEBUG(20919: 0)cothread_create:139: new stack (case 2) at 0x40c20000 [23:05] <arik> hmm [23:05] <arik> i forget what i was getting before [23:06] <omega> of course, this is the second cothread to be created, as indicated by the : 0) [23:06] <arik> right [23:07] <omega> running in gdb shows that the entire pthread state appears totally trashed [23:08] <arik> i can't get gdb to work with gst [23:08] <arik> it always complains about not being able to load libgst [23:08] Action: arik is no gdb expert [23:10] <wingo> arik: fix in GstDebugging node in the wiki [23:10] <wingo> it's a libtool bug [23:10] <wingo> if you are using uninstalled gst [23:10] <omega> I'm using uninstalled and it works <g> [23:10] <vishnu> omega: you should keep pthread state mprotect'd [23:10] <omega> ltmain.sh (GNU libtool) 1.4 (1.920 2001/04/24 23:26:18) [23:11] <wingo> yeah, debian used 1.4b [23:11] Nick change: chillywilly -> cwAWAY [23:11] <omega> vishnu: no, the pthread's pthread state. gdb claims there are no threads [23:11] <vishnu> ugh [23:11] <omega> making it oh so easy to debug [23:11] <arik> wingo: ok i really should read this wiki thing [23:11] <omega> I'm gonna create a non-gui version of this to see if it's fltk's problem [23:12] <arik> bbiab [23:12] arik (ar...@sd...) left irc: leaving [23:13] <omega> it's fltk's problem [23:13] <omega> text-mode version works perfectly [23:14] <wingo> hmm [23:15] <wingo> when you find the fix, document it :-) [23:17] ajmitch (aj...@p2...) left irc: Ping timeout for ajmitch[p22-max7.dun.ihug.co.nz] [23:18] ajmitch (aj...@p1...) joined #gstreamer. [23:32] <Uraeus> omega: you there? [23:32] Nick change: cwAWAY -> chillywilly [23:32] <omega> Uraeus: yeah [23:33] <Uraeus> omega: are you going to ALS? [23:33] steveb (st...@no...) left irc: advance! [23:33] <omega> dunno yet, at this point probably not unless I can find a cheap place to stay in the area [23:34] <Uraeus> omega: well we could share a room maybe to get it cheaper [23:34] <Uraeus> omega: also the hotel I stayed at last time I where in SF was very cheap [23:35] <ChiefHighwater> beware of cheap rooms in Oakland...they are usually cheap for a reason [23:35] <ChiefHighwater> there are two parts to Oakland...the tourist ready and the tourist hazardous [23:35] <omega> Uraeus: you're going to ALS? [23:36] <Uraeus> omega: yes, think so, my boss told me that if I wanted Oracle would pick up the bill [23:36] <omega> cool, you can bring my my Ximian monkeys <g> [23:36] <Uraeus> hehe, ok [23:37] <Uraeus> hmm, is it possible to stay in the SF city center or is Oakland way outside the city? [23:37] <omega> CHW: ? [23:38] <ajmitch> hi Uraeus, ChiefHighwater :) [23:39] <wtay> yo [23:40] <ajmitch> oh hi wtay [23:40] <ChiefHighwater> Oakland is across the bay from SF [23:40] <Uraeus> shit my mail to the gnome-foundation list was put onto LT [23:41] <omega> LT? [23:41] <Uraeus> LinuxToday [23:41] <ajmitch> Uraeus: yeah, you're famous now ;) [23:41] Action: Uraeus hates himself for sending that mail now [23:42] <vishnu> Uraeus: who are u? [23:43] <omega> Uraeus: once again, someone needs to take RMS's crack pipe away (IMO) [23:43] Apoc_ (Ap...@dy...) joined #gstreamer. [23:44] <Apoc_> Hi all [23:44] <Uraeus> vishnu: I am one of the poor dudes doing the GNOME Summaries and moderating Gnotices ;) [23:44] <Uraeus> hi Apoc_ [23:44] <vishnu> Uraeus: oh, there it is [23:44] <vishnu> Uraeus: neat [23:45] <ajmitch> Uraeus: as i said, you're a famous person that ppl look up to ;) [23:47] <omega> Uraeus: I can get there and back for $155.50 [23:47] <chillywilly> and why didn't you mention the Open Office instead of Star Office then? [23:49] <Uraeus> chillywilly: we have mentioned OO lots of times, and starOffice only once [23:51] <thomasvs> hey, here's a silly question [23:51] <thomasvs> if you're running esd, how do you tell mpg123 to play through it ? [23:51] <omega> dunno. you can run it with esddsp as a worst case [23:51] <chillywilly> RMS may be rough and such but he can be reasoned with believe me, he can....however, requesting that a GNU project not promote non-free software is not out of line [23:52] <vishnu> cw: yah, RMS sometimes says silly things to make a point. i love his technique [23:53] <ChiefHighwater> chillywilly:is it a 'free-software' promo organism or a 'Gnome' promo organism (I am truly asknig, I don't know) [23:53] <omega> I don't [23:53] Nick change: ChiefHighwater -> UberNewbie [23:55] <thomasvs> omega: esd mpg123 whatever.mp2 should work right ? I get a lib loading error [23:55] <omega> esddsp [23:56] <thomasvs> yes, sorry, that's what I meant. don't work tho [23:56] <thomasvs> agh, this is frustrating ;) [23:56] <omega> odd [23:56] <omega> filesrc location=whatever.mp3 ! mad ! esdsink <g> [23:57] <thomasvs> heh [23:57] <Uraeus> UberNewbie: is there a subway from Oakland or something into the town center? or do we need to stay in Oakland? [23:57] <thomasvs> oh wait, there was a symlink missing to the esdlib [23:57] <thomasvs> damn [23:57] <omega> doh [23:57] Nick change: UberNewbie -> ChiefHighwater [23:59] <thomasvs> woah [23:59] <thomasvs> esddsp adds LOTS of interleaved noise [00:00] --- Sat Oct 20 2001 [00:00] <omega> neat [00:00] <chillywilly> Uraeus: http://linuxtoday.com/news_story.php3?ltsn=2001-10-19-011-20-NW-CY-GN-0044 [00:01] <chillywilly> something I a sure RMS would even agree with [00:01] <Uraeus> chillywilly: yeah, I guess that is a rather reasonable approach [00:02] <omega> chillywilly: I wouldn't assume that [00:02] <omega> chillywilly: have you ever talked to him in person? [00:02] <chillywilly> yes [00:03] <chillywilly> I know him, he let GNUe use python ya know....he can be reasoned with [00:03] <omega> 'let'? [00:04] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) got netsplit. [00:04] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) returned to #gstreamer. [00:05] <chillywilly> i.e, he didn't raise a stink after we talked with him [00:06] <omega> frankly, RMS reminds me of a spoiled child who whines when he doesn't get what he wants [00:06] <vishnu> omega: dude, you're such an asshole [00:07] <taaz> rms is cool [00:07] <Uraeus> hmm, is the flamewar moving in here now? [00:09] <omega> no, it's not. everyone's stated their opinion, that's that. [00:09] <thomasvs> rms is root mean square [00:09] <thomasvs> move along, nothing to see [00:09] <taaz> blah, debian people questioning if liba52 acceptable for debian. [00:10] <taaz> odd since there are already a few packages in debian with ac3 decoding code [00:10] <vishnu> taaz: can't be too careful. ;-) [00:13] Action: omega expects that all this development will have to move overseas at some point [00:13] <thomasvs> omega: do you remember what fake X server I used to use to run gstreamer a few months ago ? [00:13] <taaz> do they have internet access in prison? I may need to know this... [00:13] <omega> nope, but if you use glib2 you don't need it [00:14] <omega> also, gtk gstreamer will deal with not having a $DISPLAY set at all [00:14] dneighbo (dne...@ma...) joined #gstreamer. [00:14] <omega> yo [00:14] <thomasvs> what, the new gtk or some changes to gstreamer gtk stuff ? [00:14] <dneighbo> yo yo yo [00:14] <omega> thomasvs: gst can deal without $DISPLAY by being careful as to what gtk stuff it initializes [00:15] <omega> or build against glib2 and avoid it entirely [00:15] <dneighbo> omega any progress made on visual 'soundlevel inputs'? [00:15] <dneighbo> and any good recommendations for 'mixing software'? [00:15] Action: dneighbo is getting ready to do a LOT more with this stuff [00:15] <omega> not yet, but wingo is getting close on the alsa plugin, so I can start messing with my hammerfall properly, and write that stuff [00:16] <dneighbo> omega : new church we started is pretty much 100% musicians [00:16] <omega> as for mixing, that's one of my near-term future projects, but I haven't put enough though into it yet to give much detail [00:16] <dneighbo> they are wanting to start a publishing company to freely distribute w/o copyright restrictions all their praise songs [00:16] <dneighbo> looking to do digital studio as well [00:16] <vishnu> dneighbo: wow, cool [00:17] <omega> dneighbo: new stuff? [00:17] <dneighbo> i was able to finally successfully not only use gstreamer but actually cut cd's that work in normal cd players for them (yeah) ;) [00:17] <omega> dneighbo: I have some neat ideas for both sound and control I/O hardware, but it's all in my head, and soldering irons and me don't usually get along too well [00:17] <dneighbo> ;) [00:18] <chillywilly> an embedded engineer who doesn't knowhow to solder? [00:18] <omega> if you know of someone who can design analog/digital circuits and wants a project... <g> [00:18] <dneighbo> i imagine soon i will be buying some pro sound and video hardware [00:18] <omega> chillywilly: I can solder, it's just not pretty [00:18] <dneighbo> actually a few of these guys might [00:18] <omega> cause what I'm thinking of isn't really that hard, from the digital side [00:18] <dneighbo> most of them work at allied signal / intel / motorola or other 'hardware' vendors [00:18] Action: chillywilly has this MIT rug warrior board...first thing I soldered [00:19] thomasvs (th...@21...) left irc: Read error to thomasvs[212-100-172-175.adsl.easynet.be]: EOF from client [00:19] <dneighbo> and are rather 'handy' [00:19] <omega> it's the mic preamp/phantom power/gainstage thing that sucks [00:19] <ChiefHighwater> details, details [00:19] <dneighbo> you mentioned before housing the storage disk in separate machine [00:19] <ChiefHighwater> we [00:19] <ChiefHighwater> we'll just leave that part off for now <g> [00:19] <dneighbo> does it need to be in diff machine or woudl antoehr box work? [00:19] <omega> in fact, if I can find the parts, I'm gonna build a siteplayer[.com] into a ethernet to stereo converter [00:19] <omega> another box would work [00:19] <omega> an 8x mic preamp [00:19] <omega> except you don't get computer gain control [00:20] <omega> which is critical for things like reasonable AGC [00:20] <ChiefHighwater> but you could just set them to unity gain [00:20] <omega> but if you have 24bit converters, you can avoid that [00:20] <dneighbo> as i have a scsi inclosure i could use that houses multi drives [00:20] <dneighbo> but you are correct that the quality is aobut that of tape w/ drive noise [00:20] Action: ChiefHighwater dreams of firewire microphones [00:20] <omega> ChiefHighwater: you have to leave enough to make sure you don't peak out, and those bits take away from your max [00:21] <dneighbo> it didnt seem so under xmms (mainly becuase listening to on low quality pc speakers) but when cut real cds and put in high end cd player NOTICEABLE drive noise :( [00:21] <omega> I think it's feasible to take a $50-75 embedded devel board with 4x SPI interfaces and hook stereo ADCs to them [00:21] <dneighbo> well im debating turning garage into my office / sound studio [00:21] <omega> then need a good clock, and analog front-end, and you're set with an 8x input or output [00:21] <omega> on ethernet.... <g> [00:21] <dneighbo> in which case we might be talking a LOT more :) [00:21] <omega> heh [00:21] <ChiefHighwater> omega:but you get good mic preamps for those few lost bits. trashed signal w/o a preamp [00:22] <omega> where are you, again? [00:22] <dneighbo> phoenix arizona [00:22] <omega> ChiefHighwater: right, that's why I said that was the hard part <g> [00:22] <ChiefHighwater> dneighbo:are you here in Portland? (/me is often out of the loop on who's who) [00:22] <omega> but if you know ppl who can do that kind of stuff... we can make some prototypes at least [00:23] thomasvs (th...@21...) joined #gstreamer. [00:23] <ChiefHighwater> omega:join my wife and I in gettnig her dad to move to Portland 8-] [00:23] <omega> I can do a prototype with a box with a hammerfall and some ethernet, make a very custom low-lat bridge between adat and ethernet [00:23] <omega> heh [00:23] Action: thomasvs should learn not to run killall X without checking the computer name [00:23] <omega> thomasvs: good idea [00:24] <omega> neat, my route to linuxshowcase.org dies at hop 4, in alternet [00:25] <omega> http://www.cirrus.com/design/products/index.cfm?DivisionID=3 [00:30] <dneighbo> omega you going to ALS? [00:30] <omega> possibly, yes [00:30] <omega> you? [00:30] <wtay> omega: bah, gtkdoc-scan refuses to find gstfilesrc because it has no .h file.. [00:30] <omega> neat [00:30] <dneighbo> cool, i wont be able to make it to this one, went to LWCE isntead [00:30] <omega> darn [00:31] <dneighbo> i will be in california though (disneyland) [00:31] <Uraeus> naptime see ya [00:31] Uraeus (csc...@c1...) left irc: Client Exiting [00:52] Shippou (no...@ot...) joined #gstreamer. [00:56] Nick change: omega -> omega_phone [01:03] harobed (harobed@AC925753.ipt.aol.com) joined #gstreamer. [01:03] Nick change: omega_phone -> omega [01:17] dneighbo (dne...@ma...) left #gstreamer. [01:23] ChiefHighwater (pa...@te...) left irc: [01:26] <wtay> omega: there are two gst_object_ref/unref implementations in gstobject.c, why is that? [01:26] <omega> hmm? that happened once during a merge... [01:27] Apoc_ (Ap...@dy...) left irc: Ping timeout for Apoc_[dyn-212-129-40-118.ppp.libertysurf.fr] [01:27] <omega> that's a good question [01:27] <omega> amazed that the compiler doesn't have a cow [01:27] <wtay> it looks like an attempt to use glib refcounting.. [01:27] <omega> dunno what the define is for either [01:27] <wtay> it'll never get compiled [01:28] <omega> right [01:28] <wtay> in the first _ref the define is defined.. [01:28] <omega> oh, the #define. yeah, I dunno [01:28] <omega> I'd remove it, or change it to #define gst_ g_ and then comment it out [01:29] <wtay> the second unref looks a bit strange too, I would be amazed if that was correct.. [01:30] <omega> oh, down there [01:30] <omega> um, considering that g_object doesn't lock afaik, something like that is indeed necessary [01:30] <omega> I think the #define .. .. above is to keep the #ifndef from triggering [01:40] <wtay> yup.. [01:48] Nick change: wtay -> wtay-tv [01:48] Action: taaz updates debs in hopes of fixing gstreamer problems... maybe its just the broken binutils ;) [01:50] <taaz> http://luxik.cdi.cz/~devik/mm.htm [01:50] <taaz> neat ;) call graph viz is cool [01:52] <vishnu> heh [02:12] wingo_ (wi...@rd...) joined #gstreamer. [02:33] Shippou (no...@ot...) got netsplit. [02:33] harobed (harobed@AC925753.ipt.aol.com) got netsplit. [02:33] chillywilly (da...@d7...) got netsplit. [02:33] thomasvs (th...@21...) got netsplit. [02:33] wingo_ (wi...@rd...) got netsplit. [02:33] wingo (wi...@rd...) got netsplit. [02:33] ajmitch (aj...@p1...) got netsplit. [02:33] evil_monk (trevo@208.141.162.68) got netsplit. [02:39] wingo (wi...@rd...) got lost in the net-split. [02:39] evil_monk (trevo@208.141.162.68) got lost in the net-split. [02:39] chillywilly (da...@d7...) got lost in the net-split. [02:39] ajmitch (aj...@p1...) got lost in the net-split. [02:39] thomasvs (th...@21...) got lost in the net-split. [02:39] Shippou (no...@ot...) got lost in the net-split. [02:39] harobed (harobed@AC925753.ipt.aol.com) got lost in the net-split. [02:39] wingo_ (wi...@rd...) got lost in the net-split. [02:41] thomasvs (th...@21...) joined #gstreamer. [02:41] Shippou (no...@ot...) joined #gstreamer. [02:41] harobed (harobed@AC925753.ipt.aol.com) joined #gstreamer. [02:41] wingo_ (wi...@rd...) joined #gstreamer. [02:41] wingo__ (wi...@rd...) joined #gstreamer. [02:41] chillywilly (da...@d7...) joined #gstreamer. [02:46] <harobed> where are codec in the source of gstreamer ? [02:47] ajmitch (aj...@p1...) joined #gstreamer. [02:47] Nick change: wtay-tv -> wtay [02:47] <wtay> harobed: you mean plugins? [02:48] Nick change: tnt-out -> tnt [02:48] <harobed> the codec and decode [02:49] <harobed> exemple for decode one mpeg ? or mp3 [02:49] <wtay> harobed: oh, plugins/mp3decode/mad/ is a decoder [02:50] <harobed> sorry, I didn't see [02:50] <wtay> plugins/mpeg2/mpeg2dec for example decodes mpeg2.. [02:50] <wtay> we usually link against a library so there is no real codec stuff in it [02:51] <harobed> ok, you don't code the codec, you use external library [02:52] <wtay> right [02:52] <harobed> do you use the ffmpeg codec ? [02:52] <wtay> yes [02:52] <harobed> where ? [02:52] <wtay> plugins/ffmpeg [02:53] <harobed> it isn't in the version 0.2.1 of gstreamer ? [02:53] <wtay> it only compiles on my machine because ffmpeg doesn't have shared libs or anything yet [02:53] <wtay> harobed: oh, could be, yes.. [02:54] <harobed> I search one software to decode one part of file in mjpeg and code this part un divx. gstreamer can do it ? [02:55] <wtay> audio and video? [02:55] <harobed> in resolution of video PAL in 25 frames/s [02:55] <harobed> yes, audio and video [02:56] <wtay> nope, we need to fix _pad_select first in order to make that work, I'm afreaid [02:57] <harobed> I would like help you in your project (gstreamer), what I do ? [02:58] <wtay> what do you want to do? [02:59] ajmitch (aj...@p1...) left irc: Ping timeout for ajmitch[p1-max5.dun.ihug.co.nz] [02:59] <harobed> one library to manipulate all video file, with the same interface [02:59] <harobed> as image magic |
From: IRC B. <wt...@us...> - 2001-10-21 04:27:25
|
******************************************************************* [03:00] <wtay> NLE like stuff? [03:00] <harobed> NLE ? what is it ? [03:00] Nick change: ShrimpZzZ -> ShrimpX [03:00] <vektor> Non-Linear Editor. [03:01] <tnt> [I'm working on NLE stuff :-) ] [03:01] <harobed> yes [03:01] <vektor> So, video editor where you can visualize many films at once and do cuts, fades, edits, composites, etc. [03:01] <vishnu> i'm doing NLE too (somewhat) [03:01] <wtay> GStreamer needs a lot of help from another lib to do that [03:01] <vektor> I've kinda been working on an NLE also. [03:01] <harobed> to, in the future, make with this library one video monting as adobe premier [03:01] Vakor (ms...@co...) joined #gstreamer. [03:02] <tnt> Well, right now, beside trying to figure out what's causing my Segmentation Fault and trying to figure out what all the functions in the plugin do, I'm working of creating transitions... is anyone else doing this? [03:03] <vishnu> tnt: i'd be happy is simple playback worked well ;-) [03:04] <wtay> me too.. and seeking.. before trying an NLE at all.. [03:05] <wtay> not to mention a scheduler that can do real chaining instead of using cothreads [03:05] <tnt> [Sorry for the delay... someone came to the door....] [03:06] <tnt> vishnu: Well, I can play a video fine. [03:06] <harobed> what is the labrary who recognize the type of file video ? and who choose the good codec to use on the video file ? [03:06] <vishnu> tnt: with A/V sync? [03:07] <wtay> harobed: it's the typedetect plugin that runs over the registered typefind functions.. [03:07] <tnt> horoded: I'm not sure... I think maybe the "filesrc" can tell you the mime type of its "source pad". [03:07] <wtay> harobed: and then the autoplugger constructs a pipeline [03:07] <tnt> tnt: I've got sdlvideosink and osssink to work fine. [03:08] <tnt> Oppsss. that was suppose to be to vishnu. [03:08] <vishnu> tnt: yah, but if you play long enough the audio lags the video by about 30 frames. at least that's what it seems like [03:08] <wtay> I am not satisfied with simple playback at all [03:08] <tnt> I wished that there was some more documentation on writing your own elements... like what each function does, etc. [03:09] <wtay> tnt: yeah.. nobody seems to be willing to write such a thing.. [03:10] <wtay> vishnu: the problem is that the PTS timestamps from the PES packets are not used ATM [03:10] <tnt> vishnu : well you could always write a new filter that have two Sink Pads (one for Audio and one for Video) and two Source Pads... and makes sure that the Audio and Video stay in step. [03:10] <wtay> vishnu: A/V sync with mpeg1 is quite solid [03:10] <vishnu> yah, PTS timestamps, right. [03:11] <vishnu> wtay: oh good. i didn't know mpeg1 was solid [03:11] <harobed> if I understand well, gstreamer is a few as mplayer ? [03:11] <wtay> tnt: no you can't [03:11] <wtay> harobed: what do you mean? [03:11] <tnt> wtay: why not? [03:12] <wtay> tnt: 'cause there are no timestamps in the case that is currently failing [03:13] thomasvs (th...@21...) left irc: [x]chat [03:13] <harobed> what is difference between gstreamer and mplayer ? [03:13] <wtay> tnt: we also need a better global clock with the audio device as a possible master [03:14] <wtay> harobed: mplayer is a media player, gstreamer is a media framework [03:14] <wtay> harobed: mplyare is not designed to encode or apply effects on media [03:15] <wtay> harobed: playback of media is actually quite simple [03:15] <harobed> what is the effects can do gstreamer ? [03:15] <harobed> s/is/are [03:16] <wtay> harobed: whatever plugin you code [03:17] <tnt> getting off topic a bit... does anyone no the purpose of 'Pad Negotiation Function'?... Is it necessary for the plugin?... When is it called? (Actually, it would be nice to know just what function ARE infact necessary for a function.) [03:17] <vishnu> mplayer is really hard to customize [03:17] <wtay> harobed: we've got a few filter plugins.. like smooth/median/volume filters etc.. [03:17] <harobed> as the adobe premier plug in ? exemple :encrust some picture in the video [03:18] <wtay> tnt: depends on what you are going to do... [03:18] <wtay> harobed: we don't have such a plugin for the moment [03:18] <wtay> tnt: I would recommend to stay away from pad negotiation :) [03:19] <harobed> but in the future this plugin can been coding ? [03:19] <tnt> wtay: Well I'm making a plugin that takes two Video input (sink) Pads and return one Video output (source) Pad. It is 'loop based' of course. Would I need it? [03:19] <wtay> harobed: sure [03:19] <tnt> wtay: "stay away.." OK... I really don't understand it yet anyways :-) [03:19] <wtay> tnt: the pad negotiation function is used to negotiate the media type between src and sink pads, I doubt you'll need it [03:20] <harobed> wtay: then, I'm very intersting by gstreamer [03:20] <tnt> OK... thanks. [03:20] <wtay> harobed: my goal is to write a NLE library [03:21] <wtay> hmm I have to sleep, cya [03:21] Nick change: wtay -> wtay-zZz [03:21] <harobed> as broadcast 2000 ? [03:21] <wtay-zZz> harobed: that's not a NLE library, it's a hacked app :) [03:21] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [03:22] <harobed> NLE is a hacked app ? [03:23] <tnt> Broadcast200 is a NLE application (not a NLE library). [03:23] <harobed> ok [03:23] <tnt> Broadcast2000 is gone anyways... the authour has taken off the GPL. [03:24] <harobed> but, gstreamer have got also one interface ? nont ? [03:24] <harobed> but, gstreamer have got also one interface ? not ? [03:24] <tnt> What do you mean by "interface"? [03:24] <harobed> interface graphic [03:25] <tnt> gstreamer is a library... there is NO GUI (Graphical User Interface) to it... people can, however, make programs/applications with GUIs and that use gstreamer. [03:25] <harobed> gui [03:25] <harobed> as gstplay [03:25] <harobed> ? [03:25] <tnt> Well, gstplay is a application that uses gstreamer. [03:26] <tnt> It's like OpenGL. [03:26] <harobed> ok [03:26] <tnt> OpenGL is a library... people can make programs/application/games that use OpenGL... but there is not a GUI that comes with OpenGL. [03:28] ajmitch (aj...@p2...) joined #gstreamer. [03:34] <tnt> So (hoping that the gstreamer guru's are listening...) would I be correct in saying that the absolute minimum set of functions needed for make a loop-based plugin are: [03:34] <tnt> 1. GType gst_???_get_type(void); [03:34] ChiefHighwater (fl...@su...) joined #gstreamer. [03:34] <tnt> 2. static gboolean plugin_init(GModule *module, GstPlugin *plugin); [03:35] <tnt> 3. static void gst_???_class_init(GstVideoMixClass *klass); [03:35] <tnt> 4. static void gst_????_init(Gst????Mix *????); [03:35] <tnt> 5. static void gst_????_loop(GstElement *element); [03:35] <tnt> Those five? [03:35] <omega> pretty much, yes [03:36] <tnt> Is there any documentation explaining the 'Pad Template' stuff. [03:36] <omega> not really ;( [03:37] <tnt> Is it a kind of 'type' for the pad? [03:37] <omega> sortof, yes [03:37] <omega> it's a set of caps that is the maximum range of types the pad can accept [03:37] <tnt> It could take more than one type ? [03:37] <omega> yes [03:40] evil_monk (trevo@208.141.162.68) joined #gstreamer. [03:42] harobed (harobed@AC925753.ipt.aol.com) left irc: Client Exiting [03:42] ChiefHighwater (fl...@su...) left irc: Leaving [03:49] Nick change: omega -> omega_tv [03:49] Nick change: tnt -> tnt-out [04:02] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [04:08] ChiefHighwater (fl...@su...) joined #gstreamer. [04:09] ChiefHighwater (fl...@su...) left irc: Leaving [05:07] Nick change: tnt-out -> tnt [05:09] <tnt> Hi, I'm trying to understand the "video/raw" mime type. Would I be correct in saying that "video/raw" is not really one 'type' but many types... and you figure out which of those may types it is by the FOURCC value? [05:12] <tnt> [Guess everyone is hiding :-) ] [05:13] <vishnu> tnt: that sounds plausible, but i don't know for sure. check the source code for videoscale? [05:13] <taaz> caps negotiation is the usual way [05:13] <taaz> i think [05:14] <Vakor> yes. Where caps are negotiated, in this particular case, through video/raw then FOURCC. Which seems a bit strange to me, but that appears to be how it works [05:14] <tnt> I'm trying to understand what's going on with the GST_PADTEMPLATE_FACTORY stuff. Part of that is that it defines the mime type "video/raw"... but also has a FOURCC number. [05:19] ajmitch (aj...@p2...) got netsplit. [05:19] wingo_ (wi...@rd...) got netsplit. [05:19] chillywilly (da...@d7...) left irc: Buffer allocation error for chillywilly[d78.as9.nwbl0.wi.voyager.net] [05:19] vektor (reet@HSE-Kitchener-ppp195088.sympatico.ca) left irc: Ping timeout for vektor[HSE-Kitchener-ppp195088.sympatico.ca] [05:20] chillywilly (da...@d7...) joined #gstreamer. [05:21] wingo__ (wi...@rd...) left irc: BitchX-1.0c18 -- just do it. [05:22] wingo (wi...@rd...) joined #gstreamer. [05:22] ajmitch (aj...@p2...) returned to #gstreamer. [05:22] Shippou (no...@ot...) joined #gstreamer. [05:25] wingo_ (wi...@rd...) got lost in the net-split. [05:31] <taaz> vishnu? [05:31] <vishnu> taaz: ? [05:31] <taaz> vishnu! [05:32] <vishnu> yo! [05:32] <taaz> so, what's up with the Xlib async replay error thread stuff? how does one go about fixing that in dvdplay.c? [05:33] ajmitch (aj...@p2...) got netsplit. [05:33] wingo (wi...@rd...) got netsplit. [05:35] <vishnu> just a sec [05:37] ShrimpX (marius@131.252.244.168) left irc: Ping timeout for ShrimpX[131.252.244.168] [05:39] ajmitch (aj...@p2...) got lost in the net-split. [05:39] wingo (wi...@rd...) got lost in the net-split. [05:41] wingo (wi...@rd...) joined #gstreamer. [05:41] ajmitch (aj...@p2...) joined #gstreamer. [05:43] <vishnu> taaz: oh, yah [05:44] <vishnu> taaz: the problem is probably happening because you connect directly to the signals emitted by videosink [05:53] <tnt> When dealing with a buffer containing data of the mime type video/raw and the fourcc type i420 (i.e., YUV... as in PAL)... What part of the buffer contains the Y, the U, and the V? [05:54] <taaz> unspecified [05:55] <tnt> So how do you get at the data to do any kind of manipulation? [05:55] <taaz> you guess? ;) [05:55] <taaz> how would you suggest this be handled? [05:55] <vishnu> tnt: how does videoscale work? [05:56] <taaz> it's probably just an issue of some caps negotiation, just need to have some policy on how to specify such things like "layout" = "interleaved" or "sequential" or whatever... [05:57] <taaz> someone who knows better will probably say the format is going to almost always be in the same format though [05:57] <tnt> vishnu: videoscale seems to call a function pointer call 'scale_cc', but I can't find the definition for it, or where it is set. [05:59] <Vakor> I would expect a single FOURCC to map to a single buffer layout. YUV is typically YYUV, for two pixels in a row. [05:59] <tnt> taaz: I would suggest it be handled with something like: GST_BUFFER_GET_Y(buff), GST_BUFFER_GET_U(buff), GST_BUFER_GET_V(buff), GST_BUFFER_Y_ACCESS(buffer, pixel_num), GST_BUFFER_U_ACCESS(buffer, pixel_num), GST_BUFFER_V_ACCESS(buffer, pixel_num)... or something like that. [05:59] <wingo> maybe those would be defined in a special video header. [06:00] <tnt> Vakor: Thanks [06:00] <taaz> uh... no [06:00] <taaz> GST_BUFFER is part of GstBuffer API [06:00] <taaz> which has nothing to do with specific data formats [06:00] <vishnu> yah, write your own inline functions [06:01] <Vakor> It would be fairly rare that you'd care about the contents of the raw decoded video data anyway, in practice. [06:01] <tnt> vishnu : I would write my own functions, but how do I access the parts of the buffer... do I treat it like an array... is the layout of the data like Vakor said: YYUV? [06:02] <vishnu> tnt: dunno [06:02] <taaz> tnt: this is not a gstreamer api sort of issue. plugins have to agree on format of data they exchange [06:03] <Vakor> you should be able to find data layouts elsewhere, for a given fourcc. They're pretty standard. [06:03] <tnt> Vakor: Well I'm trying to make transition filter (like those found in FinalCutPro, Premiere, etc), that take two video inputs and return one... I need to get at the actual data to do that. [06:03] <Vakor> yeah, you probably do :-) [06:04] <taaz> tnt: you always have access to the data in a buffer. you just have to basically ask the element giving you buffers what format they are in [06:04] <Vakor> It might be worth adding a standard cap for decoded video buffer layout, if it's not always given by the fourcc (I _think_ it will be, I don't know) [06:04] <tnt> taaz: I've been looking at http://www.webartz.com/fourcc/ and it was kind of vague about i420 (YUV)... it says: "8 bit Y plane followed by 8 bit 2x2 subsampled U and V planes."... which isn't very helpful in who the buffer are in relation to one another. [06:04] <taaz> tnt: where is this data coming from? [06:06] <Vakor> 2x2 subsampled? ok. Essentially this means you have a byte for Y for every pixel and one for each of U and V for every 4 pixels. Layout for that sort of thing is very standard, look it up in anything about computer video formats. I forget what it is offhand, though ;) [06:06] <tnt> taaz: Well, I'm hooking up a 'filesrc' to a 'mpegdec' (actually 2 of these) and sending those to the filter. [06:07] <tnt> Vakor: Ya... the class I took subject (when I was in University) didn't have a textbook :-( but maybe I have something around here that will help. [06:07] <taaz> well, for starters look at mpegdec (or do you mean mpeg2dec?) source and see what format it spits data out as [06:07] <tnt> taaz: OK, I'll go and look. [06:08] <Vakor> (well, it's more that I've never worked with that particular video layout. I've done a heap of stuff with 422 (which is just YYUV in memory layout), but that doens't help much [06:08] <taaz> mpeg2dec does non-interleaved Y' Cr Cb iirc [06:09] <tnt> taaz: mpeg2dec returns the i420 (YUV)... so it's OK. [06:09] <taaz> don't let vektor here you say YUV or you get a lecture ;) [06:09] <tnt> :-) [06:10] <tnt> Maybe vektor knows how it's layed-out in memory :-) [06:10] <vishnu> a lekture, to be precise [06:11] <tnt> :-) [06:11] chillywilly (da...@d7...) left irc: Ping timeout for chillywilly[d78.as9.nwbl0.wi.voyager.net] [06:12] Action: Vakor denies saying 'YUV' ever. I _do_ know better :) [06:12] chillywilly (da...@d1...) joined #gstreamer. [06:16] <tnt> :-) ... Oh how nice in gstmpeg2dec.c, part that writes the i420 data is one single function... 'mpeg2_decode_data', which is part of libmpeg2dec. [06:23] <tnt> OK... with i420, it is a 'planar format' (and not a 'packet format') where the Y, the U, and the V are all stored in seperate arrays... not lets see if I can find out which comes 1st, 2nd, and 3rd. [06:25] <taaz> Y' Cr Cb... didn't i just say that? [06:25] <tnt> OK, i420 has... for a NxN image, a NxN Y array, followed by a (N/2)x(N/2) U array, followed by a (N/2)x(N/2) V array. [06:26] <tnt> taaz: Huh... Cr and Cb are different than U and V. [06:27] <tnt> U = 0.493(B-Y) and V =0.877(R-Y) [06:30] <taaz> ok whatever i dunno whoop whoop! video voodoo is not my thing [06:36] <wingo> tnt: i don't know if this is too late for you, but I started a little guide to plugin writing (not near as large in scope as wtay's) [06:36] <wingo> it's at gstreamer.net/wiki/?PluginWriters [06:37] <wingo> i think i'm going to sleep now, but it's about 2/3 of the way done. [06:38] <tnt> wingo: I never saw that before... I'll take a look at it. [06:38] <wingo> i just started this evening :) [06:38] <tnt> That explains it. [06:39] <tnt> Does anyone know if i420 formated data is always sqaure (NxN)... or can it be rectangular (AxB) too??? [06:40] <tnt> (If it's always NxN it's easy to figure out the dimensions... dimension = buffer_size * 2 / 3 ... but if can be rectangular, I'm not sure how to figure it out.) [06:40] <taaz> sometimes it's circular, sometimes triangular! [06:41] <tnt> :-) [06:41] Action: taaz smacks taaz [06:41] <tnt> I'll let you write the code to figure out the dimensions them :-) [06:41] <taaz> it's not always square [06:41] <taaz> it's dependent on the output dimension [06:42] <taaz> the UV planes are just N/2 in each dimension as the Y plane [06:42] <taaz> its N,M i guess [06:42] <taaz> i'm talking out my ass right now too [06:43] <taaz> so yeah, i guess you have pic that's XxY so Y=XxY, U and V = X/2xY/2 [06:43] <taaz> that sounds right... [06:43] <taaz> maybe [06:45] <tnt> Hmmm: so how do I solve: X * Y = 2 * buffer_size / 3 ??? [06:45] <tnt> Need some more data! [06:45] <taaz> huh? [06:47] <taaz> oh.. [06:47] <taaz> i'm too sleepy to help now it seems :( [06:47] <Vakor> tnt: erm. the x and y resolution must be available. Another cap? [06:52] <tnt> Vakor: I'm guessing that it is probably in the GstProps list of strings. [06:52] <Vakor> I'd guess so. I've never looked at any video stuff in gstreamer. [06:53] <tnt> The Props stuff is a free-form list of strings... for video you would have a name "format" and then give it the value of a fourcc 32-bit-int. Maybe there is a width and height thing in there. [07:13] <wingo> tnt: it's "width" and "height" [07:14] <wingo> you can check in the xvideosink code [07:14] <wingo> i'm not a video fellow though, ymmv [07:14] <wingo> they are ints [07:14] <wingo> your best resource is the source of those who have gone before you :-) [07:15] <tnt> Kool... thanks... I was reading your new wiki document and Steve Baker's GstDynamicParams. [07:15] <wingo> is it readable to someone who's just come to gst? [07:15] <wingo> my doc that is :) [07:15] <tnt> Well, it's readable to me :-) [07:16] <tnt> Of course I've already learned quite a bit. [07:16] <wingo> true [07:17] <wingo> where are you tnt? i thought you told me a while ago, i seem to have forgotten [07:18] <tnt> I'm on the west coast of Canada, in BC. [07:18] Nick change: taaz -> taazzzz [07:19] <wingo> oh yeah. [07:19] <tnt> Where are you at? [07:20] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [07:22] <wingo> north carolina, middle of the east coast, in the states [07:26] <tnt> Oh... that's you that was from there. [07:26] BeNOW (an...@ds...) joined #gstreamer. [07:27] <BeNOW> heya. [07:27] <tnt> I think one of my cousins (and his family) is there. [07:27] vektor (reet@HSE-Kitchener-ppp3508308.sympatico.ca) joined #gstreamer. [07:27] Vakor (ms...@co...) left irc: Ping timeout for Vakor[co3016667-a.lowrp1.vic.optushome.com.au] [07:27] <tnt> BeNOW: Hey [07:28] <wingo> hi BeNOW [07:28] <BeNOW> there's a live show on now with the dj encoding with gst (osssrc!lame!icecastsend) and i'm redirecting it with gstreamer here at the server (httpsrc!icecastsend) [07:28] <BeNOW> heya guys. [07:29] <BeNOW> only one place of encoding... it's soooo great. [07:29] <BeNOW> his lame quality=7 tho, as he's only (!) got a p2 333 [07:30] <wingo> is it vektor? [07:30] <wingo> stream url? [07:30] <BeNOW> nope... haven't talked to vektor about the gig yet. [07:30] <BeNOW> http://benow.ca and click on listen. [07:30] <BeNOW> there's a couple streams. [07:31] <wingo> my mailcap is broken unfortunately [07:31] <BeNOW> erm.. mirrors/relays... one may be better than the other depending on rounting. [07:31] <tnt> What do you need to listen to it... XMMS doesn't seem to work. [07:31] <BeNOW> http://us.benow.ca:10390 [07:32] <wingo> nice [07:32] <BeNOW> gstreamer-lauch httpsrc location=http://us.benow.ca:10390 ! mad ! osssink [07:32] <wingo> :) [07:32] <wingo> so mad works well for you [07:32] <wingo> have you had funny errors recently at all? [07:32] <BeNOW> not at all since switching to mad. [07:33] <BeNOW> I changed the clock while gst was running.. it didn't like that ;) [07:33] <BeNOW> but that's it. [07:33] <wingo> heh [07:33] <BeNOW> 2 days of problem free streaming. [07:33] <wingo> wonderful! [07:33] <BeNOW> xmms never gave me that long. [07:33] <wingo> that's real nice. [07:34] <wingo> congrats :) [07:34] <BeNOW> (I have a suspicion that the frame code of mpg123 is mostly to blame) [07:34] <wingo> probably [07:34] <wingo> having isolated the component :) [07:34] <BeNOW> it's not me that deserves the congrats, is the gst devs. [07:34] <BeNOW> I just put the bits together. [07:34] <BeNOW> but, it's not pretty now. [07:34] <wingo> eh? [07:35] <BeNOW> all the tools I built for xmms are broken [07:35] <BeNOW> so I have to switch to djs myself, etc. [07:35] <wingo> sucks. [07:35] <BeNOW> next week will take care of that tho. [07:35] <wingo> i was thinking on basing my app on xmms before i heard of gst [07:35] <BeNOW> hopefully with the java bindings I can do it right from a servlet. [07:35] <wingo> or really *realized* xmms' limitations and gst's strengths [07:35] <wingo> that would be nce [07:36] <BeNOW> yeah, we started with winamp on windows.. [07:36] <wingo> nice even [07:36] <BeNOW> built telnet plugins, etc. [07:36] <wingo> you employed to do this now? [07:36] <BeNOW> but os reliability was a prob, then a file server running debian and samba, with stream still on windos [07:37] <BeNOW> no it eats all the spare time I have :) [07:37] <wingo> me too :)) [07:37] <BeNOW> then to xmms on linux, then gst on linux. [07:37] <wingo> quite the trajectory [07:37] <BeNOW> I did get lucky and land a contract that's got alot of overlap with the stuff I do here, tho. [07:38] <BeNOW> and they're into the whole open source library thang, so I don't have to fork (!) [07:38] <BeNOW> so I'm building a scheduling backend for stream management [07:38] <BeNOW> (dj's, radio rebroadcast, playlist based shows, live shows, etc) [07:39] <wingo> nice [07:39] <BeNOW> and a beefed up web framework that will tie all the bits together. [07:39] <BeNOW> if I can get the java bindings going, I should have security right into gst. [07:40] <BeNOW> ... it's been nearly 3 years that we've been streaming.... [07:41] <BeNOW> (with a break of nearly 8 months as people moved around) [07:41] <BeNOW> the one year aniversary of this incarnation is coming up in dec. [07:41] <wingo> there is a bindings project for java, they say it works but i'd imagine it takes effort [07:41] <wingo> that's pretty nifty [07:41] <BeNOW> yeah, played a bit with it... need to play more :) [07:41] <wingo> :) [07:41] <wingo> must ... play ... more [07:42] <BeNOW> hopefully gst will be fully integrated by then and can have a world wide party, without me having to babysit [07:42] <BeNOW> (I'm actually thinking of doing my masters on this stuff in a year orso) [07:43] <wingo> in school now? [07:43] <BeNOW> no, been out for nearly 4 years now. [07:43] <BeNOW> getting the urge to go back tho. [07:44] <tnt> I don't have that urge yet. [07:44] <tnt> But I've only been out for a year. [07:45] <BeNOW> yeah, takes a few years of being shafted by industry and the race for the big bucks. [07:45] <BeNOW> but, there may be a slight chance that I'm biased ;) [07:46] <BeNOW> the stream just needs dj's from europe now. [07:46] <tnt> I'll go back eventually... I want to do my PhD eventually... but I still remember what doing homework is like... and not having a life. [07:46] <BeNOW> everything is coming from mst or pst, so it's all balled up from 6pm-1am [07:46] Action: BeNOW had very little of one to start with :) [07:47] <tnt> :-) [07:47] <wingo> s/dj's in europe/dj's in europe with broadband/ :) [07:49] <BeNOW> yeah, tho alot of the web traffic comes from europe (not monitoring stream stats, yet, unfortunately) [07:50] <wingo> these are pretty sweet sounds [07:50] <BeNOW> yeah, this is ivan's first step back to the decks in a couple years. [07:51] Action: taazzzz grumbles at BeNOW who keeps poping up here and mentionnig interesting stuff streaming right as I want to go to bed. [07:51] <BeNOW> about 6 years ago he used to have a radio show in Lethbridge, Alberta (Can)... [07:51] Nick change: taazzzz -> taaz [07:51] <BeNOW> hehe, there's always something interesting. [07:51] <BeNOW> 80GB orso always requestable ;) [07:52] <wingo> :) [07:52] <BeNOW> http://benow.ca/music/Music [07:52] <BeNOW> Have wanted to make a "if the stream sucks it's your fault" intro for a while, but have decided not to offend too much. [07:53] <wingo> whose fault? the user for using mpg123? [07:53] <BeNOW> the listener for not requesting something they'd enjoy :) [07:54] ajmitch (aj...@p2...) got netsplit. [07:54] wingo (wi...@rd...) got netsplit. [07:54] BeNOW (an...@ds...) got netsplit. [07:54] BeNOW (an...@ds...) returned to #gstreamer. [07:54] wingo (wi...@rd...) returned to #gstreamer. [07:54] ajmitch (aj...@p2...) returned to #gstreamer. [07:55] <wingo> good point :) [07:55] <taaz> is this radio? [07:55] <BeNOW> nope live dj [07:56] <BeNOW> radio coming up in 5 min, [07:56] <taaz> like from the djs living room? [07:56] <BeNOW> yeah, spare bedroom, actually ;) [07:56] <taaz> heh [07:57] Action: wingo has dreams of doing this one day [07:57] Action: taaz too [07:57] <wingo> vektor can shut the hell up to. this is a preemptive strike :) [07:58] <taaz> BeNOW: try to get vektor live. he's got lots of cool gadgets to make noise with [07:58] ShrimpX (marius@131.252.244.168) joined #gstreamer. [07:59] <wingo> yeah i'd like to hear him too [07:59] <tnt> OT: When writing a plugin... did anyone else get a problem linking with the function "g_alloc"? [07:59] <taaz> he sent me a cd of live tracks... some good stuff in there [08:01] <BeNOW> this is radio [08:01] <BeNOW> yeah, for sure. [08:01] <tnt> [Never mind... I should learn to spell... it's g_malloc not g_alloc.] [08:01] <BeNOW> i'm looking forward to it. [08:02] <BeNOW> you could always work on this project ;) [08:02] <BeNOW> is GPL, etc [08:07] <taaz> fund raisers heh [08:10] <BeNOW> yeah, once a year... [08:10] <taaz> i get to answer phones during our univ "radiothon" next week [08:11] <BeNOW> commercial free university radio station, so it's worth it 1 week out of 52 ;) [08:11] Shippou (no...@ot...) left irc: Write error to Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca], closing link [08:11] <BeNOW> it's nice to hear people on the radio. [08:11] <taaz> we have it twice a year [08:11] <BeNOW> sweet. [08:12] <taaz> dude... our station doesn't have a delay though so you can't hear people that call in [08:12] <BeNOW> these are all live in the studio. [08:12] <taaz> oh.. never mind... ;) [08:12] <BeNOW> they put a call out for listeners to come in for this show... those that donate. [08:13] <BeNOW> this show's got a bit of a, erm, special crowd. [08:14] <taaz> i'll be taking donations during the dark wave/goth show... i wonder who will call in ;) [08:15] <BeNOW> gotta love non-commercial radio. [08:16] <BeNOW> CJSW is broadcasting on the internet, but with real-audio at 24k... I'm trying to do my bit to get them on at higher bitrate. [08:18] <taaz> i gotta try and get an FM show before I leave here... got an AM show now (which isn't actually AM, since hardware is dead, but broadcast over TV). they make us pions answer phones next week rather than do a show. ;) [08:18] <taaz> www.wuvt.vt.edu my station. they have realaudio feed for FM [08:19] <tnt> Hey, does anyone know if you should apply a gst_buffer_unref or gst_buffer_destroy to a Sub-Buffer??? [08:19] <BeNOW> kewl. [08:20] Vakor (ms...@co...) joined #gstreamer. [08:20] <BeNOW> heh, yeah, this feed is pulled off cable FM. [08:20] <wingo> tnt: unref i think [08:20] <taaz> we gotta hack out a realaudio src in gstreamer [08:20] <tnt> OK... [08:21] <BeNOW> yeah so I can httprasrc!radec!lame!icecastsend.. hehe. [08:21] <BeNOW> there's a ground problem in my building somewhere, so there's a bit of a hum in the bg. [08:22] <tnt> What happens if you don't unref it?... just a memory leak?... To be honest, I don't really understand when you unref something and when you destroy something... most programs I see don't bother doing either. [08:22] <wingo> you get a memory leak, yes [08:22] <wingo> you unref refcounted objects [08:23] <wingo> like buffers [08:23] <wingo> actually, unref ends up calling destroy [08:23] <taaz> you just have to follow mem management policies [08:23] <wingo> exactly [08:23] <tnt> Whenever you get something from a _new function, should you always _unref it when you are done? [08:23] <taaz> you don't always need to unref [08:23] <taaz> only if you created it somehow [08:23] <taaz> yeah [08:24] <taaz> err [08:24] Action: taaz sleepy [08:25] Action: taaz must wake up to go to surplus auction... where else can you buy 50 CGA monitors for $1? ;) [08:26] Nick change: taaz -> taazzzz [08:41] Nick change: tnt -> tnt-out [08:44] <vektor> Hiya. [08:44] <vektor> No walken :( [08:46] <wingo> hey fella. [08:47] <vektor> Hi wingo! [08:56] <wingo> i really want to finish my alsa rewrite [08:57] <wingo> it is taking too long :) [08:57] <omega_tv> BeNOW: the stream is totally tweaked.... [08:58] <wingo> tweaked? [08:58] <wingo> yeah [08:58] <wingo> oof [08:58] <wingo> i thought it was the 'music' at first [08:58] Nick change: omega_tv -> omega [08:58] <omega> nope, once the DJ started sounding like that for a while, I realized it wasn't <g> [09:06] <BeNOW> hmmm, yeah, perhaps I shouldn't be doing that while encoding ;) [09:06] <omega> yeah [09:06] <BeNOW> is better now? [09:06] <omega> it's happened a couple times, and recovered fine [09:07] <omega> "Doctor, doct<smack!>" "Don't do that!" <g> [09:07] <omega> what did you do, anyway? browse somewhere with an evil java applet? [09:08] chillywilly (da...@d1...) left irc: [09:09] <BeNOW> nah, some encoding. [09:09] <BeNOW> now encoding live takes 50-60% on a p3 750 [09:10] <omega> is that good or bad? [09:11] <BeNOW> makes me want something more ;) [09:11] <BeNOW> something always does... [09:11] <omega> heh [09:11] Hal69 (Ha...@ca...) joined #gstreamer. [09:11] Shippou (no...@ot...) joined #gstreamer. [09:12] <BeNOW> so just then was a switch back to the stream. [09:12] <BeNOW> a little pop. [09:12] <omega> missed it [09:12] <BeNOW> yeah, I've got it down pretty good. [09:12] <BeNOW> Now, I've just to automate it. [09:14] <BeNOW> neway, I'm off to bed... take it easy folks. [09:14] <omega> l8r [09:15] <wingo> cya [09:21] <wingo> boy, the thinko rate is pretty hight tonight [09:24] <wingo> ug, going to sleep, cya [09:24] Nick change: wingo -> wingo-zzz [09:35] tnt-out (tn...@h2...) left irc: Client Exiting [09:36] Zeenix (ad...@ho...) joined #gstreamer. [09:36] <Zeenix> hello all [09:46] steveb (st...@no...) joined #gstreamer. [09:48] <Zeenix> yo [09:50] Hal69 (Ha...@ca...) left irc: Ping timeout for Hal69[cable-212-10-237-24.anchorageak.net] [09:51] <steveb> hi [09:51] Hal69 (Ha...@ca...) joined #gstreamer. [09:54] Hal69 (Ha...@ca...) left irc: Ping timeout for Hal69[cable-212-10-237-24.anchorageak.net] [09:56] Hal69 (Ha...@ca...) joined #gstreamer. [10:03] Zeenix (ad...@ho...) left irc: Ping timeout for Zeenix[host-216-252-173-178.interpacket.net] [10:05] Hal69 (Ha...@ca...) left irc: Ping timeout for Hal69[cable-212-10-237-24.anchorageak.net] [10:12] Hal69 (Ha...@ca...) joined #gstreamer. [10:15] Hal69 (Ha...@ca...) left irc: Ping timeout for Hal69[cable-212-10-237-24.anchorageak.net] [10:18] Hal69 (Ha...@ca...) joined #gstreamer. [10:21] Hal69 (Ha...@ca...) left irc: [11:31] Vakor (ms...@co...) left irc: Leaving. [11:34] omega (om...@om...) left irc: sleep [12:19] BeNOW (an...@ds...) left irc: Ping timeout for BeNOW[dsl-mr-207-34-113-i106-cgy.nucleus.com] [12:21] BeNOW (an...@ds...) joined #gstreamer. [12:26] bstard (Lor...@a2...) joined #gstreamer. [13:51] ajmitch (aj...@p2...) got netsplit. [13:51] Shippou (no...@ot...) got netsplit. [13:51] steveb (st...@no...) got netsplit. [13:51] BeNOW (an...@ds...) got netsplit. [13:51] wingo-zzz (wi...@rd...) got netsplit. [13:51] ShrimpX (marius@131.252.244.168) got netsplit. [13:51] bstard (Lor...@a2...) got netsplit. [13:51] vektor (reet@HSE-Kitchener-ppp3508308.sympatico.ca) got netsplit. [13:57] wingo-zzz (wi...@rd...) got lost in the net-split. [13:57] ajmitch (aj...@p2...) got lost in the net-split. [13:57] vektor (reet@HSE-Kitchener-ppp3508308.sympatico.ca) got lost in the net-split. [13:57] ShrimpX (marius@131.252.244.168) got lost in the net-split. [13:57] Shippou (no...@ot...) got lost in the net-split. [13:57] steveb (st...@no...) got lost in the net-split. [13:57] BeNOW (an...@ds...) got lost in the net-split. [13:57] bstard (Lor...@a2...) got lost in the net-split. [14:13] thomasvs (th...@ad...) joined #gstreamer. [14:14] ShrimpX (marius@131.252.244.168) joined #gstreamer. [14:14] ajmitch (aj...@p5...) joined #gstreamer. [14:18] bstard (Lor...@a2...) joined #gstreamer. [14:37] chillywilly (da...@d1...) joined #gstreamer. [15:27] vektor (reet@HSE-Kitchener-ppp3508308.sympatico.ca) joined #gstreamer. [15:30] Nick change: wtay-zZz -> wtay [15:30] <wtay> heya [15:31] chillywilly (da...@d1...) left irc: Read error to chillywilly[d185.as6.nwbl0.wi.voyager.net]: EOF from client [15:46] wingo-zzz (wi...@rd...) joined #gstreamer. [15:46] Shippou (no...@ot...) joined #gstreamer. [15:46] steveb (st...@no...) joined #gstreamer. [15:46] BeNOW (an...@ds...) joined #gstreamer. [15:48] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [15:49] thomasvs (th...@ad...) joined #gstreamer. [15:50] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [15:52] thomasvs (th...@ad...) joined #gstreamer. [15:53] thomasvs (th...@ad...) left irc: Killed (NickServ (This nick is reserved by another user)) [16:12] ajmitch (aj...@p5...) left irc: Ping timeout for ajmitch[p57-max2.dun.ihug.co.nz] [16:38] ajmitch (aj...@p6...) joined #gstreamer. [16:39] bstard (Lor...@a2...) got netsplit. [16:42] chillywilly (da...@d7...) joined #gstreamer. [16:45] bstard (Lor...@a2...) got lost in the net-split. [16:45] Nick change: wtay -> wtay-bath [17:07] Shippou (no...@ot...) got netsplit. [17:07] steveb (st...@no...) got netsplit. [17:07] BeNOW (an...@ds...) got netsplit. [17:07] ajmitch (aj...@p6...) got netsplit. [17:07] wingo-zzz (wi...@rd...) got netsplit. [17:07] ShrimpX (marius@131.252.244.168) left irc: Buffer allocation error for ShrimpX[131.252.244.168] [17:07] BeNOW (an...@ds...) returned to #gstreamer. [17:07] steveb (st...@no...) returned to #gstreamer. [17:07] Shippou (no...@ot...) returned to #gstreamer. [17:07] wingo-zzz (wi...@rd...) returned to #gstreamer. [17:07] ajmitch (aj...@p6...) returned to #gstreamer. [17:07] Shippou (no...@ot...) got netsplit. [17:07] steveb (st...@no...) got netsplit. [17:07] BeNOW (an...@ds...) got netsplit. [17:07] ajmitch (aj...@p6...) got netsplit. [17:07] wingo-zzz (wi...@rd...) got netsplit. [17:08] BeNOW (an...@ds...) returned to #gstreamer. [17:08] steveb (st...@no...) returned to #gstreamer. [17:08] Shippou (no...@ot...) returned to #gstreamer. [17:08] wingo-zzz (wi...@rd...) returned to #gstreamer. [17:08] ajmitch (aj...@p6...) returned to #gstreamer. [17:10] ShrimpX (marius@131.252.244.168) joined #gstreamer. [17:20] thomasvs (th...@ad...) joined #gstreamer. [17:20] thomasvs (th...@ad...) left irc: Client Exiting [17:30] thomasvs (th...@ad...) joined #gstreamer. [17:30] vishnu (jo...@cx...) joined #gstreamer. [17:30] ajmitch (aj...@p6...) left irc: Ping timeout for ajmitch[p61-max10.dun.ihug.co.nz] [17:39] Nick change: wtay-bath -> wtay [17:45] <vishnu> ready for some patches?? [17:45] <wtay> sure [18:04] <vishnu> wtay: you should get email soon. please apply the patch at your earliest convinence [18:04] ajmitch (aj...@p6...) joined #gstreamer. [18:19] chillywilly (da...@d7...) left irc: [18:28] BeNOW (an...@ds...) left irc: BitchX-1.0c18 -- just do it. [18:30] bstard (Lor...@a2...) joined #gstreamer. [18:36] bstard (Lor...@a2...) got netsplit. [18:36] bstard (Lor...@a2...) returned to #gstreamer. [18:38] Action: vishnu serves up some nice, warm, melt-in-your-mouth patch cookies [18:40] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [18:41] chillywilly (da...@d1...) joined #gstreamer. [18:44] bstard (Lor...@a2...) got netsplit. [18:45] bstard (Lor...@a2...) returned to #gstreamer. [18:47] Nick change: wingo-zzz -> wingo [18:47] Action: wingo has memory corruption 'issues' in the alsasrc :-\ [18:50] <steveb> wingo: try counselling, or drugs [18:50] <vishnu> wingo: -lmcheck ? [18:52] <thomasvs> anyone know of a xawtv mailing list ? [19:00] <vektor> There's the video4linux list. [19:00] <vektor> It seems to take a fair amount of traffic on xawtv bugs. [19:05] <thomasvs> 'kay, will check [19:05] <thomasvs> invite only it seems [19:07] <vektor> Huh??? [19:07] <vektor> No no. [19:08] <vektor> www.redhat.com/lists/video4linux/ [19:08] <vektor> something like that [19:08] <vektor> search for 'video4linux list' on google. [19:17] <thomasvs> this frequency table stuff in xawtv is confusing. [19:17] <thomasvs> you should just be able to fill in the frequencies based on your cable provider's list ;) [19:41] Shippou (no...@ot...) joined #gstreamer. [20:06] bstard (Lor...@a2...) left irc: Ping timeout for bstard[a213-84-40-242.adsl.xs4all.nl] [20:14] Nick change: wtay -> wtay-snooker [20:27] bstard (Lor...@a2...) joined #gstreamer. [20:32] Nick change: taazzzz -> taaz [20:33] chillywilly (da...@d1...) left irc: Read error to chillywilly[d170.as8.nwbl0.wi.voyager.net]: Connection reset by peer [20:34] chillywilly (da...@d1...) joined #gstreamer. [20:46] wingo (wi...@rd...) left irc: Read error to wingo[rdu25-21-076.nc.rr.com]: EOF from client [20:51] tnt (tn...@h2...) joined #gstreamer. [20:52] wingo (wi...@rd...) joined #gstreamer. [20:52] <wingo> yo [20:52] <tnt> Hey [20:53] <vishnu> yo [20:53] <tnt> hello [20:53] <taaz> yo [20:54] <tnt> howdy [20:54] <vishnu> anyone want to bet on how long it takes to get my patches applied? [20:56] <wingo> heh [20:57] <wingo> no :) [21:05] ajmitch (aj...@p6...) left irc: http://www.freedevelopers.net [21:05] Action: vishnu throws darts at his gstreamer dartboard [21:06] ajmitch (aj...@p6...) joined #gstreamer. [21:11] chillywilly (da...@d1...) left irc: Ping timeout for chillywilly[d175.as4.nwbl0.wi.voyager.net] [21:20] Nick change: tnt -> tnt-helping-someone-move [21:30] harobed (harobed@AC8CFD14.ipt.aol.com) joined #gstreamer. [21:32] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [21:43] thomasvs (th...@ad...) left irc: Read error to thomasvs[adsl-63552.turboline.skynet.be]: Connection reset by peer [22:11] chillywilly (da...@d1...) joined #gstreamer. [22:14] thomasvs (th...@ad...) joined #gstreamer. [22:27] <wingo> vishnu: there? [22:30] <vishnu> yah [22:30] chillywilly (da...@d1...) left irc: [22:31] <wingo> can i sucker you into updating the statuses of the video plugins, adding entries for ones that aren't there? :-) [22:31] <wingo> i don't know what i can trade you for that :) [22:31] <wingo> maybe that big cat's eye marble i have? [22:31] <wingo> :-) [22:31] <vishnu> remind me how to login [22:32] <wingo> gstreamer.net/admin.php, then click on status [22:33] <wingo> i should probably change that password, it's a bit rude [22:36] omega (om...@om...) joined #gstreamer. [22:37] <ajmitch> hi omega [22:38] <omega> yo [22:39] <vishnu> wingo: ok, i updated the few plugins with which i am familiar [22:44] <wingo> thanks vishnu [22:44] <wingo> hi omega [22:45] <omega> wingo: what's the password for updating it? [22:47] chillywilly (da...@d1...) joined #gstreamer. [22:48] <wingo> if the whole password-remembering thing is a hindrance, we can make it just one password for all admin pages [22:48] <omega> yeah [22:49] <wingo> ok, i'll do that. [23:09] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [23:09] wayne (wa...@po...) joined #gstreamer. [23:09] <wayne> hi, i just installed gstreamer via rpm ... how do i start up the media player? [23:09] wingo (wi...@rd...) left irc: going home [23:09] <omega> gstmediaplay [23:10] evil_monk (trevo@208.141.162.68) joined #gstreamer. [23:10] <omega> it doesn't work very well though, yet, especially in the rpms [23:10] Zeenix (ze...@13...) joined #gstreamer. [23:10] <wayne> [wayne@zero]~$ gstmediaplay [23:10] <wayne> zsh: command not found: gstmediaplay [23:10] <wayne> gstreamer-* exists though [23:10] <omega> did you install the media player rpm? [23:12] <Zeenix> omega: yo [23:12] <omega> yo [23:12] <Zeenix> omega: new gst user/devel ? [23:12] <omega> Zeenix: huh? [23:13] <Zeenix> omega: wayne, new gst user/devel ? [23:14] <omega> yes [23:14] harobed (harobed@AC8CFD14.ipt.aol.com) left irc: Client Exiting [23:15] <Zeenix> omega: how many hours have wtay spent on snooker today ? [23:15] <vishnu> if my patches are applied then i'll submit more patches [23:15] <omega> dunno, I just got here [23:15] <omega> vishnu: I'm working on a reply to the plugin split proposal atm, then I'll work on the patches [23:16] <Zeenix> connected at 14Kbps, i am flying [23:16] <omega> I already did the printfa locally, though <g> [23:16] <vishnu> omega: ok, thank you [23:16] <vishnu> omega: why not commit it?? [23:16] <omega> forgot to [23:16] <vishnu> d;oh! [23:16] <omega> you need it now? [23:17] <vishnu> omega: no, it's not a big deal [23:20] <Zeenix> omega: i needed some advice(s) [23:20] <Zeenix> omega: RtpSend sends an Rtcp App packet, telling no. of packets sent, when going from Null-> Ready state, How should GstRtpRecv respond ? [23:20] <Zeenix> errr [23:20] <Zeenix> its Ready -> Null state [23:21] thomasvs (th...@ad...) left irc: Ping timeout for thomasvs[adsl-63890.turboline.skynet.be] [23:22] <omega> probably by firing a signal for the app to figure out how to deal with [23:24] <Zeenix> omega: so we should have a signal declared in our RtpRecv element & add developer should connect his func. with it ? [23:24] <omega> yes [23:24] <omega> any event that happens that doesn't strictly concern the streamed data itself should be passed up to the application in a signal [23:26] <Zeenix> in our case, would we be having some default handler [23:26] <omega> a signal never has a 'default handler' [23:27] <Zeenix> omega: oh, sorry, i just recently learned all these gtk+/gnome/glib...... stufff [23:28] wingo (wi...@rd...) joined #gstreamer. [23:29] <Zeenix> omega: but i remember some simillar thing, am i mixing any other concept with it [23:29] <Zeenix> wingo: yo [23:32] <wingo> hi zeenix [23:32] <wingo> how goes it? [23:33] <Zeenix> wingo: it? [23:33] <omega> s/it/life/ [23:33] <wingo> i dunno, life, the universe, everythin [23:33] <wingo> :) [23:33] <Zeenix> wingo: good [23:34] <Zeenix> wingo: studying communism though [23:34] <wingo> it's good to study [23:34] Action: Zeenix hides from omega [23:34] <wingo> parecon is my economic fantasy though [23:34] <wingo> parecon.org [23:35] <wingo> one must have vision to put the present in perspective :) [23:36] <Zeenix> wingo: presently, i have bought a book about the history of humanity & religiouns [23:37] <wingo> nice. [23:37] <vishnu> Zeenix: hey, wow. excellent [23:38] <omega> Zeenix: did it come with the heavy equipment necessary to haul it around? [23:38] <wingo> hah! [23:38] Nick change: wingo -> wingo-frisbee [23:38] <Zeenix> omega: no, its a medium size book [23:38] <omega> I'm sure one of the US transport choppers could help you out <g> [23:39] <Zeenix> omega: no, ATM they cant help themselves out, a helicopter crashed today [23:39] <omega> accidents happen, especially in situations like this [23:39] <Zeenix> omega: actually i was in contact with some communists a day ago [23:40] <omega> uh oh <g> [23:41] <Zeenix> omega: they recommended a book called "From Moses till Marx", in which the author proves that Moses was the actual founder of communism [23:41] <omega> pfff [23:41] <Zeenix> omega: i cant say anything without studying it [23:42] Action: omega is always very wary of books that claim to "prove" anything like that [23:43] <omega> especially when the subject at hand fundamentally can't be proved or disproved [23:44] <Zeenix> omega: i think we shouldnt be wary of anything, study it & then conclude [23:44] <omega> good ;-) [23:44] <Zeenix> omega: all we know about communism is US & islamic propaganda [23:45] <omega> explain? [23:45] <Zeenix> omega: explain what ? [23:46] <omega> please explain your previous statement, it doesn't make sense on its own [23:46] <omega> oh, er, nevermind [23:46] Action: omega is being dense [23:47] <Zeenix> omega: our knowledge about communism comes from US & islamic media( propaganda ) [23:47] <omega> right [23:47] <Zeenix> omega: i can't talk to you about it till i really study it [23:49] <vishnu> Zeenix: if the book gives you a headache then stop reading <g> [23:49] <omega> it's a sign <g> [23:49] thomasvs (th...@ad...) joined #gstreamer. [23:50] <Zeenix> vishnu: yeah, of course [23:51] <thomasvs> wingo-frisbee: did you find some of the mails re: the plugin division ? [23:51] <thomasvs> I once had a proposition which I thought was clean but most people found too complicated/layered [23:51] <thomasvs> so I tend to go for either my structure or a totally flat structure [23:52] <thomasvs> mail me if you read this in the logs ;) [23:52] <vektor> omega: you there??? [23:52] <vektor> omega: need your help. [23:52] <omega> ? [23:52] <vektor> k [23:53] <vektor> I need you to edit an email. [23:53] <vektor> or tell me if there's anything i should add. [23:53] <omega> ok, send it [23:53] <Zeenix> omega: is Zaheer an experienced developer ? [23:53] <omega> afaik, yes [23:53] <omega> but haven't seen him around for a while. I should bug him [23:54] <Zeenix> omega: when i find something bad in his code, i am not confident on that, e.g change of a gchar* to gpointer [23:55] <omega> gchar* and gpointer mean the same thing, of course, it's just whether you need to access the data in a certain way or not [23:55] <Zeenix> s/that/changing his code [23:55] <omega> you have his email addr? [23:55] <Zeenix> yes, the one mentioned in the GstRtpSend [23:55] <omega> ok, send him some email ask him [23:56] <Zeenix> oh, he is probably to busy, i did sent him some emails but he only replied to the fist one [23:56] <Zeenix> s/to/too [23:56] <omega> keep bugging him <g> [23:56] <vektor> omega: http://www.billybiggs.com/xv-email.txt [23:57] <Zeenix> s/fist/first of course [23:59] <Zeenix> whats that integer to random32 for ? [23:59] <vektor> anyone else have anything to add to the list in that URL? [00:00] --- Sun Oct 21 2001 [00:00] <omega> vektor: part of 1) and 4) would be supporting double- and multi-buffering [00:00] <vektor> Ah ok, should I add that? [00:00] <vektor> Or merge or what? [00:00] <omega> and related to that is the issue that some cards (especially the i810 and other "on-board" video chips) kinda by definition can read the video frames right out of main memory, no copy at all [00:00] <omega> dunno... [00:01] <omega> er, move that ')' 3 words down [00:02] <omega> I guess mention system-memory buffers in 1) and multibuffering in 4) [00:02] <vektor> Yes I'm doing that. [00:02] <omega> uh oh, benow.ca lost it [00:03] <omega> used to be an hourly occurance, but now he's using gstreamer, so this is the first failure in something like 3 days [00:04] steveb (st...@no...) left irc: advance! [00:05] <vektor> Ok, new version of the email up. [00:05] <vektor> omega: thoughts? [00:06] Uraeus (csc...@c1...) joined #gstreamer. [00:06] <vektor> (like on the web, I haven't posted yet). [00:06] <Uraeus> hi dudes [00:08] <omega> vektor: "libXv.a is distributed as a non-PIC .a only" [00:08] <omega> Uraeus: any news on the ALS front? [00:09] <Zeenix> Uraeus: yo [00:09] <Uraeus> omega: news? well I am waitng for hotel feedback from you until I mail a expense report to my boss for final approval [00:09] <vektor> omega: I put that at the bottom, although in less words: '6) libXv.a is distributed as a .a.' [00:09] <vektor> omega: which is being a bit inflamitory [00:09] <omega> vektor: I think the extra words are useful [00:09] <vektor> ok. [00:10] <omega> they clarify the grievance <g> [00:10] <vektor> Uh. [00:10] <vektor> Well my grievance is the lack of a shared object. [00:10] Shippou (no...@ot...) joined #gstreamer. [00:10] <vektor> But I assume you know that. [00:10] <omega> right [00:10] <vektor> So why can't I say it as I mean? [00:11] <omega> "a bit inflamitory" were your words <g> [00:11] <vektor> hahaha :) [00:11] <vektor> ok [00:11] <vektor> I'll just say that it's not PIC. [00:11] <vektor> omega: Can I cc you on this? [00:12] <omega> sure [00:12] Nick change: wingo-frisbee -> wingo [00:12] <wingo> thomasvs: no i didn't. if you would send them to me that would be nice :) [00:12] <Uraeus> omega: do you want me to try and find us a hotel? [00:13] <wingo> thomasvs: or maybe even to the list, being as i didn't find them on google and geocrawlers is broken [00:14] <omega> Uraeus: best western, 0.63mi from center, $114/n [00:15] <wingo> where are you now uraeus? [00:15] <Uraeus> wingo: at home (in Oslo, Norway) [00:16] <Uraeus> omega: that 114 that is splitt between us right? it is not their per person charge? [00:16] <wingo> damn kiddo, that's a long journey! [00:16] <omega> right [00:17] <omega> oooooh, battlebots!!! [00:17] <omega> they're filming the same week [00:17] <Uraeus> omega: ok, I mail my boss on monday then with my the cost and hopefully get her approval [00:18] <omega> MUST GO SEE BATTLEBOTS [00:18] <Uraeus> wingo: it is somewhat of a long journey yes :) [00:18] <omega> Uraeus: I want to be there by 6pm the 7th for the BOF's [00:18] <omega> and leave sat night to get home that night [00:19] <omega> so I'd be there 3 nights (7th, 8, 9th) [00:19] <Uraeus> omega: well my current flight schedule is leaving Oslo the 5th (and arriving the 5th) and staying in SF until the 12 [00:19] <omega> you going to tutorials then? [00:19] <Uraeus> omega: staying shorter makes my airplane costs skyhigh [00:19] <omega> ah [00:20] <Uraeus> omega: well mostly I am going to ALS to meet up with people :) [00:20] <omega> yeah [00:20] <omega> don't forget my monkeys <g> [00:20] <Uraeus> omega: you can probably get new ones from the Ximian people there :) [00:20] <omega> ok, conference center is at 1001 Broadway, this hotel is at 223 Broadway [00:21] <omega> oh, so now you'r holding them ransom? ;-) [00:21] <Uraeus> omega: yeah, I give them to you on the 12th and only on the 12th :) [00:21] <omega> hrm, yeah, right <g> [00:23] Action: Uraeus was hoping that omega would stay for the same amount as days as himself [00:23] <omega> I can get there earlier, definitely [00:24] <Uraeus> cool [00:25] <thomasvs> wingo: sent [00:25] <thomasvs> Uraeus: get some extra monkeys if you will ;) [00:25] <omega> problem is the closing session ends at 5pm and the flight out is at 6:55 [00:26] <omega> oooh, autoconf list gets a spam titled "Advertisement for get-rich-quick scheme!" [00:26] <omega> at least its honest [00:28] <omega> Uraeus: so do I get two sets of tickets for battlebots? [00:30] <Uraeus> omega: you will be staying until the 12th if I get you those tickets? [00:30] <thomasvs> ok, night all [00:31] thomasvs (th...@ad...) left irc: Client Exiting [00:35] <omega> Uraeus: heh [00:36] <omega> first lemme check ticket availability, and see if there some reason I *have* to be back in pdx by Sunday morning [00:36] <Zeenix> Uraeus: any sign of GAUDEC or GUADEC ? [00:37] <Uraeus> Zeenix: have not seen any concrete news yet [00:38] <omega> Uraeus: lemme create schedule page, sec... [00:38] Nick change: tnt-helping-someone-move -> tnt [00:38] <omega> http://temple-baptist.com/~omega/als-sched.txt being filled in [00:43] <omega> Uraeus: ok, load it [00:43] <Uraeus> rofl [00:43] <omega> ;-) [00:44] <Uraeus> btw, I mispokke I am leaving on the 11th (I will be back on Norway the 12th) [00:44] <omega> ok [00:44] <omega> what time? [00:45] <tnt> Hello, I seem to be getting an error with my little test program (used to test my new filter). The error looks like it occurs when I call the "gst_bin_iterate" function on the pipeline. (This function, "gst_bin_iterate", is being called from my 'Gtk idle function'... if it matters.) There error I get is: "GStreamer-CRITICAL **: file gstscheduler.c line 1374 (gst_schedule_iterate): assertion 'chains != NULL' failed." Anyone have any clue w [00:45] <tnt> hat's the cause of problem???? I should say that my pipeline has 3 threads in it. (And I've put these 3 threads in the pipeline.) [00:45] <omega> have you set the state to PLAYING? [00:46] <tnt> Yes. I set it on the pipeline (but not on the threads individually.) [00:46] <omega> hmm, what does your pipeline look like? [00:46] <tnt> I'll draw a little ASCII art diagram... it will take three lines, so hang on.... [00:47] <tnt> filesrc_1 -> mpeg2dec_1 -> queue_1 -v [00:47] <tnt> videomix -> sdlvideosink [00:47] <Uraeus> omega: not sure about the exact time [00:47] <tnt> filesrc_2 -> mpeg2dec_2 -> queue_2 -^ [00:47] <tnt> Each line is in a seperate thread. [00:47] <tnt> videomix is a loop based filter that takes 2 inputs. [00:47] <omega> tnt: ok, and it would be helpful if you split up your questions for IRC, otherwise it's hard to follow them [00:48] <tnt> OK. [00:48] <omega> so the only children of the pipeline are the threads? [00:48] <tnt> Yes. [00:48] <omega> does --gst-mask=-1 tell you anything interesting? [00:48] <tnt> hang on I'll check. [00:49] <omega> Uraeus: chances are you'll miss the finals then ;-( [00:49] <wingo> should buffer data be malloc'd or g_malloc'd? [00:49] <omega> g_malloc'd, unless you provide a freefunc [00:49] <wingo> ok [00:50] <omega> g_malloc must be paired with g_free, and g_free is the default if freefunc==NULL [00:51] <tnt> How do you get that stuff to stop scrolling so fast... "... | less" and " | more" don't seem to work. [00:51] <omega> less -R [00:51] <omega> 2>&1 | less -R [00:53] <omega> Uraeus: so there are people in Oakland you want to meet with? when? [00:55] <Uraeus> omega: no one in particular, but I expect a lot of GNOME people to be there [00:56] <omega> ok, but not during the tutorials [00:56] <omega> I'm just trying to figure out what the plan would be as far as going to battlebots [00:56] <omega> a full pass is only $132.50, so going to a large number of them isn't really that big a deal for me [00:56] <Uraeus> of battelbots? [00:56] <omega> yup [00:57] Action: Uraeus is confused [00:57] Action: omega is going to the main ALS stuff, but wants to see as much battlebots as feasible outside that [00:57] <Uraeus> omega: is battlebots some sort of turnament held during ALS? [00:57] <omega> it's totally unrelated [00:57] <omega> you've never heard of battlebots?? [00:57] <Uraeus> omega: isn't it a competition where people make programs to go up against eachother? [00:58] <omega> oh, you poor deprived soul [00:58] <Uraeus> :) [00:58] <omega> the battlebots.com videos need qt, darn [00:58] <omega> people build robots in various weight classes [00:58] <omega> remote control [00:58] <omega> with really nasty weapons [00:58] <omega> and fight them [00:59] <Uraeus> omega: have you ever looked at BEAST/BSE? [00:59] <omega> one example: take a wok (big bowl), put armored chisels around it, invert it, put it on a robot, and spin it up to several thousand RPM. [00:59] <Uraeus> hehe [00:59] <omega> drive towards other bot, watch it (in the 300-400lb class) go flying 20ft [01:00] Nick change: tnt -> tnt-back-in-a-bit [01:00] <Uraeus> omega: have you ever looked at BEAST/BSE? [01:00] <omega> think about it, 300+lbs of robot flying 6-10ft in the air, you're protected only by the 6" lexan windows [01:00] <omega> somewhat, only slightly [01:00] <Uraeus> omega: could it be usefull for us as a midi engine? [01:00] <omega> dunno [01:01] <omega> Uraeus: let's put it this way: you're going to battlebots [01:01] <Uraeus> omega: well I am mailing a bit with Tim these days, so if Beast/BSE have something we need then I could suggest it to him [01:01] <Uraeus> omega: ok [01:01] <Uraeus> ;) [01:02] <omega> Uraeus: check out the front image on battlebots.com for an idea [01:02] <omega> then go into 'meet the robots' and have a look around [01:02] <Uraeus> omega: ok and you check out beast.gtk.org and tell me if I should suggest for Tim to GStreamerise Beast/BSE [01:03] <omega> good luck <g> [01:03] <wingo> omega: alsasrc works!!!!! [01:03] <omega> oooooh [01:03] <wingo> well, i need to test multichannel now [01:03] <wingo> but it blocks now, instead of eating the cpu [01:03] <wingo> and has no hardcoded channel limitations :) [01:03] <omega> cool [01:04] Action: wingo is pleased with himself :-) [01:04] <omega> the ADAT box is off, but you can use channels 9-16 in loopback on psi, if you need to test [01:04] <wingo> ok [01:04] <omega> Uraeus: BSE looks useful in the same way arts does: constructing processing graphs... [truncated message content] |
From: IRC B. <wt...@us...> - 2001-10-22 04:27:31
|
******************************************************************* [03:05] <vishnu> bytestream patch status? [03:06] <wtay> vishnu: not applied yet.. I'm struggling with my copies of gstreamer now.. [03:06] <wtay> I've got 4 copies and I'm trying to sync one... [03:07] <vishnu> d'oh [03:10] <wingo> anyone know how to set alsa to record from line in on an ens1370? [03:22] omega_dinner (om...@om...) left irc: Ping timeout for omega_dinner[omegacs.net] [03:23] omega_dinner (om...@om...) joined #gstreamer. [03:24] Nick change: omega_dinner -> omega [03:24] <omega> wingo: set the mixer [03:25] <Apoc_> i need my bed .. cya ... [03:25] <omega> l8r [03:25] <wtay> omega: is there a doc about the timecache? [03:25] <omega> not really [03:26] Apoc_ (ap...@dy...) left irc: dodo [03:26] <wtay> what is a group? [03:26] <omega> a set of events that are closely related [03:27] <omega> i.e. as you play out, each sequential event is in the same group [03:27] <omega> if you seek, you create a new group [03:27] <wtay> oh ok [03:27] <omega> complete usage of the timecache will be difficult, and I haven't thought through it all yet anyway [03:30] <omega> any thoughts on the categorization I suggested? [03:31] <wtay> no objections.. [03:31] <omega> ok, any thoughts on how that should map to the plugin/module structure? [03:31] <wtay> cvs module? [03:32] <omega> yes [03:32] <wtay> plugins depending on external libs should go into a separate module no matter what [03:32] <omega> right [03:33] <omega> so an example in my email draft is dvd stuff [03:33] <omega> I'd think that dvdsrc, subtitle decoder, and any dvd-specific compositing stuff should go in the same module [03:33] <omega> what if the subtitle decoder eventualyl depends on an external lib? [03:33] <omega> should they still be in the same module? [03:33] <wtay> then it goes into a module on its own.. [03:33] <wtay> we need to draw a line.. [03:34] <wtay> 'cause we can go on discussing this forever.. [03:34] <omega> wtay: are you sure? dvdsrc and subtitle decoder aren't generally useful each on their own [03:34] <omega> right. what I haven't seen on the list yet this time is a list of all the plugins with categorization into modules and plugins [03:35] <omega> I'm working on some ideas, and will try to post a complete proposed structure and some rules tonight [03:35] <wtay> well. we could just say that the rule is to use common sense.. [03:35] <wtay> bah, I need to sleep :( [03:35] <omega> right, but common sense means totally different things to various people here [03:36] <wtay> you could easily drag mpeg2dec into the dvd module too then... [03:36] <wingo> i have a constant memory leak in alsasink. all i need to do after a peek is bytestream_flush, no? [03:36] <omega> yup [03:36] <wingo> hmm [03:36] <wtay> wingo: no, you must unref the buffer too [03:36] <omega> wtay: huh? [03:36] <wtay> wingo: or use peek_bytes [03:36] <wingo> i do [03:36] <omega> oh, yeah, unref the acquired buffer [03:37] <wingo> oh... [03:37] <wingo> ? [03:37] <wingo> i use peek [03:37] <wtay> I know, I saw the code :) [03:37] <wingo> ok :) [03:37] <wingo> peek_bytes instead ? [03:37] <wtay> I would do that, yes [03:37] <wtay> cya later.. [03:37] <wingo> cya [03:37] Nick change: wtay -> wtay-zZz [03:38] <wingo> so after a peek_bytes, i just flush and it's taken care of [03:38] <omega> should be, yes [03:41] <wingo> i'm still growing, only slower [03:41] <omega> hmm. another thing I'm gonna see if I can do tonight is resurect the gsttrace stuff [03:42] <omega> with which I'll instrument buffers and write some analysis tools to try to debug this stuff [03:42] <omega> try using memprof ? [03:43] <wingo> ah, i don't care for now, i'm about to head out :) [03:43] <omega> ok, if you commit it I'll look at it if I have a chance [03:44] <wingo> 10k/s or so [03:44] <wingo> ok [03:56] <vishnu> memprof doesn't like your cothreads, afaik [03:57] <omega> I've never had problems with it [03:58] <vishnu> really? heh, maybe i'll try it again. [03:58] <omega> what it has a problem with is libtool, as usual [03:59] <omega> but alias memprof='libtool --mode=execute memprof' [03:59] <omega> fixes that for me [04:00] <omega> I get lots of leaks running -register because of the XML save stuff, but it always exits right after, so.. [04:01] <omega> bleck. I get a lot of leaks in -inspect when it uses the registry. neat [04:01] <omega> looks like we need to audit the xml code used to load the registry [04:03] <omega> but no additional leaks in fakesrc ! fakesink, as expected [04:45] omega (om...@om...) left irc: Read error to omega[omegacs.net]: Connection reset by peer [05:47] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [06:50] chillywilly (da...@d1...) left irc: Ping timeout for chillywilly[d185.as8.nwbl0.wi.voyager.net] [06:55] Shippou (no...@ot...) joined #gstreamer. [07:24] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [08:14] Nick change: tnt-back-in-a-bit -> tnt [08:44] <tnt> Hello? [08:48] <vishnu> ya? [08:49] <tnt> Do you know if there are any existing filters that take 2 inputs and give 1 output... I looked but couldn't find any. [08:51] <tnt> (I seem to be getting a Segmentation Fault when I call the "gst_pad_push" function, and I'm not sure if it is my test program, or my plugin.) [08:51] <vishnu> tnt: hrm [08:52] <tnt> (If there was another filter with 2 ins, and 1 out, then I could test it to see if my test program worked... and see if there error was in my plugin or not.) [08:53] <tnt> These video transitions I want to would actually be written very quickly, if I could only get the first one to work. [08:54] <tnt> The rest would be basically a huge copy-and-paste job, with a change of a small little section for the algorithm. [08:54] <tnt> It's kinda frustrating :-( [08:56] <vishnu> tnt: hey, i sympathize. i spent like a month coming to grips with gstreamer before i got it working the way i want [08:56] <vishnu> tnt: there are still lots of rough edges. [08:56] <tnt> I keep on wondering if this is my bug... or a bug in gstreamer. [08:56] <vishnu> tnt: Erik & Wim are clueless about encapsulation, so the code is kind of crappy [08:57] <tnt> :-) [08:57] <tnt> There's times when there seems to be "race-conditions" going on in the code. [08:57] <vishnu> tnt: they gave me access to commit in cvs but revoked it recently because they didn't like my attitude ;-) [08:57] <tnt> :-) [08:58] <vishnu> tnt: yah, i looked at gstschedule a bit -- it's a total piece of crap, imho [08:58] <vishnu> still, i like the general gstreamer architecture. it's mind-boggling that there isn't any competition, i mean gstreamer is basically the only free software project of its kind [08:59] <tnt> I started coding something similar a while ago... but with the amount of spare time I have -- very little -- I've been looking to work with other people. [08:59] <vishnu> tnt: you can see i submitted patches on the mailing list this morning. i bet it will take them a few *days* to commit them. it's really frustrating [09:00] <tnt> My system leaned more towards using seperate processes, and various means of IPC (InterProcess Communication) like Shared Memory, etc. [09:00] <tnt> :-) [09:00] <tnt> I know all about frustration :-) [09:03] <vishnu> tnt: heh, well i put all my gstreamer code in a separate process so the toxic waste doesn't contaminate my project. i control things by sending stuff through pipes [09:04] <tnt> Putting stuff in seperate process is nice to, because it can take advantage of proccess migration (to other computers) found in MOSIX. [09:04] <vishnu> tnt: you've tried mosix? it sounds cool but i didn't try it yet [09:05] <tnt> I haven't tried it yet... but I'm very interested in it abilities and limitations... clustering and Multimedia can work very well together. [09:06] <tnt> But using MOSIX stuff for getting alot of computers to work together isn't necessarily the best way to do things... [09:07] <tnt> ... sometimes it is easier to just write your own custom dispatcher. [09:07] <vishnu> tnt: yah, probably. dunno. as far as your video stuff, you can figure it out yourself and submit patches. or if you can grovel enough then they will help you, but it sucks [09:07] <tnt> I would submit a patch if I had a clue what the error was,... but I don't :-( [09:09] <vishnu> how much programming experience do you have? [09:11] <vishnu> basically it's time to get more experience debugging ;-) [09:11] <tnt> I've got a BSc (Comp Sci was my major)... I know about 30 different language,... I've been out of University for 1 year... so I guess, if you count school: 5 years. [09:11] <vishnu> tnt: oh, you know enough [09:12] <tnt> I've read the code... but I don't see the error... [09:13] <tnt> It might be due to a seperate thread, or cothread, and that's hard to debug. [09:13] <vishnu> yah, so change it so it's more robust. add lots of assertions [09:13] <vishnu> #1 temporarily remove threads from your pipeline [09:13] <tnt> Have you seen the output with --gst-mask=-1 ? [09:14] <vishnu> tnt: yah, way too much output [09:14] <tnt> I've been looking at that, and it still doesn't help. [09:14] <tnt> I tracked down my error by adding g_print's in my code. [09:15] <vishnu> tnt: yah, g_print is good [09:15] <tnt> g_print, printf, cout <<, etc have always been my favourite way of debuging :-) [09:16] <vishnu> heh .. [09:30] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [09:52] Uraeus (csc...@c1...) joined #gstreamer. [09:52] <Uraeus> mornin [09:54] <tnt> Ummm... Morning... Evening... err what do you say at 1:00am? [09:54] <Uraeus> good night? ;) [09:55] <tnt> (I guess I should be saying "Good Night" :-) But I'm busy reading and coding :-) ) [09:55] <Uraeus> tnt: what are you working on? [09:57] <tnt> Write now I'm reading the spec for OpenML, trying to find info on OpenOffice's UNO, and looking for info on KParts (... I've heard KParts is a nice component architecture). [09:57] <tnt> I've taken a break from my gstreamer related coding. [09:57] <Uraeus> tnt: KParts is a embeding API not a real component architecture [09:58] <tnt> OK... [09:58] <tnt> ... work a look any ways. [09:58] <tnt> ... worth a look any ways. [09:58] <Uraeus> hehe [09:59] <Uraeus> tnt: if you want to do something component related then you should work on completing the work started by some of the OpenOffice hackers on doing an UNO - Bonobo bridge [10:00] <tnt> :-) ... I read a post at http://dot.kde.org/1002007199/1002471145/ ... this guy said good things about UNO and KParts, so I thought I'd learn more about them. [10:00] <Uraeus> tnt: http://whiteboard.openoffice.org/bonobo/index.html [10:00] <tnt> I'll read it. [10:05] <Uraeus> tnt: hmm, this Brad fellow can't have done as much research as he claims if one are to believe Iain answers [10:06] <Uraeus> but as for UNO I have heard Micheal Meeks (bonobo author) also give that high remarks [10:07] <tnt> None of these component systems ever seems to support streaming natively. [10:07] <tnt> Which is importsnt for multimedia stuff. [10:07] <tnt> The aRts project's MCOP is the only one I've ever heard of. [10:08] <Uraeus> tnt: doesn't CORBA do streaming nativly? I mean aRts used CORBA to begin with [10:09] <tnt> From what I've read about it, aRts does not use CORBA anymore because of that very reason -- because it doesn't support streaming. [10:09] <tnt> MCOP was made to replace it. [10:09] <tnt> I've yet to do any programming with aRts, but I plan to soon. [10:10] <Uraeus> tnt: a friend of my started doing something called http://gnostream.sourceforge.net/ which was supposed to do that [10:10] <Uraeus> tnt: I thought aRts stoped using CORBA cause Mico was a slowmo hog :) [10:11] <tnt> Maybe... maybe that and because it didn't support streaming natively. [10:12] <tnt> It would be nice if GObject component/object system supported streaming natively. [10:13] <tnt> But maybe that's what GstObject kinda is. [10:13] <Uraeus> hehe [10:13] <Uraeus> well problem with more features is that they tend to come with more overhead :) [10:14] <tnt> true. [10:15] <Uraeus> tnt: so what is it you want to acomplish/create? (I guess you are not messing with all these different things without a reason) [10:16] <tnt> Well, the project I'm working on is matterial. [10:16] <tnt> It's a movie compositing / NLE application. [10:16] <tnt> Originally I was planning to write something similar to gstreamer myself. [10:17] <tnt> But it's ALOT of work for someone doing it in his spare time. [10:17] <tnt> So I've been looking for others to work with. [10:17] <tnt> I'm learning about gstreamer now. [10:17] <tnt> And plan to write a simple NLE with it. [10:18] <tnt> And write the transition plugins (for gstreamer) that the NLE needs. [10:18] <tnt> But I want to look at other APIs too. [10:18] <tnt> aRts, dmSDK, etc.... [10:18] <tnt> Not to mention that I have some of my own ideas about how I think things should be done. [10:19] <Uraeus> hehe, seems everyone working in this area has some ideas of their own on how things *SHOULD* work ;) [10:19] <tnt> :-) [10:20] <tnt> Well, if everyone can work together, regardless of the differences in opinion, things would progress much faster. [10:22] <tnt> Hmmm, it seem UNO has some stuff about streaming... http://udk.openoffice.org/common/man/concept/streams.html [10:22] <Uraeus> tnt: true, but for that to happen I think one project needs to attain a clear lead over the others (which I think GStreamer is doing) and people must be willing to sacrifice some of their own *SHOULD*'s [10:23] <tnt> Ya. [10:23] <tnt> I don't mind adding things to GStreamer. [10:23] <tnt> Now if I could only get my program to work. [10:23] <Uraeus> that is a nice start :) [10:24] <tnt> :-) You wouldn't know why I get a Segmentation Fault when I call "gst_pad_push"? :-) [10:26] <Uraeus> tnt: sorry no, but if we manage to get wtay-zZz awake I bet he could help out [10:27] <tnt> I think he's in about the same Time Zone as me... so I will likely have to wait about 9 or 10 hours. [10:30] <ajmitch> hi [10:30] <tnt> Hey. [10:31] <tnt> Uraeus: We should get a map (of the world) and mark it with where everyone lives... so everyone will know when to get on IRC when they want to talk to someone. [10:31] <ajmitch> hehe [10:32] Action: ajmitch lives far far away from anyone else on irc here ;) [10:32] <tnt> Where? [10:32] <tnt> Australia? New Zealand? [10:32] <ajmitch> New Zealand [10:34] <Uraeus> tnt: no he is here with me in europe so he could come online any moment [10:34] <Uraeus> (I am in Norway and Wtay is in Belgium) [10:35] <ajmitch> Uraeus: this is wtay, tho ;) [10:35] <tnt> Uraeus: Hmmm, is it common for people to learn Englist in Norway and Belguim? [10:36] <Uraeus> tnt: yes, everyone has to learn it in school, we start at age 9 actually [10:37] <ajmitch> tnt: pity we only have to learn english here :( [10:37] <Uraeus> ajmitch: hmm, yeah sice it it wtay I guess 5-6 hours is a more accurate guess :) he is now morning bird [10:37] <tnt> Uraesus: Here, in Canada, in Elemenary school, they make us learn French... but I really don't really remember much of it. [10:38] <Uraeus> tnt: at age 13 we have to start learning french, spanish or German (most people choose french or german) [10:38] <tnt> ajmitch: Where are you from? England? [10:38] <Uraeus> tnt: I know, so you can speak with your Quebecan kin [10:39] <tnt> :-) The thing is, the Quebecers can't understand the people -- the English speakers -- that learn french from school that way. They don't teach the Quebec dialect of French. [10:40] <ajmitch> tnt: no, NZ has english as the official language [10:40] <ajmitch> tnt: NZ is a commonwealth country like canada ;) [10:40] <Uraeus> ajmitch: how large percentage is the maori population? [10:41] <tnt> ajmitch: Oppps... I misunderstood Uraesus... when he said ""tnt: no he is here with me in europe so he could come online any moment" I thought he was talking about you. [10:41] <tnt> But he was infact talking about wtay. [10:41] <Uraeus> yup [10:41] <tnt> (I know they speak English in New Zealand.) [10:42] <ajmitch> Uraeus: hmm, not sure, maybe somewhere between 10%-25% are part Maori ;) (could be more) [10:42] <tnt> They don't teach the Morai (sp!) language? [10:42] <ajmitch> Maori is being taught in schools [10:42] <ajmitch> and is quite a big part of NZ culture still [10:43] <Uraeus> ajmitch: so you can speak a little Maori? [10:44] <ajmitch> Uraeus: not much unfortunately [10:44] <ajmitch> Uraeus: i haven't learnt any since primary school [10:45] <Uraeus> ajmitch: well I guess you still will be responsible for the GStreamer Maori translation then when we get i18n support :) [10:45] <ajmitch> Uraeus: don't worry, i have friends that can speak it quite well ;) [10:45] <ajmitch> (and Samoan ;) [10:46] <ajmitch> it would mean that i'd have to do some work on gstreamer (the shock! the horror!) [10:46] <tnt> Maori and Samoan are both Polynesian languages, aren't they? So there probably quite similar. [10:46] <Uraeus> hehe, well my friend Kjartan which like me is norwegian and leads the GNOME i18n project is trying to find some people here to translate GNOME into samii [10:47] <ajmitch> tnt: yep [10:48] <tnt> What's Samii? Don't they speak Svenska in Sweden? [10:48] <ajmitch> Uraeus: what about gstreamer in gaelic? ;) [10:49] <Uraeus> tnt: yes, Samii are the 'native' population of Scandinavia whose traditional way of life is reindeer herding [10:49] <tnt> ajmitch: I'm part Scotich... so maybe I can get a hold of some of my cousins in Scottland to do it ;-) [10:49] <Uraeus> tnt: swedes speaks swedish (svensk) [10:49] <ajmitch> tnt: heh, Dunedin (where i live) is called Edinburgh of the South [10:50] <tnt> Uraesus: I see... I also thought it refered to the language. [10:50] <tnt> s/also/slways/ [10:50] <tnt> s/also/always/ [10:51] <tnt> ajmitch: Is there alot of Scottish blood in New Zealand? [10:51] <Uraeus> of course I am not sure how correct it is to coin the samii more native than us(the norse population) since we too have lived here for over 2000 years [10:52] <ajmitch> tnt: far more so down south here [10:52] <tnt> Hmmm. [10:52] <ajmitch> tnt: this city & province was founded by many scots [10:53] <tnt> During the Clearings??? [10:53] <Uraeus> ajmitch: do you have a kilt? [10:53] <ajmitch> bout 1848 or so [10:53] <ajmitch> Uraeus: not with me ;) [10:53] <Uraeus> hehe [10:53] <ajmitch> Uraeus: nah, i don't have one [10:53] <tnt> My cousin was suppose to being over my family kilt -- the Kerr Clan -- but he forgot the last time he came over. [10:53] <ajmitch> Uraeus: i've worn one during a performance for the Highland & Gaelic society here ;) [10:54] <ajmitch> they have kiltmakers here [10:54] <Uraeus> I have wondered about getting myself a traditinal suite but I am not sure what, in Oslo there is little tradition for it and being mixed norwegian/german I feel even less connected [10:55] <ajmitch> Uraeus: don't worry, i'm part dutch as well ;) [10:55] <tnt> Uraesus: :-) Ya, I'm a mutt too. [10:56] <tnt> [That should be Uraeus and not Uraesus of course.] [10:56] <Uraeus> I used to have a leatherhosen as a child, but I can't really see myself in it today [10:56] <Uraeus> and a traditional norwegian suit is very expensive, but I could use it for many occasions in norway [10:58] <Uraeus> but if I move from Norway I am not sure how often I would use it (on the other hand maybe it would be cool to show up at parties around the world in it :) [11:00] <ajmitch> hehe [11:00] <ajmitch> coming to NZ for a visit sometime? ;) [11:02] <Uraeus> ajmitch: well if I start working in China I very well might fly down sometime :) [11:02] <ajmitch> Uraeus: heh, ok ;) [11:02] <Uraeus> ajmitch: I was not joking [11:03] <ajmitch> we should hold a gstreamer meeting here next month, i think steveb will be here for a few days/weeks ;) [11:03] <Uraeus> ajmitch: hmm, I will be in Oakland, US for ALS next month so I guess that does not put me in the neighbourhood :) [11:03] <ajmitch> Uraeus: NZ is a nice place to visit, anyway, and i think i'll be here for a few years anyway :) [11:04] <tnt> How about holding it in BC, Canada :-) [11:04] <ajmitch> tnt: too far to swim ;) [11:04] Nick change: Uraeus -> Ura_brb [11:08] <ajmitch> tnt: so you have scottish heritage, i take it? [11:09] <tnt> I'm 1/4 Scotish... my Father's Mother. [11:09] <ajmitch> k [11:09] <tnt> I think most people in Canada have at least some Scottish. [11:10] <ajmitch> i'm possibly a bit more than that, not sure where one of my grandparent's family is from ;) [11:10] <tnt> Alot of places here are or were named 'New Scottland'. [11:10] <tnt> We have Nova Scotia. (Latin) [11:10] <ajmitch> Edinburgh in Gaelic is Dun Eideann, i live in Dunedin :) [11:10] <tnt> BC used to be called New Caladonia (sp!). [11:11] <ajmitch> i come from a town called balclutha (a gaelic name) [11:11] <tnt> I've heard of others... but I can't remember them right now. [11:11] <ajmitch> this city has the only 2 castles in the southern hemisphere afaik ;) [11:12] <tnt> Kool! [11:12] <tnt> We're they ones flowin in from other parts of the world (like the ones in North America), or do they originate there? [11:13] <tnt> s/We're/Were/ [11:13] <ajmitch> sorry? [11:14] <tnt> Huh? [11:14] <ajmitch> http://www.visit-dunedin.co.nz/history-scots.html [11:45] Ura_brb (csc...@c1...) left irc: Client Exiting [11:58] steveb (st...@no...) joined #gstreamer. [11:59] mef (ma...@us...) joined #gstreamer. [11:59] mef (ma...@us...) left #gstreamer. [12:00] <tnt> I'm tired. Gonna go to bed. See ya. [12:00] tnt (tn...@h2...) left irc: Client Exiting [12:52] thomasvs (th...@ad...) joined #gstreamer. [12:58] harobed (harobed@ACA79B75.ipt.aol.com) joined #gstreamer. [13:03] thomasvs (th...@ad...) left irc: Read error to thomasvs[adsl-64382.turboline.skynet.be]: Connection reset by peer [13:04] chillywilly (da...@d1...) joined #gstreamer. [13:27] Nick change: wtay-zZz -> wtay [13:28] <wtay> yo [13:28] <chillywilly> yo [13:28] <wtay> yay! 1e+02% function docs coverage (443 functions documented, 0 not documented) [13:28] <wtay> :) [13:28] <chillywilly> what is that output from? [13:29] <ajmitch> hey wtay [13:29] <wtay> gtkdoc [13:29] <wtay> hi [13:29] <wtay> never tested probably :) [13:30] <chillywilly> ah [13:30] thomasvs (th...@ad...) joined #gstreamer. [13:32] <steveb> wingo: there? [13:39] <harobed> how to generate the documentation with gtkdoc and gstreamer CVS version ? [13:41] <wtay> harobed: do make html in the docs/gst dir [13:41] <harobed> thanks [13:41] <wtay> harobed: be warned; it's not easy, you need lots of tools for it [13:42] <wtay> I'm going to upload a new version to gst.net when I'm satisfied with the docs.. [13:43] <wtay> hopefully today.. [13:43] <harobed> ok, I wait [13:43] <wtay> debian unstable? [13:43] <harobed> yes [13:44] <wtay> ii gtk-doc-tools 0.7-1 GTK documentation tools [13:44] <wtay> ii jade 1.2.1-21 James Clark's DSSSL Engine [13:44] <wtay> ii docbook-stylesheets 1.49-1 Modular DocBook stylesheets, for print and HTML [13:44] <harobed> It's installed [13:44] <wtay> ii docbook 4.1-6 SGML DTD for authors of technical documentation [13:45] <wtay> ii docbook-utils 0.6.9-3 Convert Docbook files to other formats (HTML, RTF, Postscript, PDF) [13:45] <wtay> I think that should do.. [13:48] Uraeus (csc...@c1...) joined #gstreamer. [13:48] <wtay> yo [13:48] <Uraeus> hi wtay [13:51] chillywilly (da...@d1...) left irc: Read error to chillywilly[d134.as7.nwbl0.wi.voyager.net]: EOF from client [13:52] <Uraeus> wtay: was tnt still online when you came online? [13:52] <wtay> no [13:52] <Uraeus> ah, well he was having some troubles with his GStreamer app which he needed some advice on [13:53] <wtay> yeah, I saw that.. [13:53] <harobed> gstreamer work only on linear data stream ? [13:55] <wtay> harobed: no, it should do seeking soon too (if that's what you mean with linear stream) [13:56] <thomasvs> Uraeus: this may be a stupid question, but how do you do advocacy for a project ? [13:56] <thomasvs> Ureaus: meaning, how do I attract developers without drawing too much attention ? [13:56] chillywilly (da...@d8...) joined #gstreamer. [13:56] <harobed> no, sorry, i don't speak about stream, but I speak about file video [13:57] <wtay> harobed: you mean, mix two streams? [13:57] <wtay> harobed: or audio + video playback? [13:58] <harobed> wtay: no, work on part of file [13:58] <wtay> harobed: that involves seeking, so yes [13:58] <harobed> wtay: it's the gnonlin do that ? this program is one extention of gstreamer ? [13:59] <wtay> harobed: yes, gstreamer doesn't automatically change the pipeline and such on specific events [13:59] <wtay> harobed: you need a fair amount of code on top of gst to handle that, and that's what gnonlin does [14:00] <wtay> s/does/is going to do/g :) [14:02] <Uraeus> thomasvs: well if you don't want to attempt to get new developers by massive announcements and releases etc,. I think talking to people online and trying to get them exited about the project is the way to go [14:02] <harobed> with gnonlin extension, we can for example : get the length of video file ? [14:03] <wtay> harobed: you should be able to do that with gst too. [14:03] <thomasvs> Uraeus: I don't mind releasing, on the contrary, I just want there to be lots of stuff to work on before attracting more public interest [14:03] <thomasvs> If you go slashdot without anything solid to offer people will probably not give you a second chance later on [14:03] <wtay> harobed: but length is a fuzzy concept in general.. [14:04] <thomasvs> Uraeus: and I'm new at this anyway so ;) [14:04] <harobed> length in frame [14:04] <harobed> wtay: no, frame length [14:04] <Uraeus> thomasvs: true, another alternative you could try was discretly trying to position your project on related mailing-lists [14:04] <wtay> harobed: impossible if you are streaming from a network source, hard for mpeg1/mpeg2 [14:04] <thomasvs> Uraeus: yeah, I was planning on doing that. [14:04] <Uraeus> thomasvs: assiciation with popular projects is also a good alternative [14:05] <wtay> harobed: easy for avi [14:05] <harobed> wtay: but if the source is one file ? [14:05] <thomasvs> are there lower-profile sites than slashdot around ? [14:05] <Uraeus> thomasvs: linuxtoday and lwn are alternatives [14:05] <wtay> harobed: depends on the media file, like I said, mpeg1 is kinda hard to get the number of frames from [14:05] <Uraeus> thomasvs: linuxgames.com if it has some game related uses [14:06] <Uraeus> thomasvs: gnotices if it is gnome related, dot.kde.org if it is KDE related [14:06] <Uraeus> thomasvs: talking about it in your advogato.org diary could also be a good move [14:06] <harobed> wtay: I look the file design of gnonlin project, and I don't see a pad for select one part of file source ? or moving to moment [14:07] <thomasvs> Uraeus: thanks. Some good ideas to start out with ;) [14:07] <wtay> harobed: nope, gnonlin has no real code yet, just an object structure [14:07] <wtay> harobed: the idea is that you add a media file to and specify a start/end point in the file [14:08] <harobed> wtay: excuse me, but one "pad" is one "function" for control the pluging or pipeline ? [14:09] <wtay> harobed: pads are objects on the element where data enters/leaves the element [14:09] <wtay> harobed: data can be events (seek/flush...) or buffers [14:09] <thomasvs> Uraeus: is advogato.org reasonibly well-known ? I only found about it pretty recently [14:10] <Uraeus> thomasvs: think so, most of the 'big brass' has accounts there (and I of course) [14:28] <steveb> is disksrc officially deprecated yet? [14:28] <wtay> yeah, it is IMO [14:29] <steveb> so it shouldn't be used or debugged? [14:29] <wtay> does anybody think that the bufferpool API docs should have a bufferpool example? [14:29] <wtay> steveb: I wouldn't bother, no.. [14:29] <steveb> ok [14:29] <wtay> it's highly inefficient too.. [14:29] steveb (st...@no...) left irc: Read error to steveb[node1ee10.a2000.nl]: EOF from client [14:51] steveb (st...@no...) joined #gstreamer. [15:37] chillywilly (da...@d8...) left irc: [15:47] wtay (wi...@ca...) left irc: Read error to wtay[cable-195-162-215-201.upc.chello.be]: EOF from client [15:47] wtay (wi...@ca...) joined #gstreamer. [15:47] <wtay> heh xchat crashed :) [15:47] Nick change: wtay -> wtay-zZz [15:48] Nick change: wtay-zZz -> wtay-tv [15:48] <wtay-tv> :) [15:51] <wingo> morning all! [15:51] <wingo> foggy outside [15:51] <wingo> bit pot of grits with eggs and cheese [15:51] <wingo> yum! [15:51] Action: wingo sings [16:07] <Uraeus> wingo: glad you are happy today wingo :) [16:07] Nick change: Uraeus -> Ura_away [16:08] wingo (wi...@rd...) left irc: gotta roll. [16:32] wingo (wi...@rd...) joined #gstreamer. [17:25] ajmitch (aj...@p6...) left irc: Ping timeout for ajmitch[p61-max10.dun.ihug.co.nz] [17:38] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [18:14] dobey (do...@dr...) joined #gstreamer. [18:24] <wingo> hey dobey. [18:25] <dobey> hey [18:26] <steveb> hi [18:39] Shippou (no...@cr...) joined #gstreamer. [18:41] <wingo> hi steveb, i haven't had a chance to re-test ladspa [18:42] <wingo> are you using the debian-packaged plugins? [18:42] <dobey> hrmm [18:42] <wingo> hrmm? [18:43] <steveb> wingo: have you tried that pipeline i emailed you? it works for me [18:43] <dobey> just wondering what to do today [18:43] <dobey> waiting for people in california to get online [18:43] <steveb> all my ladspa plugins are compiled from source [18:51] <wingo> i'll try that when i get home, i'm writing a paper atm [18:51] <wingo> i'm using debian plugins, that could possibly be the issue [18:51] <wingo> it's something funny due to the linker i think [18:52] <steveb> yeah, try compiling them [18:53] <wingo> will do. [18:54] <wingo> what situation was causing double-speed playback with alsasink for you? [18:54] <wingo> or, what pieline rather [18:54] <wingo> pipeline even [18:54] <wingo> i'd like a pieline :-) [18:55] <steveb> foostereo ! stereo2mono ! alsasink [18:55] <steveb> it looks like alsasink thinks its a stereo signal and outputs both channels or something [18:56] <wingo> hmm [18:56] <wingo> oh [18:56] <wingo> oh yeah [18:56] <wingo> ok, i'll fix that [18:56] <wingo> there's also a neat request pad syntax that's good for requesting one channel or another [18:56] <steveb> oh yeah, i haven't tried that yet [18:56] <wingo> if you have an 8 channel card and you want the fifth channel, request sink4 [18:57] <steveb> ha. i wish [18:57] <wingo> me too [18:57] <steveb> ;( [18:57] <wingo> omega has a box with a hammerfall [18:57] <wingo> make me drool :-] [18:57] <steveb> have you got state change handling code in alsa yet? [18:57] <wingo> just not pause [18:57] <wingo> as you saw :) [18:57] <wingo> i forgot about it [18:59] <wingo> have you thought about making guilaunch like a test bed for plugins, with sliders and such? [18:59] <wingo> cf the thread on l-a-d :) [18:59] <wingo> i have an idea! [19:00] <steveb> absolutely. thats what i'm working on right now [19:00] <wingo> is there any interest in a widget that will control all the params on a plugin, for common use in gst apps? [19:00] <wingo> i have my own in beatbox [19:01] <steveb> cant really do sliders for args yet as the range param_specs currently go into a hole in the shim [19:01] <wingo> you can see some shots of it at ambient.2y.net/beatbox/ [19:01] <wingo> it can use some improvement, but hey [19:01] <wingo> yeah [19:01] <wingo> i have to do wierd hacks with ladspa.h :-\ [19:01] <steveb> but a decent gui can be whipped up from dparams specs [19:01] <wingo> really? i really have to sit down and grok the dparam api [19:02] <steveb> I'd like to see a libgstui for common widgets [19:02] <wingo> yeah. [19:02] <wingo> where would it go? [19:02] <steveb> libs/ui [19:02] <wingo> gtk only? [19:02] <wingo> libgstgtkui? :-) [19:03] <steveb> gtk is implicit ;) [19:03] <wingo> all others be damned! [19:04] <steveb> not really, but in gstreamer CVS yes. [19:04] <wingo> maybe libgstui-gtk. [19:05] <wingo> i know that others do have an interest in other toolkits, it would be nice if contributed efforts could go into cvs... [19:05] <steveb> yeah. omega might want it in a seperate modult though - since most widgets won't apply to player apps, only editors. [19:06] <steveb> s/modult/module [19:18] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [19:28] thomasvs_ (thomas@212.100.172.175) joined #gstreamer. [19:31] Nick change: dobey -> dobey-eat [19:31] mef (ma...@us...) joined #gstreamer. [19:31] mef (ma...@us...) left #gstreamer. [19:36] vektor (reet@HSE-Kitchener-ppp3508308.sympatico.ca) left irc: too bad the scene is dead [20:18] Shippou (no...@cr...) joined #gstreamer. [20:36] Nick change: dobey-eat -> dobey [20:57] wtay-tv (wi...@ca...) left irc: Read error to wtay-tv[cable-195-162-215-201.upc.chello.be]: EOF from client [21:04] wtay (wi...@ca...) joined #gstreamer. [21:04] #gstreamer: mode change '+o wtay' by ChanServ!s@ChanServ [21:04] #gstreamer: mode change '-o wtay' by wtay!wi...@ca... [21:04] tnt (tn...@h2...) joined #gstreamer. [21:04] <tnt> Hello all. [21:04] <wtay> hi [21:04] <wingo> hello tnt [21:04] <wingo> hi wtay [21:06] <tnt> I've been trying to figure out why I've been getting a Segmentation Fault,... and having my program crash... [21:06] <wtay> ? [21:07] <tnt> I haven't been sure if it was my plugin (library), my test program, or maybe even gstreamer... [21:07] <tnt> By placing g_print's everywhere, I've found that calling "gst_pad_push" in my Loop-function is what is causing the Seg Fault... [21:08] <wtay> when, wherE? [21:08] <wtay> what plugin is it connected to? [21:08] <tnt> Anyone have a clue as to why this might happen? [21:08] <tnt> A plugin that I have made. [21:08] <wtay> to what is it connected? [21:09] <wingo> didn't you connect it to a queue, tnt? [21:09] <tnt> The connections look like this... [21:09] <wingo> i can't recall [21:09] <tnt> filesrc_1 -> mpeg2dec_1 -> queue_1 -v [21:09] <tnt> videomix -> queue_3 -> sdlvideosink [21:09] <tnt> filesrc_2 -> mpeg2dec_2 -> queue_2 -^ [21:09] <tnt> Like that above. [21:10] vektor (reet@HSE-Kitchener-ppp3508308.sympatico.ca) joined #gstreamer. [21:10] <wtay> does the connection succeed? [21:10] <wingo> what's the last few lines of the debug output? [21:10] <vektor> hmmm [21:10] vektor (reet@HSE-Kitchener-ppp3508308.sympatico.ca) left #gstreamer. [21:10] <tnt> wtay: It seems to... I get all the way in the loop-function until I call "gst_pad_push", and then it Seg Faults. [21:11] <tnt> wingo: hang on... [21:11] <wtay> tnt: you have a gdb trave? [21:11] <wtay> trace even.. [21:11] <tnt> Using --gst-mask=-1 I get... [21:12] <tnt> cothread_switch:346: exit thread #0 0 0xbf7ff967<->0x20294328 (1621707201) [21:13] <tnt> cothread_switch:350: set stack to 0xbf63fffc [21:13] <tnt> cothread_switch:353: in thread [21:13] <tnt> cothread_switch:342: endther thread #2 1 0xbf63fffc<->0xbf640000 (4) [21:14] <tnt> null)mgst_schedule_loopfunc_wrapper:35: (0,'decoder'): entering [21:14] <tnt> gst_schedule_loopfunc wrapper:38: calling loopfunc _init for element decoder1 [21:15] <tnt> gst _mpeg2dec_loop:379:MP [21:15] <tnt> And then it just froze. [21:15] <tnt> (After pressing enter I got the command line though.) [21:15] <wtay> oh [21:15] <tnt> wtay: gdb... what? [21:16] <wtay> videomix runs in its own thread? [21:16] <wtay> tnt: run with libtool gdb [21:16] <wtay> libtool gdb <app> [21:16] <tnt> wtay: In the test program, I put the first line in its own thread... [21:16] <wtay> depending on whether you use libtool that is... [21:16] <tnt> ... the last line in its own thread... [21:17] <tnt> ... and the middle one directly into the pipeline. [21:17] <tnt> wtay: OK, I'll try gdb. [21:17] <wtay> tnt: what if you put them all into a thread? [21:18] <tnt> wtay: I had the middle line in its own thread before, but that made it so nothing happened at all. [21:18] <tnt> I guess it a bug. [21:18] <wtay> so it seems, yes [21:19] <tnt> I wish there was another filter with 2 inputs and 1 output,... then I could tell if it was my plugin or my test program that had the bug. [21:19] <wtay> videomix a buffer from each of its two sinkpads? [21:20] <wtay> tnt: aggregator can do that (untested though) [21:20] <tnt> wtay: Ya, videomix is doing a gst_pad_pull for each input (sink) pad. [21:21] <tnt> wtay: aggregator... OK, I'll look it up. [21:21] <wtay> in gst/elements/ [21:21] <wtay> it has a request pad to make it create N sinks [21:21] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [21:22] <tnt> wtay: what does its single source pad give? (Just out of curiousity.) [21:23] <wtay> it just interleaves the buffers as they come in [21:23] <wingo> adder can do 2->1 as well [21:23] <tnt> OK. [21:23] <wtay> ah yes [21:23] <wtay> adder works [21:23] <wingo> you hae to request its sink pads [21:23] <tnt> wingo: addr... I'll look at that too. (Thanks) [21:23] <wingo> request "sink%d" [21:24] <wtay> tnt: a bit more complicated as it also takes care of buffer sizes etc.. [21:24] <wtay> hmm tv.. [21:24] Nick change: wtay -> wtay-tv [21:27] harobed (harobed@ACA79B75.ipt.aol.com) left irc: Ping timeout for harobed[ACA79B75.ipt.aol.com] [21:32] mef (ma...@us...) joined #gstreamer. [21:32] <mef> howzit? [21:33] <mef> wingo: did I hear right that you were trying to get gstreamer working on windows? Or was that just a joke by (hmmm).. I forget who mentioned it. [21:33] <wingo> yo [21:33] <wingo> must have been a joke! [21:33] <mef> ok [21:33] <mef> oh well [21:34] Action: wingo is not a masochist :) [21:34] <mef> too bad... we need one! [21:34] <wingo> don't we all! er... um... [21:35] <mef> I was just getting excited over the fact of having gstreamer act as a "game" on an X-box and to stream audio/video to it from a PC or the internet. [21:36] <mef> For that to work requires that gstreamer function with DirectX 8 on the stripped down W2K kernel that is used by the Xbox. [21:37] <tnt> I'm having trouble requesting a pad from aggregator... is it correct to make the request by just calling "gst_element_connect(queue_1,"src",aggregator,"sink1");"? [21:46] harobed (harobed@ACAF5310.ipt.aol.com) joined #gstreamer. [21:46] ajmitch (aj...@p6...) joined #gstreamer. [21:46] evil_monk (trevo@208.141.162.68) joined #gstreamer. [21:48] <wingo> tnt: no [21:48] <wingo> see the application developer's manual [21:48] <tnt> sink%d ??? [21:49] <wingo> the app developer's manual has a section on request pads that's pretty complete [21:49] <tnt> OK, I'll read it. [21:52] mef (ma...@us...) left #gstreamer. [22:15] <tnt> Hmmm, aggregator doesn't seem to want to produce any output (in the sdlvideo sink) :-( [22:16] <wingo> i've never looked at that element, wtay wrote it afaik [22:16] <tnt> :-( [22:16] <tnt> Would you know what this means... [22:16] <tnt> gst_buffer_new:62: creating new buffer 0x8108e30 [22:16] <tnt> gst_mpeg2dec_vo_frame_draw:189: out: -1 3 40000 [22:17] <tnt> null)mgst_pad_push:1450: (decoder2:src): entering [22:17] <tnt> gst_pad_push:1456: calling chainhandler &gst_queue_chain of peer pad queue_2:sink [22:17] <tnt> gst_queue_chain:283: [queue_2] adding buffer 0x8108bb0 of size 118272 [22:17] <tnt> gst_queue_chain:328: [queue_2] waiting for space, level is 100 [22:17] <tnt> Did it run out of memory??? [22:17] <wingo> no, i think it's full for some reason [22:18] <wingo> i've never really used a queue [22:18] <wingo> but it has an internal buffer that it can fill i think [22:18] <wingo> maybe it's waiting for a pull [22:18] <dobey> hmm [22:18] <wingo> so in short, dunno :) [22:19] <tnt> Hmmm, I wonder what wtay is watching on TV. [22:19] <wingo> something belgian :-] [22:19] <tnt> They speak French there, don't they? [22:20] <wingo> i think he's in the flemish part [22:20] <dobey> heh [22:20] <dobey> they speak french and dutch mostly [22:27] Shippou (no...@cr...) joined #gstreamer. [22:43] steveb (st...@no...) left irc: advance! [22:49] dobey (do...@dr...) left #gstreamer (eh). [22:53] Nick change: Ura_away -> Uraeus [23:02] sienap (sy...@s3...) joined #gstreamer. [23:02] <sienap> Goeiendag [23:02] <Uraeus> hi sienap [23:02] sienap (sy...@s3...) left irc: No windows for this server [23:03] sienap (sy...@s3...) joined #gstreamer. [23:03] <sienap> Dutch Militair Police are plain nazis [23:03] <sienap> so you guys know :) [23:04] <sienap> uraeus do you have any information about a release date ? [23:04] <sienap> so i can update my svgasink someday [23:04] Procule (Ac...@cn...) joined #gstreamer. [23:05] <Procule> hi there [23:05] <Uraeus> sienap: release date of new gstreamer? [23:05] <sienap> hi... there [23:05] <Uraeus> hi Procule [23:05] <sienap> uraeus well yeah.. :) [23:05] <tnt> Procule: Howdy [23:05] Zeenix (ze...@13...) joined #gstreamer. [23:05] <Uraeus> sienap: not really, but since the event stuff is now merged I think we are very close :) [23:05] <sienap> zeenix!!! [23:05] <sienap> how are you ? [23:05] <sienap> uraeus 1 month ? [23:05] <Zeenix> fine [23:05] <sienap> geez it is getting bad there huh ? [23:06] <sienap> i had email from the rights shop [23:06] <Uraeus> sienap: no less, maybe next week ar the week after [23:06] <sienap> she wasn't able to find REAL information about coming as a refugee and then start studying [23:06] <sienap> anyway she thinks the change is really little that it is going to work. [23:06] <sienap> uraeus mym y ;) [23:06] <Zeenix> sienap: ok, whose "she" ? [23:06] <Uraeus> sienap: huh? [23:07] <sienap> zeenix the girl from the rights shop. [23:07] <Zeenix> Uraeus: he is talking to me [23:07] <sienap> indeed.. query [23:07] sienap (sy...@s3...) left #gstreamer. [23:07] <Procule> I see on the website that the v4l support is not fully usuable ? [23:07] <Uraeus> Zeenix: doubt he meant the -> Uraeus mym y sentence for you :) [23:08] <Uraeus> Procule: could be some bitrot problems with it [23:08] <Zeenix> Uraeus: not that one of course [23:08] <Procule> but does it work "well" ? [23:08] <wingo> i think vektor might be familiar with it, i dunno [23:09] <wingo> he's a compatriot of yours :) [23:16] <Procule> when was version 0.2.1 released ? [23:17] <Uraeus> Procule: some time ago, 2 months maybe [23:17] <tnt> Procule: It was released on June 27th. [23:17] <Uraeus> oops, a lot more than 2 months :) [23:18] <tnt> (GStreamer really needs a new release.) [23:18] <Zeenix> Uraeus: did you get the news that i have become a plug-in writer ? <g> [23:18] <Procule> So... if I wanna make an app with gstreamer, I should use the CVS ? [23:18] <tnt> Procule: I found that I had to. [23:19] <Uraeus> Procule: absolutly, but a new release is soon to come [23:19] <Zeenix> Procule: depends [23:19] <Uraeus> Zeenix: cool that you have become a plug-in writer [23:19] <Procule> ok [23:20] <Procule> I've download the CVS version then ;) [23:20] <Procule> ll [23:20] <Zeenix> Procule: i have both of them compiled , i install the cvs when i need it, else i have 0.2.1 installed [23:21] <tnt> Procule: :-) You'll have all sorts of fun compiling the CVS version. [23:21] <Procule> really ? [23:21] Nick change: wtay-tv -> wtay [23:22] <wtay> tnt: you cannot connect aggregator to sdlvideosink since it expects vide data [23:22] <tnt> You'll need to update some things on you're system, install some things, and do some other stuff. [23:22] <wtay> s/vide/video [23:22] <tnt> wtay: What does aggregator return then? [23:23] <Procule> If I use the cvs version, I guess my app will work with the next version of gstreamer ? [23:23] <wtay> tnt: an untyped buffer most likely [23:23] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-38.brain.net.pk] [23:23] <tnt> Procule: It will make things easier, for you, in the long run. [23:24] <tnt> wtay: :-( [23:24] <wtay> tnt: whicb is why queue doesn't want to connect to sdlvideosink and the queue is not emptied [23:24] <wtay> tnt: why don't you use fkaesink? [23:24] <wtay> fakesink even [23:24] <tnt> wtay: Hmmm, why doesn't gstreamer simply NOT allow the connection then... and give some kind of error. [23:25] <wtay> tnt: it doesn't allow the connection that's why no data flow is happening [23:25] <tnt> wtay: fakesink... good idea. [23:25] <wtay> tnt: the error is another thing because an app might catch the signal and do something to solve the problem [23:25] <tnt> wtay: IMO, GStreamer needs a little better error handling. (Maybe it's coming though.) [23:25] <wtay> tnt: tso we don't spew lots of errors on caps nego failure [23:25] <wtay> tnt: yes [23:26] <tnt> wtay: I'll go give fakesink a try... and hopefully that will work. THanks. [23:27] <Uraeus> Procule: yes, and wtay here is one of the core developers so he can better answers your gstreamer questions than me [23:28] <tnt> Procule: If (probably when) you have trouble getting the CVS version working, just ask here... I recently switched over to the CVS version, and I remember all the problems and the solutions to them. [23:28] <wtay> Procule: the api should be considered stable at around version 0.5.0 mind you.. [23:28] <Procule> ./configure: line 6270: syntax error near unexpected token `PKG_CHECK_MODULES(GLIB2,' [23:28] <Procule> ./configure: line 6270: ` PKG_CHECK_MODULES(GLIB2, glib-2.0 gobject-2.0 gthread-2.0 gmodule-2.0,' [23:29] <tnt> Procule: You need to run ./autogen.sh first. [23:29] <Procule> I did [23:29] <wtay> Procule: you need pkgconfig [23:29] <wtay> Procule: version 0.8.0 at least [23:29] <Procule> I have it [23:29] <wtay> 0.8.0? [23:29] <Procule> yes [23:30] <Procule> oh [23:30] <tnt> Procule: Is it installed under /usr or /usr/local ??? [23:30] <Procule> I guess /usr/local [23:30] <tnt> Procule: Install it under /usr ... or it won't work. [23:30] <Procule> ok [23:30] <wtay> that's another error.. [23:31] <wtay> oh, maybe not.. [23:31] <tnt> Procule: You have to uninstall the version you already have with "make uninstall" first. [23:31] <Procule> arf I've erased the sources tree [23:31] <wtay> you can add an extra include path for automake too IIRC [23:31] <wtay> autoconf even [23:32] <tnt> Procule: Hmmm, that's not good... I'm don't know how to uninstall it without the source three. [23:32] <Procule> I redownloaded it [23:32] <wtay> taaz: do you remember the flag to add something to autoconf path? [23:33] <tnt> Procule: (NOTE: I'm guessing here!!!) Well, go through the steps again... ./configure , make , make install ... and then after that do a "make uninstall". [23:33] <Procule> I did a configure and a make uninstall and it worked [23:34] <Procule> I reattempting the autogen.sh [23:34] <Procule> it worked [23:34] <tnt> Procule: Now type "make" (if you haven't already). [23:35] <Procule> I think I'll eat pizza tonight [23:35] <tnt> :-) [23:35] <Procule> programmer'S favorite meal [23:35] <Procule> haha [23:35] <Procule> with a 2L of coke [23:35] <Procule> arf [23:35] <Procule> only 30 megs left [23:35] <Procule> I really need a new hard drive [23:36] <wtay> uhm.. [23:36] <wtay> du . == 150940 here.. [23:37] <Procule> arf [23:41] <tnt> wtay: Using fakesink doesn't change things... I still get the same message from '--gst-make=-1' (although it is with queue_1 this time, and not queue_2)... does this mean that there's a bug in aggregator??? [23:41] <Procule> humph [23:41] <Procule> first error in compiling [23:43] <wtay> tnt: oh, right aggregator has some args you need to set [23:43] <wtay> tnt: try sched=1 [23:43] <tnt> wtay: OK. [23:46] <Procule> aaah I'm too tired for that today [23:46] <Procule> I'll check the news [23:46] <Procule> thank you [23:46] Procule (Ac...@cn...) left irc: Butez moi avant que je mange toute le ragout !!!!!!!!!!!! [23:48] <tnt> wtay: :-( No change,... same error. [23:48] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [23:50] <wtay> tnt: I've had bad experiences with mixing too lately (that's why I built aggregator) [23:51] <tnt> wtay: Do you of anyone who has gotten mixing to work??? [23:51] <tnt> wtay: If you have, then maybe I could look at their code. [23:51] <wtay> tnt: yes, examples/mixing works AFAIK [23:51] <tnt> wtay: Thanks, I'll go look at it. [23:52] <wtay> oh, I know what's wrong... [23:52] <wtay> hmm. no I don't.. [23:52] <tnt> :-) You got my hopes up. [23:58] <tnt> wtay: Just out curiousity, is the new CVS (after the merge that omega did) more stable/usable than the one just before that? [23:58] <wtay> tnt: the same for the moment [23:59] <wtay> mot of the code isn't touched, just some APIs added.. [23:59] <wtay> s/mot/most/ [23:59] <tnt> wtay: Hmmm,... I'm wondering if getting the very freshest CVS will get rid of the problems I'm having? [23:59] <wtay> I seriously doubt iy [00:00] --- Mon Oct 22 2001 [00:00] <wtay> s/iy/it [00:00] <wtay> I'm going to play around a bit with aggregator tomorrow [00:00] <tnt> OK. [00:14] <Uraeus> good night [00:14] Uraeus (csc...@c1...) left #gstreamer (Client Exiting). [00:14] ajmitch (aj...@p6...) left irc: Ping timeout for ajmitch[p60-max3.dun.ihug.co.nz] [00:18] ajmitch (aj...@p2...) joined #gstreamer. [00:19] <wtay> I need sleep too.. cya [00:19] Nick change: wtay -> wtay-zZz [00:26] thomasvs (th...@ad...) left irc: Client Exiting [00:27] <tnt> What kind of timing is available in GStreamer??? Is there a stream based time identifier... like a from count for video... but not a timestamp. [00:50] Shippou (no...@cr...) joined #gstreamer. [01:08] harobed (harobed@ACAF5310.ipt.aol.com) left irc: Client Exiting [01:13] wingo (wi...@rd...) left irc: going home... [01:17] arik (ar...@sd...) joined #gstreamer. [01:20] <taaz> woof [01:20] <arik> howdy [01:20] <taaz> hiking is fun [01:20] <arik> yes [01:24] <ajmitch> heya arik [01:24] <arik> hey aj [01:24] <arik> sup? [01:28] <ajmitch> study ;) [01:28] <arik> heh [01:28] <arik> work ;-) [01:29] <ajmitch> i got 3 exams in the weekend ;) (well, friday, sat, monday) [01:29] <arik> ouch [01:30] <ajmitch> lotsa fun [01:30] <ajmitch> anyway, i must be off.... [01:30] <arik> alright [01:30] <arik> god luck [01:30] <arik> er good [01:30] <arik> but god would prob help too [01:30] <ajmitch> there's stuff to learn! [01:30] Action: ajmitch gags [01:30] <arik> heh [01:38] wingo (wi...@rd...) joined #gstreamer. [01:38] <wingo> yop [01:38] <arik> yo [01:39] <wingo> emacs 21!!!!!!! [01:39] <arik> emacs 21? [01:39] <wingo> at slashdot [01:40] <arik> heh [01:40] <arik> i use xemacs ;-) [01:40] <arik> still that's neet [01:55] arik (ar...@sd...) left #gstreamer. [01:55] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [02:11] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [02:13] evil_monk (trevo@208.141.162.68) joined #gstreamer. [02:24] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [02:29] <wingo> mom's gumbo. mmm... [02:30] vishnu (jo...@cx...) joined #gstreamer. [02:31] <vishnu> hey wow, my patch got applied [02:32] <tnt> vishnu: kool. [02:34] <wingo> hey vish [02:34] <wingo> (nu) [02:35] <vishnu> ? [02:35] <wingo> nothing, just being silly [02:35] Action: vishnu steals wingo's pillow [02:36] <vishnu> hahaha! [02:37] Action: vishnu gives wingo his pillow back [02:38] <wingo> it wasn't gone long enough for me to be mad! [02:38] <wingo> i'm mad about that! [02:38] <vishnu> :-) [02:38] Action: wingo is mad! [02:38] Action: wingo is not mad anymore [02:38] <wingo> :-) [02:40] <wingo> where is steveb??? [02:40] <wingo> ladspa works for me !!! [02:40] Action: wingo is pleased [02:40] <wingo> thanks steve, wherever you are :-) [02:57] Shippou (no...@cr...) joined #gstreamer. [02:57] ajmitch (aj...@p2...) left irc: Excess Flood [02:58] <vishnu> "revived _buffer_ref_by_count" -- d'oh! [02:59] <wingo> heh [02:59] <vishnu> can't you feel the love in here? ;-) |
From: IRC B. <wt...@us...> - 2001-10-23 04:27:40
|
******************************************************************* [03:00] ajmitch (aj...@p2...) joined #gstreamer. [03:08] wtay-zZz (wi...@ca...) got netsplit. [03:08] Shippou (no...@cr...) got netsplit. [03:08] ajmitch (aj...@p2...) got netsplit. [03:09] wtay-zZz (wi...@ca...) returned to #gstreamer. [03:09] mwc (ma...@ly...) joined #gstreamer. [03:11] Action: wingo needs coffee [03:11] ajmitch (aj...@p2...) returned to #gstreamer. [03:11] <wingo> sweet! leftover coffee from this morning [03:12] <tnt> wingo: I hope it's warm. [03:12] <vishnu> microwave it! [03:12] <wingo> with the halp of ms. microwave [03:12] <wingo> (her name is written on the door even) [03:12] <wingo> my house is strange :-) [03:14] Shippou (no...@cr...) got lost in the net-split. [03:25] <wingo> aha! [03:25] Action: wingo finds a bug in the ladspa plugin [03:36] mwc (ma...@ly...) left irc: [x]chat [03:44] tnt (tn...@h2...) left irc: Client Exiting [03:50] Action: vishnu gives wingo a rubber chicken prize for his debugging work [03:51] <wingo> i still haven't sorted it out, keep it for now :-) [03:51] <wingo> i think it's a shim issue [03:53] Action: vishnu puts the rubber chicken in escrow [03:56] <vishnu> it's waiting for you wingo. you must be excited now :-) [04:03] <wingo> hoo boy! [04:03] Action: wingo chomps at his bit [04:08] Shippou (no...@cr...) joined #gstreamer. [04:16] <taaz> vishnu: fix smp issues, thanks! [04:17] <vishnu> taaz: huh? [04:17] <taaz> thread stuff [04:17] <taaz> it's broke [04:17] <vishnu> taaz: yah, i know .. ? [04:17] <taaz> it's either more broke or just more likely to happen since the events merge [04:17] <wingo> anyone know why _gst_plugin_init loads an xml file? [04:17] <wingo> along with all o the plugins into memory [04:18] <taaz> to make sure the hardware still works? ;) [04:18] <vishnu> taaz: i think leaked memory was hiding the corruption ;-) [04:18] <vishnu> taaz: now there are fewer leaks so the corruption becomes more apparent :-) [04:19] <taaz> i guess that's good [04:19] <vishnu> ya, definitely [04:23] Action: wingo is in awe of the plugin system [04:25] <wingo> here's the issue. [04:25] <wingo> 1) database is reloaded [04:25] <wingo> s/reloaded/loaded [04:25] <wingo> 2) all the elementfactories are created [04:25] <wingo> 3) plugin you actually want gets loaded when you ask for it [04:26] <wingo> 4) when it sets its padtemplates in the plugin_init function they have to go and replace all of the ones that were reloaded in step 2) [04:26] <wingo> seems a hack to me [04:26] <vishnu> why is it a hack? [04:27] <wingo> the pad templates are already in the factory [04:27] <wingo> hell, the factory alread exists [04:27] <wingo> that might be the key though [04:27] <wingo> that all plugins should check to see if there's a factory already loaded for them... [04:28] <wingo> that would reduce ladspa load time considerably [04:29] <vishnu> hrm, i don't really understand that code. sounds promising though [05:12] <wingo> oh, i was linking to the wrong gst [05:13] <wingo> i'm glad that we've merged trees [05:13] <wingo> er, branches [05:14] chillywilly (da...@d1...) joined #gstreamer. [05:21] <wingo> autoplug is slow. [05:31] <wingo> taaz: what is meant by the term "dynamic autoplugger"? [05:39] tnt (tn...@h2...) joined #gstreamer. [05:51] <taaz> eh? [05:51] <taaz> oh.. [05:51] <taaz> like the current autoplugger isnt very smart. does everything up front [05:52] <taaz> so it guesses some things [05:52] <taaz> like since the mpeg2parse element can output various audio formats (mp3, ac3, pcm, etc) it will just randomly (?) chose something to hook to it. that could be wrong [05:53] <tnt> :-) [05:53] <taaz> you really want to wait until you have more info before doing the autoplugging [05:53] <tnt> Filters for [05:53] <taaz> hence the "dynamic" [05:53] <tnt> Filters for GStreamer need more meta data [05:53] <tnt> More info about what they do. [05:53] <taaz> uh.. like what? [05:53] <tnt> That way the autoplugger could make a better choice. [05:54] <vishnu> taaz: oh, so maybe the autoplugger should run "just in time" [05:54] <taaz> tnt: this is a runtime issue [05:54] <taaz> you don't know what format the data is until you get it [05:54] <tnt> Like if a filter does a 'conversion' from one format to another... or something else. [05:55] <tnt> GStreamer needs alot of things! [05:57] <tnt> You know, GstThread should, IMO, be renamed to GstThreadBin. [05:58] <vishnu> yah, probably [05:59] <tnt> I'd like to make a GstProcessBin. [05:59] <tnt> But I want to get my test program working first :-) [06:00] <wingo> anyone know the difference between "destroy" and "finalize" [06:00] <vishnu> wingo: use dispose & finalize. what is destroy? [06:01] <tnt> I think, and me stress I THINK, destroy calls finalize. [06:01] <tnt> If it is anything like Java it will. [06:02] <tnt> Has anyone seen gegl <http://www.gegl.org/> before? [06:02] <tnt> Has GStreamer made any effort to work with them> [06:03] <wingo> there are funny refcounting errors that i get when removing an element from a bin (which should unref and destory an element) [06:03] <vishnu> wingo: dispose is suppose to drop references to other objects. finalize is suppose to release memory [06:04] <wingo> tnt: it looks extra old, the web page and todo lists [06:04] <wingo> thanks vishnu [06:04] <wingo> does finalize unref the object too? [06:04] <tnt> wingo: I thought GIMP 2 was being based on it. [06:04] <wingo> dunno :-) [06:05] <vishnu> wingo: finalize is called with refcnt=0. dispose is called with refcnt=1. finalize always follows dispose [06:06] <tnt> Is there a good reference/book/whatever of all this GObject memory management stuff. [06:06] <wingo> vishnu: thanks [06:06] <wingo> tnt: GGAD [06:06] <vishnu> tnt: yah, read the source code :-) [06:06] <wingo> look in the ReferenceLibrary on the wiki [06:06] <tnt> wingo: I guess I should have read the whole thing :-) [06:07] <vishnu> most of the documentation is out-of-date for 2.0 [06:07] <wingo> true [06:07] <wingo> but the concepts are still there [06:22] Action: wingo is going to try to build against glib2 [06:23] <wingo> is it possible to use glib2 gst with an app that uses gtk1.2? [06:23] <vishnu> no [06:23] <wingo> damn. [06:23] <tnt> gst is using a "shim" though... [06:23] <wingo> does gnome2 libglade exist? [06:24] <vishnu> dunno [06:24] <tnt> gst is really using glib1.2... even though it thinks it is using glib2.0 [06:24] <wingo> i know [06:24] <tnt> In the CVS I think... but it is not done. [06:25] <tnt> On the gnome-devel mail list, they have been talking about bug fixes and code freezes... GNOME 2 is suppose to be coming really soon. [06:25] <ajmitch> you need to either use a converter script or write .glade files from hand for the newer libglade [06:25] <ajmitch> tnt: february release date apparantly [06:26] <tnt> ajmitch: the sooner the better! [06:26] <wingo> ajmitch: know of any unofficial gnome2 debs? [06:26] <ajmitch> http://lists.gnome.org/archives/gnome-2-0-list/2001-October/msg00068.html [06:26] <ajmitch> wingo: there are only really libs that are working at the moment [06:27] <wingo> right [06:27] <wingo> all i need are the libs for my app :-) [06:27] <ajmitch> hadess has done quite a bit on gnome-utils lately [06:27] <wingo> that's what i read on uraeus' news doober [06:27] <wingo> tight community, this :-) [06:28] <ajmitch> yeah [06:28] <tnt> Is Uraesus the one that does the Gnotices? [06:29] <tnt> Or the summaries? [06:29] <wingo> both [06:29] <wingo> afaik [06:29] <tnt> He does a good job! [06:31] <tnt> You know, I was thinking... what happens if someone makes a cyclic connection in a pipeline? [06:31] <tnt> Is an error raised? [06:32] <wingo> try it :) [06:33] <tnt> I can't even get my example program (with 2 in streams and 1 out stream) to work... I wouldn't know where the error was coming from :-) [06:34] <wingo> oh, the pain. i wish properties of gtk1.2 elements had bounds. [06:34] <wingo> ah well. [06:35] <tnt> wingo: Can't you force bounds on them by redefining the "Set Property" function??? [06:36] <wingo> no, what i want to do is not possible with 1.2. it cannot be done. [06:37] <tnt> wingo: maybe it is a 2.0 feature. [06:37] <wingo> it is. [06:37] <tnt> :-) ... it's a nice feature. [06:50] <wingo> so... will there be a gtkobject in gtk+2? [06:51] <tnt> wingo: Yes, but it is derived from GObject. [06:51] Nick change: taaz -> taazzzz [06:57] Nick change: wingo -> wingo-z [07:02] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [07:10] chillywilly (da...@d1...) left irc: Read error to chillywilly[d150.as8.nwbl0.wi.voyager.net]: EOF from client [07:16] kikichu (clint@63.166.193.71) joined #gstreamer. [07:22] Nick change: tnt -> tnt-zZz [07:34] steveb (st...@no...) joined #gstreamer. [07:59] mef (ma...@us...) joined #gstreamer. [08:10] kikichu (clint@63.166.193.71) left irc: Client Exiting [08:19] walken (fo...@c1...) joined #gstreamer. [08:26] Nick change: mef -> mef-zZz [08:26] Nick change: mef-zZz -> mef [08:27] mef (ma...@us...) left #gstreamer. [08:27] <tnt-zZz> mef: So are you asleep or not? :-) [10:10] uwe (user424@213.174.92.50) joined #gstreamer. [10:59] walken (fo...@c1...) left irc: l8r [11:17] sienap (sy...@s3...) joined #gstreamer. [11:22] <sienap> ehm [11:22] <sienap> Hi [11:22] <sienap> :) [11:34] <uwe> ! [11:34] <uwe> = [11:34] <uwe> + [11:35] <uwe> & [11:35] <sienap> That.. is what imeant [11:36] <uwe> sienap: bent jij nederlands? [11:36] <sienap> nee [11:36] <sienap> dat ben ik niet [11:36] <sienap> ik .. ontken [11:37] <sienap> mwha eigenlijk wel [11:37] <sienap> maar goed ;) [11:37] <uwe> prima. [11:38] <uwe> ik werk in rotterdam [11:39] <sienap> daar woont m'n vriendin [11:41] <uwe> ik woon in het zuid, ijsselmonde, maar ik ben duitser [11:41] <sienap> hehe :) [11:41] <uwe> en ik spreek niet veel nederlands :( [11:41] <sienap> anyway wat doe jij met gstreamer / [11:42] <sienap> och.. dat leer je vanzelf wel ;) [11:42] <sienap> ik spreek niet goed nederlands is het trouwens. [11:43] <uwe> ik ga in november terug, niet veel tijd voor leren / [11:43] <sienap> too bad.. [11:43] <sienap> waarom ben je in nederland ? [11:44] <uwe> we are programming (ok, try to...) a streaming flash solution with gstreamer [11:44] <uwe> i'm working for a company, then i'll write my report and finish my studies in germany [11:45] <uwe> where are you from? [11:45] <sienap> i live around nijmegen [11:45] <sienap> a streaming flash solution with gstreamer.. must be nice [11:45] <sienap> i wish someone made a flash movie codec ;) [11:47] <uwe> aha nijmegen. are you german? i hate flash. isn't be nice to do that. i understand your ";)" [11:48] <sienap> no.. i am NOT german :) [11:49] BBB (BB...@uc...) joined #gstreamer. [11:50] <sienap> hej BBB :) [11:50] <uwe> belg? fransman? [11:51] <sienap> i am plain dutch [11:51] <sienap> damn [11:51] <sienap> ;) [11:51] <uwe> ! [11:51] <sienap> nijmegen lays in the netherlands :) [11:51] <uwe> know that. my co is from there [11:52] <sienap> aah ok :) [11:59] <sienap> eos [11:59] sienap (sy...@s3...) left #gstreamer. [13:03] Nick change: ShrimpX -> ShrimpZzZ [13:10] harobed (harobed@ACAD7F6A.ipt.aol.com) joined #gstreamer. [14:31] asmod (stevec@64.5.222.2) joined #gstreamer. [14:49] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [14:55] <BBB> wargh.... [14:55] <BBB> nouja, hoi sienap (hele late reactie) [15:02] <uwe> BBB bent oog uit nederland? [15:07] <BBB> ja [15:07] <BBB> iedereen hier ofzo? :P [15:08] <uwe> mischien... [15:08] <BBB> wtay (Wim) is belgisch.... [15:08] <uwe> ik weet [15:08] <uwe> maar vaan flanderen ;) [15:08] <BBB> flanderen? [15:09] <uwe> the part where they speak dutch? [15:09] <BBB> owh, vlaa[mn]deren [15:10] <uwe> danke je. ik leer nog [15:11] <BBB> that's okay :) [15:11] <thomasvs_> ik ook [15:11] <uwe> goed! [15:12] <BBB> NL conquers the gst-world [15:12] <BBB> :D [15:12] <uwe> thomasvs_: ben jij ook vaan nederland? [15:12] <thomasvs_> nee, belgie ;) [15:13] <BBB> hum [15:13] <uwe> ich krieg echt Angst mit euch! [15:13] <BBB> ik ga ff reboeten jongens [15:13] <BBB> mijn kernel doet gemeen :) [15:13] <BBB> totzo [15:13] BBB (BB...@uc...) left irc: [x]chat [15:16] <uwe> thomasvs_: waar ben jij vaan in belgie? [15:16] <thomasvs_> Gent [15:17] <uwe> great, wtay said i should visit gent ;) [15:17] <thomasvs_> yeah, Gent is great [15:18] <uwe> when i have some time i'll try to visit it [15:18] <uwe> i was in antwerpen few weeks ago and i like it. [15:40] rambokid (timj@62.67.108.89) joined #gstreamer. [15:41] Zygo (zbl...@24...) joined #gstreamer. [15:45] evil_monk (trevo@208.141.162.68) joined #gstreamer. [15:51] Shippou (no...@ot...) joined #gstreamer. [15:54] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [15:58] BBB (BB...@uc...) joined #gstreamer. [16:00] zblaxell_ (zbl...@24...) joined #gstreamer. [16:01] Zygo (zbl...@24...) left irc: Ping timeout for Zygo[246.chipworks.com] [16:03] BBB (BB...@uc...) left irc: Ping timeout for BBB[ucu-105-116.ucu.uu.nl] [16:05] BBB (BB...@uc...) joined #gstreamer. [16:07] harobed (harobed@ACAD7F6A.ipt.aol.com) left irc: Ping timeout for harobed[ACAD7F6A.ipt.aol.com] [16:43] wingo-z (wi...@rd...) left irc: Read error to wingo-z[rdu25-21-076.nc.rr.com]: EOF from client [16:43] wingo (wi...@rd...) joined #gstreamer. [16:43] <wingo> topothemornin [16:44] <uwe> tagchen [16:46] <wingo> tag==day? what is chen? [16:47] <rambokid> nice-ification [16:47] Nick change: wtay-zZz -> wtay [16:47] <wtay> yo [16:48] Action: wtay left early from his work.. [16:48] <uwe> small day, maybe dagje for wtay ;) [16:48] <wtay> uwe: yup [16:49] <uwe> can you say that in dutch? [16:49] <wtay> yes, sure [16:49] <wtay> not as a greeting though... [16:49] <uwe> for a short day i guess ... [16:50] <wingo> heh [16:50] <wtay> or for a hectic day.. "het was mijn dagje wel.." [16:50] <wtay> actually it was... [16:50] <uwe> soory to hear that. [16:50] <uwe> whats about party in leuven? [16:51] <wtay> IMS mainframe was copletely going nuts.. [16:51] <uwe> oh shit [16:51] <wtay> uwe: what about it? [16:52] <uwe> sounds good from my side... [16:52] Action: wingo wants to be in europe :-) [16:53] <uwe> take a flight ;) [16:53] <wingo> i'll finish school first :) [16:53] <uwe> you're allowed to drink your beer on the street here! [16:53] <wingo> i know! i spent a year in spain [16:54] <wingo> i miss it :-\ [16:54] <uwe> espana! como estas? [16:54] <wtay> uwe: you can even smoke pot here now too.. [16:54] <wingo> pues bien, tio. como andas tu? [16:55] <wingo> un argentino vino aqui el otro dia y acabamos hablando en ingles ;-) [16:55] <uwe> wtay: now that. unfortunatly i don't smoke. maybe have a coekje ... [16:55] <wtay> uwe: yeah, space cake :) [16:56] <uwe> wingo: stop it please ... would like to speak spanish, but i can't. something about wine and english in argentina? [16:57] <wingo> just that an argentinian came in the other day to #gstreamer and we ended up talking in english because i couldn't help him alone :-) [16:58] <uwe> thanks [16:58] <wingo> vino is wine but also the past tense of "to come"==venir [16:59] <uwe> have a friend from honduras, will start to learn spanish [16:59] <wingo> cool. tagchen is dutch? or german? or... [16:59] <uwe> german. you will not hear that so much [17:00] <uwe> rambokid: you speak german? [17:01] Nick change: wingo -> wingo-class [17:01] <wingo-class> later [17:02] <rambokid> yes [17:02] <uwe> you're german? [17:04] <rambokid> yeah [17:04] <uwe> woher? [17:04] <rambokid> HH [17:04] <uwe> DE [17:04] <rambokid> uwe: eh? [17:05] <uwe> Dessau [17:05] <rambokid> ah [17:05] <uwe> will move to SN in december [17:05] <uwe> not so far away [17:05] <rambokid> uwe: why are you hanging out here? [17:06] <uwe> try to program with gstreamer ... [17:07] <uwe> are there good jobs in HH? [17:08] <rambokid> haven't found any [17:11] <uwe> rambokid: what are you looking for? i'll try to find something in the north next spring. HRO/HWI/HH/SN, i thought it's more easy to find something in HH [17:23] <rambokid> well, i'm only looking for freesoftware jobs [17:26] <uwe> good luck. will see you next year :) Party in SN! [17:27] <uwe> wtay: apropos ... we didn't finished the discussion ;) [17:34] <wtay> sorry, was away.. [17:34] <wtay> uwe: go on.. [17:35] <uwe> did you mean really mean that? or was it a joke? [17:36] <wtay> I'm not joking.. [17:36] <wtay> I was away though :) [17:37] <uwe> i don't know what to do next weekend [17:40] <wtay> ? [17:48] <uwe> wtay: you said something about party in leuven last week... [17:49] <wtay> uwe: there a party somewhere in leuven every day.. [17:50] <wtay> uwe: I'm well past behind the "party age" so I don't have much details.. [17:52] <uwe> sorry, i thought you were speaking about party at tayman's :( [17:56] <wtay> oh.. party at my place? hmm.. I live on a dump :) [17:57] <uwe> :) [17:57] <wtay> people could get hurt :) [18:02] <uwe> sounds funny ;) [18:03] <wtay> ah food.. [18:03] Nick change: wtay -> wtay-food [18:10] evil_monk (trevo@208.141.162.68) joined #gstreamer. [18:16] vishnu (jo...@cx...) joined #gstreamer. [18:57] uwe (user424@213.174.92.50) left irc: Client Exiting [19:08] vekky (wb...@pe...) joined #gstreamer. [19:08] <vekky> hey has ds been around? [19:08] <vekky> guess not eh [19:08] <vishnu> it's pretty quiet so far today [19:08] <vekky> YEAH WE SHOULD DO SOMETHIN ABOUT THAT EH. [19:09] <vishnu> yah, let's have a party!! [19:09] <vekky> woo hoo !! [19:09] Action: vishnu offers everyone a slice of pumpkin pi [19:09] <vishnu> pie [19:14] BBB (BB...@uc...) left irc: Ping timeout for BBB[ucu-105-116.ucu.uu.nl] [19:40] ShrimpX (marius@Photon.pdx.netroedge.com) joined #gstreamer. [19:46] ShrimpX (marius@Photon.pdx.netroedge.com) left #gstreamer. [19:57] Nick change: wingo-class -> wingo [19:57] <wingo> wtay-food: ? [19:58] <vishnu> wingo! [19:58] <wingo> vishnu! [19:58] <wingo> like omigod! [19:59] <wingo> :) [19:59] <vishnu> :-) [20:08] Uraeus (csc...@c1...) joined #gstreamer. [20:09] <Uraeus> hi [20:27] <wingo> hi [20:33] Nick change: wtay-food -> wtay [20:33] <wtay> yup [20:33] <vishnu> wtay! [20:34] <vishnu> gst_bytestream_get_status -- what do you think? [20:36] <wtay> dunno yet, it still feels odd to me.. [20:41] <wingo> hey, we need to rethink the contact page, or the links on the left [20:41] <wingo> lists need to be mroe obvious [20:41] <wingo> and why is there a list of developers, other than ego-stroking? not to discount ego-stroking, but... [20:42] <wtay> not my idea.. [20:42] <wingo> wtay, uraeus, thought? [20:42] <wingo> thoughts even [20:43] <wingo> so... i know you made it uraeus, i was really pleased to show up on there when you did it, but does it have a purpose? [20:43] <Uraeus> wingo: not really :) [20:43] <wingo> heh [20:43] <Uraeus> wingo: Hall of Fame? [20:43] <wtay> it's not even a complete list, you better remove it [20:44] <wingo> so. the thought is, "if you want to get in touch with gstreamer people, talk to the lists"? [20:45] <wingo> and or irc :-) [20:45] <wtay> yeah.. maybe [20:45] <wingo> anyway. whatever's clearest. i just don't want to inadvertantly offend someone :-) [20:46] <wtay> vishnu: why not return the event and let the plugin check for GST_IS_EVENT? [20:46] <wtay> vishnu: or maybe not.. [20:47] <wtay> vishnu: we somehow also need the bytes *before* the event occured and the onew remaining (discont event)... [20:50] <vishnu> wtay: huh, you have all the bytes [20:50] <vishnu> ? [20:50] <wtay> ok [20:50] <wtay> avail_out then.. [20:51] <vishnu> wtay: you can't return the event. that would suck. you'd have to add checks everywhere [20:51] <vishnu> yah, avail_out [20:51] <wtay> yes, but you have to check for NULL now anyway [20:52] <vishnu> wtay: sure, you have to check for NULL, but that's easy. checking for NULL and IS_EVENT is much more of a burden [20:52] <vishnu> i mean, you can still do it. just add a call to get_status after you get a NULL [20:53] <wtay> why check for NULL? [20:53] <vishnu> but generally, i think you want to bail out of your processing and handle events in the central looping component [20:53] <wtay> yup [20:53] <vishnu> wtay: you have to check for NULL because that's what happens when you run out of bytes [20:53] <vishnu> you can run out of bytes even without an event [20:54] <wtay> let's hope that doesn't happen.. that's what EOS should do.. [20:55] <wtay> I was just thinking that the API of _pad_pull can also retrun an event [20:55] <vishnu> wtay: hrm .. interesting. [20:55] <wtay> so bytestream_read and _pull look a bit more similar then [20:56] <wtay> but then again _peek_bytes should return NULL... [20:56] <vishnu> unless you scrap the peek_bytes API and only offer peek [20:57] <wtay> hmm, yes.. [20:57] <wtay> but peek_bytes has other advantages.. [20:58] <vishnu> then, instead of checking for NULL, you'd have to check for IS_EVENT everywhere. it seems like a bad trade off. we have the same number of checks and we loose peek_bytes [20:58] <wtay> yes.. I'm applying the patch [20:59] <vishnu> ok, cool. i'll do the next patch this afternooon. i should be able to finish off seek support for mpeg2parse. [20:59] <wtay> vishnu: have you tried gst_pad_send_event? [20:59] <vishnu> wtay: no, i haven't tried it yet, but that's what i'll do next. i'm just trying to get the obvious stuff behind us [21:00] <vishnu> how about applying the 64bit filesrc patch too? ;-) [21:01] <wtay> yes, that's next :) [21:01] <wtay> does it work with the shim? [21:01] <vishnu> wtay: no, i didn't add shim support. [21:02] <vishnu> wtay: if you want to go beyond the call of duty then the shim can emulate it with casts to/from double. however, i'd just support 32bit stuff under glib1.2 [21:02] <wtay> yes [21:02] <vishnu> wtay: you can hack it for int32 by adding lots of #defines :-) [21:03] <wtay> doing that now [21:03] <vishnu> wtay: super cool, thx [21:03] Nick change: vishnu -> vishnu-gym [21:04] <vishnu-gym> i gotta prepare for my boxing match with Erik ;-) [21:04] <Uraeus> vishnu-gym: I think Erik would prefer a battlebots contest :) [21:05] <vishnu-gym> hehe [21:07] harobed (harobed@ACA4BFE8.ipt.aol.com) joined #gstreamer. [21:08] <Uraeus> hi harobed [21:11] <harobed> hello [21:11] <Uraeus> harobed: managed to get the compilation problem fixed? [21:17] vekky (wb...@pe...) left irc: too bad the scene is dead [21:18] Nick change: wtay -> wtay-tv [21:19] <ajmitch> morning [21:19] <wingo> afternoon :) [21:19] <Uraeus> morning ajmitch [21:22] <ajmitch> what's everyone up to today? :) [21:22] harobed (harobed@ACA4BFE8.ipt.aol.com) left irc: Ping timeout for harobed[ACA4BFE8.ipt.aol.com] [21:23] <wingo> chillin' :-) -- doing homework due a few weeks ago [21:23] <wingo> yourself? can't be up to much yet :-) [21:23] <ajmitch> wingo: hehe, i shoudl be studying for exams in a few days ;) [21:23] harobed (harobed@ACA4BFE8.ipt.aol.com) joined #gstreamer. [21:44] BBB (BB...@uc...) joined #gstreamer. [22:05] chillywilly (da...@d7...) joined #gstreamer. [22:07] Zeenix (zeenix@203.128.13.26) joined #gstreamer. [22:08] vektor (reet@HSE-Kitchener-ppp194616.sympatico.ca) joined #gstreamer. [22:08] <wingo> reet! [22:09] <vektor> Hi all. [22:09] <vektor> I'm cvs co'ing the X source! Fun! :( [22:09] <wingo> ug! [22:09] <wingo> all this regarding your recent activities on xfree-devel? [22:09] <vektor> I hate it when someone says 'try changing this value in our source'. Soo much work. :( [22:09] <vektor> Yeah kinda. I'm tracking Xv bugs in my card's driver. [22:10] <vektor> U programs/Xserver/XpConfig/C/print/models/SPSPARC2/fonts/Helvetica-Bold.pmf [22:10] <wingo> well someone has to :-\ [22:10] <vektor> But I have to live through stuff like that. [22:10] <wingo> nast. [22:10] <vektor> It's like taking 3 million years downloading fonts or something. [22:10] <wingo> card? [22:10] Zeenix (zeenix@203.128.13.26) left irc: Ping timeout for Zeenix[203.128.13.26] [22:10] <vektor> i815 onboard video. [22:10] <vektor> It's surprisingly good. [22:10] <vektor> (except for these bugs, of course). [22:10] <vektor> I ripped out my TNT2 to use it. :) [22:11] <wingo> heh [22:24] omega (om...@om...) joined #gstreamer. [22:25] <wingo> yo [22:26] <ajmitch> hey omega [22:27] <omega> yo [22:32] <wingo> you haven't had a chance to test alsasrc have you, omega? [22:33] <omega> nope, not yet [22:33] <omega> but atm I'm trying to reconstruct a driver for my other ADAT card, so I can have one here at home [22:33] <wingo> recenstruct? [22:33] <omega> though with nothing but loopback fibers, since I only have one interface [22:33] <wingo> s/cen/con/ [22:34] <omega> the vendor of the card had a driver under development, and I have the docs and source code under NDA, but the driver is for alsa 0.4.1 and kernel 2.2.x [22:34] <omega> so I have known viable code to work from, but I'm rewriting the driver [22:34] <wingo> are you trying to do alsa 0.9 with it then? [22:34] <omega> of course <g> [22:34] <omega> but I'm just gonna get the card doing I/O first [22:34] <wingo> will it be releaseable with alsa? [22:35] <omega> afaik, that was the intent of me getting an NDA, yes [22:35] <wingo> what card? [22:35] <omega> but the card is out of production now, so it's even more likely I can release it [22:35] <omega> Sonorus STUDI/O [22:35] <wingo> is that secret? :-) [22:35] <omega> not really, why? [22:36] <vektor> anyone know how to compile the X source tree? [22:36] <wingo> just asking. [22:36] <omega> make World [22:36] <wingo> you might have to xmkmf [22:36] Nick change: vishnu-gym -> vishnu [22:38] <vektor> Yeah I don't think I got the full tree. [22:38] <vektor> xmkmf -a fails. But it's ok, I'm getting it now. [22:46] Zeenix (ze...@13...) joined #gstreamer. [22:47] <Zeenix> hmm..., seems that wtay gets addicted to tv [22:49] <wingo> we all have our vices ;P [22:51] steveb (st...@no...) left irc: advance! [22:52] zblaxell_ (zbl...@24...) left irc: Read error to zblaxell_[246.chipworks.com]: EOF from client [22:52] asmod (stevec@64.5.222.2) left irc: [x]chat [22:54] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [23:06] Nick change: tnt-zZz -> tnt [23:06] <tnt> Hello all. [23:08] rambokid (timj@62.67.108.89) left irc: Ping timeout for rambokid[62.67.108.89] [23:11] <omega> yo [23:18] <Zeenix> omega: i dont think wtay would be coding for gst anymore <g> [23:30] <Zeenix> omega: whats that argument to random32() does ? [23:31] <omega> you mean "what does the argument to random32() do?" the man page should say [23:33] <Zeenix> i dont fine a man page for that [23:33] <Zeenix> & i dont know where it belongs: glib, gtk+, gst .... [23:33] <Zeenix> s/fine/find [23:33] <omega> I have no idea where it comes from [23:33] <omega> the only thing I would expect the argument could be is a seed [23:34] <omega> as per the argument to srandom (for which there is a man page) [23:35] <Zeenix> ok, have you ever looked at algorithms to choose random nums ? [23:35] <Zeenix> i mean algorithms that choose random nums [23:36] <Zeenix> s/at/into [23:37] harobed (harobed@ACA4BFE8.ipt.aol.com) left irc: Ping timeout for harobed[ACA4BFE8.ipt.aol.com] [23:49] gtb (syln@195.68.94.88) joined #gstreamer. [23:49] <gtb> hello ! [23:51] <gtb> wingo: if you are there, what is the current status of the alsa plugins ? [00:00] --- Tue Oct 23 2001 [00:06] <taazzzz> gtb: look at the changelog on their source [00:07] <taazzzz> or better yet, just try it [00:07] <vishnu> hey, aren't you asleep? [00:07] <tnt> he's sleep talking [00:07] <gtb> taazzzz: yep [00:08] Nick change: taazzzz -> taaz [00:08] <taaz> been bizy all day [00:08] <taaz> so the deal with bytestream evet processing is what? you always check for null and if its null you check for event? [00:09] <vishnu> taaz: yah [00:09] <vishnu> taaz: you got a problem with that? shall we step outside to settle this? [00:10] <taaz> yeah yeah! [00:10] <taaz> fight fight! [00:10] <vishnu> :-) [00:10] <vishnu> i'll give ya some events buddy! [00:10] <taaz> so, care to update some plugins to see how well coding that works? [00:10] <vishnu> yah, i'm doing that now :-) [00:11] Shippou (no...@cr...) joined #gstreamer. [00:11] gtb (syln@195.68.94.88) left irc: [00:12] <vishnu> heh, works pretty well. neat. i'm feeling a patch coming on ... [00:12] <Uraeus> hehe [00:12] <Zeenix> Uraeus: yo [00:13] <Uraeus> hi Zeenix [00:13] <omega> Uraeus: so what's the deal, do I go get plane tickets and stuff? [00:14] <Uraeus> omega: I will know tommorow for sure, but its looking good [00:14] <omega> sooner the better, cause the ticket prices go up [00:15] <Uraeus> if I get go-ahead I order tickets tommorow [00:17] Action: Zeenix is enjoying reading the real History of Human Evolution beside coding for RtpRecv [00:17] <Uraeus> night everyone [00:17] Uraeus (csc...@c1...) left irc: Client Exiting [00:18] tnt (tn...@h2...) left irc: Client Exiting [00:19] tnt (tn...@h2...) joined #gstreamer. [00:19] bstard (Lor...@a2...) joined #gstreamer. [00:23] <vishnu> Zeenix: who is the author? [00:25] <Zeenix> vishnu: Sibtay Hassan, a muslim-become-communist [00:25] <vishnu> Zeenix: hrm, amazon doesn't have it [00:29] <vektor> I love it when large compiles, because they get bored, generate more code for themselves to compile. [00:31] <Zeenix> ok bi [00:31] Zeenix (ze...@13...) left irc: I pay 25 Rs( 0.5$ )/hour [00:35] <vishnu> yo, can someone evaluate the patch i just sent? [00:39] <omega> I'll check it out sometime today (investigating other stuff atm) [00:44] Action: taaz ponders better ways to handle events [00:46] <taaz> am i the only one that thinks these event switch statements could be improved? [00:46] Action: vishnu offers taaz a helmet so he doesn't hurt his head [00:46] <taaz> there are other options [00:46] <vishnu> such as? [00:47] <tnt> callbacks? [00:47] <taaz> i'm not sure they are less code [00:47] <tnt> (... I'm not even sure what the event are for...) [00:48] <taaz> yeah i was thinking more along the line of callbacks [00:48] <vektor> cool i'm compiling x drivers for cards i don't own. [00:48] <taaz> but more like an OOP design [00:48] <tnt> Is the event handling Blocking or Non-Blocking? [00:48] <vishnu> goop design? ugh [00:49] <taaz> i think all my designs are too complex though :( [00:49] <vishnu> i have enough goop, my cloths are a mess, gotta go to the cleaners [00:50] <tnt> taaz: What are you working on? [00:50] <taaz> my thesis ;) [00:50] <taaz> this gst stuff is a convienient distraction from that crap [00:50] <tnt> Ahhh... not gstreamer. [00:51] <taaz> vishnu: you really should start writing code in gstreamer coding style [00:52] wingo (wi...@rd...) left irc: going home... [00:52] <vishnu> taaz: yah, yah .. [00:57] vektor (reet@HSE-Kitchener-ppp194616.sympatico.ca) left irc: back in a sec [00:58] <vishnu> make [00:58] <vishnu> oops [01:20] ChiefHighwater (fl...@su...) joined #gstreamer. [01:31] wingo (wi...@rd...) joined #gstreamer. [01:31] <omega> yo [01:31] <wingo> yo. [01:31] <wingo> so omega. [01:31] <ChiefHighwater> ello. [01:31] <wingo> release sometime? :-))) [01:32] <wingo> hey chief. [01:32] <taaz> is there an event for a new stream? [01:32] <omega> there will be [01:33] <ChiefHighwater> he just doesn't want to release cuz it's his turn to buy the pizza 8-] [01:33] <wingo> should autoplug_to_caps work? [01:33] <taaz> i was just wondering how you can keep state around until the next stream comes along [01:33] <wingo> hah! [01:33] <taaz> eos would clear things too early, but i'm not sure what else you could do [01:33] <ChiefHighwater> wingo:I'm juat teasing him...although if he wants pizza, he knows where to go 8-] [01:34] <wingo> i'd buy him a pizza [01:34] <wingo> yeah [01:35] <wingo> ok, when release comes, i'll buy you a pizza o. [01:35] <wingo> what's your favorite? [01:35] <ChiefHighwater> wingo:where we go for pizza, the pie runs $25US [01:35] <ChiefHighwater> but it is worth it 8-] [01:35] <wingo> heh [01:35] <ChiefHighwater> right omega? [01:35] <omega> yup [01:35] <wingo> where is it? i've been to pdx [01:36] <ChiefHighwater> Flying Pie Pizzaria, near Mt. Tabor [01:36] <ChiefHighwater> best pizza in pdx [01:36] <wingo> eh. didn't get out there :-) [01:36] <wingo> i think i went somewhere in the warehouse dist. for pizza [01:36] <ChiefHighwater> come on over, and omega and I will show you our reserved table 8-] [01:36] <omega> wingo: you mean the Pearl district? [01:37] <wingo> dunno [01:37] <omega> NW portland downtown [01:37] <wingo> you have to go under a interstate, it's really wierd [01:37] <omega> right [01:37] <wingo> near a brewery (that's everywhere tho) [01:37] <omega> in the Pearl, yeah [01:38] Nick change: wingo -> wingo-eat [01:38] <ChiefHighwater> hehe..made ya hungry? [01:38] <ChiefHighwater> hremm, makin me hungry [01:39] <vishnu> how can you think about food when there are nice warm patch cookies on the mailing list? [01:39] <wingo-eat> man cannot live on patches alone :-) [01:39] <ChiefHighwater> hehe [01:46] harobed (harobed@AC90B951.ipt.aol.com) joined #gstreamer. [01:46] Action: ajmitch wonders when the next release will be ;) [01:51] ChiefHighwater (fl...@su...) left irc: Leaving [01:53] ajmitch (aj...@p2...) left irc: Ping timeout for ajmitch[p2-max4.dun.ihug.co.nz] [01:55] ajmitch (aj...@p8...) joined #gstreamer. [01:59] Nick change: wingo-eat -> wingo [01:59] <wingo> oh yes. lentil soup. [02:00] <wingo> liffe is good. [02:00] <wingo> so is life. [02:10] <taaz> what's the point of loop based identity element? [02:10] <omega> testing [02:10] <taaz> but elements like that should really be done chained right? [02:10] <omega> yes [02:11] <taaz> okay [02:11] <omega> well, in most cases [02:11] <taaz> ? [02:11] <omega> if you have a bytestream-based element, it'll be loop-based for now [02:12] <taaz> btw, i think that identity duplicate feature will really screw with events ;) [02:12] <omega> duplicate? [02:12] <taaz> like duplicating EOS events would be odd [02:12] <taaz> yeah, duplicate [02:13] <omega> must be wtay's feature [02:16] Action: BBB is away: zzz [02:16] Nick change: BBB -> BBB-zZz [02:19] <wingo> gotta roll [02:19] wingo (wi...@rd...) left irc: Client Exiting [02:45] wingo (wi...@rd...) joined #gstreamer. [02:49] <wingo> coffee. [02:59] <taaz> so are events just cast to buffers? [02:59] <taaz> so the api works [02:59] <vishnu> taaz: yah [02:59] <taaz> that's kind of a hack [02:59] <vishnu> taaz: why? [02:59] <taaz> cause if you have foo(GstBuffer *buf) it may not actually be a buffer |
From: IRC B. <wt...@us...> - 2001-10-24 04:27:38
|
******************************************************************* [03:00] <vishnu> taaz: that's what documentation is for, but yah, maybe it should be cleaned up -- funcs should take GstData * [03:00] <taaz> like if you dont check to see if its an event you'll crash or something right? [03:00] <vishnu> taaz: yah, definitely [03:01] <wingo> that's what alsa does :-\ [03:01] <taaz> i think this situation could be improved [03:01] <wingo> by improving plugins? [03:01] <wingo> only about 60 1 line fixes [03:02] <taaz> ? [03:02] <vishnu> taaz: we can ask plugins to request events explicitly and migrate the plugins gradually [03:02] <taaz> well a better short term hack is to make events subclass empty buffers [03:03] <vishnu> yuck [03:03] <taaz> so at least you dont crash in a totally crazy way [03:03] <taaz> yuck? [03:03] <taaz> err yeah wait [03:03] <taaz> I meant make buffer subclass event [03:03] <vishnu> well, i don't know. maybe it's better than crashing. ask wtay or omega, they'd have to approve it [03:04] <taaz> yeah... like consider buffers just a data event [03:04] <taaz> i can already hear whining about overhead [03:04] <wingo> that's when you want to crash tho -- before it was an infinite loop [03:04] <wingo> it's better than before [03:04] <wingo> eos i'm referring to [03:04] Action: vishnu gives wingo a rubber chicken prize [03:05] <taaz> i find it extrordinarily difficult to program OOP with this GTK/GObject stuff [03:05] <wingo> taken [03:05] <taaz> almost impossible [03:05] <vishnu> taaz: why? you mean gstreamer or gobject? [03:05] <taaz> i mean what i said [03:05] <vishnu> i mean, gstreamer is not exactly following the usual gobject patterns [03:05] <taaz> you have to jump through hoops to do anything [03:06] <vishnu> dunno, i really like gtk 2.0 [03:06] <taaz> i'm more a hardcore OO coder.. like Objective-C, Java, Python, Smalltalk, etc [03:07] <wingo> where them python bindings at? ;) [03:07] <vishnu> taaz: yah, but those are higher-level languages. i like perl too, but doing OO in C is cool [03:08] <taaz> cool? are you crazy? [03:08] <taaz> wingo: yeah yeah... i'll do it sometime ;) [03:09] <vishnu> yah, i mean, i expect it to be harder than java and it definitely offers good performance [03:09] <taaz> its not just harder, its prohibitavly difficult to code anything [03:10] <vishnu> taaz: it's not that bad. you just have to be more formal about it. [03:10] <taaz> speaking of difficult, do these property things get inherited? [03:10] <vishnu> yah [03:13] mwc (ma...@ly...) joined #gstreamer. [03:14] <taaz> so would it be a good idea to expose a "name" arg in GstObject? It's already got the set/get and field, just no property. or does that break something? [03:19] <wingo> why would you do that? just for reference... when do you need to do that to the name? you already have GST_OBJECT_NAME [03:20] <wingo> just wondering, not criticising [03:23] vishnu (jo...@cx...) got netsplit. [03:23] chillywilly (da...@d7...) got netsplit. [03:23] chillywilly (da...@d7...) returned to #gstreamer. [03:23] <taaz> just wanted to give names on -launch command line [03:24] <wingo> use brackets, i think: gstreamer-launch filesrc [ mynamehere ] location=dsasdfa ! .... [03:24] <wingo> see the source, that's iirc [03:24] <taaz> oh, silly me. i shall look into that [03:24] vishnu (jo...@cx...) returned to #gstreamer. [03:24] <taaz> still may be nice to have that as an arg for interactive tools [03:25] <taaz> maybe readonly or something [03:33] <vishnu> taaz: yah, that sounds sensical [03:36] <wingo> o mega, where art thou? ;-) [03:37] <wingo> vishnu: do you use any of gnome in redael? [03:40] <wingo> and another question, do you use libglad? [03:40] <wingo> libglade i mean [03:51] <omega> wingo: ? [03:51] <wingo> name proerty? [03:52] <omega> seems reasonable [03:52] Action: wingo has spilt coffe on his mouse pad [03:52] <omega> about a dozen lines of code, can't hurt anything [03:57] <taaz> writable? [03:57] Nick change: taaz -> taaz-brb [04:10] <tnt> Have you guys seen this: http://advogato.org/article/357.html ? [04:14] <tnt> Maybe Red Hat will take it over. [04:26] mwc (ma...@ly...) got netsplit. [04:28] <vishnu> wingo: no, i don't link with gnome. i'm trying to minimize my library dependencies .. uh somewhat [04:29] <vishnu> wingo: no, i don't use libglade [04:32] mwc (ma...@ly...) got lost in the net-split. [04:32] <wingo> ok [04:32] <wingo> is gtk2 significantly different to work with? [04:33] <vishnu> dunno, i never used gtk-1.2 [04:33] <wingo> than gtk1.2? [04:33] <wingo> oh [04:33] <vishnu> i like 2.0 a lot though [04:36] <vishnu> wingo: there's a Changes-2.0.txt file in gtk+ 2.0 that lists all the programmer visible changes, afaik [04:36] <wingo> ok, i'll check it out [04:41] mwc (ma...@ly...) joined #gstreamer. [05:00] vektor (reet@HSE-Kitchener-ppp3508492.sympatico.ca) joined #gstreamer. [05:00] <vektor> taaz-brb: when you wake up say hi [05:06] chillywilly (da...@d7...) left irc: [05:07] Nick change: taaz-brb -> taaz [05:07] <taaz> hi [05:07] <vektor> hi! [05:07] <vektor> hey taa [05:07] <vektor> z [05:07] <vektor> taaz: How would you like to become FAMOUS? [05:08] <vektor> taaz: ... with or without the capslock enabled. [05:08] <taaz> does that mean i get women, money, and power too?! [05:08] <vektor> Yeah probably. [05:08] <taaz> is it legal? [05:08] <vektor> Completely! [05:08] <vektor> Interested? [05:08] <vektor> I know you are. [05:08] <taaz> sign me up! [05:09] <vektor> Ok! [05:09] <vektor> I need you to set up the configure script stuff for movietime. [05:09] <taaz> yeah, i figured it would be something like that ;) [05:09] <vektor> Hey and by the way, audio/video sync ROCKS now in movietime. [05:10] <vektor> It will never ever go out of sync. [05:10] <vektor> (+/- a small small constant factor). [05:10] <taaz> what did you change? [05:10] <vektor> I now use mmap and do things correctly. [05:10] <vektor> I use the soundcard as a clock. [05:10] <vektor> I'm slightly wrong on the timestamps of my video frames, but it's not noticable. [05:10] <vektor> The audio sync correction made a HUGE difference. [05:11] <taaz> how do you know the soundcard is accurate? [05:11] <vektor> No more magic constants. [05:11] <vektor> Ah that's the key. [05:11] <vektor> It doesn't need to be accurate. [05:11] <vektor> I use it as my clock. [05:11] <vektor> So if it's really 48003 then we're on it. [05:11] <vektor> Anyways, you should try it out and tell me how it goes, but I'd really appreciate it if you could configure-script it up. [05:11] <vektor> I want to post it for release this week. [05:11] <taaz> so like your fingers are broken or what? [05:12] <vektor> No I have some midterms. [05:12] <vishnu> configure isn't that bad. [05:12] <vektor> vishnu: I know, I've done it before. [05:12] <taaz> too much typing fixing sync, too tired now? [05:12] <vishnu> oh huh [05:12] <vektor> taaz: Yeah something like that. [05:12] <vishnu> heh [05:16] harobed (harobed@AC90B951.ipt.aol.com) left irc: Client Exiting [05:17] <vishnu> ogg123 doesn't work ... so i tried gstreamer and it's great :-) [05:19] <taaz> vektor: so how portable is this soundcard thing? [05:19] <taaz> and how flexable? [05:20] <taaz> i guess you're just concerned with dvds... but soundcard stuff isn't a general solution :( [05:20] <vektor> Um, well supported on linux and very flexible. [05:20] <taaz> like for gstreamer or something... we may just have video [05:20] <vektor> Well, sure. [05:20] <vektor> But if you have audio you must use it as a clock. [05:20] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [05:20] <vektor> You can't use the system clock and resample your audio. [05:21] <vektor> Like, y'know, in DirectShow for example, every fliter graph has a clock source, and it is usually the soundcard. [05:21] <vektor> So, that sort of proves you can have a system where you find the local clock. [05:21] <taaz> i know jack about direct* [05:21] <vektor> Exactly which is why I'm explaining. [05:22] mwc (ma...@ly...) left irc: Ping timeout for mwc[lychee.ntu.edu.au] [05:22] mwc (ma...@ly...) joined #gstreamer. [05:22] <vektor> Anyways I believe I can simulate what I'm doing using the normal read()/write() soundcard API but it will be tricky and just as unportable. [05:22] <vektor> Multimedia applications demand certain features from the underlying API. [05:23] <vektor> Just as the little voice in my head is demanding I get some sleep :( [05:23] omega (om...@om...) left irc: Ping timeout for omega[omegacs.net] [05:23] <taaz> so are you going to copy libmpeg2 to release? [05:24] <vektor> Potentially. [05:24] <vektor> It's stupid though. Part of the point of my app is to depend on readily-available libraries. [05:24] <vektor> But I need certain features from X and to argue my point I need an example application. [05:24] <vektor> :( [05:24] <taaz> whee! [05:25] <taaz> start snail mailing walken requests for api updates ;) [05:25] <taaz> he'd get a kick out of that [05:26] <vishnu> heh [05:26] <vektor> He might assume I'd put anthrax in them. Suspicious mail. I wouldn't want to risk it. :( [05:27] <taaz> i suspect that in all my bills. so i no longer open them [05:28] <taaz> i wonder how many people actually use that excuse when their utils are cut off ;) [05:28] <taaz> so this api stuff... [05:29] <vektor> api stuff. [05:29] <taaz> is it a big performance hit to have opaque types and accesor functions for these flags? [05:29] <taaz> it's much less prone to breakage that way [05:31] <vishnu> yah, much less prone to breakage. can you talk some sense into these guys? i'm ready to resort to violence [05:31] <vektor> taaz: uh... [05:31] <vishnu> i'm getting pumped up. i'm went to the gym today [05:31] <vektor> taaz: no it isn't huge. i don't really give a fuck. i just want to get something released that i can use. [05:32] <taaz> i'm pretty sure walken wont listen to my ideas. lose an extra cycle and the world ends blah blah [05:34] <vishnu> heh [05:42] <taaz> whoop! i'm making a cool little plugin [05:43] <taaz> keeps track of stream stats [05:43] <vishnu> what is it?? [05:43] <taaz> it's cool [05:44] <taaz> if i'm really k-rad i'll add moving average calc to it too [05:44] <taaz> and rates too [05:44] <vishnu> whoa! [05:44] <taaz> i'm not sure how to do some of this in a sane way though [05:48] <taaz> man, you don't realize how many bytes a decoded stream is until you keep track of it [05:48] <vishnu> heh [05:49] <vektor> taaz: So are you going to do that configure stuff? Or should I? [05:50] <taaz> if i do it will you code something cool while you wait? ;) [05:51] <taaz> i can do it... i need a bit of time to finish this other stuff though [05:51] <vektor> ok. [05:51] <vektor> Well if you ahve a chance by thursday that would be cool. [05:53] <vektor> goodnight [05:56] <taaz> alrighty [05:57] <taaz> damnit vektor [05:57] <taaz> you're pgp signed mails make my broken pgp setup take nearly a minute to just decide i can't verify your sig [05:58] <taaz> ok, yeah, this stats element kicks booty [05:59] <taaz> where did the omega go? [06:01] <taaz> this benow track is sweet [06:04] chillywilly (da...@d1...) joined #gstreamer. [06:13] walken (fo...@c1...) joined #gstreamer. [06:14] wingo (wi...@rd...) left irc: My damn controlling terminal disappeared! [06:22] Shippou (no...@cr...) joined #gstreamer. [06:41] wingo (wi...@rd...) joined #gstreamer. [06:41] Nick change: wingo -> wingo-sleep [06:45] mwc (ma...@ly...) left irc: Ping timeout for mwc[lychee.ntu.edu.au] [06:46] chillywilly (da...@d1...) left irc: Read error to chillywilly[d128.as14.nwbl0.wi.voyager.net]: EOF from client [07:02] mef-zZz (ma...@us...) joined #gstreamer. [07:02] Nick change: mef-zZz -> mef [07:03] <mef> anyone tried to build gstreamer on windoze? [07:03] <mef> I am trying right now, but am still diddling with getting pkgconfig to compile. [07:04] <mef> that is, I am trying it using cygwin. [07:14] mef (ma...@us...) left irc: [07:24] mwc (ma...@ly...) joined #gstreamer. [07:31] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [07:42] steveb (st...@no...) joined #gstreamer. [07:52] walken (fo...@c1...) left irc: l8r [08:09] mwc (ma...@ly...) got netsplit. [08:09] vektor (reet@HSE-Kitchener-ppp3508492.sympatico.ca) got netsplit. [08:09] steveb (st...@no...) got netsplit. [08:09] Shippou (no...@cr...) got netsplit. [08:09] wtay-tv (wi...@ca...) got netsplit. [08:09] evil_monk (trevo@208.141.162.68) got netsplit. [08:09] BBB-zZz (BB...@uc...) got netsplit. [08:09] ShrimpZzZ (marius@131.252.244.168) got netsplit. [08:09] ajmitch (aj...@p8...) got netsplit. [08:09] bstard (Lor...@a2...) got netsplit. [08:09] wingo-sleep (wi...@rd...) got netsplit. [08:11] Nick change: tnt -> tnt-zZz [08:15] ShrimpZzZ (marius@131.252.244.168) got lost in the net-split. [08:15] wtay-tv (wi...@ca...) got lost in the net-split. [08:15] evil_monk (trevo@208.141.162.68) got lost in the net-split. [08:15] BBB-zZz (BB...@uc...) got lost in the net-split. [08:15] bstard (Lor...@a2...) got lost in the net-split. [08:15] ajmitch (aj...@p8...) got lost in the net-split. [08:15] vektor (reet@HSE-Kitchener-ppp3508492.sympatico.ca) got lost in the net-split. [08:15] Shippou (no...@cr...) got lost in the net-split. [08:15] wingo-sleep (wi...@rd...) got lost in the net-split. [08:15] mwc (ma...@ly...) got lost in the net-split. [08:15] steveb (st...@no...) got lost in the net-split. [08:34] ShrimpX (marius@131.252.244.168) joined #gstreamer. [08:34] wtay-tv (wi...@ca...) joined #gstreamer. [08:34] evil_monk (trevo@208.141.162.68) joined #gstreamer. [08:34] BBB-zZz (BB...@uc...) joined #gstreamer. [08:34] bstard (Lor...@a2...) joined #gstreamer. [08:34] ajmitch (aj...@p8...) joined #gstreamer. [08:34] mwc (ma...@ly...) joined #gstreamer. [08:34] steveb (st...@no...) joined #gstreamer. [08:34] vektor (reet@HSE-Kitchener-ppp3508492.sympatico.ca) joined #gstreamer. [08:34] wingo-sleep (wi...@rd...) joined #gstreamer. [09:09] Shippou (no...@cr...) joined #gstreamer. [10:03] mwc (ma...@ly...) left irc: [x]chat [11:44] xav (xav@AMontpellier-201-1-4-3.abo.wanadoo.fr) joined #gstreamer. [11:48] <xav> can't build after a cvs update: make chokes at automake (after sucking my >700Mb ram and 2Gb swap) [11:48] <xav> probably terminated by the kernel OOM killer [11:50] uwe (user424@213.174.92.50) joined #gstreamer. [11:50] <uwe> hi. [11:51] <xav> hi [11:53] <uwe> any idea for good gtk docu? [11:53] Action: xav is away: I'm busy [12:15] Action: xav is back (gone 00:22:02) [12:21] <xav> uwe: GGAD ? [12:22] xav (xav@AMontpellier-201-1-4-3.abo.wanadoo.fr) left irc: boot new kernel [13:18] harobed (harobed@ACA045DE.ipt.aol.com) joined #gstreamer. [13:25] wingo-sleep (wi...@rd...) got netsplit. [13:25] harobed (harobed@ACA045DE.ipt.aol.com) got netsplit. [13:31] wingo-sleep (wi...@rd...) got lost in the net-split. [13:31] harobed (harobed@ACA045DE.ipt.aol.com) got lost in the net-split. [13:52] Nick change: ShrimpX -> ShrimpZzZ [13:58] MM (m.m...@20...) joined #gstreamer. [14:24] wingo-sleep (wi...@rd...) joined #gstreamer. [14:24] harobed (harobed@ACA045DE.ipt.aol.com) joined #gstreamer. [14:24] xav (xav@AMontpellier-201-1-4-3.abo.wanadoo.fr) joined #gstreamer. [14:26] xav (xav@AMontpellier-201-1-4-3.abo.wanadoo.fr) left irc: Ping timeout for xav[AMontpellier-201-1-4-3.abo.wanadoo.fr] [14:40] asmod (stevec@64.5.222.2) joined #gstreamer. [14:47] manuel (user422@213.174.92.50) joined #gstreamer. [14:48] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [15:05] MM (m.m...@20...) left irc: [15:40] Zygo (zblaxell@209.217.112.246) joined #gstreamer. [15:42] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [15:43] <uwe> manuel: du hier? [15:48] <manuel> uwe: hey uwe.. long time no see! ;-) [15:49] <uwe> manuel: look to the left ;) [15:50] <manuel> uwe: ah there you r.. [16:13] wingo-sleep (wi...@rd...) left irc: Ping timeout for wingo-sleep[rdu25-21-076.nc.rr.com] [16:15] wingo-sleep (wi...@rd...) joined #gstreamer. [16:51] <uwe> manuel: let's take the bus 6:05 (or earlier ;) [16:52] Shippou (no...@ot...) joined #gstreamer. [16:55] xav (xav@AMontpellier-201-1-4-3.abo.wanadoo.fr) joined #gstreamer. [17:02] <xav> hi sleepers [17:03] <xav> any1 got automake to work correctly right now ? It sucks all my mem [17:04] <taaz> what version? [17:13] <xav> cvs [17:14] <xav> err ... debian sid: automake (GNU automake) 1.4-p4 [17:19] harobed (harobed@ACA045DE.ipt.aol.com) left irc: Ping timeout for harobed[ACA045DE.ipt.aol.com] [17:25] vishnu (jo...@cx...) joined #gstreamer. [17:25] <vishnu> wtay? [17:26] uwe (user424@213.174.92.50) left irc: Client Exiting [17:27] manuel (user422@213.174.92.50) left irc: Client Exiting [17:35] xav (xav@AMontpellier-201-1-4-3.abo.wanadoo.fr) left irc: Read error to xav[AMontpellier-201-1-4-3.abo.wanadoo.fr]: EOF from client [17:41] harobed (harobed@ACA22943.ipt.aol.com) joined #gstreamer. [18:22] xav (xav@AMontpellier-201-1-4-3.abo.wanadoo.fr) joined #gstreamer. [18:25] <xav> Hi, I'm back. had to reboot because automake really trashed my machine [18:29] <xav> do you have problems with automake too ? [18:30] <taaz> no [18:30] <taaz> did you read README? [18:30] <taaz> newer automakes don't have that mem issue [18:30] <vishnu> i use automake 1.5-1 [18:33] <xav> mmh ... I'm sure I had automake 1.5 a while ago. is it possible that apt-get downgraded automake ? [18:33] <xav> and autogen.sh didn't warn me about an upatched automake, as it used to do [18:36] <xav> ok, I see now. debian has effectively renamed automake 1.5 to automake1.5, and downgraded automake to 1.4 [18:41] <taaz> yup, just trying to make everyones life suck just that little bit more ;) [18:42] <vishnu> yah, gotta suck out the corners [19:05] <xav> gstwinenc.cc: In function `void gst_winenc_chain(GstPad *, GstBuffer *)': [19:06] <xav> gstwinenc.cc:280: `BitmapInfo' undeclared (first use this function) [19:06] <vishnu> xav: yah, i just comment out that plugin [19:09] <xav> I just added #include <avifile/image.h> and it worked [19:09] <xav> (well, it compiles) [19:09] <vishnu> oh, cool [19:11] <xav> --- gstwinenc.cc 2001/10/17 10:21:25 1.7 [19:11] <xav> +++ gstwinenc.cc 2001/10/23 17:10:23 [19:11] <xav> @@ -26,6 +26,7 @@ [19:11] <xav> [19:11] <xav> #include "gstwinenc.h" [19:11] <xav> #include <avifile/creators.h> [19:11] <xav> +#include <avifile/image.h> [19:11] <xav> [19:11] <xav> /* elementfactory information */ [19:11] <xav> GstElementDetails gst_winenc_details = { [19:11] <vishnu> xav: email to the mailing list [19:14] <taaz> eh? compiles for me i think [19:16] <xav> taaz: not for me. perhaps debian sid's avifile headers have changed [19:16] <taaz> i'm using the almost latest debs [19:16] <xav> almost ;) [19:16] <taaz> pre-sdl craziness changes [19:18] <taaz> you know: "here's a patch, i didn't test it" does not inspire confidence ;) [19:19] <xav> taaz: but that's the truth ! make didn't even finished, I just know that gstwinenc.cc is compiled ok [19:22] <xav> taaz: how can I test that ? [19:23] <taaz> libavifile0.6-dev 0.6.0.20011002-1 [19:23] <taaz> compiled fine for me with that deb [19:32] <xav> well ... you're lucky ;) [19:35] <steveb> xav: i compiled avifile-0.6.0.20011003 from source and it works fine. are you using CVS gstreamer? [19:41] <xav> yes [20:14] ChiefHighwater (pa...@te...) joined #gstreamer. [20:19] <xav> what happens if I connect an src pad which can do video/raw YUV to a sink pad which can do video/raw RGB ? [20:20] <xav> currently it seems the sink element just takes the YUV buffers as if they where RGB, but I think I don't quite understand how this works [20:27] Nick change: wtay-tv -> wtay [20:27] <wtay> yo [20:30] wtay (wi...@ca...) got netsplit. [20:30] asmod (stevec@64.5.222.2) got netsplit. [20:30] Shippou (no...@ot...) got netsplit. [20:30] evil_monk (trevo@208.141.162.68) got netsplit. [20:30] BBB-zZz (BB...@uc...) got netsplit. [20:30] BBB-zZz (BB...@uc...) returned to #gstreamer. [20:30] asmod (stevec@64.5.222.2) returned to #gstreamer. [20:30] wtay (wi...@ca...) returned to #gstreamer. [20:31] evil_monk (trevo@208.141.162.68) returned to #gstreamer. [20:36] Shippou (no...@ot...) got lost in the net-split. [20:39] <vishnu> yo [20:40] <vishnu> wtay? [20:42] <xav> how do I gdb things in tests/ ? [20:42] <xav> libtool gdb doesn't work [20:42] <vishnu> xav: you can edit the shell scripts [20:43] <vishnu> xav: just scroll down to the line that says exec and add a similar line with gdb [20:43] msterret (mst...@xl...) joined #gstreamer. [20:46] <xav> whee ... gdb eats all CPU ... and seems doing I-dunno-what after "998.376000 MHz Pentium III (Coppermine) processor detected" [20:47] <vishnu> xav: it take a long time to read all the symbols from the plugins. be sure you have gstreamer-register up-to-date [20:47] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [20:48] <xav> vishnu: yeah, that's it. thx [20:50] <xav> argh. I have a sigsev in the colorspace converter [20:50] <wtay> vishnu: yes? [20:50] <vishnu> wtay: any comments on my patch for events support? [20:51] <wtay> vishnu: looks good.. [20:51] <vishnu> wtay: cool .. [20:55] <vishnu> gst_pad_send_event (gst_element_get_pad (fv->src, "src"), [20:55] <vishnu> gst_event_new_seek (GST_SEEK_BYTEOFFSET, offset, TRUE)); [20:55] <vishnu> i tested it with pad_send_event [20:55] <wtay> and? [20:55] <vishnu> wtay: it works fine [20:56] <wtay> cool [20:57] evil_monk (trevo@208.141.162.68) joined #gstreamer. [20:57] Nick change: tnt-zZz -> tnt [20:59] <tnt> Good Morning. [20:59] <wtay> yo [20:59] <tnt> Hmmm, I guess it's actually the afternoon. [20:59] <tnt> Good Afternoon. [21:00] <tnt> vishnu: So this new patch of your basically makes it so you can "scrub" now, correct? [21:00] <vishnu> tnt: scrub? [21:01] <tnt> Scrubbing is manually moving through the Video and Audio... usually by dragging something like a scroll bar... you control where you are, and the speed you move at. [21:01] <xav> is there a way to force colorpace output to RGB ? [21:01] <xav> from gstreamer-launch [21:02] <vishnu> tnt: yah. my video player has that ability, actually [21:02] <tnt> tnt: Can it go backward too? [21:02] <tnt> vishnu: Can it go backward too? [21:02] <vishnu> tnt: backward? not smoothy but it can skip back and play forward [21:02] <tnt> (I was talking to myself there, for a minute :-) ) [21:03] <wtay> xav: nope [21:03] <tnt> vishnu: Hmmm,... to bad... you patch almost gives scrubbing then. [21:03] <tnt> Does your player have a website? [21:03] <vishnu> tnt: to play backward, you'd need to seek before every frame. it's probably possible, but tricky [21:04] <wtay> tnt: scrubbing requires a bit more work from the app [21:04] <wtay> tnt: and mpeg2dec doesn't really handle it efficiently [21:04] <vishnu> tnt: yah, my player is part of redael -- http://redael.berlios.de/ [21:05] <vishnu> tnt: actually, i haven't uploaded the recent source code yet [21:05] <tnt> vishnu: You need some screenshots :-) [21:05] <vishnu> tnt: click on documentation for screenshots [21:06] <tnt> Is it GNOME based, GTK based, or something else? [21:07] <vishnu> tnt: just gtk 2.0 and gstreamer. no additional dependencies [21:08] <vishnu> tnt: join the mailing list if you like ;-) [21:09] <tnt> Where's the archive for the 'devel' mailing list? [21:10] <tnt> Or the 'announce' mailing list? [21:10] <vishnu> tnt: there aren't any messages yet. <g> i moved to berlios recently [21:10] <tnt> Ahhh... I see the mailing lists are *very* high volume, as your webpage describes :-) [21:11] <vishnu> tnt: heh, i don't want to surprise anyone if there really is a mail flood someday [21:11] <tnt> :-D [21:12] <vishnu> tnt: generally there is more anger about too much mail than too little ;-) [21:12] <tnt> Ya... the devel mailing list for my project, matterial, is quiet sometime, and then sometimes it jumps to like 50 messages a day... some people get quite a surprise. [21:13] <vishnu> tnt: yep :-) [21:16] <ajmitch> hi all [21:17] <tnt> Hello. [21:24] Nick change: tnt -> tnt-out [21:47] Action: taaz wants to go home and play with new guilaunch... [21:50] <wtay> vishnu: applied your patch [21:50] <vishnu> wtay: thank you! now our trees are in sync :-) [21:51] <wtay> cool, and now I understand what's going on :) [21:51] <ajmitch> hehe [21:56] <vishnu> my app can use the next generation of debs for gtk+ and gstreamer without additional modifications [21:56] <wtay> vishnu: is gst feature complete now for you? [21:57] <wtay> I'm writing a TODO list now... [21:58] <vishnu> wtay: it's not feature complete at all. rather it has the absolute minimum sufficient features. ;-) [21:58] <wtay> what fdo you need more? [21:59] <vishnu> wtay: A/V sync would be nice. a time->offset mapping is the most urgent thing, for my stuff [21:59] <vishnu> wtay: i'd really prefer to seek by time instead of by offset [21:59] <wtay> vishnu: yup, clocking is essential [22:00] <wtay> that's a major todo still [22:00] <vishnu> wtay: it would also be neat to handle vcdsrc, dvdsrc and mpeg1 files. gstreamer can probably do it now, but i haven't spent the time to figure it out [22:00] <wtay> ok, those are plugins, someone should just write/rewrite them [22:01] <wtay> mpeg1 should work fine too [22:01] <vishnu> wtay: yah it's mostly a matter of testing and tweaking [22:01] <wtay> what I would like to do is send the seek event in the mpeg2dec src pad [22:01] <ajmitch> wtay: gonna try for a release in the next month or so? ;) [22:02] <wtay> which sends it to the demuxer which does the time/byte conversion using timecache [22:02] <wtay> ajmitch: "or so", yes :) [22:03] <vishnu> wtay: seek to a particular frame? dunno, maybe audio is easier? [22:03] <wtay> vishnu: time seek [22:03] <wtay> vishnu: frame == time/rate (in the naive case) [22:04] <vishnu> wtay: yah time seek. but will the seek has per-frame resolution or 48khz resolution (audio) [22:04] <wtay> vishnu: per frame I would say [22:04] <wtay> vishnu: for audio it should be sample accurate [22:05] <vishnu> wtay: well .. whatever. any kind of time->offset mapping is better than nothing [22:05] <wtay> yes [22:05] <wtay> time seeking is essential for an NLE lib I guess [22:05] <vishnu> yes, certainly [22:06] <wtay> and I would love to play some media types backwards too :) [22:06] <vishnu> presently, all my bookmarks are tied to a particular vob file. i can't rescale it or anything because that would change the bookmark offsets (!) [22:07] <vishnu> slow motion, backwards, etc. yah [22:08] Action: wtay dreams of _set_clock(- _get_clock()) [22:15] steveb (st...@no...) left irc: advance! [22:16] ajmitch (aj...@p8...) left irc: Ping timeout for ajmitch[p8-max4.dun.ihug.co.nz] [22:21] omega (om...@om...) joined #gstreamer. [22:27] msterret (mst...@xl...) left #gstreamer. [22:29] <wtay> yo [22:29] <omega> yo [22:30] Shippou (no...@cr...) joined #gstreamer. [22:30] <wtay> omega: I'm going to make the scheduler plugable [22:31] <omega> eek <g> [22:31] <wtay> hehe [22:31] <omega> any reason to do that now as opposed to later, say afer doing a release? [22:31] <wtay> no [22:32] <omega> ok, let's wait <g> [22:34] Action: omega just RMA'd another maxtor drive that's having troubles [22:34] <omega> but this [22:34] <omega> this'll give me a chance to dupe drives more easily when I build my new gateway machine later this week [22:39] <ChiefHighwater> so, what number will this release be? [22:39] <omega> 0.2.2 [22:39] <omega> with 0.3.0 coming up soon [22:40] <wtay> we need a default event handler.. [22:40] <omega> yup [22:41] <wtay> any idea as how it will work? [22:41] Action: omega reads through the -cvs mail [22:41] <ChiefHighwater> codename: mugshot...a picture taken for future reference? [22:42] <wtay> mugshot? [22:42] <ChiefHighwater> the picture police take of you to put on file when you get arrested [22:42] <omega> wtay: picture of you taken through your beer mug <g> [22:43] <omega> same thing in some cases [22:43] <wtay> ah :) [22:43] <ChiefHighwater> how about "Thumbnail"...those pics are often 2x2 [22:43] chillywilly (da...@d4...) joined #gstreamer. [22:43] <ChiefHighwater> Ello [22:43] <omega> taaz: here? [22:44] <ChiefHighwater> just playing off the idea that this release is really a snapshot so that we can do the major push for 0.3 [22:44] <omega> well, it's got a huge amount of new stuff [22:44] <ChiefHighwater> maybe "snapshot"? [22:44] asmod (stevec@64.5.222.2) left irc: [x]chat [22:44] <wtay> streamshot? [22:45] <wtay> hehe, we're actually going to have 100% API doc coverage for this release :) [22:45] <omega> cool [22:45] <ChiefHighwater> but isn't a lot of the new stuff <almost> done and really waiting for 3.0 to really get movin? [22:45] Action: wtay proud [22:45] <omega> I get a 1.8MB diff between 0.2.1 and HEAD [22:45] <omega> and I dunno if that includes all the new files or not, checking [22:46] <omega> no, it doesn't... [22:46] <wtay> I lost my cvs archives... you might need to send them to me so I can make a compilation of changes [22:46] <omega> any idea what the cvs diff flag is to do that? [22:46] <omega> lost your cvs archives? [22:46] <wtay> mail commit logs.. [22:47] <omega> ok, yeah, I can gather them [22:47] <omega> back to what date? [22:47] <wtay> 0.2.1 release date.. [22:47] <omega> looks like ~3mo ago [22:47] <omega> wow, thought it was longer ago [22:48] <omega> June 27 is last change date on configure.base for that tag [22:48] <wtay> 27 June, 2001 [22:48] <ChiefHighwater> ~4 mo [22:48] <omega> 3mo3wk [22:49] <omega> ok, rdiff shows a diff of 2.9MB [22:49] <wtay> wow [22:50] <omega> a large chunk of which is ChangeLog [22:50] <wtay> hmm [22:50] <omega> which is about 512KB [22:51] <omega> btw, that's a unidiff, too [22:52] ajmitch (aj...@p3...) joined #gstreamer. [22:52] Zygo (zblaxell@209.217.112.246) left irc: Read error to Zygo[209.217.112.246]: Connection reset by peer [22:53] <taaz> omega: here now [22:53] <taaz> what up? [22:53] <omega> taaz: did you have any neat ideas re: events and buffers last night? [22:54] <wtay> thomasvs_: how much longer do we have to look at that stupid compile error? [22:55] <taaz> omega: some, though i was busy doing other cool stuff [22:57] <omega> anything interesting yet? [22:57] <taaz> wtay: could add something about how to get timestamps from byte interface of bytestreams to TODO list [22:57] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [22:57] <taaz> wtay: though any solution i have come up with is no better than just dropping the byte interface and just using buffers [22:57] <taaz> omega: i made a cool plugin last night [22:58] <omega> oh? [22:58] <wtay> taaz: the definition of a timestamp on a buffer is the time the first byte of the buffer should be presented.. [22:58] <wtay> taaz: bytestream does that just fine [22:58] <omega> wtay: -cvs archives sent (415KB .bz2) [22:58] <taaz> gststatistics, identity-type element that keeps track of buffers/bytes/events. does some basic timing on it too so you get bandwidth reports as well [22:58] <wtay> omega: ok, I'll compile a nice "new stuff" log [22:58] <omega> 795 messages [22:59] <omega> cool [22:59] <wtay> neat [22:59] <omega> though the trace mechanism will be able to do that offline later, too [23:00] <taaz> wtay: if you use the byte interface then only way to get timestamps is to do a buffer peek too... which is really just more work than only doing the buffer peek and using the data pointer from it (i think) [23:00] <wtay> taaz: true [23:00] <taaz> i'm thinking in terms of mpeg2parse setting timestamps and wanting a52dec (or whatever) to just pass that through [23:01] <omega> bytestream could store the latest timestamp (but it would have to cycle it through as buffers are flushed out of the tail position) [23:01] <wtay> hmm, -guilaunch seriously rocks, but it's slowly turning into gsteditor.. [23:01] <omega> well, to some extent that's good, as far as the various dialogs [23:01] <omega> though he should look at what the editor has so far [23:01] <omega> I'll mail him [23:02] <ChiefHighwater> isn't he a little big? How much postage would that be? And won't he need some food? [23:02] <omega> hmm, depends on how far and how high (airmail or slow-boat) [23:03] <ChiefHighwater> and make sure they don't burn him with the rest of the white house mail [23:03] <omega> oh? [23:03] <taaz> omega: what should i do with this stats element? it's really cool and imho almost useful enough to put in gst/elements/ ;) [23:03] <wtay> taaz: isn't it possible to add the features to identity? [23:04] <taaz> wtay: no, i refuse. we have this cool pipeline stuff, let's use it. no reason to put non-identity features in identity. i question the addition of that duplicate stuff in there even. [23:05] Action: taaz will start making demands next ;) [23:05] Action: omega needs to figure out what this duplicate stuff is [23:06] <taaz> wtay: speaking of duplicate, it's not going to work too well with events [23:06] Action: wtay needs to figure out why identity should not count buffers [23:06] <taaz> wtay: and identity doesn't handle eos (ie, going to PAUSE) properly [23:06] <wtay> taaz: that's a bug in identity [23:07] <taaz> wtay: seriously, why should identity do -anything- except a super simple passthrough? [23:07] <wtay> taaz: oh, identity does a lot more these days, like different scheduling options and such [23:07] <wtay> taaz: and sleep time... [23:08] <omega> taaz: it is a testing plugin [23:08] <omega> it was written to stress-test the scheduling, primarily [23:08] <wtay> taaz: this plugin is designed to test out a lot of stuff [23:09] <omega> just like fakesrc and fakesink [23:09] <wtay> fakesrc is out most complex plugin now :) [23:09] <taaz> btw, i have no clue how to watch video with -launch. it fails any way i try it :( [23:10] <wtay> identity should also be able to copy-on-write buffers in the future.. [23:10] <omega> taaz: such as? [23:10] <wtay> taaz: filesrc location=/opt/data/DolbyDigitalBroadway.vob ! mpeg2parse video_0! queue ! { mpeg2dec ! xvideosink } [23:10] <wtay> works for me [23:11] wtay (wi...@ca...) left irc: Read error to wtay[cable-195-162-215-201.upc.chello.be]: EOF from client [23:11] <omega> whoops [23:11] <vishnu> taaz: don't use a thread [23:11] <omega> vishnu: you have any ideas as to how to *fix* threads? [23:11] <vishnu> omega: i don't think you want to hear my suggestions ;-) [23:12] <omega> why's that? [23:12] <omega> taaz: works for me too [23:12] <vishnu> how about you get an SMP machine and try to debug it? [23:12] <omega> I'm working on it. but I asked if *you* had any ideas [23:12] Action: vishnu hides under a rock [23:13] <omega> great. Xv is screwed up on my machine [23:13] wtay (wi...@ca...) joined #gstreamer. [23:13] #gstreamer: mode change '+o wtay' by ChanServ!s@ChanServ [23:13] <wtay> when it doesn't crash my X server :) [23:13] #gstreamer: mode change '-o wtay' by wtay!wi...@ca... [23:13] <taaz> ok, so like it doesn't work for me, yes i'm using the same cmd line from the man page and all [23:14] <omega> you have HEAD [23:14] <omega> ? [23:14] <taaz> of course! ;) [23:14] <omega> hmm, I'd confirm that... [23:14] <omega> not just have the source, but, um, built too <g> [23:15] <taaz> gotta go push start a car... brb [23:20] <omega> wtay: so, is that TODO for 0.2.2 ? [23:21] <wtay> default event handlers. [23:21] <omega> ok, how should those work? [23:21] <wtay> an element should return FALSE in the event handler, which would trigger the default handler for one [23:21] <omega> ok, that's only for chained and reverse events [23:22] <wtay> for elements without event handler the default handler would be triggered always [23:22] <omega> hmm? [23:22] <wtay> yes [23:22] <omega> should I look at a specific example plugin? [23:22] <wtay> not that I know of, no [23:23] <omega> well, let's take a52dec first [23:23] <wtay> we have two cases basically.. [23:24] <wtay> a) the event handler (for which the _send_event code looks good) [23:25] <wtay> b) the downstream events (which require an element flag) [23:26] Zeenix (ze...@13...) joined #gstreamer. [23:26] <Zeenix> hello all [23:26] <wtay> yo [23:27] <Zeenix> wtay: i think the RtpRecv will be ready tommorow [23:27] <wtay> cool [23:27] <Zeenix> wtay: but cant guarantee that it would work too [23:28] <wtay> omega: a52dec is almost correct, the only thing it should do is call the default handler for the events it doesn't know abourt [23:28] <vishnu> Zeenix: release early & often :-) [23:28] <omega> wtay: ok [23:28] <wtay> omega: thing is... [23:28] <Zeenix> omega: oh, i see reply from robla in my inbox [23:28] <omega> cool [23:29] <wtay> omega: the default handler should either do _push (event) or gst_pad_send_event [23:29] <omega> hmmm [23:29] <Zeenix> omega: forwarding the message to you, its small [23:29] <wtay> upstream or downstream... [23:30] <omega> well, ok, the case of a52dec... [23:30] <omega> the default case of the event switch [23:30] <omega> in this case it's looped, so we can do a pad_send_event [23:30] <Zeenix> omega: email ? sorry, i should remember it [23:30] <omega> om...@te... [23:30] <wtay> hmm.. pad_send_event is meant for upstream events.. [23:31] <taaz> can i spam you all for a sec? [23:31] <taaz> ./tools/gstreamer-launch httpsrc location=http://us.benow.ca:10390/ ! statistics [mp3] buffer_update_freq=10 ! mad ! statistics [raw] bytes_update_freq=100000 ! fakesink silent=true [23:31] <omega> only if we don't end up on someone *elses* list... [23:31] <taaz> statistics: (mp3) total: s:19.4694 buffers:400 bytes:558348 events:0 [23:31] <taaz> : (mp3) total: buf/s:20.5451 B/s:28678.2 e/s:0 B/buf:1395.87 [23:31] <taaz> : (mp3) last: s:0.570302 buffers:400 bytes:558348 events:0 [23:31] <taaz> : (mp3) last: buf/s:17.5346 B/s:18323.6 e/s:0 B/buf:1045 [23:31] <omega> [] ? [23:31] <taaz> statistics: (raw) total: s:19.9688 buffers:1342 bytes:6183936 events:0 [23:31] <taaz> : (raw) total: buf/s:67.2048 B/s:309680 e/s:0 B/buf:4608 [23:31] <taaz> : (raw) last: s:0.55719 buffers:1342 bytes:6183936 events:0 [23:31] <taaz> : (raw) last: buf/s:39.4838 B/s:181942 e/s:0 B/buf:4608 [23:31] <taaz> isn't that neat? [23:32] <omega> cool [23:32] <wtay> nice [23:32] <omega> that required mods to -launch ? [23:32] <Zeenix> omega: sent [23:32] <taaz> no [23:32] <omega> what's the [mp3] then? [23:32] <taaz> [] is in there already to name elements [23:32] <wtay> [] is an element name IIRC [23:32] <omega> oh [23:32] <wtay> arguably not needed btw [23:32] <omega> a space is allowed before it? [23:32] <taaz> that's why i asked about the GstObject name property yesterday. [23:32] <wtay> name=mp3 would work too actually [23:33] <omega> wtay: do we have that property? [23:33] <taaz> wtay: i dont think name is exposed as a property yet [23:33] <wtay> hmm.. no.. [23:33] <wtay> it should [23:33] <taaz> could/should be [23:33] <wtay> should be [23:33] <omega> should be in gstobject.c [23:33] <wtay> yes [23:34] <omega> taaz: you can add that in a couple minutes, while we figure out events <g> [23:34] <taaz> so anyway, this stats thing can print or send an 'update' signal on buffer/byte/event count thresholds... it's good for timing a 'dd' test too... gstreamer is actually not too bad as far as a dd replacement goes ;) [23:35] <wtay> omega: so, how should the default handler know about the _push? [23:35] <omega> wtay: we've gotta have several default handlers ;-( [23:35] <wtay> omega: yup.. unfortunatly.. [23:35] <omega> taaz: no, especially once we have a filesink written the same way as filesrc, with bufferpools and such [23:36] <omega> wtay: my GstDiskFormatCache idea has to be retought somewhat in light of the filesrc, esp with the possibility of writable buffers coming from filesrc.... [23:37] <omega> also, has anyone done a docs build recently that can be put on the website as a snapshot? [23:38] <wtay> I have [23:38] <taaz> they build now?! [23:38] <wtay> I'll put them online soon [23:38] <wtay> taaz: they always did [23:38] <taaz> haha [23:38] <wtay> on my PC :) [23:38] <wtay> which is simply debian unstable so... [23:39] <wtay> oh right.. you need one obscure .deb not in unstable.. [23:39] <taaz> the only reason i haven't updated the debs on the site is because i spend crazy amounts of time trying to fix docs to build at all [23:40] <taaz> and i just give up and curse at those voodoo makefiles [23:40] <wtay> make html is all you need [23:40] <wtay> s/you/I/g :) [23:40] <Zeenix> omega: so are you allowing the "rate" in the osssrc/sink to start from 4 or 5000 instead of 8000 ? [23:40] <taaz> ok, i'll try again soon [23:41] <omega> the lower limit is set by the OSS API, which I think is actually 8KHz [23:41] <wtay> omega: actually 1 API for a default handler is enough [23:41] <taaz> wtay: it's unacceptable to require any debs not in the deb arcives though :( autobuilders need to be able to build this thing at some point [23:42] <Zeenix> the osss goes to 4000 on my sys if asked too [23:42] <Zeenix> s/too/to [23:42] <omega> ok, then the caps should be changed [23:42] <wtay> taaz: gnome-doc-tools_1.0-1_all.deb [23:43] <wtay> Zeenix: you can chenge them, ther are arbitrarily choosen [23:43] <Zeenix> omega: yes & if the sound card doesnt support it, it would fall to the nearest match [23:43] <wtay> omega: ..since the only time you're going to call the default handler is for downstream events [23:44] <omega> hmm, ok [23:44] <wtay> omega: for upstream events you simply return FALSE in the handler [23:44] <omega> well, so there are two general classes of problems afaict: single dest pad, and multi dest pad [23:44] <Zeenix> wtay: check inbox plz [23:45] <wtay> omega: just loop over the pads and send the event on all of them [23:45] <omega> ok, so how does the default handler do that? does the plugin loop over them all? [23:45] <wtay> omega: in case of EOS, change the elements state [23:46] <wtay> omega: the default handler gets the pads of the plugin and loops over them.. [23:46] <omega> ok [23:46] <omega> ok, so let's make a list of where event handling needs to be added in the core... [23:46] <wtay> so, there are two default handlers internally, [23:47] <wtay> omega: look at gstpad.c:1987 [23:47] <omega> gboolean [23:47] <wtay> Zeenix: can I add this to cvs soon? [23:48] <omega> ok [23:48] <wtay> that is the upstream handler [23:48] <omega> ok, so *pad is a srcpad? [23:49] <wtay> it calls the event handler of the plugin if it has one and triggers the default upstream handler if the handler returns FALSE [23:49] <wtay> omega: yes [23:49] <wtay> omega: for now it is [23:49] <Zeenix> wtay: hmm..., lets hope it will work when i build it tommorow [23:50] <wtay> omega: then there is something like a default handler on line 1963 [23:51] <omega> right [23:52] <Zeenix> wtay: but i dont know what should i do with the rate, i wanted to make it an object var. in GstRtpRecv like port & mtu, but i think it doesnt fit there [23:52] <wtay> Zeenix: it's ok, I think [23:53] <wtay> omega: test/events/seek.c is an example of and app sending a seek event [23:54] <Zeenix> omega: i saw that RoboCar for which i'll be coding for, it seemed so primitive, wheels were made of wood [23:55] <omega> heh [23:55] Action: wtay wants guppi 0.40 [23:56] Action: Zeenix announces RH Linux 7.2 incase anybody doesnt know that already [23:56] <omega> well, apparently ximian and enigma (rh72) are quite incompatible right now, so I'll be waiting until that sorts out [23:58] <wtay> omega: gst_pad_event_default should remain static and be renamed to _event_default_upstream or something. Then gst_element_event_default should handle the downstream events [00:00] --- Wed Oct 24 2001 [00:02] <omega> so, I'm gonna make a table of what we need to do [00:02] <Zeenix> wtay: had a look at the code ? [00:03] <wtay> Zeenix: I'll do that tomorrow I you don't mind [00:03] <wtay> I need sleep soon [00:03] Action: wtay is very tired [00:03] <omega> I'll see if I can write this up before you sleep [00:04] <wtay> ok. I'm going to prepare myself... [00:04] <Zeenix> wtay: no i dont, you need a mind to mind anything <g> [00:04] Action: vishnu volunteers to sleep for wtay so wtay can stay up longer [00:04] <ChiefHighwater> hehe [00:05] <Zeenix> vishnu: plz do that for me too [00:05] <vishnu> :-) [00:05] <omega> if that were possible, we'd have sleep farms all over the place. would be nice [00:06] <taaz> job title: "senior sleeper engineer" [00:07] <Zeenix> taaz: wtay fits for this job, he could then code all the night long [00:08] <wtay> I'm willing to buy 4 hours of good quality sleep :) [00:08] <wtay> $10/hour [00:09] <vishnu> no thanks, i'll do it for free ;-) [00:09] <omega> wtay: so in a chained element, if it handles events, how do we go about forwarding events? do we want a generic forwarder? [00:10] <wtay> omega: yes [00:10] <omega> or if it handles EOS, it should have the code for sending the event on in each event case? [00:10] <taaz> ok, sorry to ask a rtfm question: what should i look at to acquire a clue about this new event system? [00:10] <taaz> how do you have a generic forwarder? [00:11] <vishnu> taaz: plugins/mpeg2/parse/mpeg2parse.c has fairly extensive event support [00:11] <taaz> does that just default to taking any event on input and forwarding to all output pads? [00:11] <wtay> taaz: yes, basically [00:11] <wtay> taaz: but it also "handles" it, like doing stuff with an EOS event [00:11] <omega> wtay: that's not what I mean.. [00:11] <taaz> maybe that should be done for caps too... ie, so no need for that sort of code in identity element [00:11] <wtay> omega: ? [00:12] <omega> for events that have been explicitly handled, do we want a function that can be called that handles sending the events on to all src pads? [00:12] <omega> or just have the element do that [00:12] <wtay> omega: oh, ok if the events handler of the element doesn't return FALSE we do nothing [00:13] <omega> wait... in the chain function [00:13] <wtay> omega: else we perform everything we would do when the element didn't have a handler [00:13] <wtay> omega: oh, right.. [00:13] <wtay> omega: but that's the same thing as with a loop based element [00:13] <omega> right [00:13] <omega> Incoming events: Need a flag to say whether an element can handle events. If flag is set, the event [00:13] <omega> pointer get passed to the chain function. If not, the scheduler has to insert a handler that can deal [00:13] <omega> with events. [00:14] <omega> for chained elements, downstream events [00:14] <wtay> yes [00:14] <wtay> the scheduler should do a IS_EVENT probably.. [00:14] Shippou (no...@ot...) joined #gstreamer. [00:15] <omega> so this isn't a handler, it's *in* the chain function [00:15] <omega> because the chain function is passed an event instead of a buffer [00:15] <wtay> yes, downstream event s don't have a handler [00:15] <wtay> s/have/require/ [00:16] mef (ma...@us...) joined #gstreamer. [00:17] <omega> yo [00:17] <wtay> but that requires an IS_EVENT check on every push :( [00:17] <mef> howzit [00:17] <wtay> hi [00:17] wingo-sleep (wi...@rd...) left irc: Ping timeout for wingo-sleep[rdu25-21-076.nc.rr.com] [00:17] <omega> wtay: right ;-( [00:17] <taaz> wtay: so? [00:17] <taaz> isn't the element going to have to do that anyway? [00:17] <mef> well... I have not made any progress on getting gstreamer to compile on my win/me laptop. [00:18] <mef> Still need to fix pkgconfig to compile using the version of cygwin I have. [00:18] Action: omega runs to find a gun with which to put mef's laptop out of its misery [00:18] <taaz> from a user standpoint (ignoring underlying impl for the moment) it would seem like you would want the following: [00:18] <omega> or a win2k or win98 disc... [00:18] <mef> Who else has attempted to get gstreamer to work under cygwin? [00:18] <omega> no one afaik [00:18] <mef> ok [00:18] <taaz> for chain elements: the normal buffer handling code we have now [00:18] <vishnu> met: what a waste of time, imho [00:18] <mef> I'll keep trying. [00:18] <taaz> plus event callbacks as well [00:19] <taaz> that way a buffer handling function could handle buffers [00:19] <mef> I know where to look to get directX based audio and video. So I am hoping to get the core compiled on windows on cygwin and then get a simple example to work. [00:19] <taaz> and event callback function(s) could handle specific events [00:19] wingo-sleep (wi...@rd...) joined #gstreamer. [00:19] <taaz> get rid of the case statements in the element code [00:19] Nick change: vishnu -> vishnu-sleep [00:20] <wtay> taaz: we discussed that and decided not to do that.. [00:20] <taaz> which would allow elemnts to easily have default implementations [00:20] <taaz> this doesnt work for loops i think [00:21] <taaz> wtay: ok, but it seems likeyou are just hacking around doing this and just making things harder for the elements and maybe saving a few checks here and there [00:21] <taaz> sorry, i really dont quite understand all this yet [00:21] <wtay> taaz: I don't think we can avoid checks anyway [00:21] <mef> On a completely different note: anyone have an idea whether gstreamer could work on EL/IX based systems -- like eCos? [00:21] <omega> possibly [00:21] <wtay> taaz: you either split buffers and events in the scheduler or in the plugin, the check is the same [00:21] <omega> if it's POSIX-ish, very likely [00:22] <taaz> wtay: exactly, so why not keep the checks in just one place rather than in every plugin? [00:22] <omega> there's a point to that, but... [00:22] <mef> ok... something worth looking into. [00:22] <omega> the problem is that in many cases the events must be handled inline with buffers [00:23] <wtay> right.. [00:23] <omega> and to do that with a separate chain/loop and event handler means lots of element-specific issues to sync between then *anyway* [00:23] <taaz> omega: that's certainly the case for loops... but chain elemetns are serialized right? [00:23] <mef> Equator's MAP-CA doesn't run Linux. They are trying to port Linux to it, but I think it might be better of if they worked on getting eCos to work on it. [00:23] <omega> heh [00:23] <wtay> taaz: right.. [00:24] Action: omega needs to spend some time dreaming up better solutions to these problems, like making the whole scheduling system event (!gstevent) driven [00:24] <taaz> know what i would do? [00:25] <taaz> eh, something like what omega is probably thinking i guess ;) [00:25] <wtay> it would unify the upstream and downstream event handler though... [00:25] <omega> wtay: for chained elements, possibly [00:25] <mef> omega: have you read about Eddie Kohler's "click modular router". [00:25] <taaz> you could have a default 'chain' handler for 'events' [00:26] <omega> mef: nope, reference? [00:26] <taaz> GstBuffers would just be a event subclass [00:26] <omega> taaz: right [00:26] <mef> http://www.pdos.lcs.mit.edu/click/ [00:26] <omega> event subclass is quite heavy [00:26] <taaz> and you put the case statement thing in the default handler [00:26] <omega> we already subclass GstData [00:26] <taaz> which for chained elements would do the dispatch to various callbacks based on type [00:26] <omega> for both events and buffers [00:27] <mef> It is primarily for building IP router like things, but there may be some ideas that will apply to gstreamer. [00:27] <omega> yeah, I'll have to read through it [00:27] <taaz> so chain element could override the default handler for custom needs, else it would be straightforward to have defaults for everthing and just override the events (including buffer handling) you need [00:27] <mef> Worth reading his TOCS paper and his dissertation on the topic. [00:27] <omega> first glance it looks quite similarly architected [00:28] <taaz> omega: define 'heavy'? like with experimental numbers ;) [00:28] <omega> wow. ok, everyone go read through that click reference [00:28] Action: Zeenix wont be able to sleep untill he see's his code doing 'physical' work [00:28] <taaz> in any case, from an OOP point of view this casting of GstBuffer to an event type is wrong [00:28] <omega> taaz: heavy as in memory [00:28] <omega> taaz: both are subclasses if GstData [00:29] <omega> we're calling it GstBuffer still because it's a typing convenience [00:29] <omega> as in, not having to add 20KB of code to cast everything around [00:29] <taaz> omega: with all due respect, that's total crap way to code [00:29] <mef> What is nice about "Click" is that they nicely defined a push and pull model. Mind you that i have not worked with gstreamer and don't know whether you have defined the same things. But heck... engineering is not about re-inventing the wheel but making it rounder. :) [00:29] <omega> taaz: you try adding GST_BUFFER(data) in every single place that needs it [00:29] <omega> and see what the code looks like then [00:30] <taaz> if you forget to check if it's an event or not, and assume a buffer, you can segfault [00:30] <omega> mef: yeah, it looks quite similar so far, I'll have to read what they do for their dataflow model [00:30] <taaz> if you get in a GstBuffer* you should reasonable be able to expect its a buffer. [00:30] <mef> glad to be of some help... [00:30] <omega> but their quick pipeline description language is quite similar to gst's [00:30] <mef> They have some neat optimizations that you might want to read into. [00:30] <omega> yeah [00:31] <mef> well... I better get back to working on my dissertation defense slides... god I can't wait to be done with that! [00:31] <omega> heh [00:31] <mef> cia [00:31] <mef> ciao [00:32] <omega> l8r [00:32] mef (ma...@us...) left #gstreamer. [00:32] <BBB-zZz> hmm..... [00:33] Action: BBB-zZz hasn't talked all day [00:33] <BBB-zZz> that's bad [00:33] <wtay> taaz has a point [00:33] <taaz> anyway, this is one reason i wrote that lat tests and now stats element. so we can test performance of new design ideas [00:33] <omega> yes, I agree that it's nasty, but passing GstData* around and casting it all the time is nastier. I know, I tried to do it [00:34] <wtay> the problem is solved with two handlers [00:34] <omega> for chained elements, yes [00:34] <taaz> i really have no basis to determine if any of this stuff makes any noticable difference in the end [00:34] <taaz> wtay: for chained elements you could actually have different handlers for each event too [00:35] <wtay> taaz: I don't like that idea [00:35] <omega> many more pointers, but no performance difference, and harder to code [00:35] <omega> more functions to write [00:35] <wtay> taaz: it doesn't scale too well and you need much more pointers [00:35] <omega> switch has to be done somewhere, might was well consolidate it into the event handler [00:36] <taaz> ok, well a convienience func/struct could be setup to do that anyway [00:36] <taaz> or inconvienience depending on your viewpoint [00:36] <wtay> I still prefer a switch in the plugin [00:37] <taaz> we can do both [00:38] <omega> there's no reason to do both [00:38] <wtay> or maybe I could live with a callback that takes the individual event fields appart and send them as arg to the callback [00:39] <wtay> if there is no registered callback it gives the raw event and you can still do the switch [00:39] <wtay> bleagh [00:39] <taaz> maybe i'll whip up a python prototype [00:39] <taaz> just to compare design ideas [00:40] <wtay> great, I'll learn python then to understand it :) [00:41] <taaz> course, pythons lack of switch statement will make it look odd... [00:41] <wtay> taaz: if..else if..else if...else if... else.. :) [00:41] <taaz> yeah i know [00:41] <taaz> there's a school of thought that OO programs have no need for switch statements [00:42] <wtay> the point of a switch is that the compiler does a binary search on the values.. [00:42] <taaz> can all be done, and be mucho more flexable, with polymorphism [00:42] <taaz> but this is low level stuff so ... [00:43] <Zeenix> i'll be right back soon [00:43] <omega> wtay: docs snapshot before you sleep? [00:43] Zeenix (ze...@13...) left #gstreamer. [00:43] <wtay> omega: tomorrow.. I'm off to bed now [00:43] <wtay> cya [00:43] <omega> bleagh [00:43] <omega> ok [00:43] Nick change: wtay -> wtay-zZz [00:43] <taaz> yeah yeah... python prototype. i'll dothat so we can maybe talk about this in some sane way [00:43] <omega> ok [00:43] <omega> I'm gonna go read the click stuff [00:50] ChiefHighwater (pa...@te...) left irc: [00:54] arik (arik@168.191.238.91) joined #gstreamer. [00:54] <arik> hi all [00:54] <ajmitch> hey arik [00:54] <arik> hey aj [00:56] sienap (sy...@s3...) joined #gstreamer. [00:56] <sienap> hoi [01:07] Zeenix (ze...@13...) joined #gstreamer. [01:07] <Zeenix> yo again [01:07] <arik> yo [01:08] <Zeenix> so should i assume that no one minds a frequency or rate var. in RtpRecv ? [01:09] <Zeenix> & a property installed for it like port & mtu [01:12] <omega> Zeenix: it should come from the caps that are negotiated with its peer element [01:12] <arik> omega: howdy [01:12] <omega> yo [01:12] <arik> i got the second phone line [01:12] <arik> finally [01:13] <omega> heh [01:14] <sienap> yo zeenix! [01:14] <arik> omega: time to start thinking about a picogui videosink [01:15] <Zeenix> omega: it doesnt seems possible ATM [01:15] <sienap> we didn't have any phone bills for around 10 months.. .. friday one came in and they calculated the other months from the current cost [01:15] <sienap> that is around 7k guilders (being something like 3k dollars) [01:15] <sienap> aka i am fucked[tm] [01:16] <omega> Zeenix: it's definitely possible. when you get new caps, you query the caps for the rate, and set that as the rate of the rtp channel [01:16] <sienap> .. some reaction ? [01:16] <omega> arik: what kind of video output support is in picogui? [01:16] <arik> omega: i'm not sure yet, i was gonna start looking today [01:16] <omega> sienap: you should be able to work out a payment plan, because it's their screwup [01:17] <Zeenix> omega: no rate on the rtpchannel [01:17] <sienap> omega for sure.. but i am still not sure how they came up with that 5k [01:17] <sienap> it is pretty obvious calculated [01:17] <Zeenix> omega: otherwise i would have done like that [01:17] <sienap> since they can't give any specifications [01:17] <omega> Zeenix: then why does the rate matter? [01:17] <Zeenix> omega: it matters for my app <g> [01:18] <sienap> anyway [01:18] <sienap> i am off [01:18] <sienap> bye :) [01:18] sienap (sy...@s3...) left #gstreamer. [01:18] <omega> Zeenix: your app should be able to query the caps of the rtpsend and get it [01:19] <... [truncated message content] |
From: IRC B. <wt...@us...> - 2001-10-25 04:27:52
|
******************************************************************* [03:13] wingo (wi...@rd...) joined #gstreamer. [03:14] <wingo> ok taaz wo why did you retract properish? :) [03:14] <wingo> s/wo/so [03:16] bstard (Lor...@a2...) joined #gstreamer. [03:44] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [03:46] vektor (reet@HSE-Kitchener-ppp3508492.sympatico.ca) got netsplit. [03:46] vishnu-sleep (joshua@24.8.191.22) got netsplit. [03:46] vektor (reet@HSE-Kitchener-ppp3508492.sympatico.ca) returned to #gstreamer. [03:46] vishnu-sleep (joshua@24.8.191.22) returned to #gstreamer. [03:51] vektor (reet@HSE-Kitchener-ppp3508492.sympatico.ca) got netsplit. [03:51] vishnu-sleep (joshua@24.8.191.22) got netsplit. [03:51] tnt-out (tn...@h2...) got netsplit. [03:51] wingo (wi...@rd...) got netsplit. [03:51] ajmitch (aj...@p3...) got netsplit. [03:51] omega (om...@om...) got netsplit. [03:51] ShrimpX (marius@131.252.244.168) got netsplit. [03:51] xav (xav@AMontpellier-201-1-4-3.abo.wanadoo.fr) got netsplit. [03:51] wtay-zZz (wi...@ca...) got netsplit. [03:51] bstard (Lor...@a2...) got netsplit. [03:51] ShrimpX (marius@131.252.244.168) returned to #gstreamer. [03:51] xav (xav@AMontpellier-201-1-4-3.abo.wanadoo.fr) returned to #gstreamer. [03:51] ajmitch (aj...@p3...) returned to #gstreamer. [03:51] wtay-zZz (wi...@ca...) returned to #gstreamer. [03:51] tnt-out (tn...@h2...) returned to #gstreamer. [03:51] wingo (wi...@rd...) returned to #gstreamer. [03:51] bstard (Lor...@a2...) returned to #gstreamer. [03:51] vishnu-sleep (joshua@24.8.191.22) returned to #gstreamer. [03:51] vektor (reet@HSE-Kitchener-ppp3508492.sympatico.ca) returned to #gstreamer. [03:51] omega (om...@om...) returned to #gstreamer. [03:58] mwc (ma...@ly...) joined #gstreamer. [03:59] bstard (Lor...@a2...) got netsplit. [04:00] bstard (Lor...@a2...) returned to #gstreamer. [04:01] omega (om...@om...) got netsplit. [04:01] omega (om...@om...) returned to #gstreamer. [04:08] <taaz> wingo? [04:10] <vektor> taaz? [04:11] <wingo> vektor? [04:11] <wingo> heh [04:11] <wingo> hi taaz. [04:12] Action: wingo enables glib2 on his gst build. [04:12] <taaz> so, my statment, do you care why i said that? [04:13] <wingo> yes. [04:14] <taaz> ok, let me commit some movietime stuff for vektor first [04:14] <vektor> yay! [04:14] <wingo> how is renaming better than versioning? you can't install (for instance) automake and automake 1.5 side by side [04:14] Action: vektor is important! [04:14] <wingo> ok [04:15] trevo (trevo@208.141.162.68) joined #gstreamer. [04:15] <ajmitch> vektor: you are? [04:15] <ajmitch> ;) [04:16] <vektor> hey, it was news to me too [04:16] <vektor> after months of walken telling me it wasn't true :) [04:17] trevo (trevo@208.141.162.68) left irc: Client Exiting [04:18] <wingo> anyone here build against glib2? [04:18] <wingo> or, anyone know how to generate gstmarshal.[ch]? [04:19] <wingo> ah, glib-genmarshal [04:20] evil_monk (trevo@208.141.162.68) joined #gstreamer. [04:20] <wingo> wierd. [04:23] <taaz> ok wingo, here's the low down: [04:24] <wingo> break it down. [04:26] <taaz> this is more common to happen to libs but apps are similar [04:26] <taaz> it's an issue of interface version [04:26] <taaz> not release version [04:27] Action: taaz trying to talk to vektor in #livid too... sorry for slow logic ;) [04:27] <wingo> those aren't the same thing in development libs? [04:29] <taaz> so there are two versioning schemes [04:30] <taaz> there is the release version like 2:0.14-3, standard debian stuff [04:30] <wingo> right [04:30] <taaz> but you quickly hit a problem [04:30] <taaz> debian only allows 1 version of any particular package name to be installed [04:31] Action: vektor waits for taaz to finish here before he can answer me in #livid :) [04:31] <wingo> ;-) [04:32] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [04:33] <taaz> so lets say your app, WingoRamaPlayer, depends on libfreakshow 0.1 API [04:34] <taaz> and a new libfreakshow 0.2 comes out with a totally different, newer, better, more joy all around, API [04:34] <taaz> but WingoRamaPlayer (WRP) depends on 0.1 [04:34] <wingo> ok, ok. but for development libs, one has to break things to move life forward, no? [04:35] <taaz> unless everyone was absurdly coordinated you can't just upgrade a lib like that [04:35] <taaz> WRP will break [04:35] <wingo> is there any way to link to a particular version of a lib? [04:35] <wingo> -lfreakshow:0.1 style [04:36] <taaz> well, yeah, WRP is linked to libfreakshow 0.1 [04:36] <taaz> so you have to keep it installed to use WRP [04:37] <taaz> but TaazManiaUltraCoolPlayer just came out that requires libfreakshow 0.2 [04:37] <wingo> but when you compile, do you always link to the whatever the .so points to? [04:37] <taaz> you can't do both... one app will not be installable [04:37] <wingo> i see. [04:37] <wingo> thanks [04:37] <taaz> ok this isn't exactly accurate [04:37] <taaz> WRP may just break silently on an upgrade ;) [04:37] <wingo> you can't have both the dev packages installed at the same time [04:38] <taaz> so anyway, the idea comes along that you have versions in the package name [04:38] <taaz> this allows you to install, for instance, libgal1 libgal2 libgal3..... libgal1230452 [04:38] <taaz> which you may have noticed [04:38] <vektor> taaz: ok, i'm going to copy libmpeg2 into my source tree now. [04:39] <taaz> this let's apps link to a particular version of a lib and the packge should remain around [04:39] <taaz> but the idea is the breakage is coming from interface changes [04:39] <vektor> taaz: shouldn't i move all my src into a src subdir? [04:40] <taaz> vektor: i don't really care ;) [04:40] <taaz> yes maybe [04:40] <vektor> taaz: come on man i want your 'opinion'. [04:40] <taaz> it's small project, doesn't matter too mcuh [04:40] <taaz> i would have done that yes [04:41] <vektor> well it isn't too late for me, is it? [04:41] <vektor> i guess cvs sucks this way, right? [04:41] <taaz> you lose revision control yes. cvs does suck like that. [04:41] <vektor> Ah well I never use revision control. [04:42] <taaz> wingo: i'm not done yet ;) [04:42] <taaz> so only time you need to change the libfooN name is when an interface changes [04:43] <taaz> if the interface is the same (however you define that) no need to change to libfooN+1 [04:43] <vektor> how do you add a directory in cvs, cvs add? [04:43] <taaz> ie, look at libc6, it's been libc6 for a while cause interfaces havent changed. even though versions have gone 2.0, 2.1, 2.2 (or whatever) [04:43] <omega> sounds to me like packages need two version numbers: interface and release. packager allows multiple packages to be installed with different interface versions, but within interface only one release ver at a time [04:44] <wingo> amen, omega. [04:44] <wingo> that's only for lib packages. [04:44] <taaz> omega: but that is what we have now [04:44] <taaz> sorta [04:44] <omega> taaz: no, it isn't [04:44] <taaz> but like here's the issue: [04:44] <omega> someone's hacked the release version (and only one part of it) into the package *name*, which is screwed up [04:45] <wingo> omega: is string_append_printf in glib cvs? [04:45] <taaz> people are using libbroken0.14 [04:45] <omega> wingo: in glib2, yes [04:45] <wingo> libmad0.14! [04:45] <omega> taaz: that's broken [04:45] <wingo> irritating. [04:45] <omega> in both senses <g> [04:45] <wingo> heh [04:45] <taaz> so even if you keep the same interface up to libbroken v 1.0 you still have libbroken0.14 [04:46] <omega> that's fine, except that's a really ugly hack. there should be a separate field for interface version [04:46] <taaz> but clearly the package maintainer intends to fuck over everyone and release libmad0.15 0.16 etc... [04:46] <taaz> omega: yes there should. it's a shame really. but the libfooN works fine unless it's used wrong [04:46] <omega> which it is [04:47] <omega> libc6 is using it wrong [04:47] <taaz> it is? [04:47] <omega> is '6' the interface version? [04:47] <taaz> yes [04:47] <omega> no [04:47] <taaz> yes [04:47] <omega> 2.2 is [04:47] <wingo> anyone know if there's a plan to package gnome-libs alphas [04:47] <taaz> is it? [04:47] <wingo> that's the same [04:47] <omega> 6 is the first part of the release version [04:47] <wingo> glibc2.2 [04:47] <wingo> libc6 [04:47] <wingo> the same [04:47] <omega> yes, it's the same [04:47] <taaz> its been libc6 since 2.1 at least [04:47] <omega> but 2.2 is the interface verson [04:48] <omega> programs built for glibc2.1 won't work with glibc2.2, or vice versa, not sure which way [04:48] <omega> I know, I've done it [04:48] <wingo> hrm. [04:48] <taaz> hmm... [04:48] <omega> so '6' is not the interface version [04:48] <wingo> right. well, it's the major version :-P [04:48] <taaz> well, it's supposed to be afaik [04:48] <omega> -rwxr-xr-x 2 root root 1236396 Apr 6 2001 /lib/libc-2.2.2.so [04:49] <omega> the interface version on my machine is 2.2.2 [04:49] <omega> it's called libc6 because it came after libc5, sorta [04:49] <omega> which itself is screwed, because it didn't, it replaced libc5 [04:49] <taaz> lrwxrwxrwx 1 root root 13 Oct 8 15:53 /lib/libc.so.6 -> libc-2.2.4.so* [04:49] <omega> cute [04:49] <omega> broken, imo [04:50] <omega> oh, I guess mine has it too. wonderful [04:50] <omega> I guess that's a carry-over from upgrading from libc5 [04:50] <taaz> ok, i don't know the specifics on libc, but anyway, i've got the theory right i think [04:50] <omega> yes, but the implementation isn't correct (not your fault) [04:51] <taaz> i mean you ldd on /bin/ stuff and it links to libc.so.6 [04:51] <vektor> taaz: this is sooo awesome!! [04:51] <vektor> taaz: you're a reasonable diety! [04:51] <vektor> s/diety/deity/ [04:51] <taaz> hardly [04:52] <vektor> ok, now time to do the horrible horrible nasty brutal disgusting thing and copy libmpeg2 into my source tree. [04:52] <taaz> sigh... [04:52] <omega> why 'bootstrap' ? [04:53] <wingo> bootstrap? [04:53] <omega> instead of autogen.sh [04:53] <taaz> eh... walken made me do it [04:53] <taaz> autogen.sh is there too [04:53] <taaz> it does what you expect... [04:54] <omega> hmm [04:54] <taaz> i just get seriously pissed off when autogen.sh runs configure for me [04:54] <taaz> like in, oh, say, GStreamer [04:54] <omega> --help [04:54] <taaz> like for deb building, I don't want to run configure twice [04:55] <omega> um, why don't you put the configure options on the autogen cmdline? [04:55] <taaz> bootstrap seperates the auto* stuff and configure stuff [04:55] <taaz> what? [04:55] <omega> how convenient, where do I find sdl 1.2.2 rpms? [04:55] <omega> ./autogen.sh --prefix=......... [04:55] <taaz> uh.. [04:55] <omega> that's what "$@" is for [04:55] <wingo> yeah it took me a while to fugure that out [04:55] <taaz> you could change the 1.2.2 to 1.2.0 ;) [04:55] <omega> I have 1.1.7 [04:55] <taaz> sorry... wasn't thinking [04:56] <taaz> well, change to 1.1.7 and see if it works ;) [04:56] <vektor> It won't. [04:56] <taaz> ok here's how I make debs: [04:56] <vektor> Of course, I say that not knowing what you want to do with 1.1.7. [04:56] <taaz> 1) run autotools on my machine [04:56] <omega> vektor: build movietime [04:56] <taaz> 2) build debs [04:56] <taaz> #2 calls ./configure with deb specific stuff [04:57] <vektor> omega: um... you sure you don't want to wait for me to finish here? ;-) [04:57] <taaz> #1 calls configure with gstreamer defaults [04:57] <taaz> so configure is run twice [04:57] <omega> vektor: where'd be the fun in that?? [04:57] <omega> taaz: then add --help to the autogen cmdline or something [04:57] <taaz> that is lame [04:58] <taaz> speaking of LAME... guy building lame debs sent me email [04:58] <taaz> has the debs up somewhere... i'll probably depend on those [05:04] <taaz> omega: you following libmpeg2 list? if you want to speak up about stadard video codec interfaces now may be the time [05:05] <vektor> um [05:05] <vektor> that dude is on crack [05:06] <vektor> libmpeg2 should be JUST an MPEG1/2 decoding library. [05:06] <vektor> and the api should reflect that [05:06] <vektor> it should not try to be general at all. [05:06] <omega> taaz: I must not be on it... fixing [05:06] <taaz> well yeah... you know what i'm saying [05:08] <omega> which thread in particular? [05:09] <taaz> hah.. geocrawler [05:09] <taaz> all the recent messages [05:09] <omega> er, right. roadmap? [05:09] <taaz> there arnt that many [05:12] <omega> vektor: which guy in particular needs to lay off the crack? [05:13] <wingo> where is the glib2 and gtk2 cvs? [05:13] <omega> afaict Arpi has the right idea [05:13] <vektor> omega: Um, whomever said that mpeg2dec's api should be general enough to extend it to mjpeg and stuff in the future. [05:14] <vektor> yes [05:14] <vektor> Arpi has EXACTLY the right idea [05:14] <vektor> see my followup where I say 'here here'. [05:14] <omega> ok, I need to finish reading the thread, then respond [05:14] <vektor> ok what does -rpath mean? [05:15] <vektor> how do i link against a particular .la in specific? [05:15] <vektor> like it would be very bad if i accidentally linked against a libmpeg2.a that was in /usr/lib [05:15] <wingo> you can put it on the command line... [05:15] <vektor> so how do i protect against that? [05:15] <vektor> so i do -l../libmpeg2/libmpeg2.la ? [05:15] <wingo> that's what i do in my source tree [05:15] <vektor> and that will work? [05:15] <wingo> no just add it [05:15] <wingo> without -l [05:15] <vektor> oh [05:15] <vektor> oh yeah [05:15] Action: vektor thwaps. [05:15] <wingo> ok so now tell me where glib2 and gtk2 cvs are ;-) [05:16] <omega> but you have to run it inside libtool [05:16] <omega> cvs.gnome.org [05:16] <wingo> glib2 and gtk2 modules? i couldn't find separate modules for them. are they tagged? [05:16] <omega> no, glib and gtk [05:16] <omega> HEAD == pre2.0 [05:16] <taaz> vektor: $(top_builddir)/libmpeg2/libmpeg2.la not .. [05:16] <omega> right [05:17] <wingo> thanks yo [05:17] <omega> definitely have a full path to it [05:17] Nick change: vishnu-sleep -> vishnu [05:17] <wingo> vish! [05:17] <wingo> nu! [05:17] <vektor> taaz: uh... [05:17] <vishnu> wow, that was a lot of sleep. i hope wtay feels better now [05:17] <vektor> taaz: if you say so [05:28] <omega> ok, I'm working on mail to libmpeg2, I'm gonna try to explain my codec API ideas again [05:28] <vektor> yay [05:28] <omega> and hopefully figure out the synth part of mpeg1l1 decoding so I can get that going.... [05:41] thomasvs_ (thomas@212.100.172.175) left irc: Ping timeout for thomasvs_[212.100.172.175] [05:45] <taaz> omega: thoughts on what to do with that stats code? [05:45] <omega> um, I guess add it to cvs for now, as a new plugin [05:45] <taaz> where? [05:45] <omega> hrm, benow.ca is looping on the same track over and over and over.... [05:46] <taaz> i just stuck it in gst/elements ;) [05:46] <omega> um, no, put it in plugins/stats/ or something [05:46] <taaz> that's not as k-rad! [05:46] <omega> well, does it have any dependencies? [05:46] <taaz> nope [05:46] <omega> you think it's generally useful to just about anyone? [05:48] <taaz> hard to say [05:48] <taaz> i think so cause i like watching stuff like that [05:51] <wingo> it looks neet to me, of about the same functionality category as fake* [05:51] <wingo> does it proxy capsnego? [05:51] <taaz> it does [05:51] <wingo> nice [05:51] <taaz> and bufferpools [05:51] <wingo> nice! [05:51] <taaz> but that's just copied straight outa identity code [05:52] <wingo> don't give away your tricks ;) [05:52] <taaz> tragically subclass identity to do this is beyond my gobject skillz [05:52] <wingo> it's not bad, yo. [05:52] <taaz> though i may revisit that [05:52] <wingo> the ggad is helpful here. [05:53] <wingo> also, s/GObject/GstIdentity generally [05:53] <wingo> more or less, anyway. [05:53] Action: wingo goes for food [05:56] Action: omega uses VNC to show his uncle how to do something under windoze <g> [06:05] <wingo> anyone know anything about libgtk-cvs? there's a deb directory in the gtk repository [06:05] <wingo> i wonder if someone has put up packages [06:05] <taaz> uh.. yeah [06:06] <taaz> like a while ago [06:06] <taaz> libgtk1.3 libglib1.3 etc [06:06] <taaz> that what you want? [06:06] <wingo> that's not up to date enough [06:06] <taaz> oh? [06:06] <wingo> no int64, no append_printf, etc [06:07] <ajmitch> it's not hard to compile gtk+ and glib :) [06:07] <wingo> i'm doing it as we type :-) [06:07] <wingo> thanks to the vicious-build-scripts, it's extra easy [06:08] <wingo> it's nice to have someone do it for me tho [06:08] <ajmitch> yep, v-b-s is useful :) [06:11] <wingo> how are exams, ajmitch? [06:12] <ajmitch> wingo: start in a couple of days ;) [06:12] <wingo> taaz: are your changes in the making of stat small enought that they can go into identity? or is that a bad idea [06:12] <wingo> good luck ;) [06:14] <taaz> $ wc gst/elements/gststatistics.[ch] [06:14] <taaz> 404 1117 13033 gst/elements/gststatistics.c [06:14] <taaz> 95 261 2241 gst/elements/gststatistics.h [06:14] <taaz> 499 1378 15274 total [06:14] <taaz> i think that's seriously lame [06:14] <taaz> why not fold a mpeg2 ecoder into idenity too, just in case [06:15] <wingo> ok! cool! [06:15] <wingo> :-) [06:16] <wingo> cvs import kitchen-sink gstreamer initial-release [06:17] <taaz> if you had never used gstreamer would you even consider looking at "identity" to find capability to keep track of buffer throughput rates? [06:18] <wingo> no. [06:18] <wingo> i'm just a-picking. [06:22] <omega> taaz: making stats a subclass of identity would probably require some minor tweaks to identity to make it more subclassable without redoing the functionality [06:23] <omega> and eventually someone'll write a filterfactory, and make most of the boilerplate unnecessary (or maybe ggad can do that) [06:23] <taaz> ggad? [06:23] <omega> er, s/ggod/gob/ [06:23] <omega> ggad [06:23] <wingo> ggggg [06:23] <omega> er, what's ggad anyway? [06:24] <vektor> hey everybody [06:24] <vektor> try movietime [06:24] <wingo> gtk and gnome application development [06:24] <vektor> it's coool. [06:24] <vishnu> what's movietime? [06:25] <vektor> vishnu: My DVD player. [06:25] <vektor> vishnu: It doesn't use gstreamer. [06:25] <omega> yet [06:25] <wingo> why not? [06:25] <vishnu> neat [06:25] <vektor> vishnu: But it does have millisecond-accurate frame blits and it's got nice sync. [06:25] <vektor> wingo: because I'm still lame. [06:25] <wingo> oh, it's ok vektor ;-) [06:26] <wingo> you can still hang out with us :-))) [06:26] Action: omega waits for sdl srpm to download [06:26] <vishnu> vektor: does it have asm optimizatons? [06:26] <vishnu> vektor: where can i download it? [06:27] <taaz> movietime.sf.net [06:27] <vektor> vishnu: you go to www.sf.net/projects/movietime and get it from cvs. [06:27] <vishnu> whoa, rad [06:28] <taaz> i'm having some file date issues [06:28] <vektor> huh? [06:28] <taaz> i dunno... [06:28] <vektor> is your clock wrong? [06:28] <wingo> vishnu: have you made pango with the vicious build scripts? [06:28] <vektor> it's 12:27 here. [06:28] <taaz> it's stuck in a loop rebuilding makefiles ;) [06:28] <vishnu> wingo: what's vicious about them? yah, i build gtk+ cvs almost every day [06:29] <omega> vishnu: heh. the scripts are called 'vicious-build-scripts' [06:29] <wingo> it can't build due to a lack of 'pango-enum-types.c' [06:29] <wingo> which apparently is generated somehow [06:29] <omega> wingo: go to that dir and run `make pango-enum-types.c` manually [06:29] <omega> maybe automake screwed up the rules [06:29] <vishnu> omega: oh, no, i just use autogen.sh [06:30] <omega> vishnu: same here, but I haven't built gtk2 in a long time [06:30] <wingo> that worked now... [06:30] <wingo> i got the same issue with gstreamer [06:30] Action: vektor sleeps [06:30] <omega> yeah, the gstmarshal bit is still screwy [06:30] <wingo> seeya vekkie. [06:30] <vishnu> omega: dude, just commit my patch, geez [06:31] <omega> oh, um, searching [06:31] <vishnu> ya want me to resend it? [06:31] <omega> sure [06:31] <omega> forgot about that [06:32] <wingo> is that a automake 1.5 only thing? [06:32] <omega> which? [06:32] <wingo> which i have, not a problem [06:32] <vishnu> hey, bootstrap is cool [06:32] <wingo> built_sources [06:32] <omega> dunno [06:33] <vishnu> vektor: why do i need sdl?? [06:35] <omega> libxosd? [06:36] <omega> have 0.7 [06:36] <omega> demuxer.c:249: too many arguments to function `mpeg2_decode_data' [06:36] <vishnu> what is libxosd? fuck, vektor, these crazy dependencies [06:37] <omega> updating [06:38] <taaz> yeah, look at gstreamer... dependency free [06:39] <vektor> vishnu: yes [06:39] <vektor> omega: um... [06:39] <omega> rebuilding with new HEAD [06:39] <taaz> omega: it's seeing the non-vekterized lib [06:39] <vektor> vishnu: libraries are for sharing code. [06:39] <vishnu> vektor: ok, how do i run it? [06:39] <vektor> vishnu: you have it all compiled? [06:39] <vishnu> ya, i guess [06:39] <taaz> i cant compile it... heh [06:40] <vektor> vishnu: then run 'vob_movietime /path/to/an/unencrypted/vob' or 'dvd_movietime /dev/dvd 1 4' to play title 1 starting at chapter 4. [06:40] <vektor> vishnu: the README file contains some key bindings. [06:40] <vektor> vishnu: run it as root to get ultra-accuracy on your frame blits, if you have a /dev/rtc device. [06:40] <vishnu> vektor: do i have to leave X? [06:40] <vektor> huh? [06:40] <vektor> no this is an X app. [06:41] <vishnu> segv [06:41] <vektor> but if you run as root make sure root has auth for the screen. [06:41] <vektor> it will segv if your display is set incorrectly [06:41] <vektor> since xosd has a bug [06:41] <vektor> like, if you can't open the display [06:41] <vishnu> vektor: i'm running it as myself [06:41] <vektor> and it segv'd? [06:41] <vektor> how did you run it? [06:41] <vektor> you gave it a vob? [06:41] <vektor> did the window pop up or just immediately segv? [06:41] <vishnu> ./vob_movietime /opt/vast/cache/d_advocate.vob [06:42] <vektor> and it just immediately segv'd ? [06:42] <vishnu> ossoutput: 1 channels, sampling rate at 48000. [06:42] <vishnu> ossoutput: Buffer size is 65536, fragsize is 4096. [06:42] <vishnu> Segmentation fault [06:42] <omega> vektor: massive frame dropping [06:42] <vektor> omega: uh.. did it detect MMX ? [06:42] <vishnu> yah [06:42] <omega> it's pegged, so probably no [06:42] <vektor> vishnu: then the mmap might have failed. [06:42] <omega> hrm, yes, it did [06:43] <omega> er [06:43] <vektor> vishnu: can you gdb that and tell me the stack trace? [06:43] <vishnu> vektor: it's a big vob. i'll try a small one [06:43] <vektor> vishnu: that won't matter. [06:43] <omega> libmpeg2 did, but a52 didn't, and no word from movietime [06:43] <vektor> omega: if libmpeg2 did then you're fine. [06:43] chillywilly (da...@d4...) left irc: Ping timeout for chillywilly[d41.as29.nwbl0.wi.voyager.net] [06:43] <vishnu> (gdb) where [06:43] <vishnu> #0 0x40257dbe in memset () from /lib/libc.so.6 [06:43] <vishnu> #1 0x0804b1fb in oss_open () [06:43] <vishnu> #2 0x0804ec29 in thread_run () [06:43] <vishnu> #3 0x4018fefa in pthread_start_thread () from /lib/libpthread.so.0 [06:43] <vektor> vishnu: ok then it looks like the mmap failed and didn't return a negative error code. [06:43] <vishnu> vektor: uh .. [06:44] <omega> vishnu: you've tested the BUILT_SOURCES patch? [06:44] <vektor> vishnu: or maybe your soundcard doens't support mmap? [06:44] <vektor> vishnu: it seems fucked up that that would segv. [06:44] <vishnu> omega: yah [06:44] <omega> ok [06:44] <omega> committed [06:44] <vishnu> vektor: dunno. how can i check? [06:44] <vektor> vishnu: what soundcard do you have? [06:44] <vishnu> vektor: i got an sbawe alsa driver [06:44] <vektor> alsa driver? [06:45] <vektor> does alsa support oss-style mmap? [06:45] <vektor> and oss-style triggers? [06:45] <vishnu> vektor: dunno. alsa has oss emulation [06:45] <vektor> see, yeah, i haven't tested with alsa. [06:45] <vektor> yeah but this is like _hardcore_ oss emulation. [06:45] <vektor> if you can try with oss that would be better :) [06:45] <vektor> at least, better for me. [06:46] <wingo> if there's a difference, it's a bug. [06:46] <vektor> omega: so, it's dropping alot of frames? is it a video-based vob? [06:46] <wingo> in the oss-emu code. [06:46] <omega> video-based? [06:46] <vektor> omega: like not from film source. [06:46] <omega> matrix [06:46] <vektor> it's on the matrix? and it's dropping frames? [06:46] <omega> bullet.vob [06:46] <vektor> turn on DMA to your drives [06:46] <vektor> dude [06:46] <vektor> I bet that's it. [06:46] <omega> using_dma = 1 (on) [06:46] <vektor> fuck [06:46] <omega> second time makes no difference [06:47] <vektor> what the fuck. [06:47] <vektor> that shouldn't happen at all. [06:47] <vishnu> vektor: welcome to beta testing :-) [06:48] <vektor> omega: ok well i need sleep i have a midterm tomorrow. [06:48] <vektor> omega/vishnu: maybe like thursday we can debug it further. :) [06:48] <omega> heh [06:48] <vektor> goodnight. [06:48] <omega> l8r [06:49] <vishnu> wow, 3:2 pulldown, adaptive deinterlacing .. sounds interesting [06:49] <vektor> vishnu: it is! [06:49] <vektor> vishnu: i have some improvements to make on the 3:2 pulldown removal though [06:49] <vektor> vishnu: so soon it should be even better [06:50] <omega> then we build gstreamer plugins out of them all [06:50] <vektor> yep yep ! [06:51] <omega> filesrc ! mpeg2parse ! mpeg2dec ! autopulldown ! adaptive-deinterlace ! xvideosink [06:51] <omega> vektor: you need a cool name for the deinterlacer [06:52] <vishnu> yah and OCR on the subtitles ;-) [06:52] Action: vektor is very asleep :) [06:56] <taaz> omega: http://crib.lehn.org:8080/~dlehn/projects/gstreamer/events/ [06:56] <taaz> that's my pseudo updated event model sorta thing [06:57] <omega> you wanna comment it? <g> [06:57] Action: omega doesn't know python [06:57] <taaz> i was hoping i didnt have to [06:57] <vishnu> omega: just pretend it's pseudo code [06:57] <taaz> pretend its pseudo code [06:57] <omega> ok, I'll look at it soon, right now I'm about 15% done with my mail to libmpeg2-devel [06:57] <vishnu> taaz: is it ev1 or ev0? [06:58] <taaz> ev0 is now ev1 is new stuff [06:58] <vishnu> taaz: what should i look at? [06:59] <taaz> ok, this is failing ;) [06:59] <taaz> the class hierarchy changed [06:59] <taaz> where methods get 'buf' its a GstBuffer, 'ev' is a GstEvent [07:01] <taaz> i was trying to make sure all methods get the proper type [07:01] <taaz> and trying to factor the switch statements into common code [07:02] <taaz> and make simple and complex elements both efficient and easy to write [07:03] <taaz> maybe i should rewrite it in Gtk-C [07:03] <wingo> or gob [07:04] <taaz> or hey, watch out, i'll just rewrite all of gst in Python with 10% the LCC of the C version [07:04] <taaz> LOC too [07:05] Action: omega klocs taaz with a C book <g> [07:05] Action: taaz 's OOP force field deflects such insignificant attacks [07:06] Action: omega goes hunting for the language geeks at OGI to lay siege to taaz with Haskell and other functional programming languages [07:07] Action: taaz is immune to functional language assults due to insufficient brainpower to comprehend such beauty [07:07] Action: vishnu moves in covertly with the C-- special forces unit [07:08] <omega> taaz: in one brain cell and out the other, eh? [07:08] <omega> ;-) [07:08] <taaz> in the one brain cell, yes [07:09] <omega> krasic threatened to teach us ocaml at one point.... I probably still should learn it... [07:09] <taaz> yo python has quite a bit of functional lang features. map and apply and so on [07:10] <taaz> so anyway, are my ideas lost in python soup? [07:10] <omega> hrm, ironic. LFO track that uses very high-frequency sounds [07:10] <omega> taaz: haven't read it yet, remember, I'm still on that codec API thing [07:11] <omega> and likely will be for several more hours.... [07:11] <taaz> sounds fun [07:11] <omega> this is gonna be a monster email... [07:11] <taaz> cc it to codecs list eh? [07:11] <omega> yeah [07:12] <omega> and gst probably [07:12] <wingo> cool [07:12] <taaz> so all 3 of us can read it there too [07:12] <omega> 6kb already and maybe 20% done so far (maybe) [07:12] <taaz> dude, multiple emails is ok [07:12] <omega> it's a single document [07:12] <omega> just happens to be being written inside an email for now [07:12] <taaz> oh... your in "document" mode ;) [07:12] <omega> yeah [07:13] <taaz> you're too [07:13] <omega> too what? <g> [07:13] <taaz> i am about 50/50 on getting your/you're right [07:13] <taaz> same with its/it's [07:13] <taaz> i suck [07:14] <vishnu> english sucks [07:14] <omega> I get your/you're right 95+% of the time, I mess up its/it's maybe 25% [07:14] <omega> english is the natural language's version of C++ <g> [07:14] Action: omega ducks [07:14] <vishnu> ducks what? [07:14] <taaz> quack [07:14] <taaz> i mean goose [07:14] <wingo> hah! [07:14] <omega> someone got it.... [07:15] <wingo> no i was laughing at the quack [07:15] <omega> oh, n/m [07:15] <omega> moooo [07:15] <wingo> gobble. [07:16] Action: omega wonders where the zookeepers are [07:16] <wingo> you don't put turkeys and cows in zoos, silly. [07:16] Action: ajmitch gets out his cattle prod [07:16] <omega> something about the inmates running the place or something... [07:16] <wingo> mooo!!! [07:17] <omega> wingo: you do when they're insane [07:17] <omega> mad cows and such [07:17] <taaz> stop making me hungry [07:18] <wingo> mad cow burger? [07:18] <omega> wingo: scary, I was gonna say the same thing, but decided not to <g> [07:18] <wingo> wild cow preserve(s)? [07:18] <wingo> sheesh we need to stop this :-P [07:19] <omega> worried about your sanity? [07:19] <vishnu> is libmpeg3 a piece of junk? [07:19] Action: wingo thinks ans decides not to answer [07:19] <omega> wingo: good move [07:20] <omega> vishnu: well, it's heroine's code, mostly copied together from libmpeg2dec, ac3dec, and other sources [07:20] <omega> so not his code, mostly.. [07:20] <vishnu> d'oh, what an idiot [07:21] <omega> yeah, he doesn't seem to get the idea of libraries either (/me carefully avoids looking at vektor <g>), he has copies of libmpeg3 in all his apps. not all the same afaik [07:22] <omega> all modified in very bizarre ways [07:22] <omega> there's been some worry about what he's going to do with b2k, as far as closing it [07:22] <omega> since it's heavily based on several people's code, including my own (libdv) [07:23] <omega> I'm gonna meet up with Jerritt Collard of LinuxFund at the next PersonalTelco meeting, since LinuxFund helped pay for b2k's development, they have a stake in this too [07:23] <vishnu> d'oh [07:24] <omega> he actually emailed me about it, and asked if I wanted to maintain a fork of it <g> [07:24] <omega> I said, um, not really <g> [07:24] <taaz> uh, he was paid for that? [07:24] <omega> $1k grant helped pay for part of development, yes [07:25] <taaz> ok, whatever, let's not gripe about it here [07:25] <vishnu> oh well, it's not that much money [07:25] <omega> yeah, I'm not too worried about it, unless it shows up commercial somewhere, in which case I'll get real interested real fast [07:25] <omega> so will walken and aaron, I assume [07:25] <omega> and OGI even [07:26] <omega> maybe [07:33] <taaz> yo so is thinking of how to clone gstreamer in python a good idea? ;) [07:33] <ajmitch> taaz: of course [07:33] <omega> depends on how long you want to stay on the *outside* of the loony-bin [07:33] <ajmitch> taaz: python is much better than those other languages ;) [07:33] <taaz> whoop! [07:33] <omega> heh [07:33] <wingo> moo! [07:33] <omega> quack [07:33] <taaz> but i'm serious here [07:33] <omega> it might be an interesting excercise [07:34] <omega> dunno why you'd use that over a native C core, assuming shims for both apps and plugins [07:34] <wingo> in futility [07:34] <omega> heh [07:34] <taaz> please don't make me school you kids with elite scripting vm powerskillz [07:34] <ajmitch> omega: at least taaz isn't trying to write it in BASIC or something ;) [07:35] <taaz> i'm nearly dont with the thinking part... i think it's doable [07:35] <omega> ya know, one of these days a 22nd century Fr[ae]ud is gonna try to figure out what kind of lunatics we were, and all they'll have is this log.... [07:35] <taaz> heh heh.. you said... heh ... log [07:35] <ajmitch> muahaha [07:35] <wingo> lol [07:35] Action: ajmitch feels pity for those poor ppl [07:36] Action: omega doesn't <g> [07:36] <taaz> this python 2.2 generator stuff is the key [07:36] <taaz> though i think it will be a tad hackish [07:38] <wingo> grumble grumble taaz grumble productive work grumble [07:38] <wingo> :-) [07:38] <ajmitch> productive work? what is that? [07:38] <wingo> not quite sure [07:39] <taaz> oh my, i'm seriously slacking here [07:39] <ajmitch> hehe [07:39] <ajmitch> we shoudl press omega for a list of things he's gonna work on before the next release ;) [07:39] <omega> pff [07:40] <taaz> what % of the doc done? [07:40] <omega> closing on 50% [07:40] <ajmitch> wtay's put his list up, what are you gonna do? ;) [07:40] <omega> 'ditto' [07:41] <ajmitch> what can we your loyal slaves do to help? ;) [07:41] <omega> um, let me finish this doc, atm <g> [07:41] <ajmitch> hehe, k ;) [07:42] <omega> and request neat music on benow.ca <g> [07:43] <omega> make me not the only one requesting stuff [07:43] <ajmitch> hehe [07:44] <ajmitch> my modem's too slow to listen [07:44] <omega> 48k? [07:44] <ajmitch> i got a 33.6k modem ;) [07:44] <omega> oh, 24k [07:44] <omega> 'lowest bitrate' <g> [07:44] <ajmitch> they have 24k now? [07:44] <omega> no 48k it seems atm [07:44] <ajmitch> yeah [07:45] <ajmitch> he said it wasn't worth going to 24k ;) [07:45] <wingo> you could remedy that omega :-P [07:45] <omega> what, rebroadcast? not for long on this dsl... [07:46] <taaz> oh man... i think i could really do this python gst thing [07:47] <taaz> crazy [07:47] <wingo> you mean a complete reimplementation, or a wrapper? [07:47] <taaz> reimpl [07:47] <ajmitch> hmm, xmms doesn't like that url [07:47] <omega> you have to get the url out of the pls [07:48] <omega> or http://benow.ca:8024/ [07:48] <taaz> i say nothing about efficiency... it will probably suck, but you never know. if 99% of the processing is in C maybe python overhead doesnt matter [07:48] <wingo> why not just work on a wrapper taaz? that would be of more than experimental value [07:48] <taaz> i'm playing [07:48] <wingo> cool [07:48] <wingo> gotta respect that [07:49] <taaz> or maybe i just want to give omega some competition so he codes faster.. yeah [07:49] <wingo> i see you shaking in your boots from here, omega [07:50] steveb (st...@no...) joined #gstreamer. [07:50] <ajmitch> hey steveb [07:50] <ajmitch> amazing, the 24k stream plays ;) [07:50] <steveb> hi [07:52] <wingo> hey steveb [07:53] <steveb> yo [07:57] Action: vishnu disappears in a puff of smoke [07:57] vishnu (joshua@24.8.191.22) left irc: using sirc version 2.211+4KSIRC/1.1 [07:59] <wingo> 'steve baker'. is that an english name? are you living in .nl? speaking norsk? [08:03] <steveb> i am from ajmitch land [08:03] <ajmitch> heh [08:03] <ajmitch> steveb: you grew up in dunedin? [08:03] <steveb> yep [08:04] <steveb> i'm going back to NZ for 2 months in December to get married [08:04] <omega> funky, you'd start to think Dunedin was the center of NZ. I know 3 people from Dunedin and zero from anywhere else in the country [08:04] <wingo> congrats! [08:04] <steveb> omega: it is ;) [08:05] <ajmitch> steveb: oh, getting married? [08:05] <ajmitch> omega: hehe [08:05] <wingo> and then, to stay in nz or go back to nl? [08:05] <steveb> then we're coming back here for at least another year [08:06] <ajmitch> hmm, i'll have to do work on gstreamer so that i can talk to steveb when he gets here ;) [08:06] <omega> heh [08:07] <taaz> hot fresh kernel order up! [08:07] <omega> .13 ? [08:07] <omega> yup [08:07] <omega> 601KB [08:07] <taaz> $ finger @kernel.org [08:07] <taaz> [kernel.org] [08:07] <taaz> The latest stable version of the Linux kernel is: 2.4.13 [08:07] <taaz> The latest prepatch (alpha) version *appears* to be: 2.4.13-pre6 [08:07] <taaz> The latest -ac patch to the stable Linux kernels is: 2.4.12-ac6 [08:07] Action: ajmitch wonders how broken this one will be ;) [08:08] <taaz> gotta love that [08:08] <steveb> maybe this is the last linus one [08:08] <omega> more -aa VM patches [08:09] <ajmitch> time to fire up the ftp client [08:09] <omega> I just built .12-ac3-pre a couple days ago... I'm fine there for now [08:09] <omega> er, -preempt [08:10] <ajmitch> dammit, the nz mirror doesn't have it yet [08:10] <ajmitch> it can wait, i'll grab it tomorrow [08:15] <wingo> ok kids, going home. it's 2:08 in the morning. blah. [08:15] <omega> heh [08:15] <ajmitch> ok [08:15] wingo (wi...@rd...) left irc: peace. [08:15] <ajmitch> i gotta go out... [08:15] <ajmitch> have fun, see y'all later ;) [08:15] <omega> l8r [09:03] mwc (ma...@ly...) left irc: Ping timeout for mwc[lychee.ntu.edu.au] [09:04] mwc (ma...@ly...) joined #gstreamer. [09:27] Action: taaz patiently waits for "the doc" ;) [09:27] <omega> I ran into a problem, researching alternate solutions [09:31] <omega> I'm waiting for a copy of glibc to finish downloading [09:31] <taaz> ok, so movietime is cool until it goes crazy [09:31] <omega> heh [09:31] <taaz> then it drops tons of audio and video [09:31] <taaz> and doesn't sync back up :( [09:31] <omega> yeah, recovery isn't its strong suit [09:32] <taaz> but while it works it's some fine playback [09:36] <taaz> i don't understand this scheduling [09:36] <taaz> looks like voodoo [09:36] <taaz> witch! [09:37] Action: taaz sleepy [09:38] apoc_ (ap...@dy...) joined #gstreamer. [09:40] apoc_ (ap...@dy...) left irc: CGI:IRC 0.4.2 [EOF] [10:25] mwc (ma...@ly...) left irc: [x]chat [10:25] Nick change: taaz -> taazzzz [11:16] Nick change: ShrimpX -> ShrimpZzZ [12:01] omega (om...@om...) left irc: Ping timeout for omega[omegacs.net] [13:26] Action: BBB-zZz is away: I'm busy [13:26] Action: BBB-zZz is back (gone 00:00:02) [13:26] Nick change: BBB-zZz -> BBB [14:18] harobed (harobed@ACAFFDC0.ipt.aol.com) joined #gstreamer. [14:58] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [15:02] asmod (stevec@64.5.222.2) joined #gstreamer. [15:25] <xav> the fn given to gst_buffer_pool_set_buffer_new_function() must to a gst_buffer_new() [15:25] <xav> but should the fn given to gst_buffer_pool_set_buffer_free_function() do a gst_buffer_unref ? [15:26] asmod (stevec@64.5.222.2) left irc: Read error to asmod[64.5.222.2]: EOF from client [15:27] asmod (stevec@64.5.222.2) joined #gstreamer. [15:45] <xav> ok, found by experimentation ;) [15:46] <xav> now, how can I easily find properties of a video file (type, duration, framerate, resolution, etc...) ? [15:47] <vektor> If gstreamer has any brains, determining framerate will be nontrivial, since it is not constant for many formats. [15:47] <vektor> Same with duration. [15:51] uwe (user424@213.174.92.50) joined #gstreamer. [15:51] <uwe> hallo. [15:52] <uwe> is anyone using redhat 7.2 and gstreamer from cvs? [15:54] Zygo (zbl...@24...) joined #gstreamer. [15:59] Shippou (no...@cr...) joined #gstreamer. [16:08] uwe (user424@213.174.92.50) left irc: Read error to uwe[213.174.92.50]: Connection reset by peer [16:10] harobed (harobed@ACAFFDC0.ipt.aol.com) left irc: Ping timeout for harobed[ACAFFDC0.ipt.aol.com] [16:13] <xav> vektor: so, if I want a simple solution for finding info for an (avi|mpeg|mov|whatever) file, what would be your solution ? [16:13] harobed (harobed@ACAFFDC0.ipt.aol.com) joined #gstreamer. [16:17] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:18] harobed (harobed@ACAFFDC0.ipt.aol.com) left irc: Ping timeout for harobed[ACAFFDC0.ipt.aol.com] [16:24] <BBB> xav: what 'info'? [16:26] wingo (wi...@rd...) joined #gstreamer. [16:26] <wingo> mornin [16:38] Nick change: taazzzz -> taaz [16:40] <wingo> yo [16:42] <xav> BBB: framerate, duration, etc... [16:47] harobed (harobed@AC8EBC38.ipt.aol.com) joined #gstreamer. [16:49] <wingo> xav: did you catch my debugging note yesterday? [16:49] <wingo> about debian libtool being broken? [16:49] <BBB> xav: doesn't libgstplay cover that? maybe you want to look at libgstplay [16:51] <BBB> dammit, I don't get it...... [16:51] <BBB> v4l is weird....... [16:54] <xav> wingo: eh ? [16:54] <BBB> wtay-zZz: is v4lsrc supposed to grab frames or should it also handle the framebuffer? [16:55] <BBB> wtay-zZz: the framebuffer stuff is basically horror and hell [16:55] <xav> BBB: libgstplay ? dunno this, I'll look [16:55] <wingo> libtool gdb gstreamer-launch and friends doesn't work, as you saw yesterday, right? [16:55] <wingo> it's a bug in debian's libtool, fixed in libtool cvs [16:56] <wingo> there's a simple fix to the problem in http://gstreamer.net/wiki/?GstDebugging [16:56] <wingo> i'm letting you know because i spent a day or so trying to figure it out [16:58] <xav> wingo: ah ! thanks for this one [16:58] <wingo> no prob, yo ;) [16:59] harobed (harobed@AC8EBC38.ipt.aol.com) left irc: Ping timeout for harobed[AC8EBC38.ipt.aol.com] [17:01] <xav> I suppose libgstplay is gstplay.c ... [17:02] <xav> now that events are there, is there a way to seek by time ? [17:03] <BBB> uhm [17:03] <wingo> see mpeg2parse, i'm told, but yes [17:03] <wingo> sending seek events [17:05] <xav> seeking to a certain timestamp ? [17:06] <xav> (not a file offset or something else) [17:08] <wingo> unfortunately, no, i don't think that's possible yet :-\ [17:08] <wingo> tnt made a proposal on the wiki [17:08] <wingo> if you scroll down the front page, it's under 'works in progress' [17:14] <taaz> wingo: deb bug #115296 has a slightly different patch [17:14] <xav> the patch doesn't apply on my /usr/bin/libtool [17:15] <wingo> libtool --version ? [17:16] <wingo> it worked for me, i'll check the deb bug [17:18] <wingo> taaz: that is slightly different. there's some voodoo in all that, i'll just go with the libtool people's solution ;-) [17:21] <xav> ltmain.sh (GNU libtool) 1.4.2 (1.922.2.53 2001/09/11 03:18:52) [17:22] <taaz> oh heh... i thought you made it up or something [17:22] <taaz> wingo: you a debian user? (sorry, i forget such things) [17:25] <wingo> yeah [17:26] <wingo> what, you think i'm some sort of heretic? ;) [17:28] <taaz> i dunno man [17:28] <taaz> so, how about you follow up to that bug and suggest the short term patch from libtool maintainers be applied [17:29] <wingo> i did ;0 [17:29] <wingo> just now [17:31] <wingo> they could just switch to libtool 1.5 in unstable [17:35] <taaz> it's not backwards compatible or something [17:37] <wingo> hum. i do have to give up the respect for package maintainers, they do a real good job, on a lot of work [17:37] <wingo> except for that taaz fellow [17:37] <wingo> bastard [17:37] <wingo> ;-) [17:38] <taaz> speaking of heretics... [17:38] <taaz> we have debian 2.2 (ie 15 month old packages) in our lab [17:38] <taaz> i suggested we upgrade to latest stuff [17:38] <taaz> guess what my advisor said? [17:39] <taaz> he has been "dissapointed" with Debian/Ximian cause it's different [17:39] <taaz> and "many of the easy-maintenence features of RedHat are gone" [17:39] <wingo> give him a box of tissues and do it anyway [17:40] <taaz> I don't have root [17:40] <wingo> that sucks [17:40] <taaz> well... i took over a machine and did it [17:40] <taaz> it's been up longer than any other machine in the lab, so ha! [17:40] Action: wingo hands taaz the cluestick [17:40] <taaz> and i upgrade it every week [17:41] <taaz> i'm shocked and disappointed with this redhat sentiment ;) [17:41] <wingo> i'm in raleigh, the local lug is redhat-rabid [17:42] Action: taaz hands back the cluestick [17:42] <taaz> at my local lug meetings they spend half their time talking about WindersXP! [17:44] <taaz> so i -still- can't build the docs [17:45] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [17:48] <taaz> guilaunch tool getting pretty cool [17:48] <taaz> or shall i say super duper cool [17:50] <wingo> i haven't been home enough to use it :-\ [17:51] <taaz> dude, you must try [17:51] <taaz> must i say! [17:52] <taaz> it really should just be folded into gsteditor [17:52] <wingo> hopefully steveb can be suckered into being the gsteditor wrangler [17:55] <taaz> so this is really trick... i've got fakesrc ! statistics ! fakesink in guilaunch. you can adjust params realtime and watch the buffer rates change [17:56] <wingo> cool! [18:00] <taaz> 8.0k gl.lo 192k gststatistics.lo [18:00] <taaz> err.. that aint right [18:00] <taaz> anyway, why is an unstripped .lo so huge? [18:00] <taaz> 189380 vs 5720 [18:01] <taaz> is 5.7k too much to add to gst/elements? [18:02] vishnu (jo...@cx...) joined #gstreamer. [18:02] <taaz> vishnu knows [18:02] <vishnu> huh? [18:03] <taaz> is 5.7k too much to add to gst/elements? [18:03] <taaz> less than that really... [18:03] <taaz> if you take out dbg stuff [18:03] <taaz> and optimize it [18:03] <vishnu> taaz: how about moving identity out of gst/elements? [18:03] <taaz> eh? why? [18:04] <vishnu> taaz: is identity useful beyond testing? is anyone going to use it in a real application? [18:04] hadess (ha...@de...) joined #gstreamer. [18:04] <hadess> hey hey hey [18:04] <vishnu> taaz: we could add a plugins/special or something and move it there [18:04] <taaz> vishnu: don't start down that path [18:04] <vishnu> taaz: huh? [18:05] <taaz> we can argue the potential usefulness of any plugin [18:05] <vishnu> well, yah, that's why they should be optional [18:06] <taaz> filesrc should certainly be optional [18:06] <taaz> right? [18:06] <vishnu> yah, what not? [18:06] <taaz> jeez [18:06] <vishnu> why not? [18:07] <taaz> filesrc stripped is 6296 [18:07] <vishnu> look at the linux kernel. almost everything can be a module or compiled in [18:07] <taaz> no no, i agree and all... [18:07] <taaz> but like for most apps this is silly [18:07] <vishnu> sure, we can have filesrc get compiled in by default, but those embedded folks might not need it [18:07] <taaz> and if you're really doing embedded work it's likely you'll strip it out by hand [18:09] <vishnu> well, you asked for my opinion .. [18:09] <hadess> if you do embedded work, you just don't install filesrc, and that's it [18:09] <wingo> hadess! [18:10] <hadess> hey wingo ;) [18:10] <wingo> eh, bbl [18:11] Nick change: wingo -> wingo-out [18:11] <taaz> this is where a cml2 sorta thing would work. have ifdef's around every plugin... can just setup things like that rather than moving plugins around [18:11] <vishnu> yah [18:18] steveb (st...@no...) left irc: Ping timeout for steveb[node1ee41.a2000.nl] [18:24] evil_monk (trevo@208.141.162.68) joined #gstreamer. [18:33] vektor_ (wb...@pe...) joined #gstreamer. [18:33] <vektor_> hi [18:49] Nick change: taaz -> taaz-away [18:56] vektor_ (wb...@pe...) left irc: too bad the scene is dead [18:58] <hadess> bye [18:58] hadess (ha...@de...) left irc: Client Exiting [19:00] steveb (st...@no...) joined #gstreamer. [19:19] ChiefHighwater (pa...@te...) joined #gstreamer. [19:26] <BBB> anyone v4l here? :) [19:35] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: Connection reset by peer [19:36] evil_monk (trevo@208.141.162.68) joined #gstreamer. [19:37] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [19:46] sienap (sy...@s3...) joined #gstreamer. [19:46] <sienap> hej hej [19:48] <steveb> hey hey [19:48] <sienap> hey hey ;) [19:54] <BBB> hey sienap [19:55] <sienap> Hey BBB [19:56] <BBB> wat weet jij van video4linux/v4lsrc? [19:57] <BBB> or are you still doing svgalibsink? [19:58] <sienap> i will eventually [19:58] <sienap> but i am busy with some other stuff right now :) [19:58] <sienap> my intrests are kinda shifting all the time [19:58] <sienap> atleast i code again :) [19:59] <sienap> anyway [19:59] <sienap> i am off [19:59] sienap (sy...@s3...) left #gstreamer. [20:00] <BBB> :) [20:33] Nick change: wtay-zZz -> wtay [20:33] <wtay> hello [20:33] <vishnu> yo! [20:33] <ajmitch> hi [20:36] <wtay> wingo-out: ? [20:36] Nick change: wingo-out -> wingo [20:36] <wingo> just got back :) [20:36] <wtay> do I need to do anything special to get the API docs online? [20:36] <wingo> no, they're not in cvs [20:37] <wtay> ok [20:37] <wingo> well, not in www cvs :-) [20:38] <wingo> i'm just finishing updating README.website and some changes to admin.php to make things better documented [20:45] <wingo> is it up? [20:45] <wingo> the docs i mean [20:45] <wtay> yup [20:46] <wtay> uhm.. sorry.. not yet [20:47] <wtay> yay, online [20:49] <wingo> nice work wtay [20:50] <wingo> yeah. lets give it 110% ;) [20:50] <wtay> thx :) [21:16] wtay (wi...@ca...) left irc: Read error to wtay[cable-195-162-215-201.upc.chello.be]: EOF from client [21:18] wtay (wi...@ca...) joined #gstreamer. [21:18] #gstreamer: mode change '+o wtay' by ChanServ!s@ChanServ [21:18] #gstreamer: mode change '-o wtay' by wtay!wi...@ca... [21:28] <wtay> wingo: we have backups? :) [21:30] <wingo> yes, actually [21:30] <wingo> every night [21:30] <wingo> and occasionally i pull them to my local machine [21:30] <wtay> cool [21:31] <wingo> the situation is under control :-) [21:31] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [21:33] <wingo> yeah, for the reference, /home/groups/g/gs/gstreamer/gstreamer.sql [21:34] <wingo> er, either for the record or for reference, i seem to have made a hybrid phrase :) [21:34] <wingo> every night that updates [21:34] <wingo> i shound make a monthly thing too [21:34] Nick change: wingo -> wingo-lunch [21:35] evil_monk (trevo@208.141.162.68) joined #gstreamer. [21:46] arik (ar...@sd...) joined #gstreamer. [21:47] omega (om...@om...) joined #gstreamer. [21:48] <arik> hey omega [21:48] <omega> yo [21:50] Action: arik is trying to figure out what video display stuff picogui can do [21:50] <arik> without knowing much about it ;-) [21:51] <arik> i think it uses SDL on linux though [21:53] <wtay> yo [21:54] <arik> yo [21:54] <ajmitch> hey arik [21:54] <arik> hey aj [22:02] <arik> omega: http://picogui.org/doc/doxygen/cli_c/html/group__pgfxprim.html [22:04] <omega> hmm, looks sparse [22:05] <arik> yes [22:05] <arik> but doable? [22:05] <arik> i really know very very little about video [22:05] <arik> should be interesting ;-) [22:05] <arik> those are the builtin gfx stuff [22:05] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [22:05] <arik> they also have a variety of video options [22:05] <arik> like sdl, fb, and ncurses [22:05] <omega> problem is that there's no definition of pgBitmap provided there (link was bogus) [22:06] <arik> hmm [22:06] <arik> let me look [22:06] <arik> http://picogui.org/doc/doxygen/cli_c/html/group__pgfxprim.html#a10 [22:07] <arik> #define [22:07] <arik> PG_WIDGET_BITMAP 4 [22:08] <arik> hmm [22:09] <arik> pghandle pgCreateBitmap(short width, short height); [22:09] <arik> pghandle pgNewBitmap(struct pgmemdata obj); [22:10] <arik> * \brief Create a new bitmap object from existing data [22:10] <arik> * [22:10] <arik> * \param obj A pgmemdata structure, as returned by a pgFrom* function [22:10] <arik> * \returns A handle to the new bitmap object created in the PicoGUI server [22:10] <arik> * [22:10] <arik> * \sa pgFromMemory, pgFromFile, pgFromStream, pgFromTempMemory, pgDelete, pgEnterContext, pgLeaveContext, pgCreateBitmap [22:11] <arik> void pgRender(pghandle bitmap,short groptype, ...); [22:11] <arik> hmm [22:13] <arik> all of that is in cli_c/include/picogui/client_c.h btw [22:14] Nick change: wingo-lunch -> wingo [22:18] asmod (stevec@64.5.222.2) left irc: [x]chat [22:21] <wingo> yes. chili and beer. mmm [22:25] <wtay> glib-genmarshal --body --prefix=gst_marshal gstmarshal.list >> gstmarshal.c [22:25] <wtay> /bin/sh: glib-genmarshal: command not found [22:25] <wtay> make: *** [gstmarshal.c] Error 127 [22:25] <wtay> :( [22:25] vishnu (jo...@cx...) joined #gstreamer. [22:25] <vishnu> vektor: your player looks great! [22:26] <vishnu> the audio is about 15 seconds ahead of the video [22:26] <wingo> wtay: maybe that needs to be in funny conditionals [22:26] <wingo> the BUILT_SOURCES patch [22:26] <wingo> BUILT_SOURCES = @BUILT_SOURCES_GLIB2@ [22:27] <wtay> if USE_GLIB2 [22:28] <wtay> that works.. [22:28] <vishnu> vektor: u there? [22:30] <arik> oooh [22:30] <arik> the dvd player? [22:31] <vishnu> arik: yah [22:31] <arik> vishnu: was it difficult to get working? (i want to try it) [22:32] <vishnu> arik: pretty easy .. [22:32] <arik> hmm [22:34] <vishnu> arik: wait till vektor is here to walk you through it [22:35] <arik> sounds like a good idea [22:36] <vishnu> vektor: wow, it's rock solid -- no dropped frames [22:36] <arik> wow, how's the speed? [22:36] <vishnu> vektor: depends on the specifics of the vob, i guess [22:36] Zeenix (ze...@13...) joined #gstreamer. [22:36] <vishnu> arik: it's eating about 60% of my dual P-III 500Mhz system [22:37] <wingo> hey zeenix. [22:37] <Zeenix> yo all [22:37] <arik> vishnu: heh [22:37] <wtay> yo [22:37] Action: Zeenix hides as he havent completed his plugin yet [22:38] Action: wtay hides as hes hasn't looked at the plugin yet [22:38] <arik> heh [22:38] Action: arik hides cause he doesn't even know what plugin it is [22:38] Action: ajmitch hides just cos he can [22:38] Action: vishnu just hides for no reason [22:39] Action: wtay hides because of a fear of being infected by these insane guys [22:39] <arik> hehe [22:39] <vishnu> vektor: wow, mplayer sucks compared to your player [22:39] <ajmitch> insane? [22:39] <ajmitch> us? [22:39] Action: arik hides because he needs to get some actual work done! [22:39] <ajmitch> vishnu: what's the player? [22:39] <ajmitch> arik: heh, so do i ;) [22:39] <wtay> ajmitch: there is no word to describe your state of mind :) [22:39] <ajmitch> arik: exam tomorrow... [22:39] <ajmitch> wtay: why thx ;) [22:39] <arik> ajmitch: hehe [22:40] Action: Zeenix sets channel name to #hidenpeople... [22:40] <vishnu> where the heck is vektor??! [22:40] <vishnu> get vektor here immediately, i have to give him some flowers [22:40] <ajmitch> arik: i expect to get at least 90% ;) [22:40] <arik> hehe [22:40] <arik> wow [22:40] <arik> ;-) [22:40] <ajmitch> arik: if you knew how easy it was... [22:40] <arik> heh [22:41] <arik> i would come and go to school thre [22:41] <arik> there [22:41] <ajmitch> arik: computer programming, they assume that you know *nothing* :) [22:41] <ajmitch> next year it's lots harder ;) [22:41] <arik> hehe [22:41] <Zeenix> omega: away ? [22:41] chillywilly (da...@d1...) joined #gstreamer. [22:42] Action: Zeenix gets his new c compiler: sdcc [22:42] <arik> seems like [22:42] <vishnu> vektor: "Any Given Sunday" is rock solid -- amazing [22:43] Zygo (zbl...@24...) left irc: Read error to Zygo[246.chipworks.com]: Connection reset by peer [22:45] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-39.brain.net.pk] [22:45] <arik> bbiab [22:45] arik (ar...@sd...) left irc: leaving [22:45] <vishnu> vektor: where are you!!!? [22:46] bstard (Lor...@a2...) joined #gstreamer. [22:46] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [22:47] <vishnu> vektor: doesn't leak memory either! [22:47] <ajmitch> vishnu: where is this amazing app? [22:48] <vishnu> movietime.sf.net [22:49] Action: omega wishes cothreads were a fundamental part of POSIX [23:07] ds_ (ds...@c7...) joined #gstreamer. [23:07] Zeenix (ze...@13...) joined #gstreamer. [23:09] <wtay> fyi: the new API docs are online [23:09] Nick change: ds_ -> ds [23:10] <omega> cool [23:10] <ds> I'm back. did anyone miss me? [23:10] <ds> I was in NY for a week [23:10] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-34.brain.net.pk] [23:10] <omega> ds: what for? [23:11] <ds> drinking and shopping, apparently [23:11] <omega> oh [23:11] <wtay> I'm still reworking the API docs though (lots of extra examples) [23:11] <omega> wtay: the object hierarchy shown in the docs includes some gtk widgets for some reason [23:11] <wtay> omega: yes, I need to run the docs with glib2 [23:12] <omega> why would that matter? the scanner should only pick up classes in the code you give it [23:12] <wtay> omega: it doesn't, it picks whatever it finds [23:12] <omega> hmm, it didn't before [23:13] <omega> I guess I need to write up some docs in various places, like the timecache [23:13] <wtay> it didn't? hmm.. [23:13] <wtay> omega: I left out trace.. [23:13] <omega> ok, I need to rebuild that [23:13] <wtay> omega: timecache could use some examples, yes.. [23:14] <omega> it needs to be finished... [23:14] <wtay> also debug has some undocumented stuff, like the typedefs of the handlers [23:14] <vishnu> heh, i thought you were going to document it before it worked ;-) [23:14] <omega> vishnu: I can document what exists, yeah [23:15] <omega> wingo: should I write a test program in the next few hours to take the 16 channels from the hammerfall and show their levels on screen? [23:18] Nick change: taaz-away -> taaz [23:23] sienap (sy...@s3...) joined #gstreamer. [23:23] <sienap> hmm [23:23] <sienap> <- new mail from the rightshop [23:23] <sienap> ooh zeenix not here [23:23] <sienap> too bad [23:23] <sienap> later [23:23] sienap (sy...@s3...) left #gstreamer. [23:24] <taaz> so why is there only int64 and not uint64 too? [23:24] <omega> where? glib2? [23:25] <taaz> i guess [23:25] <taaz> i dunno [23:25] <taaz> in gstreamer glib to shim stuff at least [23:25] <omega> dunno [23:26] <taaz> can someone test something for me? [23:26] <taaz> run guilaunch and check if it shows params from superclasses [23:26] <wtay> taaz: it doesn't [23:26] <taaz> that blows [23:26] <taaz> why not? [23:26] <wtay> not implemented [23:27] <wtay> I discovered it today [23:27] <taaz> speaking of that.. the get/set stuff is weird [23:27] <taaz> why don't they return a gboolean on success/failure? [23:27] <wtay> also -inspect doesn't show the signals/properties of the superclasses [23:27] <omega> taaz: I never figured that out [23:27] <taaz> sigh... [23:27] <wtay> taaz: what get/set? [23:27] <taaz> omega: care to ask on glib lists? [23:28] <taaz> wtay: yeah [23:28] <omega> if I had time, sure, but why don't you? [23:28] <taaz> wtay: seems like it could be smart and if one class fails get/set then it would ask the superclass [23:28] <taaz> i'm scared of new mailing lists [23:28] <omega> pff [23:28] <wtay> taaz: yes.. it's not optimal.. [23:29] <taaz> are we the first people to notice this? [23:29] <wtay> I doubt it.. [23:29] <taaz> + G_OBJECT_WARN_INVALID_PROPERTY_ID (object, prop_id, pspec); [23:29] <taaz> that does nothing afaict [23:29] Action: wtay spotted and inconsistency in the API [23:30] <omega> taaz: yeah, it prints junk to the screen [23:30] <taaz> i didnt look at what it does though... just silently ignores invalid named params :( [23:30] <taaz> omega: it didn't for me [23:30] <omega> using gtk or glib? [23:30] <omega> s/glib/glib2/ [23:31] steveb (st...@no...) left irc: advance! [23:31] <taaz> gtk [23:31] <omega> right, I don't think the... [truncated message content] |
From: IRC B. <wt...@us...> - 2001-10-26 04:27:57
|
******************************************************************* [03:03] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [03:04] omega_afk (om...@te...) joined #gstreamer. [03:04] <omega_afk> wingo-bbiab: ok, I still get error setting rate (44100): EINVAL, this time on the hammerfall [03:05] Nick change: omega_afk -> omega [03:10] Nick change: omega -> omega_dinner [03:11] <omega_dinner> wingo-bbiab: if you can commit or otherwise fix the problem while I'm eating... <g> [03:16] Nick change: taaz-away -> taaz [03:25] <vishnu> where's vektor?!! [03:32] <taaz> why you so anxious? [03:35] bstard (Lor...@a2...) got netsplit. [03:41] bstard (Lor...@a2...) got lost in the net-split. [03:50] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [04:03] bstard (Lor...@a2...) joined #gstreamer. [04:10] Nick change: wingo-bbiab -> wingo [04:10] <wingo> yo [04:11] <wingo> omega_dinner: ? [04:15] <wingo> my gst build is hosed atm as i redo for glib2 [04:25] chillywilly (da...@d1...) left irc: [04:28] Nick change: omega_dinner -> omega [04:30] <wingo> yo [04:30] <omega> so, any idea what's going on? [04:30] <omega> ** WARNING **: error setting rate (44100): Invalid argument [04:30] <wingo> well [04:30] <wingo> hrm [04:31] <wingo> try [04:31] <omega> hrm, alsasrc period_count=2 ! fakesink works [04:31] <wingo> period_frames=256 [04:31] <wingo> you set the rate manually? [04:31] <omega> no [04:31] <wingo> oh [04:31] <wingo> yeah i think i need to change the default period_count [04:31] <omega> basically the code I gave you [04:31] <wingo> back to 2 [04:31] <omega> I do that [04:32] <wingo> because you switched cards and i tested it against the non-hammerfall ;-) [04:32] <wingo> and that works? [04:32] <omega> alsasrc set state to NULL [04:32] <omega> um, testing [04:33] <omega> sbawe does the same [04:33] <omega> should I upgrade to latest alsa? [04:33] <wingo> what do you have? [04:33] <wingo> i use 0.9b7 i think [04:34] <wingo> it shouldn't change that much though, the pcm code it stable more or less [04:34] <omega> prob sept 17 CVS HEAD [04:34] <omega> updating now [04:35] <omega> but I have some files here from Oct 3 too, odd. [04:35] <omega> I made a point of not upgrading once the hammerfall worked for the first time <g> [04:36] <wingo> heh [04:36] <omega> so I made a copy <g> [04:36] <wingo> try setting the rate manually, see what that does [04:37] <wingo> rate=8000 [04:37] <wingo> rate=96000 [04:37] <wingo> whatev's [04:37] <omega> same [04:37] <wingo> also, what device are you using? [04:37] <wingo> the default? [04:37] <omega> er, "rate" not found [04:37] <omega> checking [04:38] <omega> yeah, default. atm the sbawe [04:38] <wingo> i think i called it frequency [04:38] <wingo> that was a dumb idea [04:38] <omega> we should setting on one, probbaly samplerate or something [04:38] <wingo> i like rate ;) [04:38] <omega> n/c [04:39] <wingo> so alsasrc doesn't work. is that only for the src%d stuff? [04:39] <wingo> the question is, does guilaunch compile under gtk2... hmmm [04:40] <omega> yes [04:40] <omega> this is just src%d, yes [04:40] <omega> I can try src, testing [04:40] <wingo> and you're requesting the pads now, or did you change the _connect() api? [04:41] <omega> requesting [04:41] <omega> ok, "src" works [04:41] <wingo> we need to change the api imho [04:41] <omega> sorta, it has some problems [04:41] <wingo> not the api, but what happens [04:41] <omega> ** WARNING **: xrun detected [04:41] <omega> ** CRITICAL **: file alsa.c: line 1254 (gst_alsa_get_channel_addresses): assertion `this->mmap_open == FALSE' failed. [04:41] <omega> DANGER WILL ROBINSON! [04:41] <wingo> hah! [04:41] <wingo> how about i play with that later, i have to write a paper now :-\ [04:42] <wingo> great, i managed to break what once worked [04:42] <omega> gah, this is when I have hardware access... [04:42] <wingo> ? [04:42] <wingo> oh [04:42] <omega> wednesday evening is when I'm in spitting distance of the actual hammerfall [04:42] <wingo> yeah, but shit. school is incredibly rough atm [04:42] <omega> ok, any suggestions? [04:43] <wingo> hmm. [04:43] <wingo> and you can record with arecord? [04:43] <omega> yes [04:43] <wingo> hmm [04:43] <omega> have been for 2mo now [04:43] <omega> weekly [04:43] <omega> haven't checked over all the files though [04:44] <wingo> you can try gdb, it's probably a stupid error [04:44] <omega> I dunno what to look for [04:44] <wingo> but you would have to become acquainted with alsa.c [04:44] <omega> right [04:44] <wingo> which takes a while [04:44] <wingo> i'll check on it tomorrow, or in a few hours if i do well on getting this paper done [04:45] <omega> ok, hurry <g> [04:45] <wingo> :) [04:45] Nick change: wingo -> wingo-busy [04:45] <omega> need help? <g> [04:46] <taaz> how's your doc coming along omega? [04:46] <omega> stalled atm [04:47] <wingo-busy> omega, alsasrc ! fakesink works did you say? [04:47] <omega> yes, but src0! would fail, confirming [04:47] <wingo-busy> hmm [04:47] <omega> right src0! fails with can't set rate [04:47] <wingo-busy> i'll let you know when i get gtk+-2.0 and gst built [04:47] <omega> ok [05:12] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [05:12] ds (ds...@c7...) joined #gstreamer. [05:14] <vektor> ds? [05:14] <vektor> you there? [05:15] <vishnu> vektor: dude, your movietime rocks! [05:15] <vektor> vishnu: really? [05:15] <vishnu> vektor: yah, really [05:15] <vektor> !! [05:15] <vektor> dude! you rock! :) [05:16] <vishnu> vektor: it's way better than mplayer -- totally smooth [05:16] <vishnu> vektor: i recompiled my kernel so i could use /dev/rtc , etc [05:17] <vektor> Cool !! [05:17] <vishnu> vektor: it uses about 60% of my dual P-III 500Mhz [05:17] <vektor> Yeah but that's fine. [05:18] <vektor> It's supposed to eat more CPU than your average player, but look really nice. [05:18] <vektor> Also, hmm... You should try it on some interlaced content, although my algorithm kinda sucks. There are some things I need to fix. [05:18] <vektor> ds: ping ping wake up :) [05:18] <vishnu> vektor: how do i know whether my dvd is interlaced? [05:18] <vektor> vishnu: Well, um, hmm. [05:19] <vektor> vishnu: First off, any DVD that is from video source (like a TV show) is a possibility for true interlaced content. [05:19] <vektor> Like, content where they use interlacing for higher framerate. [05:19] <vektor> Any DVD from film source (like it was shown in a theatre originally) will likely be progressive, but there are exceptions for parts of the video where they performed the 3:2 pulldown process before encoding. [05:20] <vishnu> hrm [05:20] <vishnu> there were a few problems -- the audio seems slight too slow, like maybe 47khz [05:20] <taaz> look at the extras [05:21] <taaz> or porn... most of that is interlaced, so i hear [05:21] <vektor> vishnu: Really? [05:21] <vektor> vishnu: Check mpegclock.c. Make sure the constants at the top use '90.0' instead of something else. [05:21] <ds> vektor: I'm around. cable modem is a bit flakey right now, though. [05:21] <vishnu> vektor: yah, it's weird. the alsa driver sounds fine but oss is slightly off [05:21] <vektor> vishnu: btw, for fun, try and set the constants (both) from 90.0 to 45.0 :) [05:22] <vektor> like replace the 90.0's with 45.0's. :) [05:22] <vektor> ds: erm, ok, nevermind i'll ask later. [05:22] <vishnu> vektor: yah, it's 90 .. [05:23] <ds> it seems ok, although I may just randomly disappear [05:24] <vektor> ds: what's your email so i cna put it and your name at the top of the linear fit code you sent me [05:24] <ds> ds...@sc... [05:24] <vektor> ok thanks [05:25] <vishnu> um, vektor, it's a little too slow now :-) [05:25] <vektor> vishnu: yea yea!! :) [05:25] <vektor> vishnu: you can set it a bit higher though and it will 'shitty' resample for you [05:27] <vishnu> vektor: your going to have a tough time getting this integrated into gstreamer. someone has to fix the mutex problems first [05:28] <vishnu> vektor: but, like, it's totally rad. it's like a different movie when there isn't any skew :-) [05:29] <taaz> vishnu: is that someone you? [05:29] Action: vishnu looks behind himself [05:29] <vishnu> huh? who? [05:29] <vishnu> erik doesn't like me anyway [05:29] <taaz> vishnu: you have a smp machine to test on? [05:30] <vishnu> taaz: yah, but i leave for india in a few days. i'm not sure when i get an SMP box there .. maybe early next year [05:30] <taaz> eh? india? [05:31] <vishnu> taaz: maharashtra :-) [05:31] <taaz> ? [05:31] <vishnu> near bombay :-) [05:31] <taaz> ok [05:33] <vishnu> vektor: it feels better at about 95 [05:38] <vektor> weird [05:38] <vishnu> taaz: basically, here's what i'd want to do: [05:39] <vishnu> rip out all the mutexen. figure out what data structures are shared. add one lock global. see if it works. benchmark. split the lock as appropriate [05:40] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [05:41] <taaz> that sounds hard [05:42] <taaz> i'd suggest coming up with various very simple examples that always break [05:42] <taaz> something simpler to debug than a dvd pipeline ;) [05:42] <vishnu> taaz: yah, no kidding. but you really have to understand what you're doing. otherwise you get a partial fix and then it breaks again in edge cases [05:44] <vishnu> taaz: if i'm a plugin or app developer, i never want to have to worry that gstreamer is going to corrupt memory because of sloppy locking [05:46] <taaz> well duh [05:46] <taaz> to be blunt ;) [05:46] <vishnu> taaz: so someone needs to *understand* what is going on in there. it's not debuggable [05:46] <taaz> i'll just take a wild guess: gstqueue.c [05:47] <vishnu> maybe. dunno [05:48] <vishnu> i don't want to guess. if someone is going is guess then it's not going to get fixed right. you have to *know* [05:49] <taaz> ok, i vote vektor applies some formal methodology to gstreamer. write up a mathematical description and prove it works. [05:49] <vektor> yea yea! [05:49] <vektor> i need to do my algorithms assignment. [05:49] Action: vektor idle bigtime [05:53] <vishnu> taaz: having an SMP machine to test on makes a huge difference too [05:53] <vishnu> SMP will kick your ass if you mess up :-) [05:57] <taaz> i'm very aware of this [05:57] <taaz> otherwise i wouldn't care too much [05:57] <taaz> which is why i think i'll end up debugging this :( [06:00] <vishnu> taaz: write some simple multi-threaded test proggies to get a feel for it before you dive in [06:01] Shippou (no...@cr...) joined #gstreamer. [06:01] <vishnu> SMP is diabolical :-) [06:02] omega (om...@te...) left irc: [x]chat [06:03] Action: taaz already swimming [06:12] bstard (Lor...@a2...) got netsplit. [06:12] wingo-busy (wi...@rd...) left irc: going home... [06:12] bstard (Lor...@a2...) returned to #gstreamer. [06:27] Action: vishnu gives taaz an inflatable rubber tube [06:31] ds_ (ds...@c7...) joined #gstreamer. [06:32] Nick change: ds_ -> ds [06:41] mwc (ma...@ly...) joined #gstreamer. [06:51] estesm (es...@ws...) joined #gstreamer. [06:59] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [06:59] Nick change: ShrimpZzZ -> ShrimpX [07:17] taaz (dlehn@66.37.66.32) left irc: Ping timeout for taaz[66.37.66.32] [07:17] taaz (dlehn@66.37.66.32) joined #gstreamer. [07:29] <taaz> how much debug spam should i add to gstqueue? [07:30] <taaz> i'm afraid it will get absurd [07:30] <taaz> i think i fixed it too [07:30] <taaz> of course i haven't tested it yet... [07:31] <taaz> but some of the debug code in there now is dubious.... could report false info in some cases [07:56] <taaz> ok, this code is confusing [08:15] steveb (st...@no...) joined #gstreamer. [08:36] ChiefHighwater (pa...@te...) left irc: Ping timeout for ChiefHighwater[temple-baptist.com] [09:01] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [09:04] ds (ds...@c7...) joined #gstreamer. [09:12] mwc (ma...@ly...) left irc: [x]chat [09:22] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [09:23] ds (ds...@c7...) joined #gstreamer. [09:27] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [09:54] ds (ds...@c7...) joined #gstreamer. [10:20] Nick change: taaz -> taazzzz [10:22] ChiefHighwater (pa...@te...) joined #gstreamer. [10:22] ChiefHighwater (pa...@te...) left irc: [10:39] ChiefHighwater (pa...@te...) joined #gstreamer. [10:40] Nick change: ChiefHighwater -> CHW_Omega [10:40] <CHW_Omega> Ello [10:40] <CHW_Omega> any Debian wizards awake? [10:41] <CHW_Omega> cuz we're tryin to install Debian on this here Sparc20+...sound like fun? [10:43] <CHW_Omega> ajmitch? [10:43] <CHW_Omega> proly at skewl [10:46] Action: ajmitch is here [10:47] <ajmitch> pity the other ppl in my lug aren't around, one of them has debian running on a sparc, iirc [10:57] <CHW_Omega> well, were downloading stuff now to try to install it on this machine [10:57] <CHW_Omega> it had cd and 2 hard drives, but i think we're gonna use floppies...cuz that's what we know how to do [10:58] <CHW_Omega> s/had/has [10:58] <CHW_Omega> Omega would know more, but he is afk atm [11:02] <ajmitch> yeah, floppies should work [11:07] Nick change: ShrimpX -> ShrimpZzZ [11:14] estesm (es...@ws...) left irc: Client Exiting [11:43] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [11:45] CHW_Omega (pa...@te...) left irc: Ping timeout for CHW_Omega[temple-baptist.com] [12:00] omega (om...@te...) joined #gstreamer. [12:01] <ajmitch> hi omega [12:01] <omega> yo [12:02] CHW_Omega (pa...@te...) joined #gstreamer. [12:02] Nick change: CHW_Omega -> ChiefHighwater [12:04] <ajmitch> have any luck with installing? [12:04] <ChiefHighwater> the debian server must be behind like a 14.4 modem or something [12:04] <ChiefHighwater> taking forever to pull stuff down [12:05] <ChiefHighwater> so no install attempts yet [12:05] <ChiefHighwater> and it's 3 am, so install attempts tonight [12:05] <omega> wimp <g> [12:05] <ajmitch> haha [12:05] Action: ajmitch shoudl sleep, fun exam tomorrow ;) [12:05] <ChiefHighwater> at 3.5k/second we'd be here for a week [12:06] <ChiefHighwater> fun? [12:06] <omega> neat. hedrick's patch for ide on 2.2.19 doesn't apply to 2.2.19 [12:06] <ajmitch> ChiefHighwater: an exam on computer programming (using java) [12:07] <ajmitch> fun only because it's dead easy ;) [12:09] <ChiefHighwater> well, I been learnin about sparcs tonight [12:09] <omega> um, ok, this is gonna have to wait [12:09] <ajmitch> heh [12:09] <omega> stupid kernel stuff [12:09] Action: ajmitch goes off to bed... [12:09] <ajmitch> hmm, so early, just after 11pm... [12:10] <ajmitch> ah well ;) [12:10] <ChiefHighwater> we built a dual proc (150s?) machine with 192 MB of ram (with 9bit bytes...that's cute) triple SCSI, dual video and onboard sound [12:10] <ChiefHighwater> what did I miss, omega? [12:10] Action: omega reboots the gateway ChiefHighwater and omega are behind.... [12:10] <ChiefHighwater> txs [12:10] <omega> heh [12:10] <omega> l8r all [12:11] omega (om...@te...) left irc: init 6 [12:11] <ChiefHighwater> hrm, what's this button do [12:14] ChiefHighwater (pa...@te...) left irc: Ping timeout for ChiefHighwater[temple-baptist.com] [13:22] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [13:25] bstard (Lor...@a2...) joined #gstreamer. [14:49] asmod (stevec@64.5.222.2) joined #gstreamer. [15:08] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [15:53] Zygo (zbl...@24...) joined #gstreamer. [16:08] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:14] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: Connection reset by peer [16:17] bstard (Lor...@a2...) joined #gstreamer. [16:27] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: Connection reset by peer [16:27] bstard (Lor...@a2...) joined #gstreamer. [16:30] dobey (dobey@141.154.95.104) joined #gstreamer. [16:43] Nick change: taazzzz -> taaz [16:48] sienap (sy...@s3...) joined #gstreamer. [16:48] <sienap> Boe.. [16:48] Nordic (ma...@26...) joined #gstreamer. [16:48] <dobey> ... [16:49] <sienap> dobey! [16:49] <dobey> hi :/ [16:49] <sienap> what is ? [16:50] <dobey> <- sad [16:51] <sienap> why ? [16:52] <dobey> because people are assholes, and the stupid furniture people got my mattress mixed up with someone elses [16:52] <sienap> grr [16:53] <dobey> and again, i'm lame [16:53] <sienap> you're not [16:53] <sienap> what happened which makes you think you are ? [16:54] <dobey> n/m [16:54] <sienap> ? [16:58] sienap (sy...@s3...) left #gstreamer. [17:04] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [17:07] harobed (harobed@AC95F86E.ipt.aol.com) joined #gstreamer. [17:08] harobed (harobed@AC95F86E.ipt.aol.com) left irc: Client Exiting [17:11] wingo (wi...@rd...) joined #gstreamer. [17:11] harobed (harobed@ACA234E3.ipt.aol.com) joined #gstreamer. [17:11] Nick change: wingo -> wingo-class [17:36] ajmitch (aj...@p3...) left irc: Ping timeout for ajmitch[p35-max4.dun.ihug.co.nz] [17:44] sienap (sy...@s3...) joined #gstreamer. [17:45] sienap (sy...@s3...) left #gstreamer. [17:53] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [18:04] ajmitch (aj...@p4...) joined #gstreamer. [18:16] vishnu (jo...@cx...) joined #gstreamer. [18:23] Nick change: Nordic -> nordic|zz [18:24] Nick change: taaz -> taaz-away [18:28] nordic|zz (ma...@26...) left irc: [x]chat [18:29] dobey (dobey@141.154.95.104) left #gstreamer (eh). [18:46] sienap (sy...@s3...) joined #gstreamer. [18:46] sienap (sy...@s3...) left #gstreamer. [18:48] Nick change: wingo-class -> wingo [18:48] <wingo> yo [18:49] Shippou (no...@ot...) joined #gstreamer. [18:51] steveb (st...@no...) got netsplit. [18:51] vishnu (jo...@cx...) got netsplit. [18:51] steveb (st...@no...) returned to #gstreamer. [18:52] vishnu (jo...@cx...) returned to #gstreamer. [18:53] wtay-zZz (wi...@ca...) left irc: Read error to wtay-zZz[cable-195-162-215-201.upc.chello.be]: EOF from client [18:57] _gst_newt_ joined #gstreamer. [19:19] <wingo> anyone here? [19:19] <taaz-away> no [19:19] <wingo> heh [19:21] Nick change: taaz-away -> |7442| [19:21] <|7442|> i don't understand the queue code [19:21] <|7442|> i think that's what's breaking smp but i'm not sure [19:21] <|7442|> there are strange hacks in gstthread.c related to queues too... [19:22] <|7442|> poor design imho [19:22] <|7442|> too bad i dont understand the issues involved [19:25] <wingo> yeah. me too. also, i don't have smp :-\ [19:25] <|7442|> care to test something for me? [19:25] <wingo> my gst build is hosed atm [19:28] <wingo> but ok [19:28] <|7442|> -launch fakesrc silent=true ! queue ! {statistics buffer_count_freq=100 ! fakesink silent=true} [19:28] <|7442|> ie, something to just use a queue [19:28] <|7442|> that stats thing is just to see how many buffers you've processed roughly before it locks [19:28] <wingo> when gst gets built i'll do that (slow machine, it takes a while) [19:28] _gst_newt_ joined #gstreamer. [19:29] <|7442|> often dies around 2-4k with no debug output, or varies alot more with --gst-mask=-1. [19:30] <|7442|> i tried to fixup queue, and think i made it more correct but that didn't help [19:30] <wingo> vishnu's most likely right on this one, smp needs serious attention [19:39] <|7442|> I managed to convince myself my queue fixups were deadlock free... when makes me think the scheduler is borked :( Ie, not consuming buffers on output for some reason. queue fills up and it just sits there... [19:39] Nick change: |7442| -> |7442|lunch| [19:50] harobed (harobed@AC8B16C1.ipt.aol.com) joined #gstreamer. [20:00] wtay (wi...@ca...) joined #gstreamer. [20:00] #gstreamer: mode change '+o wtay' by ChanServ!s@ChanServ [20:00] #gstreamer: mode change '-o wtay' by wtay!wi...@ca... [20:07] vishnu (jo...@cx...) left irc: Ping timeout for vishnu[cx959399-a.santab1.ca.home.com] [20:14] harobed (harobed@AC8B16C1.ipt.aol.com) left irc: Ping timeout for harobed[AC8B16C1.ipt.aol.com] [20:17] sienap (sy...@s3...) joined #gstreamer. [20:26] bstard (Lor...@a2...) joined #gstreamer. [20:26] BBB_ (BB...@uc...) joined #gstreamer. [20:29] BBB_ (BB...@uc...) left irc: [x]chat [20:29] BBB-zZz (BB...@uc...) left irc: Ping timeout for BBB-zZz[ucu-105-116.ucu.uu.nl] [20:33] swivel (swivel@168.103.157.45) joined #gstreamer. [20:38] <wtay> hi [20:39] sienap (sy...@s3...) left #gstreamer. [20:39] <swivel> hi [20:59] Nick change: |7442|lunch| -> taaz [21:00] <taaz> yo wtay [21:00] <wtay> yo [21:00] <taaz> do you understand the scheduling stuff? [21:01] <wtay> somewhat [21:01] <wtay> I'm hacking on it now [21:01] <taaz> it's broke for smp ya know [21:02] <wtay> ah that.. [21:02] <taaz> ah yes [21:02] <taaz> i can't run anything that uses threads right now with any sort of reliability [21:02] <taaz> i'm sure it's just luck it doesnt hurt non-smp [21:02] vektor_ (wb...@pe...) joined #gstreamer. [21:02] <vektor_> Hi! [21:02] <wtay> taaz: you think it's the queue? [21:03] <wtay> yo [21:03] <vektor_> Oh sorry... [21:03] <vektor_> Yo yo! [21:04] <vektor_> wtay: Hey dude try movietime. [21:04] <vektor_> wtay: There's a configure script. [21:04] <wtay> vektor_: I saw the code :) [21:04] <wtay> vektor_: never managed to get it working.. [21:04] <wtay> vektor_: does it play vobs too? [21:04] <taaz> i think the queue issues are partly to blame [21:04] <wtay> taaz: the queue uses the object lock to lock itself [21:05] <taaz> is that bad? [21:05] <wtay> taaz: the scheduler uses the object lock to lock objects [21:05] <vektor_> wtay: It plays vobs yes, it's much better now with the configure script (and the copied-in libmpeg2). [21:05] <taaz> i saw that and didn't know what to make of it [21:05] <taaz> hmm.. [21:05] <wtay> taaz: looks like an candidate for the trouble you get [21:05] <taaz> the problem is the queue fills up and the pipeline stops [21:05] <wtay> try to move the lock into the queue itself [21:06] <taaz> which means it's not reading values out [21:06] <taaz> but ... but... [21:06] <taaz> that logic in the queue is kind of crazy [21:06] <wtay> howso? [21:07] <taaz> there's some strange checks i dont understand [21:07] <wtay> like? [21:07] <taaz> and btw, it totally doesn't handle events properly [21:07] <taaz> are those buffer flags supposed to just disappear soon? [21:07] <taaz> like EOS, FLUSH, etc? [21:07] <wtay> yeah [21:07] <wtay> yes [21:08] <taaz> ok, other thing: the stuff in gstthread.c that calls the queue conditions for no apparent reason [21:08] <taaz> that is really breaking encapsulation [21:08] <wtay> taaz: on state changes it needs to unlock the queue [21:09] <taaz> i don't understand that stuff [21:09] <wtay> it should be rethought at some point [21:09] <taaz> that point is now ;) [21:09] <wtay> why now? [21:09] <taaz> cause it's broke [21:09] <taaz> something is broken [21:10] <taaz> maybe not that... [21:10] <wtay> I'm not convinced that's the problem [21:10] <taaz> problem is i just dont understand what it's doing right now. so i'm probably 'fixing' the wrong thing [21:11] <swivel> whats the problem? [21:11] <wtay> the "magic" only happens on state changes so you don't have to bother with that [21:11] <swivel> threading deadlocks on smp? [21:11] <taaz> why does thread have to signal queue conditions on state changes? [21:11] <taaz> swivel: yes [21:11] <wtay> taaz: 'cause the queue might be waiting [21:11] <wtay> taaz: it needs to unlock it [21:11] <swivel> pthread_cond_signal? [21:11] <swivel> () [21:12] <taaz> eh? [21:12] <swivel> by signal queue conditions do you mean pthread_cond_signal() ? [21:12] <taaz> i need a state diagram for this ;) [21:12] <swivel> or is this some glib interface for threads [21:12] <taaz> swivel: look at gstqueue/thread.c [21:12] <swivel> i dont use gstreamer ive just read about it [21:12] <wtay> taaz: I would think the thread would simply change the state of the queue and the queue would unlock itself [21:12] <swivel> but i know pthreads pretty well ;) [21:13] <wtay> swivel: gst/gstqueue.c and gst/gstthread.c need some code reviews :) [21:13] <swivel> sounds like a pain [21:13] <wtay> swivel: show up how well you know pthreads :) [21:13] <wtay> s/up/us/ [21:13] <swivel> wtay: is it pthreads? [21:13] <wtay> yup [21:14] <swivel> send me source tarball [21:14] <swivel> or give me url [21:14] vektor_ (wb...@pe...) left irc: too bad the scene is dead [21:14] <wtay> sec.. [21:14] <swivel> ive never used the thing so... i hope this wont be like trying to fix a bug in mozilla. [21:14] <wtay> http://download.sourceforge.net/gstreamer/gstreamer-0.2.1.tar.bz2 [21:15] <swivel> k wgetting [21:15] <wtay> swivel: it will :) [21:15] <swivel> i dont have an smp box here though so testing will be difficult [21:15] <swivel> if it only happens on smp [21:15] <wtay> yup, same here [21:15] <swivel> i can get access to an 8 way xeon tho ;) [21:15] <wtay> taaz: can you try to move the object lock to the queue? [21:15] <swivel> but it would have to wait til tonight, that box is kerberized [21:16] <ajmitch> hi all [21:16] <wtay> yo [21:17] <taaz> wtay: yeah i can try that [21:17] <taaz> wtay: but i need to understand what this code is supposed to do. on a state change what is supposed to happen? [21:18] <taaz> and what parts of the queue need to be protected by the object lock vs a queue lock? [21:18] <wtay> taaz: you don't need to fix the state change stuff for now [21:18] <taaz> oh? [21:19] <wtay> taaz: the state change stuff is only htere when stopping the pipeline [21:19] evil_monk (trevo@208.141.162.68) joined #gstreamer. [21:19] <taaz> well ok but it needs to happen [21:19] <taaz> err... happen properly [21:19] <taaz> if pipeline is stopped what happens to stuff in the queue? [21:19] <wtay> yes, but first try to get reliable data passing [21:19] <wtay> taaz: it goes away [21:20] <taaz> and are threads waiting on read/write to q supposed to just return null or what? [21:20] <swivel> strange [21:20] <wtay> taaz: yes, the queue is unlocked and the it detects the state change [21:20] <wtay> taaz: it then just stops [21:20] <taaz> uh... i don't think it detects anything right now [21:21] <taaz> we need state diagrams man! i should write up some graphviz .dot files for it [21:21] <swivel> this is confusing [21:21] <swivel> i dont like this :P [21:21] <wtay> taaz: it does, thats what the GST_STATE_PENDING stuff is all about in the queue [21:21] tnt-out (tn...@h2...) left irc: Client Exiting [21:21] <swivel> how come in some places stuff like g_cond_wait() and g_cond_signal() ar eused [21:21] <swivel> and elsewhere pthreads functions are used directly [21:22] <wtay> swivel: glib only wraps some functions of pthreads [21:22] <wtay> swivel: the locking is in glib but the pthread_* isn't [21:22] <swivel> wtay: I thought the point of glib's threading interface was so you didnt use pthreads, so your threaded apps could work on platforms that dont have pthreads available [21:22] <wtay> swivel: at least not in glib1.2 [21:23] <wtay> swivel: once we move completely to glib2 we should use the glib functions, yes [21:23] arik (ar...@sd...) joined #gstreamer. [21:23] <swivel> imo if you're going to use any pthreads functions directly you should use all pthreads functions [21:23] <taaz> wtay: so is it safe to actaully fix the queue? [21:23] <arik> yo [21:23] chillywilly (da...@d1...) joined #gstreamer. [21:23] Action: taaz asks with a <g> [21:23] <arik> vektor: you here? [21:23] <wtay> taaz: fixing is always safe :) [21:23] <taaz> well, i renamed parts of it to make sense [21:24] <wtay> that's ok if it makes sense to other people too :) [21:24] <taaz> ie, "fullcond" "emptycond" should really be queue->not_full/not_empty [21:24] <taaz> signaling "fullcond" is confusing because you acutally mean it's not full anymore [21:24] <taaz> on first look i thought you signaled when it became full and didnt know why [21:24] <swivel> um [21:24] <swivel> heh [21:25] <swivel> i have a feeling this code has race conditions scattered [21:25] <swivel> if g_cond_wait() is anything like pthread_cond_wait() [21:25] <arik> heh, i think all of libgstplay is one giant race condition waiting to happen [21:25] <wtay> swivel: it is [21:25] <taaz> yo, i like rewrote some of this to make more sense [21:25] <arik> i think i will spend today moving the current gstplay code to the new lib and then update cvs [21:25] <taaz> but it's still broken [21:26] <wtay> taaz: just try to make it work on your machine before commiting stuff [21:26] <swivel> you almost always have to use pthread_cond_wait() in a acquire lock; while(foo) wait_cond(); do something; release lock; [21:26] <taaz> wtay: clearly [21:26] <swivel> to use it properly / race-free [21:26] <taaz> wtay: are these queues always one reader, one writer, or can their be multiple readers/writers? [21:27] <taaz> s/their/there/ [21:27] <wtay> swivel: yup, that's the while (queue->level_buffers >= queue->size_buffers) { [21:27] <wtay> taaz: one reader/one writer [21:27] Nick change: arik -> arik|work [21:28] <taaz> wtay: i have a patch to make it slightly more clear/correct.. but as i say, it doesn't fix my problems. [21:28] <wtay> taaz: it's just a rename of some fields? [21:28] <swivel> what is gst_queue_chain() supposed to do? [21:28] <taaz> no, slight adjustment of cond/signal stuff too [21:29] <wtay> swivel: that's where the buffer is queued [21:29] <wtay> swivel: _get is used to take a buffer off the queue [21:29] <swivel> k [21:29] <wtay> taaz: "slight adjustment"? [21:30] <taaz> you want a patch? [21:30] <wtay> ok [21:30] <taaz> err... i need to redo some stuff, i kinda killed the code with some misunderstandings ;) [21:30] <wtay> hehe, details.. [21:31] <taaz> need to run off for a bit... will fix up later [21:31] Nick change: taaz -> taaz-away [21:34] <arik|work> woah [21:34] <arik|work> loads of changes to gst since i last updated [21:36] arik|work (ar...@sd...) left irc: leaving [21:49] ds (ds...@c7...) joined #gstreamer. [21:49] nash (wa...@13...) joined #gstreamer. [21:51] harobed (harobed@AC82F8DB.ipt.aol.com) joined #gstreamer. [21:52] arik (ar...@sd...) joined #gstreamer. [21:54] ChiefHighwater (pa...@te...) joined #gstreamer. [22:02] Nick change: ShrimpZzZ -> ShrimpX [22:21] arik (ar...@sd...) left irc: leaving [22:21] Nick change: taaz-away -> taaz [22:22] asmod (stevec@64.5.222.2) left irc: [x]chat [22:24] <ChiefHighwater> Ello taaz. We were looking for you last night as we attempted a Debian instal on a sparc 20+ 8-] [22:25] <taaz> eh? [22:25] arik (ar...@sd...) joined #gstreamer. [22:25] <arik> vektor: you here? [22:25] <ChiefHighwater> Omega and I got our hands on this sparc and we'er gonna set it up as a test bed of sorts [22:25] Nick change: arik -> arik|food [22:25] <wingo> nice [22:25] <arik|food> bbiab [22:25] <taaz> is that a good machine? [22:26] <ChiefHighwater> it's a free machine [22:26] <taaz> ie, fast [22:26] <taaz> ? [22:26] <ChiefHighwater> dual proc, dual vid [22:26] <ChiefHighwater> i think they are 150 procs [22:26] <swivel> you're installing linux on it [22:26] Action: swivel sighs [22:26] <ChiefHighwater> yes, we are [22:26] <swivel> solaris would be a more interesting test for gstreamer ;) [22:27] <swivel> on sparc [22:27] <swivel> smp nonetheless [22:27] <ds> solaris would be nearly impossible for gstreamer [22:27] <swivel> ds: why is that? [22:27] <ds> solaris's shared library loader sucks [22:27] <swivel> i ported my project to solaris on sparc in one night with the help of a friend (whos machine it was) [22:27] <swivel> ds: umm, we didnt have any problems with the dlopen that was available [22:27] <wingo> shouldn't g_module help out with that? [22:27] <ds> did it involve 100 shared libraries? [22:28] <swivel> not quite, more like ~10 dynamically loaded (dlopen) modules for audio/video/file/playback/playlist functionalities [22:28] <swivel> and dynamically linked libraries at startup, like libgtk gdk etc. [22:29] <ds> that's pretty good. [22:29] <swivel> works quite well on solaris [22:29] <swivel> better than linux, the threading was very nice. [22:29] <swivel> pthreads support on solaris is up there [22:29] <ds> I've had horrible problems with solaris and shared libs [22:29] <swivel> this was on an ultra10 creator3d [22:30] <swivel> ill check what version of solaris it is [22:30] <ds> what's wrong with pthreads on linux? [22:31] <swivel> nothing is wrong with it, pthreads support on solaris is just good, they have very well optimized systems for their hardware [22:31] <swivel> they have platform specific releases of the system libraries [22:32] <swivel> my app performed much better on the solaris box than it did on a dual xeon of much higher clock speed [22:32] <swivel> but that could just be XSun being faster and better graphics hardware [22:33] <ds> probably [22:33] <swivel> shit i cant login to the box, they are fixing his heat in the office [22:33] <swivel> so the box is down atm [22:33] <swivel> solaris 8 [22:34] <swivel> ds: now getting the app to work on freebsd, that took two nights [22:34] <swivel> which is somewhat amusing, considering it was freebsd on x86 too heh [22:34] <swivel> but there was no gnu configure or any of that which would have made it painless [22:34] <ds> much newer that what I used. I keep forgetting that old experience is no (or worse than no) experience [22:34] <swivel> oh and usleep() on solaris is most definately not MT-SAFE [22:34] <swivel> so dont use it [22:35] <swivel> i had to fix that before it worked on solaris [22:35] <swivel> we would get random SIGALRM when using usleep in multiple threads. [22:36] <swivel> switched to nanosleep [22:36] chillywilly (da...@d1...) left irc: [22:37] <ds> was that because the SIGALRM was picked up by the wrong thread? [22:39] <swivel> im not sure why, i just looked at the man page of usleep() after narrowing down where it was happening, and found that it wasnt safe [22:39] <swivel> switched to nanosleep and the problem was gone, usleep() on linux works though. [22:40] <swivel> but switched to nanosleep exclusively regardless [22:41] <taaz> can't you just fix the code? ;) [22:41] <swivel> fix what? the code that uses sleep? [22:42] <taaz> sleep [22:42] <swivel> you mean make usleep() mt-safe? [22:42] <taaz> yeah [22:42] <swivel> on solaris?! [22:42] <taaz> exactly [22:42] <swivel> its not supposed to be mt-safe [22:42] <taaz> wrap it [22:42] <taaz> whatever.. [22:43] <swivel> at least solaris had a man page that said it wasnt, unlike some other os's [22:46] <ds> it depends on the implementation. The glibc implementation is thread-safe [22:46] <ds> but the standard doesn't require it [22:46] <swivel> its not required for usleep to be thread safe [22:51] Zeenix (ze...@13...) joined #gstreamer. [22:51] <Zeenix> hi [22:54] <wtay> yo [22:55] <Zeenix> wtay: dont ask anything, i didnt get any time for coding today [22:55] <Zeenix> that nash came to my house & destroyed all my time :) [22:57] <wtay> rtp looks good [22:59] omega (om...@om...) joined #gstreamer. [22:59] <wtay> yo [22:59] <Zeenix> yo [22:59] <omega> yo [23:00] <Zeenix> omega: you misguided me yesterday :) [23:00] <omega> ? [23:00] <Zeenix> "Signals are typically associated with a class function pointer which is invoked every time the signal is emitted; if non-NULL, the pointed-to class function serves as a default handler for the signal" [23:01] <Zeenix> i remember you said "signals dont have any default handlers" [23:02] <wtay> Zeenix: reread that sentence [23:02] <Zeenix> wtay: which one ? [23:03] <wtay> omega: I have a CD here with 5 tracks.. what do I do? [23:03] <wtay> <Zeenix> "Signals are typically associated with a class function pointer which is invoked every time the signal is emitted; if non-NULL, the pointed-to class function serves as a default handler for the signal" [23:03] <omega> nothing, the package arrived <g> [23:03] <omega> finally [23:04] <wtay> you mean I screwed up a black CD for nothing? <g> [23:04] <omega> I warned you not to burn anything until I heard from her again... [23:04] Action: wtay slaps omega with the CD :) [23:04] <omega> now you've got a nice music disc to listen to though ;-) [23:05] <wtay> true.. [23:06] wingo-home (wi...@rd...) joined #gstreamer. [23:06] <omega> yo [23:06] <wingo-home> yo [23:06] <wingo-home> i'm checking into the alsa doober [23:07] <wingo-home> hey omega [23:07] <wingo-home> should all of the class->set_property = setfunc be before the property installations? [23:08] <wingo-home> for glib2 [23:08] <omega> um, dunno [23:08] <omega> that rings a bell as an issue, but I think I found a different problem [23:08] <wingo-home> i get GRuntime-Criticals [23:08] <wingo-home> which makes sense, it seems [23:09] <omega> dblcheck the paramspecs, if those are odd, it's pretty fragile [23:09] <wingo-home> it's just a warning though [23:09] <omega> it's a fatal warning if you try to use the param [23:09] <wtay> hmm [23:09] <omega> at least it was when I ran into something in that bit of code [23:09] <wingo-home> hrm. [23:10] <wtay> maybe that's how the default values is supposed to be set.. [23:10] <wingo-home> introspection works [23:10] <wingo-home> wtay: ? [23:10] <wtay> nevermind.. [23:10] <omega> wtay: not afaik [23:10] <wingo-home> ok [23:10] <wtay> what's the point of a default value [23:11] <wingo-home> hmm [23:11] <wtay> other than info for introspection that is [23:11] Zygo (zbl...@24...) left irc: Read error to Zygo[246.chipworks.com]: Connection reset by peer [23:11] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [23:11] <wingo-home> in glib 1.2 it's set in the instance init [23:11] <wingo-home> in glib2 hopefully the paramspec stuff will do that [23:11] <wingo-home> one would hope [23:11] <wtay> wingo-home: exactly, and it can only do that it the _set is done before adding the properties [23:12] <wingo-home> really? [23:12] <wingo-home> ok [23:12] <wingo-home> oh [23:12] <wingo-home> ok, you're making sense [23:13] <wingo-home> for once :-))) [23:13] <wtay> I would think it would call the _set_property function when adding the property to set the default value [23:13] <wingo-home> yeah [23:13] <wtay> maybe.. [23:13] <wingo-home> let's check on that.. [23:15] <wingo-home> still getting an error, hrm [23:15] <arik|food> back [23:15] Nick change: arik|food -> arik [23:17] <wtay> omega: I understand now why there cannot be a separate event handler in the loop based case.. (doh) [23:17] <omega> heh [23:17] <omega> not for downstream at least [23:17] <wtay> yup [23:18] steveb (st...@no...) left irc: advance! [23:19] Action: arik is working on gtmediaplay [23:20] <taaz> why? [23:20] <arik> why? [23:20] <arik> oh, you are talking to omega [23:20] <taaz> wtay: we all want to know... why? [23:20] <wtay> arik: feeling masochistic today? [23:21] <arik> wtay: hehe, something like that ;-) [23:21] <arik> trying to merge the newest libgstplay with the old gstmediaplay so that at least something usable will be in the next release [23:21] <wtay> taaz: because that's how the scheduler works [23:22] <wingo-home> hah! it's in gst_object_class_init [23:22] <wtay> arik: I have a feature request... [23:22] <arik> wtay: oh? [23:22] <wtay> arik: we currently have libgstmediaplay which actually is a bit too large [23:22] <wtay> arik: I want a lib with only the playobject in it [23:22] chillywilly (da...@d8...) joined #gstreamer. [23:23] <arik> wtay: not sure i completely understand [23:23] <wtay> arik: libgstmediaplay also has the glade stuff and playlists in it [23:23] <arik> ah yeah, i understand [23:23] <omega> need libgstplay that has no dependencies besides libgst [23:23] <arik> no [23:24] <omega> and some specific plugins, of course [23:24] <arik> libgstplay does have no other deps [23:24] <arik> he is talking about libgstmediaplay [23:24] <arik> that can be done [23:24] <omega> hrm, separate? [23:24] <arik> yep [23:24] <arik> libgstmediaplay is the play object [23:24] <omega> why? [23:24] <wtay> ok [23:24] <arik> the actual program widget [23:24] <arik> wtay: um, how would you get rid of the glade? construct the gui manually? [23:24] <wtay> libgstplay is just a widget (GtkBox) [23:24] <omega> hmmm, um. how much code is that really? and why would it be a separate lib and not a component? [23:24] <arik> i mean, why is that a good thing [23:25] <wtay> arik: of course [23:25] <wtay> arik: I don't want the gui [23:25] <arik> ah [23:25] <arik> so you just want the routines that call libgstplay? [23:25] <omega> wtay: libgstplay has no gtk at all, sounds like libgstmediaplay or libgstplayer or something would, which I don't get [23:25] <wtay> for bonobo-media that is [23:25] <arik> why don't you just use libgstplay directly wtay? [23:26] <wtay> there is no libgstplay, that's the point [23:26] <omega> gstplay should be about 500 lines of gui (no libglade) and some code to talk to libgstplay [23:26] <arik> omega: right, i agree [23:26] <arik> omega: well and calls to playlist code [23:26] <arik> or not [23:26] <arik> i mean that could be in the main file [23:26] <wtay> wait.. [23:26] <arik> wtay: there is a libgstplay [23:26] Action: arik thinks we are all confused [23:26] swivel (swivel@168.103.157.45) left #gstreamer. [23:27] <wtay> arik: there is only libgstmediaplay [23:27] <wtay> arik: it has all the gui/glade stuff in it [23:27] <arik> wtay: no [23:27] <omega> wtay: currently, yes, and basically all the gstplay/* code pretty much goes away [23:27] <wtay> in your copy or mine? <g> [23:27] <arik> um [23:28] <arik> in both i thnk [23:28] <arik> there is libgstplay which has the gst pipeline [23:28] <wtay> lib_LTLIBRARIES = libgstmediaplay.la [23:28] <arik> wait [23:28] <arik> oh! [23:28] <arik> yeah [23:28] <arik> you're right [23:28] <wtay> ah [23:28] <arik> it used to be called libgstmediaplay [23:28] <arik> i've renamed it ;-) [23:28] <wtay> ok [23:28] <arik> heh [23:28] <arik> i was so confused there for a minute [23:28] <wtay> even removing the gtk dep would be cool too [23:28] <arik> but libgstmediaplay doesn't have any gui code in it [23:28] <arik> yeah [23:28] <arik> i've done that [23:29] <wtay> arik: uh? [23:29] <arik> the new libgstplay doesn't require _anything_ besides gst [23:29] <wtay> gstmadiaplay has all the gui in it [23:29] <arik> the one i am about to commit [23:29] <wtay> ok [23:29] <arik> gstmediaplay has gui [23:29] <omega> do we even need libgstmediaplay in the long run? just need libgstplay and gstmediaplay [23:29] <arik> yes omega [23:29] <wtay> yes [23:29] <arik> there is no libgstmediaplay anymore [23:29] <arik> it's now libgstplay [23:29] <wtay> gstplay should be as small as possible [23:29] <omega> ok, </confusion> [23:29] <arik> and gstmediaplay is a probram [23:29] <arik> er program [23:29] <omega> both [23:29] <arik> libgstplay is very small [23:30] <arik> one sec [23:30] <wtay> libgstmediaplay is a lib too [23:30] <omega> wtay: not for long [23:30] <wtay> and gstmediaplay is a program using the lib [23:30] <arik> yes [23:30] <arik> 823 lines [23:30] <arik> total [23:30] <arik> for the lib [23:30] <wtay> so, two libs and 1 app [23:30] <arik> oh wait [23:30] <arik> that's not right [23:30] <arik> no [23:30] <arik> one lib [23:30] <arik> one app [23:30] <arik> the lib is 377 lines of code [23:31] <arik> that's the newest version [23:31] <arik> with no deps [23:31] <arik> other then gst and glib [23:31] <arik> which comes with gst [23:31] <wtay> cool [23:31] <arik> there is still some gtk stuff in it actually [23:31] <arik> signal stuff [23:31] <wingo-home> apparently, one must install_property with a paramspec that has a G_PARAM_CONSTRUCT in the flags [23:31] <wtay> then gstmediaplay is an app with a lot of gui stuff? [23:31] <arik> but i will be removing it [23:31] <wingo-home> for it to automagically initialize it [23:31] <arik> yes, gstmediaplay is an app with lots of gui stuff [23:32] <arik> that calls libgstplay [23:32] <wtay> arik: ok [23:32] <omega> arik: remember, moving to glib2 means that signals are part of glib/gst [23:32] <arik> wtay: once i commit this should be a lot simpler [23:32] <arik> omega: they are? [23:32] <omega> of course, they're part of GObject [23:32] <arik> omega: are they now g_signal_connect then? [23:32] <omega> yes [23:32] <arik> ooh [23:32] <arik> i know very little about glib2 [23:32] <wtay> arik: yes, you can use them now [23:32] <arik> just that we move from GtkObject to GObject [23:32] <wtay> no gtk needed [23:32] <arik> wtay: ok [23:32] <arik> wtay: sweet [23:33] <wingo-home> wtay: got that? regarding props initialization... i'm going back to school (aikido time) [23:33] <arik> wtay: ok, so once i commit you should be able to see what i'm talking about, should be later today [23:33] <omega> arik: write to the glib2 API, the shim will deal with gtk for you, within reason [23:33] <arik> omega: right [23:33] <wingo-home> l8r folks [23:33] wingo-home (wi...@rd...) left irc: Client Exiting [23:33] <omega> if it doesn't, ask, and we'll figure out whether it should or shouldn't be shimmed at all anyway [23:33] <arik> sounds good [23:33] <arik> right [23:33] <arik> ok [23:33] <arik> i should have gstmediaplay using the new libgstplay in a few hours at most [23:33] <arik> but first i'm gonna remove all the gtk_signal stuff [23:33] <wtay> we use the g_signal shims quite succesfully now [23:34] <arik> what about gtk_arg_get_infO? [23:34] <arik> er gtk_objcet [23:34] <arik> it becomes g_object_arg_get_info? [23:34] <wtay> uhm, that's the tricky stuff [23:34] <arik> is there any sample code for any of this? [23:34] <wtay> look at -inspect for that [23:34] <arik> alright [23:34] <wtay> -launch actually implements the xid lookup [23:35] <arik> hmm [23:35] <arik> i don't need that cause i don't have video in the new libgstplay yet [23:35] <arik> should be simple to add eventually though [23:35] <wtay> oh [23:35] <arik> it was a bit complicated to do both at the same time [23:36] <wtay> ah great.. -launch doesn't do video with glib2 [23:36] <omega> right [23:36] <arik> heh [23:36] <arik> wonderful [23:37] <wtay> so. libgstmediaplay would be used by an app that wants the gui stuff too [23:37] <omega> no, libgstmediaplay's gui stuff will be folded into gstplay [23:38] <wtay> no lib? [23:38] <omega> because once libgstplay is done, there won't be anything but gui code in there, and not much at that [23:38] <arik> um what? [23:38] <arik> now i'm confused again [23:38] <arik> libgstplay is just the pipeline [23:38] <arik> the app has all the gui code [23:38] <wtay> but what if I want an app that has nice play/pause/stop buttons .. [23:38] <omega> yes [23:38] <arik> they you right them yourself wtay and hook it to gstplay ;-) [23:38] <wtay> do I have to reimplement them in my app? [23:38] <omega> ok, summary: [23:38] <omega> there are 2 pieces: [23:39] <omega> libgstplay: libgst-based (only) library to handle the pipeline aspects of a player [23:39] <omega> gstmediaplay: uses libgstplay and gtk/gnome to put a GUI on top of libgstplay [23:39] <omega> no libgstmediaplay, it goes away [23:39] <arik> right [23:39] <wtay> ok... [23:39] <arik> yes [23:39] <arik> and [23:39] <arik> we could make libgstmediaplay a component eventually [23:39] <arik> in a shell [23:40] <wtay> as a component? [23:40] <arik> and then you could just add that to your program [23:40] <omega> bonobo [23:40] <arik> yes [23:40] <wtay> ok, I agree then [23:40] <arik> ok ;-) [23:40] <wtay> but the point is that the component and gstplay would share lots of code [23:40] <omega> not that much... [23:40] <wtay> like the slider and the buttons [23:40] <omega> yeah, about 20 lines <g> [23:41] <arik> no [23:41] <arik> the button and slider would not be in libgstplay [23:41] <wtay> af course not [23:41] <omega> the *component* would have them [23:41] <omega> as would gstmediaplay [23:41] <arik> ah [23:41] <arik> no [23:41] <omega> but it's so little code that there's no need to have a lib [23:41] <arik> there would be no gstmediaplay [23:41] <wtay> omega: 20 lines out of the 30 <g> [23:41] <arik> it would just be a shell with a component [23:41] <omega> huh? [23:41] <omega> oh [23:41] <arik> the program would use the component [23:41] <omega> ok, whatever [23:41] <arik> heh [23:41] <arik> this is how we did it in eog [23:42] <omega> arik: write this up to the list, so people stop getting confused. [23:42] <arik> omega: alright, as soon as i commit i will [23:42] <omega> why? [23:42] <arik> um [23:42] <wtay> ok, that was basically the idea for libgstmediaplay but I mispackaged it.. [23:42] <arik> cause i'm working ;-) [23:42] <omega> this is a long-term terminology/stategy [23:42] <omega> ok [23:42] <arik> heh [23:43] <arik> alright [23:43] <arik> execpt for the arg stuff [23:43] <wtay> btw, the playlist should be a component too [23:43] <arik> wtay: yes [23:43] <arik> i now have no gtk in gstplay [23:44] Action: wtay is going to do a QT wrapper too someday [23:44] <arik> heh [23:44] <arik> why bother ;-P [23:44] <omega> the playlist needs to be a totally separate project, really, because it needs to be done *right* for a change [23:44] <arik> well [23:44] <omega> maybe create a simple one for now, but the component API will have to change drastically later [23:44] Action: wtay starts xfig [23:44] <arik> i mean i think hadess is doing a really good job [23:45] <arik> he just needs to eventually componatize it [23:45] <omega> but is it a component? [23:45] <arik> not yet [23:45] <arik> but it could pretty easily be [23:45] <arik> afaik [23:45] <omega> I'll have to look at it, but there are a *lot* of things that need to be in a good playlist [23:45] <arik> i've played around with it quite a bit [23:45] <omega> xmms just doesn't cut it [23:45] <arik> yeah there are [23:45] <arik> no, no it doesn't [23:45] <arik> hadess seems to be on the right track though [23:45] Action: omega beats taaz with a clue-by-four and tells him to write up his playlist ideas [23:45] <arik> mostly cause he is just copying itunes i think ;-) [23:46] <omega> hrm, not always a great idea [23:46] <omega> qt player 4.0 was one of the worst UI disasters of last several years [23:46] <arik> um [23:46] <arik> yes [23:46] <arik> heh [23:46] <arik> where in the inspect code does it deal with the arg stuff? [23:47] <arik> bascially i have one function in gstplay that needs to be changed [23:48] <arik> hmm [23:48] <arik> g_object_get_property? [23:48] <arik> oh wait [23:48] <arik> i don't need this function at all [23:48] <arik> it's just for printing out on the console [23:48] Action: arik removes it [23:49] <arik> hmm [23:49] <arik> or is it [23:49] Action: arik tests [23:50] <arik> OK! [23:50] <arik> er [23:50] <arik> ok! [23:50] <arik> there are now 0 deps in libgstplay on anything but gst [23:50] <arik> and of course glibc ;-P [23:50] <omega> gotta remove that <g> [23:51] <arik> hehe [23:51] <arik> i'll get right on it! [23:52] <omega> grrr, Uraeus has been ignoring my pleas for confirmation about ALS [23:52] <arik> heh [23:52] <arik> bad! [23:53] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-33.brain.net.pk] [23:53] nash (wa...@13...) left irc: Ping timeout for nash[13-15.brain.net.pk] [23:53] <wtay> http://gstreamer.net/player.png [23:54] <omega> right [23:54] Zeenix (ze...@13...) joined #gstreamer. [23:54] <arik> i can't wait till the dsl is hooked up [23:54] <arik> looks good wtay [23:57] Action: taaz rubs bump on head... playlist what? [23:57] <arik> heh [23:57] <omega> when you first showed up in this channel, you said you had a host of ideas for a really good playlist, and you were gonna write them up Any Day Now(tm) [23:58] <taaz> yeah, situation is still the same [23:58] <arik> hehe [23:58] <omega> great [23:58] <arik> right it! [23:58] <arik> write it! [23:58] <omega> heh [23:58] <arik> do both! [23:58] <omega> cause a playlist is quickly coming onto the radar [23:59] <taaz> freeamp is pretty cool [23:59] <taaz> and there are many dj like playlist makers [23:59] <arik> ick! [23:59] <arik> i _hate_ freeamp's playlist [23:59] <taaz> ? [00:00] --- Fri Oct 26 2001 [00:00] <arik> i hate it [00:00] <arik> it is totally unusable [00:00] <omega> can't be worse than xmms's [00:00] <taaz> i think cplay's playlist is fairly ok for single song purposes [00:00] <arik> i think that really, you should look at rhythmbox omega, hadess is on the right track imho [00:00] <omega> ok [00:00] <taaz> main thing i wanted to add was groups of tracks [00:01] <arik> groups of track? [00:01] <arik> s [00:01] <taaz> so you can insert full albums in the play list [00:01] <omega> right, I want hierarchical playlists [00:01] <arik> hmm [00:01] <arik> talk to hadess ;-) [00:01] <arik> i really am just gonna have to rewrite gstmediaplay [00:01] <arik> it will be easier then merging it [00:01] <taaz> ok, do it [00:02] <arik> yes [00:02] <arik> i am ;-) [00:02] <taaz> add right-click popups... those are cool [00:02] <arik> for what? [00:02] <taaz> whatever ;) [00:02] <arik> haha\ [00:03] <arik> ;-) [00:04] <wtay> and mouse movement hints.. <g> [00:04] <arik> haha [00:04] <arik> ;-) [00:04] <arik> tooltips galore [00:04] <arik> i think emacs promotes carpal tunnel [00:04] <omega> tooltips larger than the whole window, like in E [00:05] <wtay> speech recognition, it should jump to the song I'm humming [00:05] <omega> doh [00:05] <arik> hahaha [00:05] <arik> those are both good features [00:05] <arik> maybe i should just write this with evas *grin* [00:06] <taaz> hmm.. can gstmediaplay actuall grab user events fromt he video window? [00:06] <arik> why not ;-) [00:06] <omega> no, we have to figure out how to do that [00:06] <arik> ah that [00:06] <arik> yes [00:06] <arik> that is still not possible [00:06] <taaz> ok, i was just dreaming [00:06] <arik> really i think that the current xid method of video should eventually be replaced [00:07] <arik> not sure by what though ;-( [00:07] <wtay> I was wondering if the events couldn't be sent to the parent window instead [00:07] <omega> possibly [00:07] <arik> why can't we write a native gtk videosink? [00:07] nash (wa...@13...) joined #gstreamer. [00:08] <omega> arik: because you can't do Xv in a gtk widget [00:08] <omega> at least without writing a whole widget for it [00:08] <arik> hmm [00:08] <wtay> a gtk videosink is just a small wrapper around xvideosink [00:09] <wtay> not unlike the gstplay widget [00:09] <wtay> currently.. [00:09] <arik> right [00:09] <omega> specifically it's a gtksocket that you attach the xid to [00:09] <arik> but the current socket approcah is not a really good idea [00:09] <arik> right [00:09] <wtay> why not? [00:09] <arik> cause it causes all sorts of problems like not being able to get events ;-P [00:10] <arik> it's just a bit hacky [00:10] <arik> anyone remember when nautilus first had openoffice embedded in it? [00:10] <wtay> that's how bonobo components are implemented [00:10] vishnu (jo...@cx...) joined #gstreamer. [00:10] <arik> the way we did that was to run openoffice then get the xid then push it in to a socket in a nautilus window [00:11] <arik> it's just a bit gross (but that's just my opinion, if there isn't any other way i will shut up ;-) [00:11] <wtay> there is nothing wrong with it IMHO [00:11] <arik> heh [00:11] <arik> alright [00:11] <omega> well, except we don't know how to get events passed up still [00:11] Action: arik backs off gracefully [00:12] <wtay> hmm no.. [00:13] <wtay> but that's only because my lack of X knowledge.. [00:13] <arik> heh [00:13] <arik> ask owen [00:13] <arik> or someone [00:13] <arik> in #gnome [00:13] <wtay> shouldn't be hard I guess [00:13] <wtay> yeah [00:13] <wtay> sleep even <g> [00:13] <wtay> cya later [00:13] Nick change: wtay -> wtay-zZz [00:13] <arik> heh [00:13] <arik> later [00:14] <omega> l8r [00:14] <Zeenix> wtay was much in hurry for his bed today [00:19] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [00:20] <Zeenix> omega: the rtcp uses the udp socket, do you think its a good idea to use udp socket for control ? [00:21] <omega> no, because that's not in the RTP spec. if you need a control channel, use rtcp or something [00:22] <Zeenix> i think it is in the RTP spec., can be wrong [00:22] <omega> no, rtp is only for data flow and basic channel setup, not control flow [00:22] <Zeenix> repeating: the rtcp uses the udp socket, do you think its a good idea to use udp socket for control ? [00:22] <arik> ok [00:22] <arik> it compiles [00:22] <arik> now i get to watch it do nothing ;-) [00:23] <omega> oh, rtcp. dunno, does it specify which protocol should be used in the rtcp spec? [00:23] <Zeenix> yes, in the RFCs on the community.roxen.com [00:23] <arik> specifically segfault on startup [00:23] <omega> if it says which protocol to use, then use it. I would assume rtcp is supposed to be used over tcp [00:26] <Zeenix> maybe the librtp makes the socket behave a bit responsible like TCP [00:27] <omega> no idea [00:32] <arik> hmm [00:32] <arik> how did i end up fixing that random core dump problem? [00:33] <omega> ulimit -c 0 [00:37] <vishnu> i've been off-line all day because my dsl modem got unplugged. d'oh! [00:38] <arik> heh [00:39] <arik> ugh [00:39] <arik> why is this segfaulting [00:40] <taaz> how should queue handle eos event? [00:41] <vishnu> taaz: can't the queue ignore eos? [00:41] <omega> no, the queue in general queues both buffers and events, but deals specifically with some events like FLUSH [00:41] <taaz> i'm asking this.. [00:42] <taaz> yeah, flush i can do, but does eos set state or something? [00:42] vektor (reet@HSE-Kitchener-ppp3508492.sympatico.ca) left irc: My damn controlling terminal disappeared! [00:42] <Zeenix> omega: the event system is on ? [00:42] <omega> Zeenix: partially [00:42] <vishnu> taaz: probably just pass it on at the right point [00:42] <omega> it does set state, see the filesrc for an example [00:42] <omega> we're not totally settled on how to do that still, but that's close [00:43] <vishnu> yah, but eos is not an out-of-band for the queue [00:44] <omega> eos and most other events are just dealt with like buffers by the queue, except some have to be dealt with specially on either end [00:44] <taaz> well... i dont want to set the state until it actually gets to other end [00:44] <omega> a FLUSH coming in needs to wipe the queue. an EOS going out needs ... [truncated message content] |
From: IRC B. <wt...@us...> - 2001-10-28 05:27:47
|
******************************************************************* [03:00] <arik> i was using volume [03:00] <arik> trust me [03:00] <wingo> ok :) [03:22] Nick change: tnt -> tnt-out [03:23] wtay-zZz (wi...@ca...) left irc: Read error to wtay-zZz[cable-195-162-215-201.upc.chello.be]: EOF from client [03:25] wtay-zZz (wi...@ca...) joined #gstreamer. [03:34] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [04:18] bstard (Lor...@a2...) joined #gstreamer. [04:20] <arik> ok [04:20] <arik> well i have officially completely fucked up my eazel-hacking [04:20] <arik> ah well [04:23] <arik> dinner [04:23] <arik> back later [04:23] arik (ar...@sd...) left #gstreamer. [04:55] chillywilly (da...@d1...) left irc: Read error to chillywilly[d148.as12.nwbl0.wi.voyager.net]: EOF from client [05:08] chillywilly (da...@d9...) joined #gstreamer. [05:15] ShrimpX (marius@Photon.pdx.netroedge.com) left irc: going home [05:16] <taaz> wingo? [05:18] Nick change: tnt-out -> tnt [05:20] <wingo> still here ;) [05:21] <taaz> done any python gui programming? [05:21] <wingo> nope [05:21] <taaz> ok... trying to decide to do an app in tk or gtk bindings [05:22] <taaz> not gst stuff though ;) [05:22] <wingo> heh [05:22] <wingo> gtk is prettier :-) [05:22] <taaz> my advisor has been insisting on a "flashy" demo for a project we're doing. I'm getting sick of his asking so i'm going to make it ridicously flashy [05:23] <wingo> what's your thesis? [05:24] <taaz> a framework for data-driven run-time reconfigurable applications [05:25] <wingo> what does that mean? (if you have time and/or desire to tell me ;) [05:26] <taaz> do you know what FPGAs are? (ie, Xilinx, Altera, etc projects) [05:27] <dobey> Pin Grid Arrays? [05:28] <taaz> field programmable gate arrays [05:28] <dobey> oh [05:28] <taaz> hardware logic and routing resources in a large mesh. can program it to do whatever you want [05:30] <taaz> so work with this hardware you can reprogram to do different stuff at runtime [05:30] <wingo> ok [05:30] <taaz> for highly regular algorithms that can process alot of data in parallel configureable hardware is a big win [05:31] <taaz> but sometimes an application wants to change around what the hardware is doing at runtime [05:31] <taaz> which is the "run-time reconfigurable application" part [05:32] <taaz> please don't ask me to tell you why this is a good idea... it's more a research thing IMHO ;) [05:33] <taaz> some people think things like configurable routers are a good idea. so like based on the input stream of ip packets you may do different things depending on the destination [05:33] <taaz> (just an example application) [05:33] <taaz> so you have some limited (ie, in theory inexpensive) peice of hardware you can configure to process for one host at a time [05:33] <taaz> or one protocol or whatever [05:34] <taaz> so you need to swap the hardware configuration based on the data [05:34] <taaz> which is the "data driven" part of my stuff [05:35] <taaz> and writing control and high level app stuff for this all kinda sucks and are one-off hardware specific apps [05:35] <wingo> is it fun too? [05:36] <taaz> so i'm playing around with ideas on how to make the process easier... software/hardware codesign, ability to target different hardware within reason, ease of writing the high level stuff so you can concentrate on other stuff, etc etc [05:36] <taaz> well... in theory i'm working on all this [05:36] <taaz> but i've gotten continuously distracted doing other stuff [05:37] <taaz> like getting ready to demo some hardware and apps for it... which should be using my framework but i've been swamped just getting the hardware to function (it's kind of a piece of crap) [05:37] <taaz> yeah, so that's my life [05:41] <taaz> my stuff is fun to work on... it's the other 90% of my time i don't enjoy ;) [05:41] <taaz> and i -still- can't figure out this smp locking! argh! [05:42] <taaz> do recent gdb's produce reliable stacktraces in threaded apps? [05:42] <taaz> i really can't tell... [05:46] <taaz> yeah, i think i'll use pygtk... [05:46] <taaz> so when my advisor yells at me and says to make it C i'll have a chance [05:48] <taaz> plus then i'll get some experiance so i can do a cool jobby job on a python gst app ;) [05:52] Nick change: ShrimpZzZ -> ShrimpX [05:53] <ShrimpX> python rocks [06:07] chillywilly (da...@d9...) left irc: Read error to chillywilly[d98.as15.nwbl0.wi.voyager.net]: EOF from client [06:10] <mef> taaz: you may want to look at the products that a company called "consystant" is building. They have the hardware/softwre codesign stuff down pretty nicely. It is based on work done by Ken Hines and Ross Ortega as part of the PhD thesis work. [06:11] <taaz> we should -really- just rebuild the registry if it's writable and things change... i mean, it detects it, it should do something like fix it [06:16] omega (om...@om...) joined #gstreamer. [06:16] <taaz> mef: i'll look into it [06:17] <taaz> typical... i put in debug printfs around the signal emit funcs and it doesn't lock anymore [06:20] <omega> 150MHz HyperSparc is slow ;-( [06:20] <omega> 2 will be better, but I have to get a SMP kernel built first [06:22] <wingo> that's neat taaz. /me was away for a bit :) programming close to the metal... or silicon... [06:23] <omega> huh? [06:23] <taaz> omega: read the logs later [06:23] <omega> grrr [06:24] <taaz> you snooze you lose! [06:24] <omega> I was rebuilding my gateway so it won't keep crashing on me [06:24] <omega> I built a naked server this morning, then the case arrived while I was gone [06:25] <omega> trashed K6-2/100 to Duron/750, 32MB to 256MB [06:25] <omega> new disk within a day or 4 [06:26] <wingo> so, wrt default events implementations... [06:26] <wingo> i've added events to some chain and loop elements [06:26] <omega> ok [06:26] <wingo> loop can't have defaults, right [06:27] <omega> defaults? [06:27] <wingo> default implementations [06:27] <omega> of what? [06:27] <wingo> event handling routines, er, /me thinks [06:27] <wingo> hum. [06:27] <omega> there are no defaults yet anyway, still working on that [06:27] ShrimpX (marius@131.252.244.168) left irc: brb [06:27] <wingo> right, that's what i was thinking about [06:28] <taaz> i should start a branch and fix the event system to my ideas ;) [06:28] <wingo> let me commit what i have to show you (it works pretty well, and the bonus is a fixed segfault in alsasink ;) [06:28] <omega> taaz: if you want to, sure [06:28] <wingo> just on stereo2mono, speed, alsa [06:28] <omega> there we go, logs [06:28] <wingo> delete, delete! [06:29] <omega> er logs don't go far enough [06:29] vektor_ (ve...@cr...) joined #gstreamer. [06:29] <vektor_> yoyoyo. [06:30] <wingo> yo [06:30] Last message repeated 1 time(s). [06:30] <omega> yo [06:30] <omega> . [06:31] <wingo> omega: is that the proper event handling in stereo2mono? the log mail should arrive any minute now [06:32] dobey (do...@cv...) left #gstreamer (eh). [06:33] <omega> sorta. the defaults system will probably be different from that, but that's sorta the idea [06:33] <omega> though for filters like that, really, you shouldn't have to change them at all [06:33] <omega> because we need to have a flag for a pad or element that says "handles events" [06:33] <wingo> ah [06:33] <omega> if not set (default), they'll be dealt with elsewhere [06:34] <vishnu> does vcdsrc work? [06:35] <omega> it did. haven't touched it for a year++ [06:35] <wingo> dunno, i just fixed a typo in it the other day [06:35] <omega> I have one vcd hidden somewhere in my room.... [06:35] <vishnu> hrm [06:36] <wingo> i have a question about autoplug_to_caps. [06:36] <omega> it's designed only to work on VCDs that bypass the usual ISO stuff, i.e. have 2352-byte sectors [06:36] <omega> wingo: wtay... I dunno anything about that <g> [06:36] <wingo> damn ;) [06:37] <wingo> so, also about dynamic pipelines. know anything about that? [06:37] <omega> um, yeah, I wrote the only one in use (autoplugger) [06:37] <wingo> make a pipeline -> run it -> pause it -> change it a little bit -> run it again [06:37] <wingo> that works? [06:38] <omega> it's supposed to [06:38] <wingo> i'll do some more testing [06:38] <omega> but right now each person to do that exposes some further bug. which is good, I expect more bugs [06:38] <omega> but the autoplugger still work afaik [06:39] <omega> er, maybe not [06:39] <omega> oh, wait, something is holding /dev/dsp ;-( [06:40] <wingo> filesrc ! autoplugger ! osssink ? [06:40] <omega> yeah [06:40] <taaz> anyone have some spare cycles I could borrow? [06:40] <omega> um, something broken [06:40] <omega> taaz: to do what [06:40] <omega> ? [06:40] <taaz> run a pipeline in gdb for a long time [06:40] <omega> heh [06:40] <omega> you don't have cycles? [06:40] <taaz> to see if it dies on non-smp [06:40] <omega> oh [06:40] <omega> um. what does it use? [06:41] <taaz> fakesrc, fakesink, queue, stats [06:41] <omega> I can probably do that, as soon as I build gst on my athlon [06:41] <taaz> mine just died at ~91700 buffers [06:41] <taaz> err..locked [06:42] <omega> grabbing cvs now [06:42] mef (me...@us...) left #gstreamer. [06:42] mef (me...@us...) joined #gstreamer. [06:42] <mef> howzit? [06:43] <mef> made little progress with getting autogen.sh to work on cygwin. [06:43] Action: omega has a new gateway finally [06:43] <wingo> yo [06:43] <mef> Maybe I should read how autoconf, automake and all that stuff really works. [06:43] <mef> Also, getting glib and gtk for windoze is non-trivial. [06:43] <omega> heh [06:44] <mef> But I still think it would be neat to see gst working on W2K. [06:44] <omega> yup [06:44] <omega> I can't install w2k in vmware because I didn't spend $300 on it. [06:44] <mef> As I mentioned to "tnt" yesterday, an XBOX is basically a W2K kernel with DirectX 8. [06:44] <omega> vmware express is NOT worth $50 IMO [06:46] <mef> right now aclocal is barfing, becuase it cannot find AM_DISABLE_STATIC, AM_PROG_LIBTOOL, AM_PATH_GLIB, and AM_PATH_ESD. [06:46] <wingo> maybe each elementfactory should have some sort of flag, "can autoplug" or something [06:46] <wingo> i ust want to prevent ladspa plugins from being tried [06:46] <omega> wingo: huh? autoplug is a feature of the caps [06:47] <omega> it won't be tried because it doesn't convert caps [06:47] <wingo> it's tried [06:47] <omega> then the caps are wrong [06:47] <wingo> and an O(N^2) at least operation [06:47] <omega> tried for what? [06:47] <wingo> cost, something. let me check it out again [06:47] <omega> mef: need to check everything related to what would be /usr/share/aclocal [06:48] <wingo> actually, i have to port my app to gnome 2 to get that to work [06:48] <omega> taaz: pipeline? [06:48] <wingo> i'm in the middle of the process [06:49] <taaz> omega: you should probably apply a patch i'm using [06:49] <omega> ok, send it and the main() [06:50] <mef> omega: autogen states that I should add the contents of /usr/local/share/aclocal/libtool.m4 to aclocal.m4. What does that really mean? [06:50] <omega> ignore it, that's what aclocal is supposed to do [06:50] <mef> hmph. [06:50] <taaz> omega: are you using gtk1.2? [06:50] <omega> but aclocal is looking in one path, and your libtool files are installed in a different one [06:50] <omega> taaz: for this build, yes [06:50] <omega> I can get glib2 if necessary [06:50] <taaz> no, it locks in gtk for me... [06:51] <omega> is it a -launch or a separate program? [06:52] <taaz> launch [06:52] <taaz> but run it in gdb [06:53] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [06:53] <taaz> oops.. patch relative to ./gst/ [06:54] <omega> ok, so I can immediately see that the enter()/leave() around gtk_main() is fatal for any gtk app [06:54] <taaz> fatal? [06:54] <taaz> i added that to try and fix things ;) [06:55] <omega> yes, the first time the app does a gdk_thread_enter(), it will lock solid [06:55] <mef> omega: I added a -I /usr/local/share/aclocal and now all it is looking for is AM_PATH_GLIB and AM_PATH_ESD, which is fine as I haven't properly installed them yet. [06:55] <omega> mef: cool [06:55] <taaz> ok, well, just take that out then.. doesn't matter for this test i guess [06:55] <mef> omega: you are right. [06:55] <mef> gurk [06:56] <mef> whoops... [06:56] <taaz> omega: after it's built, fire up -launch in gdb and: [06:56] <omega> you need a log? [06:56] <taaz> r fakesrc silent=true ! queue max_level=1 ! {statistics [S] buffer_update_freq=10 ! fakesink silent=true} --gst-mask=0x300 [06:56] <taaz> noo... the log could be very very long [06:56] <omega> oh, I'm gonna do this on the console of that machine [06:56] <omega> rather than over the wireless network <g> [06:57] <omega> I have an 80gb drive ;-) [06:57] <taaz> just let me know if it locks. maybe hit ^c after a while and see what the last stats buffer count is [06:58] vishnu (jo...@cx...) joined #gstreamer. [06:58] <taaz> you could use screen, just switch to another console while it's spewing, doesn't use any bandwidth that way [06:58] <taaz> you're welcome to log it... [06:58] <omega> on an athlon 1.4, how long would you expect this to run before locking, if it's gonna lock? [06:59] <vishnu> vcdsrc works fine [06:59] <taaz> i have no idea [06:59] <wingo> that's good [06:59] <taaz> i've never had it work for more than a few minutes [06:59] <vishnu> sweet [07:00] <wingo> taaz: vcdsrc or the thread thing? [07:00] <taaz> thread thing [07:00] <omega> vishnu: that's based on code I hacked up on the plane back from DC one time [07:00] <vishnu> omega: good job :-) [07:00] <omega> taaz: still going, it pauses for a tiny bit every few seconds though [07:00] <taaz> pauses? [07:01] <omega> maybe 1/20th sec, just enough to be barely visible as stuff flies by [07:01] <omega> it's not regular at all [07:01] <taaz> i have no idea... maybe just some stdio buffer flushing or something [07:01] <vishnu> ah ha! [07:02] Last message repeated 1 time(s). [07:02] <wingo> eh? [07:02] <omega> taaz: could be [07:03] <taaz> omega: if it's still running you could check if there are any mem leaks i guess ;) fun with top [07:03] <vishnu> i'm an idiot [07:03] <omega> taaz: yeah, leaking, probably in the debug stuff [07:03] <omega> not fast [07:03] <omega> 18M [07:03] <omega> 18.5M [07:05] <omega> taaz: threads are sorta balanced, between 1:2 and 1:1, varying [07:05] <omega> keeps going, and going, and going [07:06] <wingo> vishnu: how did you get that mpeg1 file? [07:06] <vishnu> wingo: i ripped a vcd. [07:06] <taaz> ok, i disabled the emits... lets see how far it gets [07:06] <omega> doh [07:06] <vishnu> wingo: oh, the bad one? from my brother :-/ [07:07] <vishnu> my brother isn't a genius yet :-/ [07:07] <wingo> heh [07:13] Action: wingo builds libgnomeui-2.0 [07:14] <omega> taaz: I'm gonna kill it and see how far it got [07:15] <omega> 387630 [07:15] <wingo> kill the pig! [07:16] <taaz> my latest run is at 99k and no problems yet [07:16] <taaz> so i blame gst_signal_emit... now, what to do about this... [07:17] <taaz> gtk_signal_emit i mean [07:18] <taaz> maybe we need a gst_ varient that does locking or something :( [07:18] <wingo> this is no time for the blame game, taaz ;-) [07:20] <taaz> ok, running it mask=0 is working fine... that usally died in like 5k buffers [07:20] vektor_ (ve...@cr...) left irc: Read error to vektor_[cr288451-a.flfrd1.on.wave.home.com]: Connection reset by peer [07:20] <taaz> i'm >1M now [07:21] <taaz> increasing the stats printout will make it fly too... [07:21] <taaz> printout freq that is [07:21] <taaz> i can't figure out from looking at gtk 1.2 code why this is locking [07:23] <taaz> i get 13k buf/s with infrequent dbg prints btw [07:23] <taaz> 13.3k [07:25] <omega> about 34k with 10k frequency [07:25] <omega> /s [07:26] <omega> HAH! [07:26] <omega> this is funny [07:26] <omega> reading Tom's Hardware Guide on Flask 4.01 [07:26] <omega> er, 4.1 [07:27] <taaz> ? [07:27] <omega> "In the past, most movies were way to quiet after encodinng. This [volume control] feature helps increase the volume by 180%" [07:27] <omega> we all know why that was a problem, don't we.... [07:27] <vishnu> heh [07:27] <omega> cause ac3dec had a bug, and that bug got propagated all the way through, and *still* exists in Flask [07:28] <vishnu> d'oh [07:28] <omega> so you increase volume by 180% and lose 1 to 1.5 bits of useful audio resolution [07:28] <taaz> and probably clip too [07:29] <omega> of course [07:29] <vishnu> i hate clipping [07:29] <omega> and flask 4.1 lost the deinterlacing capability [07:29] <omega> which vektor will have a cow about [07:29] <taaz> where do gtk people hang on on irc? [07:30] <omega> #gnome [07:30] <taaz> not on this server i take it... [07:30] <omega> no [07:30] <omega> irc.gnome.org [07:31] <vishnu> oh! [07:31] <omega> 2 million cycles with no debug, no lock [07:31] <omega> oops, 2.5 [07:31] <omega> er, 3 [07:31] <taaz> 8.5M here [07:32] <taaz> without emit [07:32] <taaz> pesky emit [07:32] <omega> 4 [07:32] Action: omega rapidly catches up to taaz [07:32] <mef> after some sifting it seems that people who know a lot more about glib and gtk are attempting to build a cygwin version of them. [07:32] <omega> heh [07:32] <omega> 5 [07:32] <omega> code problems, or build problems? [07:33] <mef> I am likely to hang low on that front, as there is no need for duplicate efforts. Particularly, given that this is so new to me and I can't fully wrap my head around those details. [07:33] <omega> 7 [07:33] <omega> taaz: you're gonna lose <g> [07:33] <mef> So... I am going to move on to my next pet project: has anyone used gst on a RIO digital audio receiver (aka RIO HOME)? [07:33] <omega> *on* ? [07:35] <omega> 10M ;-) [07:35] <mef> RIO HOME has an ARM processor and runs Linux. So has someone used gstreamer as an audio player on that device? [07:35] <omega> nope [07:35] <omega> I know gst works on an ARM though [07:35] <mef> but people have used it on ARM/Linux based systems, right?! [07:35] <mef> oh... ok [07:35] <omega> but do you know if it uses offboard mp3 decoding or anything? [07:36] <mef> it does not [07:36] <omega> hmm, which chip does it use and at what speed? [07:36] <mef> someone got the MAD mp3 decoder to work on it. [07:36] <omega> ok [07:36] <omega> vorbis is a lot heavier than mp3 right now, which sucks ;( [07:36] <mef> but I have no idea how good it sounds... it is a 80 or 90Mhz ARM proc. [07:36] <omega> one of these days I'm gonna look into using libcodec stuff for it.. [07:37] <omega> 80 or 90?? way to slow for even MAD, at least for a StrongARM [07:37] <omega> need at least 120-130MHz to play a normal mp3 [07:37] <mef> I see... [07:37] <omega> with mad [07:37] <omega> I know, I've timed it <g> [07:37] <taaz> 13.5M [07:37] <mef> another silly question: people are using gstreamer to stream audio over the net, right? [07:37] <omega> 16M [07:38] <omega> yeah, icecast stuff [07:38] <mef> Is there also a gstreamer based client? [07:38] <omega> see benow.ca, it runs gstreamer 24x7 these days [07:38] <omega> well, there aren't any apps yet, so no ;-) [07:39] <omega> oooh, and benow.ca's got a new DJ live right now too [07:39] <omega> new to the stream at least [07:39] <mef> my thought is to use gstreamer on the PC to translate MP3, Vorbis, etc. into PCM and ship that over the LAN to the RIO HOME. [07:40] <mef> Let the PC do the lifting! [07:40] <omega> yeah [07:40] <omega> I've got a siteplayer[.com] that I want to see if it's powerful enough to handle UDP->SPI to a DAC at full rate [07:40] <taaz> i want an iPod... wonder if it works in linux. would be cool if it just worked as a normal firewire hd. [07:40] <omega> reminds me, I should go ask... [07:41] <mef> And... if one could get gstreamer to work on a Windoze box, then presumably one could then build a WMA based plugin and also listen to that format. [07:41] <ajmitch> hmm, anyone got gstreamer working on windows yet? ;) [07:42] Action: ajmitch is thinking of having to use windows, but using gstreamer stuff [07:42] <omega> mef: you mean convert it to another format <g> [07:42] <mef> omega: PCM would be fine. Shouldn't saturate a HPNA link. [07:42] <omega> yeah, 1.5Mbps [07:42] <ajmitch> omega: what would be needed for windows? i'd want to play videos, so might use SDL [07:43] <mef> Sorry... I meant HPNA 2.0 -- closer to 6Mbps [07:43] Action: ajmitch is working with some ppl to throw together a multimedia cd for playing at expos, etc [07:43] <mef> amazingly enough the PCI cards for this cost $20 and works like a charm in my house. [07:44] <omega> wow, they finally got the price down to a reasonable level. good. [07:44] <omega> you have an intel one? [07:44] <mef> ajmitch: I started to work on compiling gstreamer under cygwin. [07:45] <mef> ajmitch: there are others out there working on getting glib and gtk+ to work under cygwin, which will make that part of life easier. [07:46] <ajmitch> fun [07:46] <taaz> so the very little info i get in #gnome is that gtk is not thread safe [07:46] <mef> ajmitch: I am going to wait until early next week to see what kind of progress they have made, but I wouldn't mind collaborating with you or anyone to get gst working on windoze. [07:46] <taaz> that's just great [07:46] <taaz> suggestions? [07:47] <mef> omega: I got it from netgear.com store's. The PCI card is now $21, but was $19 and the USB version is $30. However, NO LINUX DRIVERS! [07:47] <omega> for the USB one? [07:47] <ajmitch> i really really don't want to use windows, but being realistic it's the only way (i'll of course have linux binaries on the cd ;) [07:47] <wingo> any way of checking whether gst was built under glib2 or glib1.2? [07:48] <mef> omega: all [07:48] <wingo> for configure.ac checks [07:48] <omega> mef: I haven't looked, but the Intel-based (21145) chips likely could be supported by now [07:48] <mef> anyone know of a HPNA 2 PCI card that works under linux? [07:48] <mef> Clearly the RIO people have one, but I don't see them releasing their work. [07:48] <omega> I worked on hacking the tulip driver to work, and got them to work, but only with some nasty stuff to force it to always use the hpna port [07:49] <wingo> taaz: that's irritating. is gobject thread safe? [07:49] <omega> dunno if the 21145 does hpna2 these days [07:49] <omega> wingo: that's the same thing [07:49] <omega> signals are a gobject thing [07:49] <omega> unless you asked about gtk1.2 and not gobject2.0, taaz? [07:49] <mef> hmmm [07:50] <wingo> that' [07:50] <wingo> s what i was referring to [07:50] <wingo> so, any idea on 1.2/2.0 linkage detection for gst at run time? [07:50] <wingo> i could do gross ldd | grep hacks [07:50] <taaz> i asked about gtk1.2 [07:50] <omega> wingo: huh? [07:51] <omega> taaz: gtk1.2 is irrelevant in ~3mo, ask about gobject [07:51] <taaz> i can't try gobject until the debs come out that are recent enough for gst to compile ;) [07:51] <wingo> how do i know whether gst was compiled with gtk1.2 or glib2.0? [07:51] <omega> taaz: pff, you heard of 'tar'? [07:51] <taaz> omega: nope [07:51] <wingo> me neither [07:51] <omega> goes with 'feathers' in situations like this [07:51] <wingo> heh [07:52] <vishnu> hehe [07:52] <wingo> let me try again: how do i know whether gst was compiled with gtk1.2 or glib2.0? ldd | grep | cut hacks? [07:52] <taaz> wingo; why does it matter? [07:52] <vishnu> wingo: yah, ldd should work [07:52] <omega> wingo: when do you need to know this? at gst-src compile time? plugin compile time? app compile time? [07:53] <wingo> app compile time [07:53] <omega> gstreamer-config and gstreamer.pc have -DUSE_GLIB2 [07:53] <wingo> i need glib2 for a gnome2 app (no?) [07:53] <omega> and both need to be hacked to give out the proper dependencies in that case [07:53] <omega> we can/should change USE_GLIB2 to GST_USE_GLIB2 probably [07:54] Action: ajmitch wonders if he shoudl try building a deb of glib2 from cvs ;) [07:54] <omega> ajmitch: only if you want taaz to finally get to work on glib2 sometime next calendar year [07:54] <ajmitch> hehe [07:54] Action: omega hands ajmitch more RAM [07:54] Action: ajmitch only has a paltry 384MB [07:54] <omega> pff [07:54] Action: omega hands ajmitch a better proc [07:55] <ajmitch> that would help ;) [07:55] <omega> what proc do you have again? [07:55] <ajmitch> 400MHz k6-2 [07:55] <omega> ouch. better than psi though [07:55] <omega> psi has the hammerfall [07:56] <wingo> i have a 300 mhz k6-2 [07:56] <wingo> i beat you ajmitch ;) [07:56] <omega> ow [07:56] Action: ajmitch bets that this deb will *not* build [07:56] <wingo> gst builds are s l o w [07:57] <ajmitch> newest file in the debian/ dir is nov 18 1998 ;) [07:57] <vishnu> mpeg1 passes my random seek test :-) [07:57] <wingo> ow! [07:57] <wingo> heh [07:57] <omega> you can get a k7s5a (good board) with a Durno 800 + HSF for $95 [07:58] <omega> er, Duron [07:58] <wingo> no, i want a durno [07:58] <omega> that like errno? [07:58] <vishnu> what's a durno? [07:58] <wingo> dur! [07:58] <omega> no! [07:58] Action: ajmitch will (hopefully) get a new system soon [07:59] <ajmitch> what decent graphics card can do tv out under linux tho? [07:59] <omega> pff, who knows... [07:59] <vishnu> i think my older brother has a durno system [07:59] <taaz> omega: does most of that patch look ok? a few debug things need to come out but most of it makes things cleaner i think [07:59] <omega> lemme scan it [07:59] <taaz> more work needs to be done in some spots [08:00] <taaz> like events in leaky queue situation [08:00] <taaz> and the high/low watermark signals are not even used [08:00] <taaz> or implemented [08:00] <omega> they will be used in many circumstances [08:01] <vishnu> agreed. any video player should really buffer frames [08:02] <omega> and high/low watermark events are critical for QoS handling and rebuffering if necessary [08:02] Action: vishnu hears the magic QoS word [08:02] <taaz> vishnu: not always. embedded procs may have known behavior and not need to buffer like a general purpose box [08:03] <omega> may, but almost never does [08:03] <taaz> anyway, yeah, it's just another TODO item ;) [08:03] <vishnu> yah .. i got tunnel vision .. i know :-) [08:03] <taaz> right now i'm more concerned with non-locking gtk solution [08:08] <taaz> ok so now i'm supposed to look at CamelObject in Evolution cause they had to rewrite GtkObject or something [08:08] <omega> doh [08:09] <vishnu> taaz: just backport glib 2.0, no? [08:09] Action: omega is having all sorts of neat ideas for a DJ app [08:09] <taaz> 02:02AM <fejj> taaz: a global lock around signal emits won't help [08:09] <taaz> 02:02AM <fejj> because gtk might emit it's own signals [08:09] <taaz> who's bright idea was it to use gtk? [08:09] <taaz> <g> [08:10] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [08:11] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [08:14] arik (ar...@sd...) joined #gstreamer. [08:14] <arik> hi [08:14] <wingo> taaz: is gobject threadsafe? [08:14] <wingo> hey [08:14] <arik> hey [08:14] <ajmitch> hey arik [08:15] <taaz> wingo: i dunno... no one answered in #gnome :( [08:16] <wingo> muahaha. /me steals work from the gimp :-) [08:16] <arik> omega: damn this is complicated building [08:16] <omega> heh [08:19] <arik> oh and i've utterly failed at my first attempt to get gnome2 [08:21] <taaz> omega: your sparc running yet? [08:21] <omega> yeah, but UP [08:21] <omega> can't switch to new kernel unless I'm there, which will be Sun. morning [08:21] <taaz> oh.. ok [08:22] <taaz> i want you to have SMP problems too so you want to fix them ;) [08:22] <taaz> this kinda sucks really really bad [08:23] <taaz> really really [08:23] <taaz> $ wc camel-object.* [08:23] <taaz> 1159 3287 29840 camel-object.c [08:23] <taaz> 147 530 5240 camel-object.h [08:23] <taaz> 1306 3817 35080 total [08:23] <omega> ooh, have fun [08:23] <arik> hahahahahaha [08:23] <taaz> i can't figure that crap out [08:24] <arik> talk to fejj [08:24] <arik> or danw [08:24] <taaz> yeah fejj pointed me to it [08:24] <omega> oh, um, 110M [08:24] <arik> wc -l [08:24] <taaz> i'm really not motivated to fix this. i just assumed someone had thought of this sort of thing before now [08:24] <omega> 33915.85/sec [08:25] <arik> what are you talking about? [08:25] <taaz> gtk is not thread safe [08:25] <taaz> oh, omega running a test for me [08:25] <arik> yes [08:25] <arik> gtk is never going to be thread safe [08:25] <taaz> i get about 2k buffers on an SMP machine before it locks up in gtk. he has 110M [08:25] <arik> not till 3 or something [08:26] <taaz> is gobject threadsafe? [08:26] vektor_ (ve...@cr...) joined #gstreamer. [08:26] <arik> vektor_: hi [08:27] mwc (malcolm@138.80.48.20) joined #gstreamer. [08:27] <taaz> no one knows if gobject is threadsafe? [08:28] <omega> most of it is, I know, but I dunno about signals [08:28] <omega> but that seems to be one of timj's major focii (focuses?) [08:28] <arik> heh [08:29] <arik> glib will be threadsafe if tim wants it to be [08:29] <arik> but afaik owen and havoc said gtk wouldn't be till 3.0 [08:31] <taaz> well, i can't think of any workarounds right now. this CamelObject thing is a tad too complex to grok at 2:30am. [08:32] <arik> yes [08:32] <taaz> at the moment i believe we are just plain fucked. excuse my french. [08:32] <arik> heh [08:33] <omega> taaz: 57760/sec without threads, just fakesrc ! stats ! fakesink [08:34] <taaz> oh jeez... hang on [08:34] <taaz> that 13.3k thing was with a queue max of 1 ;) [08:34] <omega> right [08:35] <taaz> hmm... it slows down with 100? [08:35] <omega> neat [08:36] <taaz> we should look into lock-free fifo algorithms [08:37] <omega> wtay has one, in gstmemchunk [08:37] <omega> er, it's a lifo [08:37] <omega> but probably can make a fifo easily enough [08:37] <omega> problem is there's no CAS primitive that's portable [08:38] <omega> queue of 100 is faster for me (33K vs 30K) [08:38] <taaz> what's CAS? [08:38] <omega> compare-and-swap [08:38] <taaz> ah [08:38] <omega> has to be an atomic asm instruction [08:38] <omega> probably better to call it CaS [08:40] <taaz> btw, -guilaunch is great for this pipeline, adjust params at runtime is fun [08:40] <omega> yeah [08:40] vektor_ (ve...@cr...) left irc: Read error to vektor_[cr288451-a.flfrd1.on.wave.home.com]: Connection reset by peer [08:40] <omega> though I tried it with lame, and it doesn't work [08:41] <omega> because lame has to be set up with params up front, and hacking it to reset them just frags lame [08:41] <omega> API flaws suck [08:44] <wingo> i want guilaunch for gtk2 ;) [08:44] <wingo> i tried earlier today [08:44] <wingo> but gtktree is 'deprecated' now [08:45] <wingo> so a no-go [08:45] <arik> heh [08:45] <arik> just switch to ctree [08:46] <wingo> i have to switch for my own app too [08:46] <wingo> aaa! i just committed some crap code [08:47] <wingo> damn [08:47] <omega> oops, don't do that [08:47] <omega> doh. you don't know JACK! <g> [08:47] <omega> or do you? [08:47] <arik> gtktree sucks [08:47] <arik> ctree sucks [08:47] <arik> etable prob sucks in the end [08:47] <arik> the new tree prob sucks [08:47] <wingo> i do ;) [08:47] <arik> tree's seem to be very hard to get right [08:48] <wingo> i have the plugin half-made [08:48] <wingo> compiles though [08:48] <wingo> i guess i'll take out the cruft that got added... [08:53] <wingo> sleep well, folks... [08:53] Nick change: wingo -> wingo-zzz [09:00] <ajmitch> so quiet now... [09:00] <arik> yes [09:01] <arik> LAJDLKSDJLKSJSLKDJASLKJDLASDJASL [09:01] <arik> does that help? [09:01] <ajmitch> nope :) [09:01] <arik> hehe [09:02] Action: ajmitch needs to actually do some coding after exams [09:02] <arik> hehe [09:02] <arik> that would be [09:02] <arik> new ;-) [09:02] <ajmitch> :P [09:02] <arik> and unsusual behavior ;-P [09:02] <arik> i need to um [09:02] <arik> get the new gstplay into cvs for freaks sake [09:03] <ajmitch> i haven't exaclty done any coding lately ;) [09:03] <arik> hehe [09:03] <arik> yes [09:06] ChiefHighwater (fl...@su...) left irc: Leaving [09:26] <arik> dum de dum [09:26] <arik> if i were any more bored, i would be dead [09:27] <ajmitch> hehe [09:27] <ajmitch> why are you bored? i thought you'd be hard at work getting gstplay into cvs [09:27] <arik> heh [09:27] <arik> i should be [09:27] <arik> i really should be [09:27] <arik> but [09:28] <arik> i'm not ;-) [09:28] <arik> the current gstmediaplay does very very little [09:28] <arik> the new gstplay lib is nice though [09:28] <ajmitch> k [09:28] <ajmitch> then you shoudl be working on the new gstmediaplay ;) [09:28] <arik> hehe [09:28] <arik> i would [09:29] <arik> but building gnome2 [09:29] <arik> is not a fun idea right now [09:30] <ajmitch> hehe [09:30] <ajmitch> having problems, or is it just slooow? [09:30] <arik> um [09:30] <arik> well [09:30] <arik> i want to use eazel-hacking to do it [09:30] <arik> and my eazel-hacking is now very very broken [09:30] <arik> for somewhat complicated reasons [09:30] <ajmitch> why eazel-hacking? [09:31] <arik> cause it makes everything much much easier [09:31] <ajmitch> v-b-s works for most normal people ;) [09:31] <arik> and i got addicted to it [09:31] <arik> at eazel [09:31] Nick change: taaz -> taazzzz [09:31] <arik> v-b-s? [09:31] <ajmitch> vicious-build-scripts [09:31] <arik> oh [09:31] <arik> ew [09:31] <arik> ;-) [09:31] <ajmitch> seems to be the accepted way to compile easily ;) [09:31] <arik> heh [09:31] <arik> george is good [09:31] <ajmitch> yeah [09:31] <arik> but ramiro is beter ;-P [09:31] Nick change: mef -> mef-zZz [09:32] <ajmitch> george seems weird ;) [09:32] <arik> hehe [09:32] <arik> george is very very weird [09:32] <arik> fun drinking buddy though [09:32] <ajmitch> haha [09:33] Action: ajmitch compiles something far more important than gnome [09:33] <arik> oh? [09:34] <ajmitch> freeciv ;) [09:34] <arik> hah [09:34] <arik> i never have tried that [09:35] <ajmitch> fun game [09:35] <arik> heh [09:35] <arik> i don't really get into computer games [09:36] <ajmitch> it's about the only one i play [09:36] <arik> ah [09:37] <ajmitch> just something to do when bored & can't be bothered thinking ;) [09:37] <arik> hehe [10:03] omega (om...@om...) left irc: sleep [10:25] steveb (st...@no...) joined #gstreamer. [10:26] <ajmitch> hi steveb [10:26] Nick change: tnt -> tnt-zZz [10:27] arik (ar...@sd...) left #gstreamer. [10:32] <steveb> hi [10:41] <steveb> ajmitch: are there any usable free ISPs in NZ ATM? [10:41] <ajmitch> hmm, usable... [10:41] <ajmitch> not really, afaik [10:41] <ajmitch> telecom decided to be nasty with interconnection agreements or something [10:42] <ajmitch> so zfree stoppped accepting new customers, i4free shutdown... [10:45] <steveb> bummer [10:46] <ajmitch> yeah, i know [10:46] <ajmitch> it's why i'm using ihug right now [10:49] <steveb> i guess i'll have to blag internet accounts off my relatives [10:50] <ajmitch> hehe [11:37] mwc (malcolm@138.80.48.20) left irc: Ping timeout for mwc[138.80.48.20] [14:20] <steveb> wtay-zZz: awake? [14:20] Nick change: wtay-zZz -> wtay [14:20] <wtay> yeah [14:21] <steveb> i think afsrc (audiofile) can be changed to read from a sink pad instead of a file... [14:21] <steveb> since audiofile has a 'virtual' file interface as well as a real file one [14:22] <wtay> oh, cool [14:22] <wtay> go for it I would say [14:23] <steveb> the only thing is, audiofile needs a length callback, so it needs to now how many bytes will be coming from the sink pad [14:23] <wtay> oh [14:23] <steveb> and the easiest way i could think of was a length caps prop on filesrc [14:24] <wtay> I would say an event would be better.. [14:24] <steveb> yeah, or a SEEK_END event which returns the file length [14:24] <wtay> EVENT_SIZE_ANNOUNCE or something.. [14:25] <steveb> so is that 2 events? one to ask for the size and another to announce it? [14:26] <steveb> or is it just announced as soon as its discovered [14:26] <wtay> steveb: yes [14:26] <wtay> 1 event [14:26] <steveb> and could other events announce how many frames / samples they have if they know? [14:27] <wtay> yup, other fields in the size announce event I would think [14:32] <steveb> that would make a 'reverse' element feasible ;) [14:32] <wtay> :) [15:14] MisterP (pk...@ic...) joined #gstreamer. [15:15] <MisterP> what's the URL for gstreamer? [15:16] MisterP (pk...@ic...) left #gstreamer. [15:16] <steveb> gstreamer.net [15:16] <steveb> too slow :( [15:43] Uraeus (csc...@c1...) joined #gstreamer. [15:43] <Uraeus> heyho [15:43] <wtay> howdy ho [15:44] <Uraeus> wtay: did Arik make any more progress on his GNOME2 port of gstmediaplay last night? :) [15:44] <wtay> dunno [15:44] <steveb> isn [15:44] <steveb> isn't it technically a gtk2 port? [15:45] <Uraeus> steveb: AFAIK he was planing on doing a port to gnome-libs2 also [15:54] <Uraeus> wtay: any more news from our German friends after you where there talking to them? [15:54] <wtay> nope [16:49] Nick change: taazzzz -> taaz [17:03] Nick change: wtay -> wtay-bath [17:22] Nick change: taaz -> taaz-football [17:46] vishnu (jo...@cx...) joined #gstreamer. [17:48] <Uraeus> hi vishnu, lots of patches submitted today I see :) [17:49] Nick change: Uraeus -> Ura_away [17:50] <vishnu> Uraeus: yes, a few, i guess [18:16] Jacmet (ja...@bo...) joined #gstreamer. [18:36] Jacmet (ja...@bo...) left irc: Ping timeout for Jacmet[bones.kulnet.kuleuven.ac.be] [19:02] Ura_away (csc...@c1...) left #gstreamer (Client Exiting). [19:44] Nick change: tnt-zZz -> tnt [19:59] Shippou (no...@cr...) joined #gstreamer. [20:00] Nick change: tnt -> tnt-out [20:24] Nick change: mef-zZz -> mef [20:25] Nick change: wingo-zzz -> wingo [20:25] mef (me...@us...) left #gstreamer. [20:25] <wingo> yawn! [20:25] <vishnu> yawn [20:26] <wingo> yo. [20:29] <ajmitch> morning ppl :) [20:30] <wingo> morning, kiwi ;) [20:31] <ajmitch> and an early morning it is too... [20:31] <wingo> that's not pejorative at all, is it? the word kiwi [20:31] Action: ajmitch can't believe he's up at 7:30am on a sunday [20:31] <ajmitch> nope. kiwi is fine [20:31] <wingo> sunday?!? [20:31] <wingo> heh. [20:31] <ajmitch> i got an exam tomorrow that i really really need to study for ;) [20:32] <wtay-bath> yo [20:32] Nick change: wtay-bath -> wtay [20:32] <wtay> gotta go again :) [20:33] Nick change: wtay -> wtay-snooker [20:33] <ajmitch> hehe [21:07] rambokid (tj...@hm...) left irc: xchat exiting.. [21:07] Nick change: wingo -> wingo-away [21:07] omega (om...@om...) joined #gstreamer. [21:07] <steveb> yo [21:08] <omega> yo [21:08] <ajmitch> yo [21:09] <ajmitch> bbl [21:14] <steveb> omega: wtay-snooker and i discussed creating a new event (EVENT_SIZE_ANNOUNCE) which tells downstream elements how many bytes/samples/frames they have to process [21:14] <omega> I was thinking this would be part of the caps [21:14] <steveb> wtay prefers an event [21:14] <omega> cause it's data-specific [21:14] <omega> why? [21:15] <steveb> 14:24<steveb> the only thing is, audiofile needs a length callback, so it needs to now how many bytes will be coming from the sink pad [21:15] <steveb> 14:24<wtay> oh [21:15] <steveb> 14:24<steveb> and the easiest way i could think of was a length caps prop on filesrc [21:15] <steveb> 14:25<wtay> I would say an event would be better.. [21:15] <steveb> 14:25<steveb> yeah, or a SEEK_END event which returns the file length [21:15] <steveb> 14:25<wtay> EVENT_SIZE_ANNOUNCE or something.. [21:15] <steveb> 14:26<steveb> so is that 2 events? one to ask for the size and another to announce it? [21:15] <steveb> 14:27<steveb> or is it just announced as soon as its discovered [21:15] <steveb> 14:27<wtay> steveb: yes [21:16] <omega> um, ok, you're talking about file-size, not frame-size [21:16] <steveb> in that case, yes. but it would be good if other elements could say how many frames/samples they have to offer [21:17] <steveb> 14:27<steveb> and could other events announce how many frames / samples they have if they know? [21:17] <steveb> 14:28<wtay> yup, other fields in the size announce event I would think [21:17] <steveb> 14:33<steveb> that would make a 'reverse' element feasible ;) [21:17] <steveb> i don't mind if its caps or an event [21:31] <vishnu> i think an event would be ok [21:34] <omega> ok, I just found a bug in the queue's event handling... doesn't shut itself down. then again, it can't really. suck [21:35] <omega> er, yes it can [21:36] <omega> oh, maybe it would help if I applied taaz's events patch to it... [21:37] <ajmitch> :) [21:51] steveb (st...@no...) left irc: Ping timeout for steveb[node1ee0e.a2000.nl] [21:52] <ajmitch> sigh, windows XP poll on www.pcworld.co.nz shows many ppl wanting to upgrade to winxp either ASAP or soon. go vote for linux, ppl ;) [21:58] Nick change: taaz-football -> taaz [21:59] <ajmitch> hi taaz [22:02] <taaz> yo [22:04] Action: omega got power-on-lan to work on his desktop, but only with an offboard card ;-( [22:04] <omega> and ext3 isn't working at all ;-( [22:05] msipkema (msi...@kf...) joined #gstreamer. [22:06] msipkema (msi...@kf...) left #gstreamer. [22:11] Zeenix (ze...@ps...) joined #gstreamer. [22:12] <Zeenix> hello [22:19] Zeenix (ze...@ps...) left irc: Ping timeout for Zeenix[psh5-119.brain.net.pk] [22:23] evil_monk (trevo@208.141.162.68) joined #gstreamer. [22:33] Nick change: taaz -> taaz-away [22:36] Zeenix (ze...@13...) joined #gstreamer. [22:36] <taaz-away> omega: not sure if those pre/post emit debug calls are of much use except for debugging [22:37] <omega> taaz-away: yeah [22:37] Action: taaz-away will be back later... [22:37] <Zeenix> why my emails dont reach gstreamer-devel list ? [22:38] <omega> no idea [22:38] <Zeenix> omega: i just completed the rtp plugins [22:38] <omega> cool [22:39] <Zeenix> omega: RtpSend seems to work just fine but i dont know for what reason does the RtpRecv gets hanged in select(...) call [22:39] <Zeenix> omega: it seems to be the last error to make it working i think [22:42] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-12.brain.net.pk] [22:43] Zeenix (ze...@ps...) joined #gstreamer. [22:44] <Zeenix> repeat last messages for me, if any [22:47] <Zeenix> omega: there were many bug, i killed many of them today but i got tired, maybe tommorow i'll be able to make it work [22:47] <Zeenix> omega: sure i am not talking to myself ? :) [22:48] <omega> yeah, you're talking to your self <g> [22:49] <Zeenix> omega: i get the following error on mailing to gst ML: [22:49] <Zeenix> 550 rejected: Cannot accept mail from <pro...@li...> because your server doesn't have a postmaster account: Response from linuxfreemail.com [22:49] <Zeenix> any clue [22:49] <Zeenix> ? [22:50] <Zeenix> ----- The following addresses had permanent fatal errors ----- [22:50] <Zeenix> <gst...@li...> [22:50] <Zeenix> (reason: 550-Envelope sender verification failed [22:50] <omega> sounds like you need to email pos...@li.... er, oops. no such account ;-( [22:50] <omega> someone screwed up that your mail server [22:51] <Zeenix> i added wim...@ch... to the CC & it doesnt complain about it [22:51] <Zeenix> it only does that on gst ML [22:51] <omega> right, the sourceforge mail server is being picky about who what domains it accepts mail from [22:52] <omega> as of today, your domain doesn't have a postmaster account, so it won't accept mail from it [22:52] <Zeenix> i send emails to pl...@ya..., it works fine fo that [22:52] <omega> that's because that's not handled by mail.sourceforge.net [22:53] <Zeenix> so who should i complain to: SF or my mail server [22:53] <omega> your mail server [22:53] <Zeenix> ? [22:54] <Zeenix> omega: i am on it [22:55] <omega> wow, your mail server is SLOW [22:56] <omega> I mean REALLY SLOW [22:56] <Zeenix> omega: what, you are mailing me ? [22:56] <omega> no, I'm confirming that there's no postmaster account [22:56] <omega> I'm getting waits of several minutes [22:56] <omega> hrm, maybe a transient [22:56] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [22:57] <omega> 550 5.1.1 pos...@li...... User unknown [22:57] <Zeenix> omega: the server it uses( MailGate ) is still under-developement, but it had been long ago, i think they are not working on it anymore [22:57] <omega> that is illegal according to all the SMTP RFCs [22:59] <Zeenix> i'll bug him hard [23:00] <Zeenix> omega: if i open an account on this name ? [23:00] <omega> huh? [23:01] <Zeenix> i mean i open an account with user name: postmaster [23:01] <omega> um, no. [23:01] <omega> postmaster is defined by the RFCs as being REQUIRED to go to the admin of the site [23:02] <Zeenix> omega: i emailed the webmaster [23:03] <Zeenix> omega: should i send you the new plugins ? [23:03] <omega> no, keep sending it to wtay, he knows the history [23:04] <Zeenix> omega: do you think it'll be good idea to send it to zaheer ATM ? [23:04] <omega> probably [23:07] <Zeenix> omega: my teacher says "questions are the symptoms of intelligence", keeping this in view, i must be the most intelligent person on this channel <g> [23:07] <omega> heh [23:09] steveb (st...@no...) joined #gstreamer. [23:09] <ajmitch> wb steveb [23:09] <Zeenix> yo [23:11] <steveb> hi [23:12] <Zeenix> anybody ever heard of .ihx file extension ? [23:19] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [23:37] Zeenix (ze...@ps...) left irc: Ping timeout for Zeenix[psh5-121.brain.net.pk] [23:40] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: Connection reset by peer [23:41] Shippou (no...@cr...) joined #gstreamer. [23:42] Zeenix (ze...@ps...) joined #gstreamer. [23:44] evil_monk (trevo@208.141.162.68) joined #gstreamer. [23:47] <Zeenix> anyone having sienaps email ? [00:00] --- Sun Oct 28 2001 [00:11] Uraeus (csc...@c1...) joined #gstreamer. [00:11] <Uraeus> hi [00:11] <Uraeus> omega: I have now booked my airplane [00:11] <Zeenix> yo [00:11] <Uraeus> hi Zeenix [00:12] <omega> Uraeus: ok, times/dates? [00:12] <ajmitch> hi Uraeus [00:12] <Uraeus> omega: leaving 5th november 0645, return 12 november 1545 [00:12] <Uraeus> hi ajmitch [00:12] <Uraeus> omega: I am flying with KLM [00:13] <omega> are those Oakland times or Oslo times? [00:13] <Uraeus> first one is Oslo time, second one is Oakland time [00:14] <omega> when do you arrive in Oakland? [00:14] <Uraeus> the airplane lands in SF 13.40 local time [00:14] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [00:14] <omega> ok [00:14] <omega> are you booking the hotel or should I? [00:15] Nick change: wingo-away -> wingo [00:15] <wingo> yo [00:15] <Zeenix> Uraeus: any news of GUADEC ? [00:15] <Uraeus> I didn't get to book it today, but they said it would be no problem booking the hotel on monday [00:15] <Uraeus> Zeenix: no, not yet [00:15] <omega> ok, but you're doing that? [00:15] <Uraeus> omega: yes, I will do that [00:15] <omega> ok [00:16] <omega> ok, so we need to get battlebots tickets asap [00:17] Action: Uraeus hopes they are not expensive :) [00:17] <omega> day pass for last day is $40 [00:17] <omega> day pass for prelims is $15 [00:18] <omega> so, http://www.temple-baptist.com/~omega/als-sched.txt [00:18] <omega> I'd say go to at least the full first and last days [00:18] <omega> would be about $65 total (w/tax) [00:19] <Uraeus> sounds reasonable [00:19] <omega> ok, so I'll start by getting those tickets now [00:20] <omega> bah, I should do that from a windoze machine with a printer, in case it's unable to give me the page later [00:20] evil_monk (trevo@208.141.162.68) joined #gstreamer. [00:20] <Uraeus> wb evil_monk [00:21] Action: Zeenix hate snooker as wtay love it [00:22] <Zeenix> its 3:21 am & i am not feeling well these days, need to sleep [00:23] <Zeenix> bi [00:23] Zeenix (ze...@ps...) left irc: I pay 25 Rs( 0.5$ )/hour [00:23] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [00:25] evil_monk (trevo@208.141.162.68) joined #gstreamer. [00:26] <omega> Uraeus: OK, day passes for both 6th and 11th? [00:26] <Uraeus> omega: ok, sounds fine [00:29] steveb (st...@no...) left irc: advance! [00:39] <omega> Uraeus: afaict I got the *last* ticket (not tickets) for the final rounds [00:40] <Uraeus> omega: hmm, so in other words I will be barred from attenting? [00:40] <ajmitch> oh [00:40] <omega> the last show, yes [00:40] <omega> I have two day passes for the first day [00:40] <ajmitch> that would be annoying [00:40] <omega> but can only get one ticket for the last show [00:41] <omega> the etix system has a 8min holding time on tickets, and they lapse (else I wouldn't have gotten this one), so I'm trying several times in case one frees up [00:41] <Uraeus> ok, good [00:41] <omega> I can try calling them [00:42] <omega> but this doesn't cover the first half of the day [00:42] <omega> except this is saturday. great [00:42] <Uraeus> hehe, well I guess I can find something else to do in SF in any case [00:43] <omega> you still want to go to the AM show? [00:43] <Uraeus> sure [00:43] <omega> ok, got those. $25+tax each [00:46] <omega> calling [00:47] <omega> not in office [00:47] <Uraeus> omega: well if I enjoy the first day there then we just get me a ticket for the last day from a ticket shark [00:48] <omega> if there is such a thing [00:49] <omega> it does appear that I managed to snag the *last* ticket [00:49] <omega> if you want to keep trying: https://www1.etix.com/ticket/servlet/onlineSale?action=selectPerformance&performance_id=22610&cobrand=etixOnline [00:49] <omega> enter 1 in 'internet adult' and see what happens [00:49] <omega> if you get it, buy it ASAP [00:49] <omega> because if you don't, it's gone [00:50] <ajmitch> popular event? [00:50] <omega> it seems, yes [00:52] <omega> ok, als-sched.txt has list of tickets obtained [00:52] <Uraeus> ;) [00:53] <ajmitch> so it's only the sunday afternoon that you're missing? [00:54] <omega> yeah [00:54] <omega> that's the Finals [00:54] <omega> all others are elimination rounds [00:54] <ajmitch> k [00:55] <Uraeus> omega: couldn't we risk that the bots are to damaged to do battle when we reach the finals? [00:55] <omega> they've done this ~5x before [00:55] <omega> so yes and no [01:00] <omega> might also get tickets *just* before the show [01:00] <omega> but have to have a net connect and printer to do that [01:00] <omega> you have a cell modem? <g> [01:00] <Uraeus> no :) [01:01] <Uraeus> well actually I do, but not the right subscription [01:01] <omega> we might avoid the printer by having them scan the laptop LCD <g> [01:01] <Uraeus> hehe [01:02] <omega> need to get my plane ticket now too [01:02] <omega> and I should see if Dell can get me a new battery here in time, or if I have to ship it to someone down there [01:04] <Uraeus> omega: actually the price had increased for me to, with 9 kroner :) [01:04] <omega> heh [01:05] <omega> piddly ~$1.20, I think you can handle it <g> [01:09] <omega> Uraeus: ok, best flight for me would arrive at 1305 the 5th [01:09] <omega> so I can wait for you in the airport [01:09] <Uraeus> cool [01:09] <omega> and I'll see if I can align the departure as well [01:10] <omega> well, choices are 1220 and 1715 ;-( [01:10] <Uraeus> heh, both are kinda a bit unaligned it seems :) [01:10] <omega> yeah [01:11] <ajmitch> brb [01:11] ajmitch (aj...@p5...) left irc: http://www.freedevelopers.net [01:11] <omega> hmmm [01:11] <omega> how much time would you want in advance for your 1545 flight? [01:14] <Uraeus> don't know, if security is strict I guess I need a couple of hours [01:15] <omega> ok, according to etix the final show is sold out [01:16] Action: omega wishes etix had more info online, like total number of tickets, total sold, total in-flight, etc. [01:19] <omega> darn, can't even sneak Uraeus in as a child 12 or under [01:19] <Uraeus> omega: the other way around might work however :) [01:20] <omega> hmm, ;-) [01:20] <omega> ok, so if I were to get the 1715 flight, I'd be sitting in the airport for a while [01:20] <omega> if I get the 1220, you'd be sitting in the airport for a while [01:20] <omega> or we just split up [01:21] <Uraeus> I guess we could just split up, at that time we are probably quite tired of eachother anyway :) [01:21] <omega> heh [01:22] ajmitch (aj...@p2...) joined #gstreamer. [01:22] <ajmitch> i'm back ;) [01:23] <Uraeus> oh no [01:23] <omega> ok, then in that case I think I'll get the 1920 flight [01:23] <omega> er, wait. that goes through Reno. [01:23] <omega> n/m [01:24] <omega> hah. flight 996. that's the one I was on when I was routed through Oakland on the way home once [01:33] Nick change: taaz-away -> taaz [01:33] <omega> Uraeus: you have flight numbers? [01:34] <Uraeus> no, not yet but I get that on monday for you [01:34] <omega> ok, I'll need to know where to meet you [01:35] <omega> ok, Southwest at least claims that you should be there at least 1.5hrs beforehand for domestic flights [01:36] <omega> so you probably want to be there 2-2.5hrs ahead [01:36] <omega> http://www.southwest.com/travel_center/travelAdvisory.html [01:37] <omega> read through that page carefully so you know what to expect in US airport as far as security and carryon limits (STUPID!) [01:38] <omega> do we want to set up a BOF? [01:38] <Uraeus> omega: I am just packing my traditional norwegian battleaxe so it should be ok [01:39] <Uraeus> omega: that could be cool [01:39] <omega> ah [01:39] <omega> ok, um, when [01:39] <omega> ? [01:39] <omega> I'd say thursday or friday [01:40] <omega> 9pm or later if friday (to avoid overlap) [01:40] <taaz> heh... reading log... pretend i was here when you started talking about Oakland vs Oslo time: I just got Oakenfold's Live in Oslo cd ;) [01:40] <Uraeus> thursday then, earlier the better since people are usually more tired late in the week [01:41] <Uraeus> taaz: hehe [01:41] <omega> hrm. Thursday 8pm and 9pm is "New X Tech from X.org: "MAS": a Media Application Server [01:41] <omega> we need to attend that (2hrs) [01:41] <omega> so maybe schedule for 10pm Thur? [01:41] <Uraeus> ok [01:42] <omega> scheduling [01:43] <omega> led by both of us? [01:45] <omega> ok, mail sent [01:47] <omega> ouch. there went $263.25 in the space of an hour <> [01:47] <omega> <g> [01:47] Action: Uraeus nods [01:50] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [01:52] Action: taaz wills the crowd to find a fix for gtk unthrededness [01:52] Uraeus (csc...@c1...) left irc: Ping timeout for Uraeus[c128s9h5.upc.chello.no] [01:52] Action: ajmitch will bbl [01:53] <taaz> only fixes i can think of are very extensive [01:54] <taaz> and all involve complex locking polices and are very prone to bugs as far as maintenence goes [02:18] bstard (Lor...@a2...) joined #gstreamer. [02:21] Nick change: wtay-snooker -> wtay [02:21] <wtay> yo [02:22] <taaz> yo! [02:24] <wtay> omega: you are busy? [02:24] <omega> nope [02:24] <wtay> ah [02:25] <wtay> I'd like to finish the event stuff.. [02:25] <omega> yeah [02:25] <omega> brb [02:25] <wtay> I've been trying a few things.. [02:25] Action: wtay get more ice cream [02:30] <omega> never put a plate on a hot burner [02:30] <omega> they explode [02:30] <omega> as my Mom just found out [02:33] <wtay> I added a default handler.. [02:33] <omega> ok [02:33] <wtay> in gst_scheduler_chain_wrapper I check for an event [02:34] <wtay> I added a GST_ELEMENT_EVENT_AWARE flag too [02:34] <omega> ok, so most elements don't need modification [02:34] <wtay> if there is an event and the element is not event aware it calls the default handler [02:34] <omega> er, most chained elements that is [02:35] <wtay> but I only added it in the chain_wrapper... [02:35] <omega> ok [02:35] <wtay> but then lots of other issues.. [02:35] <wtay> the main problem is with threads.. [02:36] <wtay> we should try to get fakesrc ! queue ! { fakesink } do EOS.. [02:36] <omega> working on it [02:36] <omega> taaz [02:36] <omega> taaz's patch helped a lot [02:36] <omega> but still causing scheduler problems [02:39] <wtay> eek, lots of conflicts in gstqueue.. [02:39] <omega> taaz changed it a fair amount [02:40] <wtay> yeah, pretty similar to the changes I did locally [02:40] <taaz> sorry ;) [02:41] <taaz> still need to handle leaky events properly [02:41] <wtay> and in the cleanup^code you cannot unref events [02:41] <taaz> will just spam the console now if it tries to drop them [02:41] <taaz> oh yeah [02:41] <taaz> so, i think events should be refcounted [02:41] <omega> ok, so the current situaton with fakesrc!queue!{fakesink} is... [02:41] <wtay> yeah, [02:42] <wtay> taaz: GstData should probably be refcounted [02:42] <omega> with a local patch to cause the queue to PAUSE on EOS iff the queue's manager is the same as the peer's manager [02:42] <taaz> dude... like i've been on phone forever trying to order a damn pizza [02:42] <omega> I get EOS all the way down the pipe [02:43] <wtay> omega: the thread locks here [02:43] <omega> however, when bin_iterate fails for the thread (because there is nothing to schedule, because everything is PAUSED), the interlock wedges [02:43] <omega> DEBUG(25371: 0)gst_thread_change_state:428: [thread0] sync(25385): waiting for thread to stop spinning [02:43] <wtay> ok [02:43] <omega> even though [02:43] <omega> DEBUG(25385: 0)gst_thread_main_loop:586: [thread0] sync-main(25371): iteration failed, something very wrong, spinning to let parent sync [02:44] <wtay> yup, same here [02:44] <omega> before that [02:44] <omega> now, right after that, I get 25371 trying to go from PLAYING to NULL [02:44] <omega> who's doing that? [02:45] <wtay> also the main thread performs nothing in the iterate, wasting cycles [02:45] <omega> well, not with my change [02:45] <wtay> ok [02:45] <omega> --- gstqueue.c 2001/10/27 20:28:30 1.18 [02:45] <omega> +++ gstqueue.c 2001/10/28 00:45:37 [02:45] <omega> @@ -273,6 +273,12 @@ [02:45] <omega> GST_DEBUG_ELEMENT (GST_CAT_DATAFLOW, queue, "flushing queue\n"); [02:45] <omega> gst_queue_locked_flush (queue); [02:45] <omega> break; [02:45] <omega> + case GST_EVENT_EOS: [02:45] <wtay> must be -launch then shutting down the pipeline [02:45] <omega> + GST_DEBUG_ELEMENT (GST_CAT_DATAFLOW, queue, "queue has EOS\n"); [02:45] <omega> + // we should only pause here if our manager is the same as our peer [02:45] <omega> + if (GST_ELEMENT(queue)->manager == GST_ELEMENT(GST_PAD_PARENT(GST_RPAD_PEER(pad)))->manager) { [02:45] <omega> + gst_element_set_state(GST_ELEMENT (queue), GST_STATE_PAUSED); [02:45] <omega> + } [02:47] <wtay> oh, I change state in the _get function [02:48] <wtay> uhm [02:48] <omega> yeah, it should be one or the other [02:48] <omega> but setting it in _get is a nop atm, so.. [02:49] <omega> problem is that DECOUPLED elements exist as a semi-scheduled element in at least two contexts [02:49] <omega> but we really can only set the state when the context matches the manager's context [02:50] <taaz> ok, how should i discuss my event/buffer ideas? email to list and get no response? <g> talk here? beat people over the head with my python pseudo code? [02:50] <omega> you talked about doing a branch, to demo, is that feasible? [02:50] <taaz> and btw, i really really really think this gtk not being thread safe is a show stopper for the who... [truncated message content] |
From: IRC B. <wt...@us...> - 2001-10-29 05:27:56
|
******************************************************************* [03:00] <omega> why there need to be global structures for signal emission in the first place I have no idea [03:00] <omega> I think someone else needs to write glib3 [03:04] <taaz> go forit [03:05] <omega> hah, one of these days [03:05] <taaz> item #1023 on the todo list... [03:05] <omega> heck, it's fewer lines of code than gst <g> [03:07] gcasey (gc...@20...) joined #gstreamer. [03:07] <omega> yo [03:07] gcasey (gc...@20...) left irc: Client Exiting [03:13] <wingo> so... what's the status then? can glib2 be hacked to provide a global signal emission mutex, or did you say that it has one? [03:13] <omega> it looks like it already has one [03:13] <omega> I think a major point of glib2 is to make everything MT safe [03:14] <wingo> that's good [03:15] <wingo> for whenever i buy my next machine ;) [03:17] <omega> this is so cool. I can power on my desktop with a command on my gateway (ether-wake) [03:17] <wingo> oooo [03:18] <omega> but I can't get the kernel to power the machine off at shutdown, there's something wrong with later 2.4.x [03:18] <taaz> you turn your machines off? [03:19] Action: wingo has a noisemaker in the corner of his room, 24-7 [03:19] <omega> yeah, because they're in my room [03:19] <omega> that's why I'm after a silent machine... [03:19] <omega> ultimate solution is a cold-water jack in the wall [03:20] <omega> power, ether, water [03:20] <ajmitch> hehe [03:20] <ajmitch> i'd love that too [03:24] <omega> bleagh. redhat's 2.4.3-12 kernel powers off just fine. but the exact same apm options in my 2.4.12-ac3 kernel don't work [03:25] <wingo> hey. [03:25] <wingo> i was athinkin [03:25] <wingo> about default param settings for gst [03:25] <wingo> does that make sense? [03:26] <wingo> stored in an xml registry, or directory [03:26] <wingo> per-element xml files [03:27] <omega> possibly [03:33] <wingo> ok, so how should gst-launch know when to stop when the eos event travels down the pipeline? [03:33] <wingo> the eos signal? [05:54] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [06:16] Nick change: wingo -> wingo-z [06:24] arik (ar...@sd...) joined #gstreamer. [06:24] <arik> lo [06:26] mef-zZz (me...@us...) joined #gstreamer. [06:26] Nick change: mef-zZz -> mef [06:26] <mef> howzit? [06:26] <arik> heh [06:27] <mef> ajmitch: yesterday you mentioned an interest in windows version of gst. Are you still? [06:27] <ajmitch> oh, i guess ;) [06:28] <ajmitch> hi arik, mef ;) [06:28] <arik> hey aj [06:28] Action: arik is more bored then any human has ever been, ever [06:28] <arik> well maybe not that bored [06:28] <arik> but i'm certainly not un-bored [06:28] <ajmitch> nah, can't be that bored [06:28] <ajmitch> can't be as bored as i was in the exam yesterday ;) [06:29] <arik> hehe [06:29] <arik> prob not [06:29] <arik> but you got to leave [06:29] <arik> i don't get out till january [06:29] <ajmitch> i got another exam tomorrow [06:29] <ajmitch> and the last one in 2 weeks [06:29] <arik> ouch [06:29] <arik> still it ends soon [06:29] <ajmitch> yep [06:29] <ajmitch> then i get to have a little break ;) [06:30] <arik> hehe [06:30] <arik> well at least you have your friends around you [06:30] <ajmitch> yep [06:30] <arik> i'm all alone [06:30] <ajmitch> :( [06:31] <arik> it sucks [06:31] <mef> http://groups.yahoo.com/group/gimpwin-dev/message/1781 provides a new version of pkg-config, glib, gtk, and the like for win32. There are people working on a cygwin version of the same. [06:31] <arik> and i've failed misserable [06:31] <arik> in building gnome2 [06:31] <ajmitch> arik: where did it fail? [06:31] <mef> when that happens, then work on win32-gst could commence. [06:31] <arik> aj, in the planning stages :-) [06:31] <ajmitch> arik: you're on redhat? [06:32] <arik> yep [06:32] <ajmitch> there are rpms [06:32] <arik> heh yeah [06:32] <arik> 56.6 modem :-) [06:32] <arik> i may do that though [06:32] Action: ajmitch is on a 33.6k modem ;) [06:32] <arik> ouch [06:32] <arik> that really really sucks [06:33] <ajmitch> yep ;) [06:33] <arik> heh [06:33] <arik> i ordered dsl [06:33] <ajmitch> shoudl get DSL next year [06:33] <arik> who knows when it will be installed [06:33] <arik> 10 days or so or more [06:33] Nick change: mef -> mef-zZz [06:36] Nick change: arik -> arik|afk [06:36] <arik|afk> brb [06:36] <ajmitch> k [06:39] <arik|afk> back [06:39] <ajmitch> cya later ppl [06:39] Nick change: arik|afk -> arik [06:39] <arik> later aj [06:39] Action: ajmitch is leaving now ;) [06:39] <arik> heh [06:57] Shippou (no...@cr...) joined #gstreamer. [07:29] arik (ar...@sd...) left irc: leaving [08:04] chillywilly (da...@d8...) joined #gstreamer. [08:23] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [09:07] omega (om...@om...) left irc: sleep [09:42] swivel (swivel@168.103.157.45) joined #gstreamer. [09:42] <swivel> bitches [09:42] <swivel> sup [09:42] <swivel> hrm, dead :P [09:42] swivel (swivel@168.103.157.45) left #gstreamer. [09:43] <ajmitch> heh, fool [09:51] chillywilly (da...@d8...) left irc: Read error to chillywilly[d89.as29.nwbl0.wi.voyager.net]: EOF from client [10:34] steveb (st...@no...) joined #gstreamer. [11:15] _Kirby_ (ki...@e2...) joined #gstreamer. [11:26] Zeenix (ze...@13...) joined #gstreamer. [11:49] Uraeus (csc...@c1...) joined #gstreamer. [11:49] <Uraeus> morning [11:49] <Zeenix> yo [11:50] <Uraeus> hi Zeenix [11:51] <Zeenix> The Rtp plugin seems to work in a day or two [11:51] <Zeenix> ugghhh..., bad english [11:52] <Zeenix> it seems that the Rtp plugins will really start to work in a day or two [11:52] <Uraeus> Zeenix: great, you said you where feeling bad yesterday, something serious? [11:52] <Zeenix> no, a better now [11:52] <Zeenix> no, better now [11:53] <Uraeus> Zeenix: gotten any further on your studies of communism? [11:53] <Zeenix> Uraeus: yeah [11:54] <Uraeus> Zeenix: what books are you reading? Das Kapital? [11:54] <Zeenix> Uraeus: i saw the "Theory of Creation" of babilonians( dont know spelling ) today, it seems that all it was inherited by all thre religions including mine [11:55] <Zeenix> Uraeus: at the moment, i am reading the "Tombs of the Past" [11:55] <Zeenix> Uraeus: its in Urdu, writen by Sibt-e-Hasan, a muslim-become-communist [11:56] <Uraeus> Zeenix: Theory of Creation was written in old Babylon? [11:56] <Zeenix> Uraeus: yes of course [11:57] <Zeenix> Uraeus: in several different places [11:58] <Zeenix> Uraeus: after that i'll read his book, about which many people say that it answers all the question/objections by capitalists on communism [11:58] <Uraeus> Zeenix: ah, ok I thought it was a book not the theory of creation :) [11:59] <Uraeus> Zeenix: did you manage to trade in my modem with one that works for you? [11:59] <Zeenix> Uraeus: no there are 6 chapters in it about each major civilization's theory of creation [11:59] <Zeenix> Uraeus: not uptill now [12:00] <Uraeus> Zeenix: hope it contains the norse theory of creation [12:00] <Zeenix> Uraeus: i said **major** civilizations <g> [12:00] <Uraeus> hmm ;) [12:01] <Zeenix> Uraeus: i heard that Norway is quit a socialist kink of country [12:01] <Uraeus> don't remember the details anymore, but I think the norse theory was that the gods had slain a giant troll and its rotting carcass became the world [12:01] <Uraeus> yes, we have had a social democratic government for 80% for the last 50 years [12:02] <Uraeus> so we have a large public sector [12:02] <Zeenix> Uraeus: seen wtay today ? [12:03] <Uraeus> Zeenix: since I came here after you how could I :) [12:03] <Uraeus> Zeenix: what kind of connection are you on now? [12:03] <Zeenix> Uraeus: no, i dont know whether you came before me for some time or no [12:04] <Zeenix> Uraeus: dialup [12:04] <Zeenix> Uraeus: Your modems seems to have been targetted only for your country, doesnt accept anolog phone line at all [12:05] <Uraeus> Zeenix: is it true that Islam has the death penalty for women who marries non muslims? [12:05] <Zeenix> Uraeus: yeah, for anyone who leave's the religion [12:06] <Zeenix> Uraeus: a friend & cousin of prophet burnt people to death for this even [12:07] <Zeenix> Uraeus: what? you are in love with a muslim girl? :) [12:07] <Uraeus> Zeenix: no, but a friend of mine just married a turkish girl, but she had already converted to budhism [12:08] <Zeenix> Uraeus: yeah, i think budhism is the most peaceful religion on earth [12:09] <Uraeus> her family accepted it luckily, her father is a turkish politician who is a memember of the party founded by Ataturk [12:09] <Uraeus> Zeenix: yes, it seems rather non-agressive [12:09] <Zeenix> Uraeus: they even teach people to love the lions [12:10] <Zeenix> Uraeus: my familly dont & will never accept [12:12] <Uraeus> Zeenix: maybe Pakistan needs its own Ataturk [12:12] <Uraeus> he was even a socialist if I am not mistaken [12:12] <Zeenix> Uraeus: Mr. President is much like him but there is a very big problem [12:13] <Zeenix> Uraeus: pakistan is not a nation, It was created in the name of religion [12:13] <Zeenix> Uraeus: so if you put religion aside, you put pakistan's border with India aside [12:14] <Zeenix> Uraeus: & the army & religious extremists will never let this happen [12:14] <Zeenix> Uraeus: Mr. President is also from the Army [12:14] <Uraeus> I see so few people indentify with or feel loyalty towards Pakistan the country just Pakistan the religious entity [12:14] <Uraeus> hmm [12:15] <Zeenix> Uraeus: there are people here who realizeses the facts & this number is increasing [12:15] <Uraeus> I saw this interview with some of the people of the nothern alliance in Afganistan where they said that the difference between them and the taliban was that the taliban fought for Allah first and Afganistan second while they fought for Afghanistan first and Allah second [12:16] <Zeenix> Uraeus: like i have convinced a strict fundamentalist friend on my thoughts [12:16] <Uraeus> the nothern alliance fighters also saaid they would much prefer that their daughters married non-muslim foreigners than taliban supporters [12:16] <Zeenix> Uraeus: he comes in the channel sometimes with nick: nash [12:17] <Uraeus> ok, I talk to him next time he comes in then [12:17] <Zeenix> Uraeus: but dont tell him that you know much about his thoughts [12:18] <Zeenix> Uraeus: he keeps his profile very very low [12:18] <Uraeus> Zeenix: actually when I was younger and a active member of the norwegian conservative party I activly worked for the party supporting cutting all support for the church and severing all ties between state and religion, I basically still holds this view today [12:19] <Zeenix> Uraeus: i'll say we have to do better than that against religions, they want us to go back to the old ages, its something very dangerous for all of us [12:20] <Uraeus> Zeenix: the church is under pretty strict political control in norway, for instance it is the politically ellected government which selects new bishops [12:20] <Zeenix> Uraeus: ok we should stop talking against all religions on gst, omega wont like that [12:27] Jacmet (ja...@bo...) joined #gstreamer. [12:28] _Kirby_ (ki...@e2...) left #gstreamer. [12:29] Nick change: Uraeus -> Ura_away [12:33] <Jacmet> hi [12:34] <Zeenix> yo [12:35] <Jacmet> Zeenix: I did see your email, but I have been ++BUSY the last many weeks with moving to Belgium.. [12:36] <Zeenix> Jacmet: my email ? [12:36] <Zeenix> Jacmet: about that logo [12:36] <Zeenix> ? [12:36] <Jacmet> yes [12:36] <Zeenix> Jacmet: no prob. [12:36] <Zeenix> Jacmet: from where ? [12:38] <Jacmet> Zeenix: denmark [12:39] <Zeenix> Jacmet: i've started another project too, a roboCar [12:39] <Jacmet> robocar? [12:40] <Zeenix> Jacmet: hah, more hardware than software [12:40] <Jacmet> Zeenix: ok - a computer controlled car? [12:40] <Zeenix> Jacmet: you can say that [12:40] <Zeenix> s/computer/microcontroller [12:41] <Jacmet> completely autononymous? [12:41] <Zeenix> Jacmet: have you coded on low-level ? [12:41] <Jacmet> or however it's spelled ;) [12:41] <Jacmet> Zeenix: yes, x86 and motorola assembler [12:41] <Zeenix> Jacmet: self-controlled you mean [12:42] <Zeenix> Jacmet: i am trying [12:42] <Zeenix> Jacmet: but i dont have much mem in that, its an intel MCS51 compatible [12:43] <Jacmet> how is it going to orient it self? [12:43] <Zeenix> Jacmet: the code & data mem. can be only upto 64k each, i have never coded for low level, so i dont know whether these will be enough or not [12:44] <Zeenix> Jacmet: i have much to discus about the car's behavious with that hardware responsible person [12:45] <Jacmet> Zeenix: ok, what's the project for? university/work/fun? [12:45] <Zeenix> Jacmet: all of them [12:45] <Jacmet> ok ;) [12:45] <Zeenix> Jacmet: specially fun, i always wanted to do physical work out of my code [12:45] <Jacmet> heh [12:46] <Zeenix> Jacmet: i found a very nice compiler for this( sdcc.sf.net ) [12:47] <Zeenix> Jacmet: but i expected a raw binary file after the assembling & linking phase, but it gives me .ihx file [12:47] <Jacmet> ok, so it's not going to be in asm? [12:47] <Zeenix> Jacmet: no [12:47] <Jacmet> sorry, I don't really know much about microcontrollers [12:47] <Zeenix> Jacmet: but you do know about asm & C [12:48] <Jacmet> a bit yes [12:49] <Zeenix> Jacmet: if i send you those output files, maybe you could gues what that compiler creates in the end [12:49] <Zeenix> ? [12:50] <Jacmet> hmm, I might just have a look at sdcc.sf.net first.. [12:50] <Zeenix> Jacmet: you can accept DCC ? [12:50] <Jacmet> I don't think so.. [12:52] <Jacmet> it looks like a nice compiler.. [13:00] <Zeenix> so [13:05] <Zeenix> Jacmet: are you still at sdcc.sf.net [13:05] <Zeenix> ? [13:06] <Jacmet> yes, but I cannot seem to find any real documentation.. [13:07] <Zeenix> Jacmet: yeah, but the source doest have it [13:07] <Zeenix> Jacmet: download the source tar ball & its in it [13:07] <Jacmet> ok - what's the problem exactly then? [13:08] <Zeenix> Jacmet: wait a min. [13:10] <Zeenix> Jacmet: about the output files, the docs says: [13:10] <Zeenix> sourcefile.asm - Assembler source file created by the compiler [13:10] <Zeenix> sourcefile.lst - Assembler listing file created by the Assembler [13:10] <Zeenix> sourcefile.rst - Assembler listing file updated with linkedit [13:10] <Zeenix> information, created by linkage editor [13:10] <Zeenix> sourcefile.sym - symbol listing for the sourcefile, created by the [13:10] <Zeenix> assembler [13:10] <Zeenix> sourcefile.rel - Object file created by the assembler, input to [13:10] <Zeenix> Linkage editor [13:10] <Zeenix> sourcefile.map - The memory map for the load module, created by the [13:10] <Zeenix> Linker [13:10] <Zeenix> sourcefile.ihx - [13:11] <Zeenix> The load module in Intel hex format (you can select [13:11] <Zeenix> the Motorola S19 format with -out-fmt-s19) [13:11] <Zeenix> sourcefile.cdb - An optional file (with -debug) containing debug [13:11] <Zeenix> information [13:12] <Jacmet> ok, so the .ihx format is what you need to put on an (e)eprom, right? [13:12] <Zeenix> Jacmet: thats what i am in doubt [13:12] <Zeenix> Jacmet: the file command tells me that its a text file [13:13] <Zeenix> Jacmet: & i can have a good pic. of it when i open it in vi [13:13] <Jacmet> Zeenix: yes, it's a bit like base64 encoding afaik, must eprom burner programs support it (otherwise they probably support the s19 format) [13:14] Nick change: Ura_away -> Uraeus [13:14] Nick change: wtay-Zz -> wtay [13:14] <wtay> yo [13:14] <Jacmet> I don't know the particular microcontroller you are using, but I did a project 2 years ago with a motorola microcontroller - there we just burned a .s19 file to an eeprom [13:14] <Jacmet> hi wtay [13:14] <Zeenix> wtay: yo [13:15] <Jacmet> and Uraeus [13:15] <wtay> Jacmet: bones? kulnet :) [13:15] <Zeenix> wtay: check inbox first [13:15] <Uraeus> yo dudes [13:15] <wtay> Zeenix: I'm trying to get thr rtp stuff working now [13:15] <Jacmet> wtay: yeah, I moved to Leuven yesterday ;) [13:15] <Zeenix> wtay: i already did much of it [13:15] Action: wtay waves at Jacmet [13:15] <Jacmet> wtay ;) [13:15] <wtay> Zeenix: I'm using cvs librtp and it doesn't compile [13:16] <Zeenix> wtay: i think this time i am not getting the RTCP packets [13:16] <wtay> Zeenix: but it's an easy fix. what version are you using? [13:16] <Zeenix> wtay: i can't tell ATM [13:17] <Zeenix> wtay: get the stable one, i think so [13:17] <Zeenix> wtay: i seems that gphone people have not been very active since last few months [13:18] <Zeenix> wtay: got my latest mail? i sent it an hour or two ago [13:19] <wtay> hmm [13:19] <Zeenix> wtay: got it ? [13:19] <wtay> yeah [13:20] <Zeenix> wtay: the RtpRecv simply segfaults now after getting the first RTP( not RTCP ) packet [13:21] <wtay> it looks like rtp_packet_send and rtcp_compound_send require a size as a fourth arg.. [13:21] <wtay> Zeenix: I couldn't find a stabe version of librtp.. [13:21] <wtay> Zeenix: shall I fix it? [13:21] <Zeenix> wtay: its in included in the gphone i think [13:22] <Zeenix> wtay: fix what ? [13:22] <wtay> Zeenix: the compile error? [13:23] <Zeenix> wtay: warnings, right ? [13:24] <wtay> errors [13:24] <wtay> rtp.c: In function `rtp_send': [13:24] <wtay> rtp.c:70: warning: passing arg 3 of `rtp_packet_send' from incompatible pointer type [13:24] <wtay> rtp.c:70: too few arguments to function `rtp_packet_send' [13:24] <Zeenix> wtay: i dont have any compile error & it was because i was passing the gobject of gtk+ call [13:24] <wtay> you have another version of librtp [13:25] <Zeenix> wtay: wait a min. [13:26] <wtay> ok, gphone 0.5.2 has the librtp you use [13:27] <Zeenix> i think so [13:27] <Zeenix> i had some rpms from rpmfind.net AFAIK [13:27] <Zeenix> s/AFAIK/AFAI remember [13:30] <Zeenix> wtay: you'll be here in the night: after some 7-8 hours ? [13:31] <wtay> Zeenix: yeah [13:31] <Zeenix> wtay: played enough snooker yesterday ? [13:32] <wtay> Zeenix: never enough, Zeenix :) [13:33] <wtay> Zeenix: ah, yes. you need to use the gobject methods in a plugin [13:34] <Zeenix> wtay: make it like that, i didnt know how to [13:34] <wtay> Zeenix: I usually just rip it out of some other plugin :) [13:34] <Zeenix> wtay: its the signal code that's using gtkobject's functions [13:35] <Zeenix> wtay: i am always too aftraid to look into new codes, i get lost in that & dont remember any more why i am here [13:36] <wtay> Zeenix: it's confusing, I must admit.. [13:36] <Zeenix> wtay: what my code, i wont admit <g> [13:38] <wtay> gtk_*, g_*, I mean :) [13:39] <Zeenix> wtay: yeah, my code cant be, as i have borrowed much of the code from outside [13:39] <wtay> not to mention how confusing librtp is... [13:40] <Zeenix> oh yes, blame gphone, they have messed up many things [13:40] <Zeenix> wtay: they should have given some sort of sequencing stuff first of all [13:41] <wtay> hm, it compiles now.. [13:42] <Zeenix> wtay: check it, meanwhile i'll make a cup of tea for myself [13:42] <Zeenix> wtay: & pray i dont get disconnected in the meanwhile [13:42] <wtay> oh? you mean it has to work too? <g> [13:43] <Zeenix> wtay: yeah, rtpsend does send rtp packets [13:43] <Zeenix> & the rtprecv just recv a single one & gets seqfaults [13:44] <wtay> ok, send seems to work.. [13:44] <Zeenix> see if it also send rtcp packet [13:44] <Zeenix> it should do that when on newcaps [13:45] <wtay> I'm using fakesrc, so no newcaps is called [13:48] <Zeenix> got the rtprecv to segfault [13:48] <Zeenix> ? [13:48] <wtay> not here.. nothing happens in rtprecv [13:49] <Zeenix> wtay: strange, i used to recieve the RTP packets but dont know what is used to do with it [13:49] <wtay> oh.. [13:49] <wtay> maybe I should send buffer with >0 bytes :) [13:50] <Zeenix> wtay: & then it started to segfault on the first RTP packet recieved [13:50] <wtay> ok, same here [13:50] <Zeenix> wtay: segfault [13:50] <Zeenix> ? [13:51] <wtay> ** ERROR **: rtp_packet_read: recvfrom: Bad address [13:51] <wtay> aborting... [13:51] <wtay> Aborted [13:51] <Zeenix> wtay: no, not here [13:51] <Zeenix> wtay: make sure you are using latest one i sent you [13:52] <wtay> I am [13:54] <Zeenix> wtay: why the diff. between your sys & mine always because of this recvfrom [13:54] <wtay> uhm.. rtp_receive looks wrong.. [13:54] <Zeenix> ? [13:54] <wtay> no idea [13:54] <Zeenix> wtay: your linux distro & redhat seems to be incompatible in this recvfrom function [13:55] <wtay> I doubt it [13:56] <Zeenix> wtay: my app using udpsrc used to give error on recvfrom but your sys didnt [13:59] <wtay> it probably because my librtp version is different [14:00] <Zeenix> wtay: so you should get as many ver. as you can & check it on each one <g> [14:00] <Zeenix> wtay: or wait till i come next time & tell you [14:01] <wtay> bleagh they seriously changed the API of librtp [14:01] <Zeenix> wtay: where are you looking at, librtp.sf.net ? [14:01] <wtay> yup, cvs [14:03] <Zeenix> i think you should ATM use the one packed with gphone [14:05] <wtay> there is no librtp in gphone, it's compiled into libghone [14:06] <Zeenix> so my rtp libs should work on you sys if i send you [14:06] <Zeenix> ? [14:07] <wtay> ok, send them to me then.. [14:07] <Zeenix> wtay: you know, can access ext2fs from windooz [14:07] <Zeenix> s/can/cant [14:08] <Zeenix> wtay: so i'll go now, feel free to make any changes but keep the one i sent you saved with you [14:08] <Zeenix> s/saved/unchanged [14:09] <wtay> ok [14:09] <Zeenix> bi [14:09] Zeenix (ze...@13...) left irc: I pay 25 Rs( 0.5$ )/hour [14:46] BBB (BB...@uc...) joined #gstreamer. [14:58] bstard (Lor...@a2...) got netsplit. [14:58] bstard (Lor...@a2...) returned to #gstreamer. [16:00] BBB (BB...@uc...) left irc: [x]chat [16:01] BBB (BB...@uc...) joined #gstreamer. [17:17] Uraeus (csc...@c1...) left #gstreamer (Client Exiting). [17:19] wingo-z (wi...@rd...) left irc: going to skool [17:46] wingo (wi...@rd...) joined #gstreamer. [17:46] <wingo> yo [17:47] Uraeus (csc...@c1...) joined #gstreamer. [17:47] <Uraeus> hi again [17:48] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [17:49] <wingo> hey Uraeus. [17:51] <Uraeus> hi wingo [17:51] <Uraeus> wtay: have you looked at taaz's thread safety problem? [17:57] msipkema (msi...@kf...) joined #gstreamer. [17:57] <msipkema> so what's this channel about? [17:58] <wingo> gstreamer! [17:58] <wingo> gstreamer.net [17:59] <msipkema> what on earth is gstreamer? [17:59] <msipkema> i am looking for linux audio developers... [18:00] <wingo> well, you can bother to check out the web site ;-) [18:00] <wingo> who's doing the looking? [18:01] <msipkema> i am looking at the website right know. i am not interested in gstreamer i guess :) but i would like to talk to linux audio developers. [18:01] <wingo> i do some work. [18:02] <wingo> i wrote the alsa plugin for gstreamer, a resampler, [18:02] <wingo> and have my own mixing project at ambient.2y.net/beatbox/ [18:02] <wingo> i don't know if that makes me a linux audio developer, buy hey :) [18:02] <wingo> s/buy/but [18:02] <Uraeus> msipkema: are you looking for the LAD mailing-list? [18:03] <msipkema> Uraeus: i just subsribed to that. i am looking to talk to audio developers on developing a midi api for linux. [18:03] <wingo> oh, that was your message [18:03] <Uraeus> msipkema: well this is a place where there are a lot of audio developers [18:03] <wingo> l-a-d is the best place imho [18:04] <wingo> we don't have adequate midi support yet within gst [18:04] <Uraeus> so if you want to work on midi msipkema we would love to have your help [18:04] <msipkema> wingo: that's what i thought. i'm hoping there will be some people interested in bringing good midi to linux. [18:05] <wingo> muse is good [18:05] <wingo> have you checked that out? [18:05] <wingo> i can't remember the url [18:05] <Uraeus> msipkema: well if you are willing to start I am sure there are people here and on our mailinglist who would be willing to help out with adding good midi support for gstreamer [18:06] <msipkema> i have looked into the sequencers that are available for linux. but i'm not content with the linux midi api. it needs support for scheduling and it needs IPC midi. [18:06] <wingo> you are familiar with the alsa sequencer code? [18:07] <wingo> perhaps you should ask on alsa-devel [18:07] <wingo> unfortunately its api is not well documented yet [18:07] <msipkema> wingo: not really, but i also don't like a sequencer device. i think the event scheduling should be in the application mostly and in the driver. [18:07] <msipkema> alsa is too low level. [18:07] <wingo> whatev's. alsa-devel does have a lot of expertise though [18:08] <wingo> i'm not a midi fellow ;) [18:09] <msipkema> wingo: well, i will join alsa-devel. i don't think they give midi a high priority. and also audio and midi aren't really related. they should probably remove midi from alsa entirely. [18:09] <wingo> it's hw related. [18:09] <wingo> you need drivers. [18:09] <wingo> they are the advanced driver people. [18:10] <wingo> they *do* care about midi. [18:10] <wingo> judging by the traffic there, anyway. [18:10] <msipkema> wingo: most professional midi interfaces are stand alone. [18:10] <Uraeus> wingo: lots of care but no code maybe :) ? [18:10] <wingo> what do you mean by professional midi interfaces? [18:10] <wingo> there's code, yo [18:10] <wingo> just not documented all that well [18:11] <wingo> if you are doing serious midi stuff in linux, you need to use alsa. [18:11] <wingo> the oss interface sucks. [18:12] <msipkema> i want a high level midi api that can use alsa for drivers. but also i'll write a "driver" (more like a server) for emagic serial midi interfaces (and probably steinberg usb if i know how it works). [18:13] <Uraeus> msipkema: that sounds great [18:14] <msipkema> i am an ex beos user. any of you used beos? seen their midi support? [18:14] <wingo> all i can do is point you to where i know there's more expertise, and that's at alsa-devel and lad ;) i'm not ashamed to admit my midi-ignorance :-) [18:14] <Uraeus> msipkema: I have just tested it briefly nothing serious [18:14] <wingo> msipkema: no, unfortunately [18:15] <msipkema> it wasn't perfect, but something like that would be nice to have for linux. [18:16] <Uraeus> msipkema: how hard would it be making something like that for linux you think? [18:17] <msipkema> Uraeus: not that hard i think. the hardest would be to get good timing. with beos you could get a very accurate 'system_time()' and also schedule on that very accurately, i.e. much more than millisecond accurate. [18:18] <msipkema> is that possible from user space with linux? [18:18] <wingo> with low-latency kernel patches and the alsa sequencer api, yes [18:19] <wingo> latencies of about 2 ms are achievable [18:19] <wingo> jitter very low [18:19] hadess (ha...@pc...) joined #gstreamer. [18:19] <wingo> yo [18:19] <hadess> heya [18:19] <hadess> yo wingo [18:19] <wingo> hey, i'm porting me app to gnome2 :-) [18:19] <wingo> you too? [18:20] <hadess> i'm just porting the gnome-utils for now [18:20] <Uraeus> hadess: msipkema here is looking for people to implement world class midi for linux, and me and wingo volunteered you :) [18:20] <wingo> a task meritorious of many thanks :-) [18:20] <hadess> Uraeus: no way ;) [18:21] <Uraeus> hmm [18:21] <hadess> i don't even have a soundcard that supports midi [18:22] <Uraeus> hadess: I thought audio and midi support where an Apple stronghold? [18:23] <hadess> audio yeah, midi is supported by their drivers, but it's not a hardware support [18:24] <msipkema> i think macos x has very good midi support. 1 ms input to output latency with 200 microseconds jitter. [18:24] <msipkema> usually the serial port is used for midi on macs i think, with an adapter. [18:25] <hadess> serial port... old macs then ;) [18:25] <Uraeus> msipkema: hadess is our resident linux-on-mac expert :) [18:27] dobey (do...@cv...) joined #gstreamer. [18:27] <hadess> s/expert/user/ ;) [18:27] <dobey> hadess: !!! [18:27] <msipkema> are you guys familiar with openml? the ust/msc model? [18:27] <hadess> nope [18:27] <hadess> hey dobey [18:27] <hadess> dobey: got the special keys to work ? [18:27] <dobey> hadess: no, sf was down so i couldn't get hotkeys [18:27] <dobey> hadess: i'll try again today sometime [18:28] <Uraeus> msipkema: think the people here who have looked at openml (I have not) think it is not very impressive (in comparisson to our own stuff that is) [18:28] <dobey> hadess: i've been working on encompass ;-) [18:28] <hadess> dobey: cool, i've been sleeping and trying to get this "minimise all windows" proggie to work or so [18:29] <dobey> i got way too much to do today :-/ [18:29] <dobey> a program to minimise all windows? [18:29] <hadess> yeah, somebody has been asking for it, just click, and you get all the windows that appear in the window list to minimise [18:30] <dobey> on all desktops? [18:30] <hadess> i don't know yet, i'm still reading the little library in gnome-core [18:31] <dobey> (custom-set-keymap (quote global-keymap) (iconify-workspace-windows . "M-C-Down")) [18:31] <dobey> add that to ~/.sawfish/custom then do M-C-Down and it iconifies all the windows on the current desktop [18:31] <hadess> i don't use sawfish [18:31] <dobey> oh [18:32] <dobey> i'm sure window manager X has some sort of ability to do the same [18:32] <dobey> if not, it needs to ;-P [18:32] <Uraeus> have you guys seen the latest Procman, it is absolutly great [18:32] <hadess> yeah, but you can do that in a window manager independant way as well [18:32] <dobey> Uraeus: latest as in gnome1? [18:32] <Uraeus> dobey: yes, the 0.99 release [18:32] <dobey> yeah, if you use xatoms [18:32] <dobey> Uraeus: no, does it AA the icons yet? [18:33] <Uraeus> dobey: guess he plans to put out a 1.0 then port to GNOME 2.0 [18:33] <Uraeus> dobey: no it doesn't [18:33] <dobey> it's been ported to gnome 2 already [18:33] <hadess> dobey: the gwm.h or so that's in gnome-core does that without you having to poke with the Xatoms [18:33] <dobey> hadess: ah [18:33] <dobey> hadess: ok :-) [18:34] <dobey> Uraeus: someone already ported it to gnome2 and made it use the new GtkTreeView thingy or whatever it is [18:34] <dobey> Uraeus: andersca did it maybe? [18:34] <hadess> dobey: kevinv did it himself [18:34] <dobey> oh, ok [18:34] <hadess> and i fixored glibtop ;) [18:34] <dobey> hehe [18:35] <dobey> i'm fixxoring encompass this weekend [18:35] <hadess> i need to rebuild my gnome2 platform [18:35] <dobey> and i need to get a shower and go buy some food [18:35] Nick change: dobey -> dobeyshwr [18:35] <dobeyshwr> bbiaf [18:38] <BBB> opinions asked, guys [18:38] <Uraeus> BBB: opinions on what? [18:39] <BBB> v4lsrc - I can build it based on v4l/bttv 0.8.x (current development/future kernel (in a few months) or v4l/bttv 0.7 (current kernel) [18:39] <BBB> which shall I do? [18:39] Action: BBB thinks he'd go for 0.8.x [18:39] <wingo> good call :) [18:40] <BBB> good thing is, 0.8.x can open devices multiple times so I can then completely separate the "view-video" and the "record video" [18:40] <BBB> (just like for v4lmjpegsrc) [18:40] <wingo> nice [18:40] <wingo> is that different than the digital camera interface? (/me is ignorant) [18:41] <Uraeus> BBB: think you are making the right choice since GStreamer will probably not be out with a 1.0 before v4l 0.8 [18:42] <hadess> ooh, damn, it will be easy [18:42] <wingo> s/v4l 0.8/2100/ [18:42] Action: hadess hacks [18:42] <wingo> hack! [18:42] <BBB> uhm [18:43] <BBB> the digicam/v4l is a bit different, afaik [18:43] <BBB> but v4l is bttv-centric anyway [18:43] <wingo> ah. [18:43] <BBB> it's just about streaming capture here (for v4lsrc) [18:43] <BBB> probably webcam-streaming-capture too... [18:47] Action: BBB starts coding v4lsrc [18:51] msipkema (msi...@kf...) left #gstreamer. [18:52] Shippou (no...@cr...) joined #gstreamer. [19:00] Nick change: dobeyshwr -> dobey [19:12] <wtay> uh? [19:12] Action: wtay was away [19:12] <wingo> yo [19:12] <wtay> yo [19:31] vishnu (jo...@cx...) joined #gstreamer. [19:31] <vishnu> yo [19:31] <wtay> aye [19:32] <vishnu> r my mad changes mad or sane? [19:32] <wtay> uhm.. sane I suppose.. [19:32] <vishnu> heh [19:54] <Uraeus> wtay: to bad you was away, we always found the midi expert we have been looking for [19:56] <wtay> I doubt it.. [19:57] <wingo> he was not interested in gst. which is fine. [19:57] <wtay> yeah :) [19:57] <Uraeus> wingo: well not that you tried selling gst to him :) [19:59] <wtay> we don't do midi [19:59] <Uraeus> wtay: but we want to :) [19:59] <wtay> do we? [19:59] <Uraeus> don't we? [20:00] <wtay> dunno, I don't know midi [20:01] <Uraeus> me neither, but I thought the consensus was that we wanted midi, but it was not on the top of the current developers agenda [20:01] <wtay> right [20:06] <ajmitch> morning [20:09] <wtay> evening [20:10] <wtay> ajmitch: another day behind that brainless computer? <g> [20:10] <ajmitch> wtay: no, another exam ;) [20:14] <wtay> what about? [20:15] <ajmitch> physics [20:15] <wtay> oh [20:16] <ajmitch> yeah, lotsa fun [20:17] <wtay> Newton? electricity? [20:18] <ajmitch> nope, that was last semester [20:18] <ajmitch> thermodynamics, optics, quantum stuff this semester... [20:18] <wtay> eek [20:19] <BBB> kewl [20:19] <wtay> but you get to see lots of rainbows then ;) [20:19] <BBB> quantum stuff is kinda interesting in some way [20:19] <ajmitch> yeah [20:22] <ajmitch> after this, it's nearly 2 weeks till my next exam ;) [20:25] bstard (Lor...@a2...) left irc: Ping timeout for bstard[a213-84-40-242.adsl.xs4all.nl] [20:26] <wingo> ajmitch: yeah, mechanics test tomorrow, quantum on tuesday for me :-\ [20:27] <ajmitch> wingo: lucky you ;) [20:27] <ajmitch> wingo: how many years you been at college/university/whatever? [20:28] <wingo> this is #5, and the last one :-))) [20:28] <wingo> the 'victory lap' ;-) [20:28] <ajmitch> hehe, this is my 1st year at uni, only 2 or 3 more to go ;) [20:28] Nick change: tnt-out -> tnt [20:29] <tnt> Good morning. [20:29] <wtay> wingo: victory lap rules :) [20:29] <BBB> who's administrating http://gstreamer.net/dev/whoswho.php [20:29] <BBB> ? [20:30] <Uraeus> BBB: well I and wingo does I guess [20:31] <ajmitch> oh, Uraeus is here... ;) [20:31] <BBB> can you add that I'm from the mjpeg-tools? :P [20:31] <tnt> Uraeus: You should add where everyone if from... that way you know 'when' to connect to IRC to talk to them. [20:32] <ajmitch> heh [20:32] <ajmitch> but some ppl keep weird hours for their timezone... [20:32] <BBB> lol [20:32] <Uraeus> ok, so where is everyone from then? [20:32] <BBB> GMT ;) [20:33] <tnt> ajmitch: Well... it could be used as a 'suggested' time to log on. [20:33] <wingo> north carolina, EST [20:33] <wtay> Uraeus: I'm from timbuktu, you should know that <g> [20:33] <BBB> dennis (Sienpa) and I are from Holland [20:33] <BBB> sienap [20:33] Action: ajmitch is from NZ, and is currently at GMT+1300 ;) [20:33] <wingo> EST=GMT-0500 iirc [20:33] <wingo> damn, 18 hours ajmitch :) [20:34] <ajmitch> wingo: yeah, it's more like a 6hr difference tho (just ignore what day it is ;) [20:34] <wingo> i try, i really do :) [20:34] <ajmitch> haha [20:35] Action: wtay slaps ajmitch for reminding him tomorrow is monday.. [20:35] <ajmitch> it's already monday here :P [20:36] <Uraeus> darn, well I don't really have the time or opportunity to fix that list now [20:36] <wtay> arrhhgg [20:36] <wingo> cvs -d :ext:gstreamer.sf.net:/cvsroot/gstreamer co www/dev/whoswho.php :-) [20:36] Action: ajmitch plans to take a nice study break following this exam... [20:39] <ajmitch> dammit i should get onto gst app coding [20:39] <tnt> Does anyone know how GST_DESTROYED and GST_FLOATING work (... what exactly do they mean)... if they are explained in the 'App Dev Manual' (like half the stuff I ask about :-) ) then just let me know. [20:39] Action: ajmitch would like some python bindings tho ;) [20:40] <wtay> tnt: it's not documented AFAIK [20:40] <wtay> tnt: they have the same sementics as the GTK_* stuff [20:41] <tnt> wtay: So basically, it is floating until you put it in a bin? [20:41] <dobey> hrmm [20:41] <wtay> tnt: yes [20:41] <wtay> hi dobey [20:41] <dobey> hey wtay [20:41] <tnt> wtay: OK. [20:41] <wtay> dobey: how's life? [20:41] <ajmitch> hey dobey [20:42] <dobey> it's ok i guess, i'm still breathing [20:42] <dobey> hey ajmitch [20:43] <dobey> hacking on encompass a bit [20:46] <Uraeus> seems linuxvideo.org is down, anyone know where I can find latest versons of mpeg2dec and ac3dec? [20:46] <tnt> There at SourceForge. [20:47] <Uraeus> ac3dec changed name didn't it? [20:47] <tnt> Ya... I think it is a52 ... I think. [20:48] <tnt> Uraeus: Go to <http://libmpeg2.sourceforge.net/> and <http://liba52.sourceforge.net/> [20:52] <Uraeus> hmm, no new releases from either, well I guess we go with the same libs as for 0.2.1 this time too then [20:54] <ajmitch> Uraeus: planning a release or something? ;) [20:54] BeNOW (an...@ds...) joined #gstreamer. [20:54] <Uraeus> wtay: I am correct that shitowax hasn't updated the quicktime plugin in some time right? so I should stick with the same OQ code as we did the last time? [20:54] <wtay> yo [20:54] <Uraeus> ajmitch: yes [20:54] <wtay> Uraeus: yes, although we might rewrite it for the next release [20:55] <wtay> Uraeus: the plugin, that is.. [20:55] <Uraeus> wtay: ok, I try mailing Shitowax then and see if I can get a commitment from him [20:55] <Uraeus> wtay: they have had what I think is a 1.0 release of OQ since our last release [20:56] <BeNOW> heya [21:00] <Uraeus> hi BeNOW [21:01] <BeNOW> como esta? [21:02] <BeNOW> estoy pretty darn good here... [21:02] <BeNOW> been working on new broadcast backend. [21:04] <wtay> BeNOW: again? <g> [21:04] <BeNOW> heh. [21:04] <BeNOW> yeah, again. [21:04] <wtay> just kidding.. :) [21:05] <BeNOW> actually, still staying with gst, of course, just writing higher level stuff to allow it to plug into the station schedule, etc. [21:05] <BeNOW> ie. when a show is scheduled, events are triggered and the stream is fed from the dj via gst, etc. [21:06] <BeNOW> http://canvas.benow.ca/docs/new/api/ [21:06] <wtay> BeNOW: maybe I should try to set up the same stuff you have on my machine and try to get the bindings in a suable state [21:07] <BeNOW> Yeah, umm, it's a bit of a pig for the stuff I'm using now, but it will be significantly easier in a couple weeks. [21:07] <Uraeus> wtay: should I make the avifile dep, or do ffmpeg give us what we need? [21:08] <BeNOW> I still do plan on using the java bindings, but I'm in 'Get it going as quick as possible' mode... so starting with a wrapped process. [21:08] <wtay> BeNOW: of course [21:08] <wtay> Uraeus: avifile [21:08] Action: Uraeus goes to download latest 'suckyfile' [21:10] dobey (do...@cv...) left irc: Ping timeout for dobey[cv652321-a.brkln1.ma.home.com] [21:23] Action: tnt leaves computer for a while [21:33] Action: wtay download the loki demo launcher [21:35] <wingo> BeNOW: what's your new mtbf? i listened to benow all last night with xmms (and had some truly strange dreams) and it was still going in the morning [21:36] <wingo> which is a novelty for me :-) [21:37] sienap (sy...@s3...) joined #gstreamer. [21:37] <sienap> Hej all! [21:37] <wingo> drunknap! [21:38] <wingo> er, sienap [21:38] <wingo> :) [21:38] <sienap> he geez! i am innocent! [21:38] <sienap> i would never drink! .. i shouldn't.. [21:38] <wingo> er.. won't again.. er.. that's not true either.. ;) [21:39] <sienap> he.. lies! [21:39] <wtay> yo [21:39] <sienap> Yo wim! [21:40] <BeNOW> heya [21:40] <sienap> hadess: how is rhythmbox development going ? :) [21:41] <hadess> sienap: i'm hacking on small stuff these days, gnome-utils and some other stuff [21:41] <sienap> aah ok.. which means rhythmbox is death ?.. [21:41] <sienap> dead that is [21:41] <hadess> i need to finish the gnome2 port of the utils, and then i'll go back to rhyhthbox [21:41] <hadess> certainly not [21:42] <sienap> happily! .. [21:42] <sienap> i personally think gstreamer should do a release again someday soon [21:42] <sienap> it will also attract new people [21:42] <sienap> and force me to fix svgasink [21:42] <Uraeus> sienap: I am making new RPMS of the deps as we speak :) [21:42] <hadess> i still need wtay and omega to fix this probably i had... [21:43] <wingo> problem you mean? [21:43] <wingo> what was it? [21:43] <hadess> problem i had, yeah, man i can't type [21:43] <wtay> gstthread doesn't allow a state change in its own context [21:43] <hadess> nod [21:43] <hadess> but i made a "show desktop" application for gnome ;) [21:43] <wingo> yeah. [21:44] <wingo> nice :-) [21:44] <sienap> ureaus the current distribution is stable ? [21:44] <sienap> i won't be able to code this week since i am busy at school.. [21:44] <sienap> for just a week :) [21:44] <Uraeus> sienap: no, but I thought that if I made a full set of deps for the new release then omega and wtay would feel the pressure :) [21:45] <wtay> Uraeus: you must be kidding.. [21:45] <hadess> wingo: http://hadess.net/files/gnominimise.tar.gz if you care :) [21:45] <sienap> ureaus bastard :) [21:45] <sienap> wim i personally think that a new release would attract a lot of new developers.. [21:45] <Uraeus> and if it doesn't work then Ijust do a set of snapshot RPMS [21:45] <Uraeus> :) [21:45] <sienap> or atleast some needed attention from the everyone else than #gstreamer [21:46] <wtay> I personally think that we should not attract more developpers until things actually work [21:46] <hadess> yeah [21:46] <Uraeus> wtay: well the world waits for noone [21:46] Action: Uraeus feels a little proud of being so philosophical in his reply to wtay [21:46] <sienap> wim hmm both visions can be counted valid [21:47] <wtay> Uraeus: noone waits for a waiting world.. :) [21:48] <sienap> >:) [21:48] <ds> if you build it, they will come [21:48] <Uraeus> wtay: hehe, well you have a point that it could be bad to oversell ourself if we are not ready to deliver [21:48] <BeNOW> I like the new status pages, btw.x [21:48] <Uraeus> ...if I make binaries they will come.... [21:48] <wingo> thanks BeNOW [21:48] <Uraeus> .....now exactly who are they....... [21:49] <wtay> Uraeus: problem is that we ripped some code out but didn't replace it with something that works [21:49] <wingo> the undead!!!! [21:49] <wtay> yet, that is.. [21:50] <wingo> also, attracting the undead isn't in our best interests [21:50] <wingo> at least not in mine :) [21:50] <Uraeus> wtay: ok, I will do a compile and see what works after I have finished my library update, but I want to make gstreamer RPMS for my 'gnomehide expansion pack' [21:50] Action: hadess goes to compile some scrambled eggs [21:50] Nick change: Uraeus -> theNecromancer [21:50] <wingo> haha [21:51] Action: theNecromancer casts a raise dead on wtay to see of he starts to walk [21:51] Action: theNecromancer feels his power growing [21:51] Nick change: theNecromancer -> theLich [21:52] Action: wtay eats theLich's brains [21:52] <wingo> ay, it's too early to start getting silly [21:52] Nick change: theLich -> brainless [21:53] Nick change: brainless -> Uraeus [21:53] <wingo> wb [21:53] <Uraeus> thanks [21:56] steveb (st...@no...) left irc: advance! [21:58] <ds> wtay: what indent settings do you use? [21:59] evil_monk (trevo@208.141.162.68) joined #gstreamer. [21:59] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [21:59] <wtay> ds: -br -bad -cbi0 -cli2 -bls -l100 -ut [22:03] evil_monk (trevo@208.141.162.68) joined #gstreamer. [22:04] <wingo> wtay: when do you run indent? [22:04] <wtay> wingo: sometimes, when the code looks too ugly.. [22:04] <wtay> not always.. [22:04] <wingo> what's your editor? [22:05] <wingo> is it desired that all of gst be indented the same (including the plugins)? [22:05] <wtay> wingo: I use vi [22:06] <wtay> wingo: you should keep the coding standards of the original author [22:07] <wtay> for new plugins I guess it doesn't matter [22:07] <wingo> cool [22:10] Zeenix (zeenix@203.128.13.35) joined #gstreamer. [22:10] <sienap> hej! [22:13] Zeenix (zeenix@203.128.13.35) left irc: Ping timeout for Zeenix[203.128.13.35] [22:15] harobed (harobed@AC9957F1.ipt.aol.com) joined #gstreamer. [22:16] Shippou (no...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [22:17] Nick change: wingo -> wingo-food [22:20] Uraeus (csc...@c1...) left #gstreamer (Client Exiting). [22:27] <BeNOW> I have a friend who's making gst and the make is crapping out on avifile [22:27] <BeNOW> he's got it installed, but it's probably not as gst expects. [22:27] <BeNOW> (i've had this prob myself) [22:27] <sienap> BeNOW i've got the same problem [22:27] <BeNOW> any way around it? [22:28] <sienap> remove avi(something) everywhere in the makefile [22:28] <sienap> the makefile that is in the plugin dir that is [22:28] <sienap> avi is just plain borked [22:28] <sienap> :) [22:28] <BeNOW> fark.k, thanks :) [22:28] <wtay> or use avifile CVS [22:28] <BeNOW> this is fresh from cvs [22:28] <sienap> but that ofcourse.. are lies! :) [22:28] <BeNOW> co gstreamer [22:28] <sienap> BeNOW the avifile CVS! [22:29] <BeNOW> ahh, k. [22:29] <sienap> like in avifile itself [22:29] <BeNOW> hmm. [22:29] <wtay> right [22:29] <wtay> avifile has a tendency to change the API every week or so.. [22:29] <sienap> geez my head is stack smashing [22:29] <sienap> and i have to learn [22:29] <sienap> a lot.. [22:30] <sienap> let's check cvs out as well.. [22:30] <sienap> wim how is the current cvs working ? [22:31] <wtay> same as 0.2.1 mostly [22:32] <wtay> much faster though [22:32] <sienap> mwha i didn't have any good experience with cvs lately [22:32] <sienap> anyway let's check out and hope it works a little agian [22:32] <sienap> again [22:33] <sienap> did you hear anything from parapraxis lately ? [22:33] <wtay> nope [22:33] <sienap> hm.. i did mail him some days ago but didn't get any reply till now.. [22:34] <sienap> too bad [22:50] Zeenix (ze...@13...) joined #gstreamer. [22:51] <wtay> yo [22:51] Jacmet (ja...@bo...) left irc: Ping timeout for Jacmet[bones.kulnet.kuleuven.ac.be] [22:51] <Zeenix> i am still not sure if i could remain connected this time [22:51] <sienap> he [22:52] <Zeenix> wtay: the RtpSend is gioning all well AFAIK [22:52] dobey (do...@cv...) joined #gstreamer. [22:52] <Zeenix> wtay: now the RrpRecv doesnt seem to see the RTCP packet [22:52] Jacmet (ja...@bo...) joined #gstreamer. [22:52] <wtay> Zeenix: nope [22:52] <wtay> Zeenix: did you see the cvs checkings I did? [22:53] <sienap> dobey / jacmet! [22:53] <Zeenix> wtay: no, i am talking of my librtp, i am sending it to you.... [22:53] <dobey> hi sienap [22:55] <dobey> what's up? [22:56] Jacmet (ja...@bo...) left irc: Ping timeout for Jacmet[bones.kulnet.kuleuven.ac.be] [22:56] <Zeenix> wtay: sent [22:57] <Zeenix> wtay: now sending the latest rtp plugin... [22:57] <dobey> Insert Coin to Continue [22:57] <ds> plonk [22:58] <Zeenix> wtay: sent [23:00] <sienap> inserting coin [23:00] <sienap> Hej dobey! [23:00] <sienap> how are you ? [23:01] <dobey> good [23:02] <Zeenix> wtay: plz check that am i giving the right fds to fd_set in get() ? [23:03] <Zeenix> anybody seen the person with nash as nick here today ? [23:03] Nick change: hadess -> hds-afk [23:04] <sienap> dobey.. cool [23:04] <dobey> how are you? [23:04] <sienap> mwha ;) [23:04] <sienap> fux0ring school up again :) [23:04] <dobey> hrmm [23:05] <ds> wtay: remember that patch I had for audioscale that I never checked in? It's ready to go now, plus adds a couple of new scaling methods [23:05] <wtay> cool [23:06] <ds> shall I check it in then? [23:06] <wtay> yup [23:06] <wtay> I need a scaler for librtp :) [23:06] <sienap> using rtp for network stuff ? [23:06] <Zeenix> sienap: yes [23:06] <sienap> great stuff[tm] [23:07] <Zeenix> sienap: mulitmedia + network infact [23:08] <wtay> Zeenix: I did quite a lot of cleanups, lemme send you the code I have.. [23:08] <wtay> Zeenix: also I have some questions.. [23:08] <Zeenix> wtay: you changed the one i just sent you ? [23:09] <wtay> Zeenix: nope, the one before that [23:09] <wtay> Zeenix: it works perfect here now. aside from some minor stuff [23:10] <Zeenix> wtay: started to work after the librtp files i sent you ? [23:10] <wtay> Zeenix: nope, I fixed it this afternoon [23:11] <wtay> sent [23:11] <wtay> Zeenix: also, the payload type has a very specific meaning.. [23:11] <wtay> Zeenix: like GSM is gsm at 8Khz [23:12] <wtay> Zeenix: also I would move the RTCP handling out of the plugin, into the app [23:12] <wtay> Zeenix: the RTCP stuff is perfect to add/remove people from the conference call etc [23:13] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-49.brain.net.pk] [23:13] <wtay> doh [23:13] Zeenix (ze...@13...) joined #gstreamer. [23:13] <ds> ok, checked in. [23:13] <wtay> ah [23:13] <Zeenix> wtay: repeat last messages if any [23:14] <wtay> Zeenix: nope, I fixed it this afternoon [23:14] <wtay> Zeenix: also I would move the RTCP handling out of the plugin, into the app [23:14] <wtay> Zeenix: also, the payload type has a very specific meaning.. [23:14] <wtay> Zeenix: like GSM is gsm at 8Khz [23:14] <wtay> Zeenix: the RTCP stuff is perfect to add/remove people from the conference call etc [23:14] <ds> right now, the sinc method is slow, slow. I have ideas to make it faster, but being right is better [23:14] <Zeenix> wtay: no [23:14] <wtay> not nessesarily in that order :-) [23:15] <Zeenix> wtay: The gsm has a rate in it that is translated seperatelly then the payload [23:15] <Zeenix> s/payload/payload_t [23:16] <wtay> Zeenix: ? not according to RFC 1890 [23:16] <Zeenix> wtay: no, i meant what i am doing [23:17] <wtay> Zeenix: yes, but that needs a RTCP negotiation first [23:17] <Zeenix> wtay: what were you trying to say, payload type has a very specific meaning or it should be like this [23:17] <Zeenix> wtay: yes of course [23:18] <Zeenix> wtay: that should be handled by the app, by any means they want it to happen [23:18] Shippou (no...@ot...) joined #gstreamer. [23:18] <Zeenix> wtay: we are using rtcp for caps negociation [23:18] <Zeenix> only [23:18] <wtay> ok [23:18] <Zeenix> wtay: & if the peer element goes to NULL state [23:18] <wtay> but RTCP is much more [23:19] <Zeenix> wtay: yes [23:19] <Zeenix> wtay: no body is stopping the app programmer to use that [23:19] <wtay> some parts of it, like the add/leave could be used by the plugins [23:19] <Zeenix> wtay: later [23:19] Nick change: wingo-food -> wingo [23:20] <Zeenix> wtay: ATM i cant get the RtpRecv get the RTCP packet [23:20] harobed (harobed@AC9957F1.ipt.aol.com) left irc: Client Exiting [23:20] <Zeenix> wtay: did you checked the fd_set in the latest code ? [23:21] <wtay> but the RTCP messages are sent out fine, no? [23:22] <wingo> what's the goal of audioscale? [23:22] <Zeenix> wtay: yes here too, but not recieved [23:22] <wtay> Zeenix: I think the type is not set correctly in rtpsend.c [23:23] <wtay> Zeenix: 'cause I get this Unsupported RTCP Packet errors.. [23:23] <Zeenix> wtay: it is [23:23] <Zeenix> wtay: what are you connecting it with [23:23] <wtay> gsmenc [23:23] <ds> wingo: resampling audio at a different rate [23:23] <Zeenix> wtay: you shouldnt get this [23:24] <Zeenix> wtay: i connected it with gsmenc & it told me correctly [23:24] <wtay> ds: does it sound good? [23:24] <wtay> ./gstreamer-launch filesrc location=/opt/data/south.mp3 ! mad ! stereo2mono ! gsmenc ! rtpsend [23:24] <wtay> ../../tools/gstreamer-launch rtprecv ! gsmdec ! osssin [23:24] <wingo> ds: i wish that [23:25] <wingo> 1) i would have known that before i made speed a long time ago [23:25] <wingo> 2) [23:25] <wingo> that somehow the work could be integrated [23:25] <ds> wtay: longer filter length sound better, of course, but I can't tell a difference at about 16 [23:25] <wingo> it's stupid that we made two things that do the same thing [23:25] <ds> i thought speed just changed the numerical sampling rate [23:26] <wingo> sampling rate on what? [23:26] <wingo> oh [23:26] <ds> i.e., you want to change pitch, change the rate [23:26] bstard (Lor...@a2...) joined #gstreamer. [23:26] <wingo> is that not essentially the same operation as resampling? [23:26] <wingo> except w/o setting caps [23:27] <sienap> bstard! [23:27] <wingo> i have to run, back in a bit [23:27] Nick change: wingo -> wingo-afk [23:27] <ds> I thought speed, e.g., took something recorded at 44100 hz and played it at 45khz, or such. [23:27] <sienap> lordsmegma.. geez :) no comment. [23:27] Zeenix2 (ze...@13...) joined #gstreamer. [23:27] <wingo-afk> ds: try test/testspeed [23:27] <ds> maybe I should look at it [23:28] <Zeenix2> i was disconnected since my last message [23:28] <wtay> Zeenix2: I get a 204 and then an unknown payload_t [23:28] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-61.brain.net.pk] [23:28] Nick change: Zeenix2 -> Zeenix [23:28] <Zeenix> wtay: a 204 ? [23:29] <wtay> RTCP_APP [23:29] <Zeenix> wtay: strange, i used to get an RTP packet & then a segfault [23:30] dobey (do...@cv...) left irc: Ping timeout for dobey[cv652321-a.brkln1.ma.home.com] [23:31] <Zeenix> wtay: it seemed that select couldnt detect change on rtcp socket [23:31] <wtay> Zeenix: that was before I fixed your code from yesterday, yes [23:31] dobey (do...@cv...) joined #gstreamer. [23:31] <wtay> Zeenix: hmm, the code you just sent me still has the bug [23:32] <Zeenix> wtay: so what is the current problem, the rtpsend doesnt sense the correct payload_t [23:32] <wtay> Zeenix: wait.. let's sync first [23:32] <wtay> Zeenix: you sent me code yesterday.. [23:32] <wtay> Zeenix: you asked me to find out was was wrong.. [23:33] <Zeenix> yes [23:33] <wtay> Zeenix: rtp_receive in rtp.c doesn't copy the data into the buffer was the problem [23:33] <wtay> Zeenix: that was the problem with the segfault [23:33] <Zeenix> wtay: sizeof problem, i am not experienced with C [23:34] <Zeenix> wtay: ? [23:34] <wtay> Zeenix: not really, you have to allocate space for GST_BUFFER_DATA and memcpy the rtp packet into it [23:34] <wtay> which is a flaw in rtp btw.. [23:35] <Zeenix> wtay: oh [23:35] <Zeenix> wtay: btw means between ? [23:35] <wtay> by the way :) [23:36] <wtay> anyway, I changed a couple of things and commited it to cvs [23:36] <wtay> and I mailed you a copy of that code [23:37] <Zeenix> this was the story of the code i sent you yesterday [23:37] <Zeenix> ? [23:37] <wtay> yes, but I see the same problem in the code of today [23:37] <ds> wingo-afk: uh, yeah. it's the same [23:38] <Zeenix> wtay: ok, but you didnt made any bigger change, right ? [23:38] <wtay> not really, lemme tell you what I changed: [23:38] <wtay> 1) added payload_to_caps method to convert a payload type into caps [23:38] Shippou (no...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [23:38] Action: Zeenix admires wtay for keeping a log of what he does [23:39] <wtay> 2) renamed recieve to receive [23:39] <wtay> 3) modified the padtample factory [23:39] <wtay> 4) slightly changed rtp_receive so that it retruns a Rtp_Packet [23:40] <wtay> 5) properly allocate and copy the packet data into a buffer [23:40] <wtay> then some interesting changes... [23:41] <wtay> 6) if there is no RTCP message with app specific data, we use rate==8KHz as the audio rate for GSM and 44100 for L16 [23:41] <wtay> some small things: [23:41] <wtay> 7) used the PLUGIN_FEATURE api required for cvs gstreamer [23:42] <wtay> oh yes, the mime type is audio/x-gsm, not gsm/x-audio [23:42] <Zeenix> yeah i did that [23:42] <wtay> 8) used the glib2 signal API [23:43] <wtay> that's it.. nothing major, it sorta worked out of the box after fixing the buffer thingie [23:44] <wtay> I didn't touch the RTCP code, so the bug is still there I guess [23:44] <Zeenix> wtay: & replaced my name with Wim Taymans .... <g> [23:44] <wtay> naah :) [23:45] <Zeenix> wtay: "if there is no RTCP message with app specific data, we use rate==8KHz as the audio rate for GSM and 44100 for L16" [23:45] <Zeenix> wtay: how do we know whats connected with our src pad [23:45] <Zeenix> ? [23:45] <wtay> Zeenix: that's in the newcaps function [23:45] <wtay> of rtpsend [23:46] <Zeenix> yes, your sentence seems to be about RtpRecv [23:46] <Zeenix> if there is no RTCP message with app specific data [23:47] <Zeenix> part of your sentence [23:47] <wtay> I'm assuming that a compliant RTP app that is sending data with the GSM payload type is going to use 8KHz but won't do the app specific RTCP messages [23:47] <wtay> Zeenix: ah yes [23:47] <wtay> Zeenix: did you get my mail? [23:48] <Zeenix> wtay: yes but just one [23:48] <wtay> ok, that's good [23:48] <Zeenix> wtay: i'll talk to the hardware person about a modem soon [23:49] vishnu (jo...@cx...) left irc: using sirc version 2.211+4KSIRC/1.1 [23:49] <wtay> Zeenix: rtcp_receive looks wrong too.. [23:49] <Zeenix> wtay: ? [23:49] <wtay> Zeenix: it allocates a struct on the stack and returns [23:49] Action: ds growls [23:49] <wtay> s/$/it/ [23:50] <Zeenix> can see a $ in you message [23:50] <ds> someone put i386 asm in test/memchunk/gstmemchunk.c [23:50] <Zeenix> s/can/cant [23:50] <wtay> Zeenix: it looks like java :) [23:50] <wtay> ds: so? [23:51] <wtay> Zeenix: $ == EOL [23:52] <wtay> ds: or does it compile no matter what arch? [23:52] <ds> wtay: so it doesn't compile on ppc [23:52] <wtay> is it compiled by default? [23:52] <Zeenix> wtay: it should use g_malloc & return the pointer which the calling func. should free after using it, right ? [23:52] <ds> yes [23:52] <wtay> Zeenix: yup [23:52] <wtay> ds: oops :) [23:53] dobey (do...@cv...) left irc: Ping timeout for dobey[cv652321-a.brkln1.ma.home.com] [23:53] <ds> =) on the same subject, plgins/capture/qcam is i386-specific, roo [23:53] <ds> s/roo/too [23:54] <wtay> ds: uhm.. [23:54] <wtay> ds: but it should compile, no? [23:55] <ds> no, since inb() and outb() are not defined [23:55] <wtay> oh [23:55] <Zeenix> wtay: "I didn't touch the RTCP code, so the bug is still there I guess", which bug ? [23:55] <wtay> I didn't know that [23:56] <wtay> Zeenix: I think the one in rtcp_receive [23:56] <wtay> Zeenix: the payload type cannot arrive correclty since it was allocated on the stack [23:57] dobey (do...@cv...) joined #gstreamer. [23:57] <ds> hmmm... it seems as though qcam is accessing ports directly instead of using raw printer port access [23:57] <Zeenix> wtay: ok [23:59] <wtay> ds: yes.. [00:00] --- Mon Oct 29 2001 [00:00] <Zeenix> wtay: is 64k code mem enough space for low-level codes [00:00] <Zeenix> ? [00:00] <wtay> Zeenix: you have no idea what can be put into the 64k memory of a commodore 64 :) [00:02] <Zeenix> wtay: that compiler( sdcc.sf.net ) i found for MCS51 microcontrollers outputs a file at the end, about which they say: [00:02] <Zeenix> the payload type cannot arrive correclty since it was allocated on the stack [00:03] <Zeenix> ugghhhh..., errr [00:03] <Zeenix> wtay: ignore that, i copy/pasted the wrong thing [00:03] <Zeenix> the say: sourcefile.ihx - The load module in Intel hex format [00:03] <Zeenix> s/the/they [00:04] <Zeenix> wtay: but the file util tells that its text file & it seems like one to me too [00:05] <wtay> I wouldn't know [00:05] <wtay> maybe it is hex expanded code [00:05] <Zeenix> can you gues, if i sen... [truncated message content] |
From: IRC B. <wt...@us...> - 2001-10-30 05:27:57
|
******************************************************************* [03:12] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [03:20] wingo (wi...@rd...) left irc: going home... [03:27] mwc (ma...@ly...) left irc: [x]chat [03:35] evil_monk (trevo@208.141.162.68) joined #gstreamer. [03:58] <taaz> ds? [03:58] <ds> yep [03:59] <ds> Do you mind if I upload some of the debian packages used for gstreamer dependencies? I was thinking mainly of a52dec [04:00] <ds> mpeg2dec, too [04:00] <ajmitch> when will gstreamer be in sid, taaz? ;) [04:02] <taaz> um... err.. [04:02] <ajmitch> hehe [04:02] <taaz> so like i want to change something first [04:02] <taaz> and i've got another issue [04:02] <taaz> i asked about this on the debian-mentors list the other day [04:02] <ajmitch> ok [04:03] <ds> such as? [04:03] <taaz> i think the dev packages should be versioned [04:03] <taaz> cause right now they conflict with versioned (theoretical) liba ;) [04:04] <taaz> so like rename to liba52-0-dev or something [04:04] <taaz> starts to look silly but i think it's needed [04:04] <taaz> and the other thing is that someone brought up the issue of netbsd letter from dolby [04:04] <ds> I wouldn't worry about it (liba) [04:04] <taaz> so netbsd doesn't allow this ac3 code [04:05] <taaz> but like there are already a number of other packages in debian that already have ac3 decode code [04:05] <taaz> anyway, i've been really busy... been intending to ask -legal about it [04:06] wtay-zZz (wi...@ca...) left irc: Ping timeout for wtay-zZz[cable-195-162-215-201.upc.chello.be] [04:08] wtay-zZz (wi...@ca...) joined #gstreamer. [04:08] wingo-z (wi...@rd...) joined #gstreamer. [04:08] <wingo-z> yop [04:09] Nick change: wingo-z -> wingo [04:09] <wingo> yowsa [04:35] <taaz> wingo: were you sleeping? [05:12] <wingo> naw [05:12] <wingo> just re-joined [05:12] <wingo> that was my old nick [05:12] <wingo> although i might become that nick in a bit ;) [05:12] <ajmitch> heh [05:14] <ajmitch> irc is so quiet today... [05:16] <wingo> that's so you notice things more, silence is a mirror -- we're doing this on purpose ;) [05:17] <ajmitch> lol [05:17] Action: ajmitch just remembered there's a LUG meeting tonight... [05:27] <tnt> What software is being used to log thes IRC sessions? [05:29] Action: tnt thinks everyone might be sleeping... [05:38] chillywilly (da...@d4...) joined #gstreamer. [05:44] arik (ar...@sd...) joined #gstreamer. [05:44] <arik> hi [05:45] <tnt> Hello arik. [05:46] <arik> howdy [05:59] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [06:01] arik (ar...@sd...) left irc: leaving [06:09] JOSF (jo...@li...) joined #gstreamer. [06:09] <JOSF> hi [06:10] <tnt> Hello. [06:11] Nick change: wingo -> wingo-z [06:52] chillywilly (da...@d4...) left irc: Ping timeout for chillywilly[d44.as12.nwbl0.wi.voyager.net] [06:52] chillywilly (da...@d1...) joined #gstreamer. [06:57] JOSF (jo...@li...) left irc: See you soon or not. [07:00] chillywilly (da...@d1...) got netsplit. [07:00] BBB-[away] (BB...@uc...) got netsplit. [07:01] BBB-[away] (BB...@uc...) returned to #gstreamer. [07:01] chillywilly (da...@d1...) returned to #gstreamer. [07:18] Nick change: taaz -> taazzzz [07:47] mef-zZz (me...@us...) left irc: [07:48] steveb (st...@no...) joined #gstreamer. [07:51] walken (fo...@c1...) joined #gstreamer. [07:51] <walken> hi [07:52] <tnt> Hey walken. [07:52] <walken> hi [08:25] <tnt> Hello?... anyone listening? [08:38] omega (om...@om...) joined #gstreamer. [08:39] <walken> hi omega [08:39] <omega> yo [08:39] <omega> walken: I have a response to your roadmap stuff, but got stalled on a problem. lemme trim that section and send it [08:40] <walken> yes... [08:40] Action: walken doing some work on libmpeg2 right now [08:40] <walken> heh [08:40] <walken> girlfriend eats time [08:40] <walken> not that I complain though :) [08:45] <omega> ack, 3 simultaneous thought streams [08:45] <omega> (here, benow irc, and the mail) [08:45] <walken> what do you have comments on ? [08:45] <omega> the whole API thing [08:45] <walken> the api, or some other stuff ? [08:45] <omega> I'v written down my ideas at a high level on API design [08:46] <walken> ok [08:46] <walken> cool [08:46] <walken> I'm looking at api ideas :) [08:46] <walken> most probably I'm going for a lower level API since most people rewrite their own... [08:47] <walken> I may have to also do a higher level api for people who dont want to mess up with things though [08:47] <omega> right, stay tuned <g> [08:47] <walken> hehe [08:55] <omega> ok, almost done [08:55] <omega> I'll get to other issues in another mail... [08:57] <omega> sent [08:58] <walken> not received yet [08:58] <omega> I have it [08:58] <omega> 11.4K [08:58] <omega> I pulled 2.5K out cause it was broken [08:59] <walken> oh - in libmpeg - ok [08:59] <omega> yeah [08:59] <walken> I was looking in inbox :) [08:59] <omega> doh [09:03] <tnt> omega: Do you know what software is being used to log this IRC channel? [09:03] <omega> eggdrop afaik [09:03] <tnt> Thanks. [09:05] <omega> walken: too busy reading? <g> [09:07] <walken> I just finished reading it :) [09:07] <walken> well its mostly nice :) [09:08] <omega> mostly? [09:08] <walken> the "lowlevel api" will be higher level than the lowest you propose I think [09:08] <omega> hmm, there's a reason I suggest that low-level an API [09:08] <walken> like my idea of a lowlevel api would be mainly the various parse_* functions and the slice_process function basically [09:09] <tnt> what's the policy on what timestamp a filter should attach to its output? [09:09] <omega> theoretically whatever the input was [09:09] <tnt> What if there were two inputs with two different timestamps? [09:10] <omega> then you have a problem <g> [09:10] <tnt> :-) [09:10] <tnt> Where is the time stamp originally gotten from? [09:11] <omega> depends [09:11] <tnt> What if the input is from a file? [09:11] <tnt> Then how is the time stamp generated? [09:12] <omega> depends on quite a few things [09:12] <omega> generally the parser will set the timestamps [09:12] <tnt> Is the time stamp always the same, for the same file... or is it different everytime the parser is run? [09:12] <omega> it had better be the same <g> [09:13] <tnt> OK... I think I understand its use. [09:13] <omega> timestamp is definitely a stream-time thing [09:14] <walken> omega. can you elaborate on lower-level interfaces ? [09:14] <omega> um, in what sense? [09:15] <walken> why do we need access to the MC and IDCT and stuff functions [09:15] <omega> because what if you want to write a program to make use of that info? [09:15] <omega> or more likely, what if you have a radeon or i810 with hw accel? [09:16] <walken> hmmm [09:16] <omega> for instance, if recoding to mpeg4... [09:16] <omega> I want the motion vectors from the mpeg2, since they're a pretty good hint [09:16] <omega> esp from a DVD [09:16] <omega> (one hope) [09:16] <omega> er, hopes) [09:16] <walken> ok - so whats needed is more like a way to set these callbacks ? [09:16] <omega> which callbacks? the bitstream stuff? that's dealt with by my bitstream library [09:17] <omega> there's really only one: "need more data!" [09:17] <walken> no, I meant idct callback [09:17] <omega> oh, um, no [09:17] <walken> have the app say "call this function whenever you need to do an idct" [09:18] <omega> ok, lemme write some example code.... [09:19] <walken> http://advogato.org/ <- doh, its looking special today [09:20] Action: omega has trouble loading it [09:22] <omega> heh, robla is the Real guy [09:24] <walken> hmmm [09:25] <walken> what browser do you use ? [09:25] <omega> ns4.78 [09:25] <walken> I'm still using netscape too.. [09:25] <walken> I was considering switching to either mozilla or the other one... I always forget its name [09:25] <omega> galeon? [09:25] <walken> galeon :) [09:32] arik (ar...@sd...) joined #gstreamer. [09:32] <arik> lo [09:33] <omega> yo [09:33] <arik> hey [09:33] <arik> what's up? [09:33] <omega> discussing codec API issues with walken [09:33] Action: arik is still unable to get eazel-hacking working and thus unable to build gnome2 [09:33] <arik> ah [09:33] <arik> neet [09:34] <omega> killall -9 netscape again ;-( [09:34] <omega> at least I have a 'kns' script to limit typing [09:34] <arik> heh [09:34] <arik> i should do that [09:34] <arik> i just end up ctrl+r [09:34] Action: omega hope she'll never need kmoz [09:34] <arik> and then searching for netscape [09:34] <omega> s/hope she/hopes he/ [09:35] <arik> heh [09:35] <arik> i was wondering [09:35] <omega> stupid space bar is lagging [09:35] <arik> ha [09:35] <arik> er ah [09:35] <omega> er, no, early [09:35] <omega> ok, n/m <g> [09:35] <arik> hehe [09:35] <arik> as soon as i get gnome2 building [09:35] Action: omega waits [09:35] Action: omega turns blue [09:35] <arik> i'm gonna rewrite gstmediaplay with it and then commit that as a new module [09:35] <arik> heh [09:35] Action: omega keels over [09:36] <arik> hahaha [09:36] <arik> i really want to get gstplay in cvs [09:36] <omega> any chance it'll compile against gtk1.2 ? [09:36] <arik> um [09:36] <arik> i doubt it [09:36] <arik> not if i write it for gtk2 :-) [09:36] <arik> why? [09:37] <omega> any idea when gtk2 will be released? [09:37] <arik> heh [09:37] <arik> um [09:37] <arik> sometime around the first of the year i think [09:37] <arik> i could be wrong :-P [09:37] <BBB-[away]> 1.3.10 is available...... [09:37] <arik> when gnome2 comes out i suppose [09:38] <arik> but i mean [09:38] <arik> noone uses gstmediaplay anyway [09:38] <arik> and i would have to rewrite it for that eventually [09:38] <arik> might as well do it now, no? [09:38] <omega> yeah [09:38] <BBB-[away]> <arik> noone uses gstmediaplay anyway <-- ? I do.... [09:38] Nick change: BBB-[away] -> BBB [09:38] Action: BBB is back (gone 09:14:29) [09:38] <arik> that way hopefully it will be pretty rock solid for gnome2 [09:39] <arik> BBB: yeah. but you are prob willing to build gnome2 to use it :-) [09:39] <BBB> uhm....... [09:39] <arik> heh [09:39] <arik> maybe not :-) [09:39] <BBB> can I install gnome1.4 and gnome2 together on 1 system? [09:39] <arik> yes [09:39] <BBB> then I'll install gnome2 [09:39] <BBB> ;) [09:39] <arik> hehe [09:40] <arik> omega: i could just port the current gstmediaplay over to the new lib and commit then do the rest [09:40] <arik> but... [09:40] <arik> the problem with that is that every time i try to do that [09:40] <arik> i end up with an app that segfault on startup [09:40] <arik> and i end up having to rewrite most of it anyway [09:40] <arik> i've started that like 3 times [09:40] <arik> and got frustrated [09:40] <arik> :-) [09:45] <omega> heh [09:46] <omega> arik: didn't someone say you had to have rh7.2 or something to put gnome2 on at the same time? [09:46] <omega> or did the 70MB ximian update I did deal with that? [09:47] <arik> i have no idea [09:47] Action: omega wonders why red-carpet is getting 5-8KB/sec from just about everywhere [09:47] <arik> i'm using eazel-hacking which lets you put it on the same system through mirrors and hidden switches [09:47] <arik> i just know that you _can_ install the havoc rpms and have both on the same system [09:48] <arik> (well actually, by creating a seperate prefix and path and ld* and installing everything gnome in it) [09:48] <tnt> arik: Are you talking about gnomehide? [09:48] <arik> tnt: yeah [09:48] <tnt> There's a new mailing list for that: gno...@gn... I think. [09:48] <arik> ah [09:49] <arik> i don't really pay any attention toit [09:57] <omega> walken: sent some examples to the list [10:02] <walken> hmmm [10:02] <walken> I'm having trouble mapping the current structure of the code to that though :) [10:02] <walken> like... [10:02] <walken> say you have skipped macroblocks [10:02] <omega> yeah, I said it was rough <g> [10:03] <walken> :) [10:03] <omega> look at libmpeg2 in codecs.org for a better chunk of code written that way [10:03] <omega> I suppose I should have mentioned that [10:03] <omega> however, libmpeg isn't *quite* the style I'm proposing, but it's really close [10:05] arik (ar...@sd...) left irc: sleeep [10:05] Action: omega wonders what this dancing thing is all about... [10:05] Action: walken gets ready to sleep instead [10:06] <omega> walken: you're in range of Oakland, right? [10:07] <walken> I'm in redwood city and working in palo alto [10:07] <omega> but in range? [10:07] <walken> oh yeah [10:07] Action: omega has little knowledge of geography down there [10:08] <walken> (about 1h drive) [10:08] <omega> ok, message coming down pipe on gst-devel [10:09] <omega> in fact, I'll be soliciting recommendations on how to get from Oakland to Treasure Island totally via public transportation..... [10:15] _gst_newt_ joined #gstreamer. [10:15] chillywilly (da...@d1...) left irc: [10:15] ajmitch (aj...@p4...) got netsplit. [10:15] wtay-zZz (wi...@ca...) got netsplit. [10:15] walken (fo...@c1...) got netsplit. [10:21] ajmitch (aj...@p4...) got lost in the net-split. [10:21] walken (fo...@c1...) got lost in the net-split. [10:21] wtay-zZz (wi...@ca...) got lost in the net-split. [10:25] _gst_newt_ joined #gstreamer. [10:29] steveb (st...@no...) got netsplit. [10:29] steveb (~st...@no...) returned to #gstreamer. [10:30] BBB_ (BB...@uc...) got netsplit. [10:30] ds_ (ds...@c7...) got netsplit. [10:30] omega (om...@om...) got netsplit. [10:30] BBB_ (~BB...@uc...) returned to #gstreamer. [10:31] omega (~omega@216.99.212.251) joined #gstreamer. [10:32] Nick change: tnt_ -> tnt [10:32] <omega> ok, I'm making noise about it [10:33] <tnt> NickServ is back up/ [10:34] <omega> yeah, I noticed ;-( [10:34] <omega> I'm making noise in #openprojects about this [10:36] ds_ (~ds...@c7...) got lost in the net-split. [10:36] <omega> ok all, /msg nickserv identify [10:36] <omega> then /mnsg nickserv set kill off [10:39] <omega> ok, they finally seem to get what the problem is... [10:42] #gstreamer: mode change '+o omega' by ChanServ!Cha...@se... [10:42] <omega> ok, that still works... [10:43] #gstreamer: mode change '-o omega' by omega!~omega@216.99.212.251 [10:56] Nick change: BBB_ -> BBB [11:04] omega (omega@216.99.212.251) got netsplit. [11:04] BBB (BB...@uc...) got netsplit. [11:04] steveb (st...@no...) got netsplit. [11:04] wingo-z (wi...@rd...) got netsplit. [11:04] wtay-zZz (wi...@ca...) got netsplit. [11:04] BeNOW (an...@ds...) got netsplit. [11:04] taazzzz (dlehn@66.37.66.32) got netsplit. [11:04] tnt (tn...@h2...) got netsplit. [11:04] taazzzz (dlehn@66.37.66.32) returned to #gstreamer. [11:04] omega (~om...@om...) joined #gstreamer. [11:04] BeNOW (an...@ds...) returned to #gstreamer. [11:04] BBB (~BB...@uc...) returned to #gstreamer. [11:04] steveb (~st...@no...) returned to #gstreamer. [11:04] wtay-zZz (~wi...@ca...) returned to #gstreamer. [11:04] wingo-z (~wi...@rd...) returned to #gstreamer. [11:05] <omega> wb... [11:05] Action: omega sleeps [11:05] omega (om...@om...) left irc: Client Quit [11:06] tnt (tn...@h2...) returned to #gstreamer. [11:30] _gst_newt_ joined #gstreamer. [11:30] wingo-z (wi...@rd...) got netsplit. [11:30] wtay-zZz (wi...@ca...) got netsplit. [11:30] ajmitch (aj...@p4...) got netsplit. [11:30] BBB (BB...@uc...) got netsplit. [11:30] taazzzz (dlehn@66.37.66.32) got netsplit. [11:30] taazzzz (dlehn@66.37.66.32) returned to #gstreamer. [11:31] BBB (~BB...@uc...) returned to #gstreamer. [11:31] wtay-zZz (~wi...@ca...) returned to #gstreamer. [11:31] ajmitch (aj...@p4...) returned to #gstreamer. [11:31] wingo-z (~wi...@rd...) returned to #gstreamer. [12:00] jarl_ (LordSmeg@213.84.40.242) got netsplit. [12:01] jarl_ (~Lor...@a2...) joined #gstreamer. [12:09] Shippou (ir...@ot...) joined #gstreamer. [13:24] jarl_ (Lor...@a2...) got netsplit. [13:24] jarl_ (~Lor...@a2...) returned to #gstreamer. [13:35] _gst_newt_ joined #gstreamer. [14:33] Last message repeated 1 time(s). [14:33] asmod (~stevec@64.5.222.2) joined #gstreamer. [14:36] Shippou (ir...@ot...) got netsplit. [14:36] BBB (BB...@uc...) got netsplit. [14:36] ajmitch (me...@p4...) got netsplit. [14:36] BBB (~BB...@uc...) returned to #gstreamer. [14:36] ajmitch (~me...@p4...) returned to #gstreamer. [14:42] Shippou (ir...@ot...) got lost in the net-split. [14:43] BBB (BB...@uc...) got netsplit. [14:43] asmod (stevec@64.5.222.2) got netsplit. [14:43] wtay-zZz (wi...@ca...) got netsplit. [14:43] wingo-z (wi...@rd...) got netsplit. [14:43] BBB (~BB...@uc...) returned to #gstreamer. [14:44] wtay-zZz (~wi...@ca...) returned to #gstreamer. [14:44] vektor (reet@HSE-Kitchener-ppp3506142.sympatico.ca) joined #gstreamer. [14:44] wingo-z (~wi...@rd...) returned to #gstreamer. [14:49] asmod (~stevec@64.5.222.2) got lost in the net-split. [14:54] asmod (~stevec@64.5.222.2) joined #gstreamer. [15:02] jarl_ (Lor...@a2...) got netsplit. [15:03] asmod (stevec@64.5.222.2) got netsplit. [15:03] wtay-zZz (wi...@ca...) got netsplit. [15:03] wingo-z (wi...@rd...) got netsplit. [15:03] BBB (BB...@uc...) got netsplit. [15:03] asmod (~stevec@64.5.222.2) returned to #gstreamer. [15:08] jarl_ (~Lor...@a2...) got lost in the net-split. [15:09] BBB (~BB...@uc...) got lost in the net-split. [15:09] wtay-zZz (~wi...@ca...) got lost in the net-split. [15:09] wingo-z (~wi...@rd...) got lost in the net-split. [15:10] _gst_newt_ joined #gstreamer. [15:14] asmod (stevec@64.5.222.2) joined #gstreamer. [15:37] vektor (reet@HSE-Kitchener-ppp3506142.sympatico.ca) joined #gstreamer. [15:37] <vektor> boo! [15:47] wingo (wingo@24.25.21.76) joined #gstreamer. [15:47] <wingo> yo [16:09] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:32] Zygo (zbl...@24...) joined #gstreamer. [16:34] taaz (dlehn@66.37.66.32) joined #gstreamer. [16:39] Shippou (ir...@ot...) joined #gstreamer. [17:07] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [17:09] evil_monk (trevo@208.141.162.68) joined #gstreamer. [17:11] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [17:17] evil_monk (trevo@208.141.162.68) joined #gstreamer. [17:18] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [17:19] evil_monk (trevo@208.141.162.68) joined #gstreamer. [17:28] sienap (sy...@s3...) joined #gstreamer. [18:23] <wingo> snack break... [18:33] steveb (st...@no...) joined #gstreamer. [18:41] <wingo> steveb: did you see that email from steve harris on l-a-d? [18:45] steveb (st...@no...) left irc: Ping timeout for steveb[node1ee11.a2000.nl] [18:45] steveb (st...@no...) joined #gstreamer. [19:02] Uraeus (csc...@c1...) joined #gstreamer. [19:02] <Uraeus> hi dudes [19:11] <wingo> hey Uraeus [19:23] ds_ (ds...@c7...) joined #gstreamer. [19:24] Nick change: ds_ -> ds [19:34] ajmitch (me...@p4...) joined #gstreamer. [19:34] <Uraeus> hi ajmitch [19:34] <ajmitch> hi [19:36] BBB (BB...@uc...) left irc: Ping timeout for BBB[ucu-104-70.ucu.uu.nl] [19:42] <steveb> wingo: on int to float? yeah. I've been waiting for omega's optimised int/float conversion routines to materialise [19:43] Uraeus (csc...@c1...) left irc: Ping timeout for Uraeus[c128s9h5.upc.chello.no] [19:47] asmod (stevec@64.5.222.2) left irc: [x]chat [19:52] asmod (stevec@64.5.222.2) joined #gstreamer. [19:53] steveb (st...@no...) left irc: Ping timeout for steveb[node1ee11.a2000.nl] [19:55] Uraeus (csc...@c1...) joined #gstreamer. [20:13] Nick change: Uraeus -> Ura_gym [20:14] <sienap> http://www.nat.org/evolution2.php3 hahahaha!!!!!! :) [20:17] <Ura_gym> sienap: heh, yeah I have seen it, I even asked Steve to include it in this weeks summary [20:19] <sienap> hehe .. [20:19] <sienap> gym [20:19] <sienap> :) [20:19] <ajmitch> hmm, quiet here.... [20:19] <ajmitch> even with sienap ;) [20:20] <ds> . [20:20] tnt (tn...@h2...) joined #gstreamer. [20:20] <ds> should bitrates be expressed in bits/sec or kbits/sec? [20:20] <tnt> Good morning. [20:21] <ajmitch> morning [20:21] Action: ajmitch guesses bits/sec, but is uninformed & knows nothing about it :) [20:29] <wingo> hey all. [20:29] <wingo> tnt, i responded on the wiki doober. [20:29] <tnt> wingo: OK, Thanks... I'll go look at it. [20:31] <sienap> <- always quiet.. the sienap talk a lot thingy is a lie. [20:31] <sienap> anyway [20:31] <sienap> bye ;) [20:31] Nick change: wingo -> wingo-lunch [20:31] sienap (sy...@s3...) left #gstreamer. [20:48] omega (om...@om...) joined #gstreamer. [20:48] <ajmitch> hey omega [20:49] <omega> yo [20:51] <ds> hey [20:51] <ds> let's ask omega, he knows everything =) [20:51] <omega> uh oh [20:51] <ds> should bitrates be expressed in bits/sec or kbits/sec? [20:51] <omega> probably kbits [20:52] <ds> right now, various plugins disagree [20:54] <ds> I'm in favor of bits/sec, because of the additional precision, but kbits/sec is "easier" [20:54] Nick change: wtay-zZz -> wtay [20:54] <wtay> yo [20:55] bstard (Lor...@a2...) joined #gstreamer. [20:55] <tnt> ds: Would it be possible (and convinient) to add a 'unit' to the bit rate... telling everyone whether it is in bits/sec kbits/sec or something else? [20:56] <tnt> wtay: Howdy [20:56] <wtay> yo [20:56] <omega> possibly, or maybe have both bitrate and kbitrate? [20:56] <ajmitch> hi wtay [20:56] <wtay> hi ajmitch [20:57] <tnt> omego: Having both kinda seems like a waste... unless it is justified by some 'speed' concern. [20:57] <wtay> omega: today is a good day to finish events.. [20:58] <omega> heh, 'today is a good day to die' [20:58] <tnt> You know... I think there was actually a proposal to having a general method of 'adding units' to numbers (and other values)... [20:58] <tnt> ... in the Wiki somewhere. [20:58] <omega> in glib? [20:59] <omega> oh [20:59] <tnt> I wrote something in C++ to do just that... but it does it a compile time. [20:59] <tnt> I don't know how useful it would be. [21:00] <tnt> The algorithms used in it would be the same though. [21:01] <tnt> Algorithms for generating 'new units' after a mathimatical operation -- like adding, subtrating, dividing, etc... [21:01] <tnt> ... putting the unit into a canonical form... [21:01] <tnt> ... and various other stuff. [21:11] Action: taaz votes bits/s as uniform standard for all plugins [21:11] Action: ajmitch already voted bps earlier ;) [21:11] <omega> yeah, is probably more generally useful [21:12] <tnt> This stuff should really be documented somewhere for others to see. [21:12] <taaz> tnt: it is, the source [21:12] <omega> it should be in an appendix or even major chapter of the plugin writers' guide [21:13] <taaz> or maybe add a "Standard Units" appendix to the manual [21:13] Action: wtay votes bps before dying [21:14] <taaz> wtay: did you see my pseudo event structure rearrangement? [21:15] <wtay> taaz: wiki? [21:15] <taaz> er.. no, on my machine [21:15] <wtay> URI? [21:15] <taaz> i could wikize it... [21:15] <taaz> http://crib.lehn.org:8080/~dlehn/projects/gstreamer/events/ [21:16] <taaz> i think it's hard to understand, sorry [21:16] <taaz> ev0 is current way, ev1 is a new way [21:17] <taaz> quick summary: drop GstData, make GstEvent super class of events, GstBuffer is just a type of event [21:17] apoc_ (ap...@dy...) joined #gstreamer. [21:17] <apoc_> hi all [21:18] <tnt> Hey apoc_. [21:18] <wtay> taaz: I see.. [21:19] Nick change: wingo-lunch -> wingo [21:19] <wingo> apoc_: are you the mikmod fellow? [21:19] <wtay> taaz: one flaw, the chain function is a pad function [21:20] <apoc_> yes [21:20] <wingo> do you want to update the status page? [21:20] <taaz> then i have some stuff in there exploring idea of default event handlers [21:20] <taaz> which is easy for chained elements. but for loops you pretty much always have to do the switch(event_type) thing [21:20] <taaz> the python is untyped so buf implies GstBuffer and ev implies GstEvent [21:20] <taaz> changing chained elements to accept GstEvent in this model would fixup current issue of crashes when you expect a GstBuffer and don't check if it's actuall a GstEvent [21:21] <taaz> wtay: yeah, it's not written properly for pads.. i was just trying to get some ideas down. [21:22] <wtay> your ain idea seems to be the event-buffer hierarchy [21:22] <wtay> s/ain/main [21:22] <taaz> yeah [21:22] <wtay> why(s that? [21:22] <taaz> it's currently wrong [21:22] Action: wtay has trouble typing [21:24] vektor (reet@HSE-Kitchener-ppp3506142.sympatico.ca) left irc: back in a sec [21:24] <taaz> the chain funcs that get GstBuffer and check if it's a GstEvent then case is very poor design IMHO. I understand why it's like that since events are hacked on, but unifing it so everything coming in is an event seems more appropriate [21:24] <wtay> taaz: why not GstData? it can still be extended that way.. [21:25] <omega> in practice it's the same thing, even [21:25] <taaz> GstData is kind of empty right now [21:25] <taaz> and just buf and events use it [21:25] <taaz> i was trying to simplify [21:25] <omega> but there's an advantage to using GstData: an element can be force to set a flag saying it can deal with events [21:25] <omega> otherwise the gst code handles the events, and the element in unaware of events at all [21:25] <wtay> taaz: it could hold the refcounting [21:25] <wtay> omega: yes [21:26] <taaz> ok, well can we get it so types are correct? any func that gets a GstBuffer should be able to assume it's a GstBuffer or a GstBuffer subclass. that is not the case right now [21:27] <omega> taaz: as I pointed out, changing it to GstData makes a *huge* mess of the code [21:27] <wtay> taaz: it will be when I commit my changes :) [21:27] <taaz> ok, can we change languages so this isn't a problem? ;) [21:27] <taaz> wtay : wath changes? [21:27] Action: taaz can't type either [21:27] <wtay> taaz: element flag indicating it can deal with events [21:28] <wtay> taaz: event function and chain function to deal with both types [21:28] <wtay> but it's still not the way I want it... [21:29] <taaz> the event union thing is kinda silly too.. why not just use event subclasses? you need to switch on type anyway, seems silly to allocate extra data space for events when its not needed [21:31] <taaz> i guess that means more typecasting and code mess... but by choosing to use Gtk you pretty much decided to do that in advance, shouldn't make ugly design just cause object model coding is nasty... [21:31] <wtay> taaz: the union has the advantage that we can use one memchunk [21:32] <wtay> taaz: as opposed to X memchunks + code to set them up [21:35] <wtay> IMO subclassing GstData is good, that's how I would do it in a real OO language too [21:35] <wtay> but I wouldn't use the union [21:35] <taaz> i kind of like buffers being a type of event but whatever... [21:36] <omega> nothing saying we can't do that even with the element flag [21:36] <taaz> i think worrying about memchunks for events is premature optimization [21:36] <wtay> if they are events, the event type should be GST_EVENT_BUFFER [21:36] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [21:36] <taaz> wtay: exactly... [21:36] <omega> but the typing of the chain function would be up for grabs then [21:36] <taaz> eh? [21:36] <omega> well, the chain func would have to use GstEvent * in the params list [21:37] <taaz> well, if there are going to be seperate buffer vs event chain funcs this doesnt matter as much [21:37] <omega> no matter what, even if the element doesn't handle events itself [21:37] <wtay> taaz: that's what GstData does [21:37] <omega> taaz: true [21:37] <taaz> so are events going to be refcounted? i think they have to be since they may be multicast in tee like elements [21:38] <omega> yeah [21:38] <wtay> yup [21:38] <omega> so either gstdata or gstevent needs to hold the refcounting stuff [21:38] <wtay> the original idea was to copy them in a multicast situation, but that's no important.. [21:39] <wtay> my problem is with subclassing GstEvent is that GstEvent becomes GstData then [21:40] <taaz> wtay: well.. that was what i was trying to do. ;) is there some other purpose for GstData? [21:40] <wtay> GstEvent is going to be pretty empty becuqse the real event info is sublassed from GstEvent [21:40] <wtay> and the union is just an implementation detail really [21:41] <wtay> taaz: refcounting + typing of the data (buffer/event) [21:44] <wtay> I feel we have two possibilities: [21:44] <wtay> 1) we make separate functions for event handlers and chain handlers [21:44] <wtay> 2) we make all functions return/accept GstData [21:45] Action: wingo votes for 2 [21:46] <wtay> I vote for 1 [21:46] <taaz> does #2 mean any func could return/accept either events or buffers? [21:46] <wingo> what about loop elements? [21:46] <wtay> taaz: yes, in general [21:46] <omega> yeah, this is only for chained elements... [21:46] <wtay> 1 would make it possible to define a default handler [21:46] <omega> well, either one could handle a default [21:46] <wtay> omega: not sure yet [21:46] <omega> but in different ways [21:47] <wtay> setting the event handler function automatically makes the plugin event aware [21:47] <wtay> and is a bit cleaner [21:47] <wingo> right, it's wierd that you would pull something that was not pushed in the first place [21:48] <omega> ? [21:48] <wingo> you get an event now on pad_pull [21:48] <wingo> oh [21:48] <wingo> that's loop stuff [21:48] <wingo> n/m [21:48] <wtay> omega: for loop based element maybe we could implement a dispatcher [21:48] <omega> meaning? [21:48] vektor (reet@HSE-Kitchener-ppp3506142.sympatico.ca) joined #gstreamer. [21:49] <wtay> omega: uhmm.. nevermind, it wont work :( [21:49] <taaz> so what does a default event handler do? [21:50] <wtay> taaz: it forwards the events to all src/sink pads and does things like setting the element to EOS etc [21:50] <wingo> paused you mean? [21:50] Nick change: Ura_gym -> Uraeus [21:50] <apoc_> yo uraeus ! [21:51] <wtay> wingo: yes [21:51] <Uraeus> hi apoc_! [21:51] <taaz> wtay: perhaps there should be defauls for buffer pool and caps proxying too [21:51] <taaz> ie, and element would default to something like identity [21:51] <wingo> taaz: yes!!! [21:52] <taaz> s/and/an [21:52] <wtay> taaz: yes [21:52] <wingo> should that have to do with object hierarchies, that elements default to their parent's implementation? [21:53] <taaz> wingo: gtk model is like that, instead of method overriding its sort of like setting a callback will override a default implementation [21:53] <taaz> or at least that's how i see it [21:53] Zeenix (ze...@13...) joined #gstreamer. [21:53] Action: taaz thinks gtk programming is waaay too hard [21:53] <wtay> possibly [21:54] <Zeenix> hi [21:54] <wingo> hey [21:54] <wtay> taaz: it isn't, the concepts map 1-to-1 to other "real" OO languages [21:54] Action: wingo likes gtk :) [21:55] <Zeenix> wtay: i synced with your changes & the one you recommended today, nothing more ATM [21:55] <Uraeus> taaz: if you want it even easier use GOB [21:55] <Zeenix> wtay: but i get errors like: [21:55] <Zeenix> gstrtprecv.c: In function `gst_rtprecv_plugin_init': [21:55] <Zeenix> gstrtprecv.c:397: warning: implicit declaration of function [21:55] <Zeenix> `gst_plugin_add_feature' [21:55] <Zeenix> gstrtprecv.c:397: warning: implicit declaration of function [21:55] <Zeenix> `GST_PLUGIN_FEATURE [21:56] <Zeenix> & [21:56] <wtay> Zeenix: yeah, that's CVS gstreamer [21:56] Nick change: Uraeus -> Ura_shower [21:56] <Zeenix> wtay: i am using the CVS tarball [21:56] <taaz> Uraeus: what's the point of GOB? Why not just use a real OO language if you get to the point of code writer intermediate languages [21:57] <wtay> Zeenix: then it should work [21:57] <Zeenix> wtay: i doubt i have the latest one [21:57] <Zeenix> wtay: are these changes very new to gst [21:57] <Zeenix> ? [21:57] <wtay> not really.. a few months old I guess [21:57] <Ura_shower> taaz: well Qt uses a pre-processor and I guess you think C++ is a real OO language [21:57] <wingo> are you working on lib docs, wtay? [21:58] <wtay> wingo: not anymore [21:58] <wingo> are the lib docs up-to-date? [21:58] <wingo> the bytestream, etc lib docs [21:58] <wtay> yes [21:58] <wtay> oh [21:58] <Ura_shower> wtay: are the devhelp compatible? [21:58] <wtay> no, not the libs [21:59] <wtay> Ura_shower: certainly not! [21:59] <Zeenix> wtay: what was that web page containing latest cvs gst tarball ? [21:59] <Ura_shower> wtay: why not? [22:00] <wtay> Ura_shower: I dunno [22:00] <Ura_shower> hehe, ok now I really need to get into the shower :) [22:00] <wtay> devhelp sucks [22:00] Action: wtay doesn't know what devhelp is [22:00] <wingo> but it sucks, whatever it is ;) [22:01] <wtay> yes [22:01] <omega> Ura_shower: you don't have a waterproof laptop?? [22:01] <wtay> Ura_shower: a garbage bag works pretty well :) [22:01] <omega> dare I ask? [22:02] <Zeenix> wtay: ? [22:02] <wtay> just put the laptop in a see-through garbage bag :) [22:03] <wtay> pretty waterproof.. [22:03] <omega> 'mostly' ? <g> [22:03] <wtay> don't accidently put it outside though... [22:03] <Zeenix> repeating: what was that web page containing latest cvs gst tarball ? [22:03] <wtay> Zeenix: gstreamer.net? [22:05] <Zeenix> wtay: no [22:05] <Zeenix> wtay: got it [22:07] <Zeenix> wtay: its the file named gstreamer-cvsroot.tar.gz ? [22:07] <wtay> I would vote for sticking with the current events API and make the code actually work, then think about changing the API [22:08] <wtay> Zeenix: sounds good [22:08] <Zeenix> wtay: ok, i'll do that, its more easy :) [22:11] <Ura_shower> wtay, wingo: devhelp (http://devhelp.codefactory.se/) is a documentation viewer for development docs, and it is integrated into both Anjuta and gIDE [22:13] <omega> what about scrollkeeper? [22:13] <Ura_shower> wtay, wingo: screenshot of devhelp in action inside gIDE (http://primates.ximian.com/~dave/gide8.png) [22:13] <Ura_shower> omega: think scrollkeeper is used in the backend [22:13] <omega> ok [22:14] <wingo> well, it looks like they have the glib, etc references in devhelp [22:14] <wingo> so yes, we're compatible :) [22:14] <wingo> devhelp still sucks ;) [22:14] <wtay> it's for wimps :) [22:14] Action: Ura_shower smacks wingo on the head with his shampoo bottle [22:14] <wingo> yeah, Uraeus is a wimp ;) [22:14] <wtay> real men use the source :) [22:14] Nick change: Ura_shower -> Uraeus [22:15] <Uraeus> if real men use the source, what do you guys use then? [22:15] <omega> whoops. see http://maxtor.com/ [22:16] Action: omega smacks Uraeus [22:16] <wtay> Uraeus: the assembler dump :) [22:16] <omega> heh [22:17] <wingo> ok, i'm going to come forward and say that devhelp looks neat [22:17] <wingo> as long as i can still use emacs and not have to submit to an ide :) [22:17] <wtay> seriously it is cool, I tried it a while ago [22:18] <wtay> I think it needs some xml description of the topics [22:18] <Zeenix> wtay: i think the one installed on my sys was from the tarball having name: gstreamer-200108111625.tar.bz2 [22:18] <Zeenix> s/was/is [22:20] <wingo> use cvs :) [22:21] <Zeenix> wingo: wincvs? :) [22:21] <wingo> heh [22:21] <wingo> i forgot :) [22:23] <omega> http://gstreamer.net/cvs/snapshots/gstreamer-20011029.tar.bz2 [22:24] <Zeenix> omega: any idea how old is that gstreamer-200108111625.tar.bz2 ? [22:24] <omega> um, it's a month and a half old, according to the timestamp [22:26] <Zeenix> omega: & it wasnt having the gst_plugin_add_feature(..) & GST_PLUGIN_FEATURE functions ? [22:27] <omega> dunno, wtay knows when he added that [22:27] asmod (stevec@64.5.222.2) left irc: [x]chat [22:28] <omega> wingo: I've added that dir to the website, can you create it in cvs and add a .cvsignore for the .bz2's ? [22:28] <wingo> what dir? [22:29] <wtay> Zeenix: I guess it wasn't in there, since you have errors.. [22:29] <omega> cvs/snapshos/ [22:30] <Zeenix> wtay: seems you are very busy ATM [22:30] <wingo> okeydoke [22:30] <omega> also have a snapshot script that can be committed in there [22:31] <omega> its at sf:~omega/work/snapshot.sh [22:31] <wtay> Zeenix: it was added at 21 aug [22:31] <wingo> so... will that be the only thing committed in there? [22:31] <omega> yeah [22:31] <wingo> ok [22:32] <omega> probably want to have both 'gstreamer' and 'gstreamer-*.tar.bz2' in the .cvsignore [22:32] <Zeenix> omega: i remember there was a webpage on gst.net/snapshots not gst.net/cvs/snapshots ? [22:33] <wingo> not anymore :) [22:33] <omega> well, it moved <g> [22:35] Nick change: wtay -> wtay-brb [22:35] <wingo> ok, updated, committed, and the web site is updated too [22:36] <wingo> i can add that skript to my sf crontab if you like [22:36] <omega> sure [22:36] <omega> though then we're gonna want to look at cleaning it on some policy [22:37] <taaz> Uraeus: sorry, was distracted: Qt preprocessor stuff looks like crap (though I haven't actually used it before), and no I don't consider C++ an OO language ;) [22:37] Action: taaz shouldn't be starting flame wars.. [22:38] <Uraeus> taaz: the DVD stuff you did a little while back, does it mean I can playback DVD's with GStreamer now? [22:39] <wingo> ok, it's now in my crontab [22:39] <Zeenix> any have used linux telephony API ? [22:39] <taaz> Uraeus: sorta [22:39] <Uraeus> taaz: sorta? meaning it plays but no subtitles etc.? [22:39] <taaz> Uraeus: it should play just as good as movietime or libdvdread example player since that's where i stole the code from [22:40] <taaz> Uraeus: no... no subtitles or menus or nav or anything [22:40] <taaz> Uraeus: but in theory it will play a full movie. i can't test it cause gtk unthreadsafeness or something locks my test apps after 2 seconds [22:41] <taaz> Uraeus: try ./test/dvdplay /dev/dvd 1 1 1 (or whatever chapter the movie starts at) and see if it works ;) [22:41] <Uraeus> taaz: ok [22:42] <wingo> so, has anyone with smp tried the thread-breakage tests with glib2? [22:43] <omega> I still don't have a SMP machine to use (sparc is misbehaving) [22:43] <taaz> nope, i should since i bitch about it so much [22:43] <omega> so until taaz tries it, no [22:44] <ajmitch> hmm, ext3 seems to have it's issues [22:45] <ajmitch> htdig is doing it's (big) sort/merge on one ext3 partition, which is causing mp3s to pause for a coupld of seconds at times (they're on the other ext3 partition) [22:47] <omega> ouch [22:47] <ajmitch> a bit of a pain ;) [22:49] <Zeenix> omega: ordinary ints are 2 bytes, right ? [22:50] ds (ds...@c7...) joined #gstreamer. [22:50] <omega> Zeenix: it depends on the architecture. on x86 yes, on others no [22:50] <omega> er, on x86 no [22:51] <omega> sorry, they're 4 bytes, but some older DOS-based compilers had 2-byte ints [22:51] <Zeenix> omega: i mean i would use g_htons on gint vars ? [22:51] <omega> yes [22:51] <Zeenix> guint too [22:51] <Zeenix> ? [22:52] <Zeenix> uh..., i wonder why wtay changed them to g_htonl calls [22:53] <omega> oh, um, htons.... well, you can use them, yes, but you'd better treat them as shots [22:53] <omega> shorts [22:54] <apoc_> cya all [22:54] <Zeenix> no, i said i am using gint & guint, so i g_htonl is correct [22:54] <Zeenix> s/i/i gues [22:54] <omega> yes [22:56] <Zeenix> & i was using G_BYTE_ORDER in the NEWCAPS call which wtay changed to G_BIG_ENDIAN, any idea why ? [22:57] <wingo> isn't all internet traffic big endian? [22:57] <apoc_> wingo : can you give me the URL to administrate status [22:58] <Zeenix> wingo: you are informing me or asking me ? [22:59] <wingo> Zeenix: asking/informing? i'm not sure is what i'm saying [22:59] <wingo> i gotta roll though [22:59] Nick change: wingo -> wingo-out [23:02] <apoc_> bye all [23:02] apoc_ (ap...@dy...) left #gstreamer (dodo). [23:03] chillywilly (da...@d1...) joined #gstreamer. [23:19] steveb (st...@no...) joined #gstreamer. [23:20] Nick change: wtay-brb -> wtay [23:22] <Zeenix> wtay: that G_BIG_ENDIAN is a mistake ? [23:22] <wtay> no, the RFC says the audio is in network endianness, which is BIG_ENDIAN [23:23] ShrimpX (marius@Photon.pdx.netroedge.com) joined #gstreamer. [23:24] BBB (BB...@uc...) joined #gstreamer. [23:24] <Zeenix> wtay: one thing you forgot to do was translating the members of Rtcp_APP_message back by ntohl() [23:26] <wtay> Zeenix: yeah, I didn't touch the RTCP stuff [23:30] Action: BBB has a proposal for a new v4lsrc/v4lmjpegsrc/v4lmjpegsink [23:30] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [23:30] steveb (st...@no...) left irc: advance! [23:37] Action: bstard is away: weg [23:37] Zygo (zbl...@24...) left irc: Read error to Zygo[246.chipworks.com]: Connection reset by peer [23:39] <wtay> I have to sleep.. cya [23:39] Nick change: wtay -> wtay-zZz [23:42] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [23:57] Uraeus (csc...@c1...) left #gstreamer (Client Exiting). [23:59] Zeenix (ze...@13...) left irc: I pay 25 Rs( 0.5$ )/hour [00:00] --- Tue Oct 30 2001 [00:12] gtb (syln@195.68.94.2) joined #gstreamer. [00:12] gtb (syln@195.68.94.2) left irc: [00:17] evil_monk (trevo@208.141.162.68) left #gstreamer (Client Exiting). [00:48] wingo-out (wingo@24.25.21.76) left irc: going home, yo [00:58] Shippou (ir...@ot...) joined #gstreamer. [00:58] arik (ar...@sd...) joined #gstreamer. [00:59] <arik> howdy [00:59] <tnt> Hey [01:00] <ajmitch> hey arik [01:02] <arik> hey aj [01:08] bstard (Lor...@a2...) left irc: Ping timeout for bstard[a213-84-40-242.adsl.xs4all.nl] [01:12] Nick change: arik -> arik|afk [01:12] <arik|afk> nrn [01:12] <arik|afk> brb [01:16] arik|afk (ar...@sd...) left irc: Changing server [01:20] ds (ds...@c7...) joined #gstreamer. [02:01] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [02:01] arik (ar...@sd...) joined #gstreamer. [02:01] <arik> omega: you around? [02:01] <omega> yeah [02:01] <arik> wanna help me with perforce? [02:01] <arik> ;-P [02:01] <omega> um, I've done about 4 commands with it [02:01] ajmitch (me...@p4...) left irc: Ping timeout for ajmitch[p47-max2.dun.ihug.co.nz] [02:02] <arik> heh [02:02] <omega> checkout, and update [02:02] <omega> er, sync, whatever they call it [02:02] <arik> i can't figure out how to get this working at all, they want me to do all these complicated things to get my code to rr [02:02] <arik> heh [02:02] <arik> yeah [02:02] <omega> at the very least you have to figure out how to get a tunnel set up. ask Eric about that [02:02] <arik> i think i have hat [02:03] <arik> eric emailed me loads of stuff [02:03] <arik> although it is complicated by ssh being so damn slow over modem :-( [02:03] <omega> heh [02:04] <arik> brb [02:04] arik (ar...@sd...) left irc: leaving [02:04] ajmitch (me...@p5...) joined #gstreamer. [02:17] Vito` (vi...@oo...) joined #gstreamer. [02:20] ShrimpX (marius@Photon.pdx.netroedge.com) left irc: home [02:25] Vito` (vi...@oo...) left irc: reboot [02:26] mwc (ma...@ly...) joined #gstreamer. [02:34] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [02:42] ds (ds...@c7...) joined #gstreamer. [02:46] bstard (Lor...@a2...) joined #gstreamer. [02:51] arik (ar...@sd...) joined #gstreamer. [02:51] <ajmitch> hi arik [02:51] Action: arik is using zenirc to see what it's like [02:51] <arik> hey aj [02:51] <arik> vektor: are you here? [02:52] arik_ (ar...@sd...) joined #gstreamer. [02:52] <arik_> man [02:52] arik (ar...@sd...) left irc: Started wasting time elsewhere [02:52] <arik_> when is vektor here [02:52] Nick change: arik_ -> arik [02:53] <arik> i want to try out his dvd stuff [02:54] <tnt> arik: vektor seems to be more active in the #livid channel... you might want to try there. [02:54] <arik> hmm [02:54] <arik> alright [02:55] <vektor> vektor is where? [02:55] <ajmitch> dunno, where is vektor? [02:55] Action: ajmitch can't see him ;) [02:55] <arik> hehe [02:56] <arik> he is invisible! [02:56] <tnt> arik: He's talking on #livid right now! [02:56] <arik> heh [02:56] <arik> weird! |
From: IRC B. <wt...@us...> - 2001-10-31 05:28:03
|
******************************************************************* [03:05] Vito` (vi...@oo...) joined #gstreamer. [03:06] Vito` (vi...@oo...) left irc: Read error to Vito`[ool-18bb5cc2.dyn.optonline.net]: Connection reset by peer [03:26] arik (ar...@sd...) left irc: leaving [03:33] omega (om...@om...) left irc: _Cancel10x14 [03:34] Shippou (ir...@ot...) joined #gstreamer. [03:45] evil_monk (trevo@208.141.162.68) joined #gstreamer. [04:29] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [04:55] <tnt> Hello, if anyone is out there listening, I'm looking for comments and critisizms on: [04:55] <tnt> http://gstreamer.net/wiki/index.php?SynchronizationProposal [04:55] <tnt> ... a Synchronization Proposal. [05:08] Action: tnt goes and watches TV [05:22] Nick change: chillywilly -> cw-awk [05:22] Nick change: cw-awk -> cw-afk [05:29] Nick change: cw-afk -> chillywilly [06:56] chillywilly (da...@d1...) left irc: Ping timeout for chillywilly[d155.as4.nwbl0.wi.voyager.net] [06:56] chillywilly (da...@d1...) joined #gstreamer. [07:37] omega (om...@om...) joined #gstreamer. [07:37] <tnt> hey. [07:37] <omega> yo [07:37] <tnt> Did you look at the Synchronization Proposal at all? [07:37] <omega> nope, where? [07:38] <tnt> http://gstreamer.net/wiki/index.php?SynchronizationProposal [07:40] <ajmitch> hi omega [07:43] ds_ (ds...@c7...) joined #gstreamer. [07:44] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [07:51] chillywilly (da...@d1...) left irc: [08:13] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [08:37] mwc (ma...@ly...) left irc: [x]chat [08:39] Nick change: tnt -> tnt-zZz [08:44] steveb (st...@no...) joined #gstreamer. [08:44] arik (ar...@sd...) joined #gstreamer. [08:44] <arik> yo [08:50] arik (ar...@sd...) left #gstreamer. [09:14] Shippou (ir...@ot...) joined #gstreamer. [09:19] walken (fo...@c1...) joined #gstreamer. [09:23] Nick change: tnt-zZz -> tnt [09:30] Nick change: tnt -> tnt-zZz [09:55] walken (fo...@c1...) left irc: ircII EPIC4-1.0.1 -- Are we there yet? [10:04] steveb (st...@no...) got netsplit. [10:04] taaz (dlehn@66.37.66.32) got netsplit. [10:04] ds_ (ds...@c7...) got netsplit. [10:05] taaz (dlehn@66.37.66.32) returned to #gstreamer. [10:05] ds_ (ds...@c7...) returned to #gstreamer. [10:05] steveb (st...@no...) returned to #gstreamer. [10:13] omega (om...@om...) left irc: sleep [10:47] taaz (dlehn@66.37.66.32) left irc: Ping timeout for taaz[66.37.66.32] [10:49] taaz (dlehn@66.37.66.32) joined #gstreamer. [11:11] ds_ (ds...@c7...) left irc: Ping timeout for ds_[c799921-a.pinol1.sfba.home.com] [12:54] Action: bstard is away: Weg [12:55] Action: bstard is back (gone 00:00:01) [14:13] asmod (stevec@64.5.222.2) joined #gstreamer. [14:33] asmod (stevec@64.5.222.2) left irc: [x]chat [14:49] manuel (user422@213.174.92.50) joined #gstreamer. [14:54] asmod (stevec@64.5.222.2) joined #gstreamer. [15:05] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [15:58] Zygo (zbl...@24...) joined #gstreamer. [16:08] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:08] Shippou (ir...@ot...) joined #gstreamer. [17:18] apoc_ (ap...@dy...) joined #gstreamer. [17:20] <apoc_> hi all [17:43] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [17:49] manuel (user422@213.174.92.50) left irc: Client Exiting [18:03] apoc_ (ap...@dy...) left irc: CGI:IRC 0.4.2 [18:22] asmod (stevec@64.5.222.2) left irc: [x]chat [18:24] asmod (stevec@64.5.222.2) joined #gstreamer. [18:42] Nick change: tnt-zZz -> tnt [18:46] Shippou (ir...@cr...) joined #gstreamer. [19:29] wingo (wi...@rd...) joined #gstreamer. [19:29] <wingo> word up [19:29] <tnt> hey wingo. [19:29] <wingo> hi [19:29] <tnt> The finished version of the Synchronization Proposal is up. [19:30] <wingo> cool, i'll check it out [19:30] <tnt> (I also replied to you last reply.) [19:34] <wingo> there's a "discontinuous" event that is supposed to be emitted on discontinuities [19:35] <wingo> so i don't see the need for checking counters all that tightly [19:35] <wingo> i have to go do some crap right now though [19:35] <tnt> Ahhh... where can I learn more about these events??? [19:35] Nick change: wingo -> wingo-afk [19:35] <wingo-afk> the source :-\ [19:35] <wingo-afk> mpeg2parse [19:35] <wingo-afk> filesrc [19:35] <wingo-afk> xvideosink? i dunno [19:36] <tnt> Hmmm... OK... that will be fun ;-) [19:36] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [19:58] asmod (stevec@64.5.222.2) left irc: [x]chat [20:07] ChiefHighwater (pa...@te...) joined #gstreamer. [20:07] <ajmitch> hey ChiefHighwater [20:08] <ChiefHighwater> Ello ajmitch 8-] [20:08] Action: ajmitch just got up... [20:09] <ChiefHighwater> umm, what time is it for you? [20:09] <ajmitch> Wed Oct 31 08:09:25 NZDT 2001 [20:10] <ChiefHighwater> so, you're up early compared to me and esp Omega [20:11] <ajmitch> yep :) [20:11] <ajmitch> had to get up early to say goodbye to some friends [20:23] asmod (stevec@64.5.222.2) joined #gstreamer. [20:24] asmod (stevec@64.5.222.2) left irc: [x]chat [20:25] ChiefHighwater (pa...@te...) left irc: Ping timeout for ChiefHighwater[temple-baptist.com] [20:30] asmod (stevec@64.5.222.2) joined #gstreamer. [20:33] asmod (stevec@64.5.222.2) left irc: [x]chat [20:34] asmod (stevec@64.5.222.2) joined #gstreamer. [20:41] asmod (stevec@64.5.222.2) left irc: [x]chat [20:42] omega (om...@om...) joined #gstreamer. [20:42] <tnt> hey [20:42] <omega> yo [20:42] <omega> anyone know of a new windoze virus? [20:43] <omega> cause ChiefHighwater and I are at Threatcon Delta because a virus mangled a drive at the church office and spewed mail all over the place [20:43] Nick change: wtay-zZz -> wtay [20:43] <wtay> yo [20:43] <wtay> I received one [20:44] <wtay> it removed the boot sector suposedly [20:44] <ajmitch> ouch [20:45] <wtay> there are two other viri, one removing .dll files.. [20:45] <omega> ok, this disk has no partition table [20:46] <omega> but if that's all it does, or just .dll's we can recover data [20:46] <omega> if it wipes the FAT, we're hosed [20:46] <ajmitch> you can recover a hosed partition table thankfully [20:46] <omega> yes, esp on a windoze box that only has one partition.... [20:46] <ajmitch> even on a box with a few ext2 partitions it can be done easily ;) [20:47] ds (ds...@c7...) joined #gstreamer. [20:47] Action: ajmitch likes GNU Parted [20:47] <omega> relatively, yes, given a linux box to use <g> [20:47] <wtay> I just received a warning from my father (still not understanding what linux is :)) [20:47] Action: omega needs to pack to head over there and deal with this ;-( [20:47] <ajmitch> i should see an email from my mother soon then ;) [20:47] <omega> l8r all [20:47] omega (om...@om...) left irc: fdisk /mbr [20:47] <wtay> later.. [20:49] asmod (stevec@64.5.222.2) joined #gstreamer. [20:50] <ajmitch> hope it's not this one: http://securityresponse.symantec.com/avcenter/venc/data/w32.elkern.3326.html [20:51] <wtay> doesn't seem to be... [20:52] <ajmitch> that's good [20:53] <wtay> do I care? <g> [20:53] <ajmitch> well, you should ;) [20:54] <ajmitch> this is omega's sanity on the line here ;) [20:55] Action: wtay wishes omega good luck.. [20:55] Nick change: tnt -> tnt-out [21:22] <taaz> hmm? viri? i got an .exe today... already in the bit bucket though... [21:23] <ds> didn't you try to run it first? [21:23] <ds> I always try to run every .exe file I get [21:24] <taaz> didn't work... invalid elf header or something ;) [21:24] <ds> =) [21:24] <ajmitch> heh [21:24] <ds> I wonder if you could make one file with trick headers that would run on both Linux and Windows [21:28] Zeenix (ze...@13...) joined #gstreamer. [21:29] <Zeenix> yo all [21:29] <wtay> hi [21:29] <Zeenix> wtay: The Rtp pipeline working..... [21:30] <wtay> cool [21:30] <Zeenix> wtay: hurray [21:30] <Zeenix> wtay: there are some probs still though [21:30] <Zeenix> wtay: like the newcaps in rtpsend doesnt get called [21:31] <wtay> what element did you connect it to? [21:31] <Zeenix> wtay: osssrc [21:32] <wtay> oh [21:32] <Zeenix> wtay: gsmenc too but it wasnt working [21:32] <wtay> ah, yes, osssrc uses G_LITTLE_ENDIAN [21:32] <Zeenix> wtay: so should rtpsend/recv ? [21:33] <wtay> rtp states that it should do BIG_ENDIAN [21:33] <wtay> so, you need a converter [21:33] <wtay> gsm should work fine [21:33] <Zeenix> wtay: ughhhh..... [21:33] <wtay> if it does caps, that is.. [21:33] <Zeenix> no [21:34] <Zeenix> wtay: the newcaps never get called [21:34] <wtay> gsmenc doesn't even have a newcaps function [21:35] <Zeenix> wtay: i mean the newcaps in rtpsend [21:35] <wtay> Zeenix: yup, osssrc uses LITTLE_ENDIAN [21:35] <wtay> Zeenix: rtpsend doesn't accept that (see the padtemplate) [21:35] <Zeenix> wtay: ok [21:36] <wtay> so the newcaps function is never called [21:36] <wtay> if you change the endianness to LITTLE_ENDIAN then it should get called [21:36] <Zeenix> wtay: so i should be now upto making a endian coverter element [21:37] <Zeenix> ? [21:38] <wtay> Zeenix: you don't want to send raw audio over the network [21:38] <wtay> Zeenix: not with your connection :) [21:38] <Zeenix> wtay: no, but there could be mad people who would like to do that [21:38] <wtay> Zeenix: then I guess you should create one, yes [21:38] <wtay> but I would go with gsm for now.. [21:39] <Zeenix> wtay: ok [21:39] <wtay> It's not curcial I suppose, I mean, It could be added later [21:40] <Zeenix> wtay: yes, but the connecting the gsmdec doesnt help either, newcaps never get called [21:40] <Zeenix> the rtpsend's newcaps i mean [21:41] <wtay> right, gsmenc needs to set caps... [21:41] <Zeenix> errr [21:41] <Zeenix> wtay: gsmenc i meant [21:41] <wtay> oh, It seems like it does... [21:42] <Zeenix> wtay: i would try again today, i havent got the chance to play enough with it yet [21:42] <wtay> it works here.. [21:42] <Zeenix> wtay: my cousin was here & we went for some movies [21:43] <wtay> . [21:44] <Zeenix> wtay: throw the command line for the pipelines you used here: [21:47] <Zeenix> wtay: i meant you should do that not me [21:49] <Zeenix> wtay: ? [21:51] wingo-afk (wi...@rd...) left irc: going home, l8r. [21:52] <Zeenix> looks like rtprecv put another HD of wtay to fire :) [22:08] <Zeenix> wtay: ??????? [22:09] asmod (stevec@64.5.222.2) left irc: [x]chat [22:12] Action: wtay was on the phone [22:13] <Zeenix> wtay: phone, you have two ph. lines or you dont use dialup net. ? [22:13] <wtay> Zeenix: I have a cable modem [22:14] <wtay> Zeenix: -launch filesrc location=/some/mp3 ! stereo2mono ! gsmenc ! rtpsend [22:14] <wtay> Zeenix: rtprecv ! gsmdec ! osssink [22:14] <wtay> uhm [22:15] <wtay> Zeenix: -launch filesrc location=/some/mp3 ! mad ! stereo2mono ! gsmenc ! rtpsend [22:15] <wtay> it alo needs a resampler to scale the audio rate down to 8KHz [22:15] <Zeenix> worked cool ? [22:15] <wtay> yeah, pretty good [22:16] omega (om...@te...) joined #gstreamer. [22:16] <Zeenix> yo [22:16] Action: omega has fun trying to clean up after a virus ;-( [22:16] <wtay> the only problem is rate control on the sender side [22:16] <Zeenix> wtay: prob. ? [22:16] <ajmitch> hi omega [22:17] <Zeenix> wtay: is the RTCP not behaving good ? [22:17] <wtay> Zeenix: the sender pumps out RTP packets at the fastest possible rate, overloading the network [22:17] <wtay> Zeenix: I never saw RTCP working.. [22:17] <wtay> omega: which one :) [22:18] <wtay> omega: try antibiotics <g> [22:18] <Zeenix> wtay: you recommended that day, we should through the rtcp stuff to the app developer ? [22:18] <wtay> Zeenix: some RTCP messages, yes [22:19] <wtay> Zeenix: like, BYE and SDES [22:19] <Zeenix> wtay: what about cleaning rtprecv/send from all rtcp stuff [22:20] <wtay> Zeenix: not sure about that... [22:20] <Zeenix> wtay: me neither [22:21] <wtay> well, BYE and SDES cannot be handled by the plugin anyway [22:23] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-42.brain.net.pk] [22:24] steveb (st...@no...) left irc: advance! [22:24] Zeenix (ze...@13...) joined #gstreamer. [22:25] <Zeenix> oh thats the last thing i got: [22:25] <Zeenix> wtay: what about cleaning rtprecv/send from all rtcp stuff [22:25] <Zeenix> <wtay> Zeenix: not sure about that... [22:25] <Zeenix> <Zeenix> wtay: me neither [22:25] <Zeenix> *** Disconnected [22:26] <Zeenix> wtay: & i suppose you havent fixed that rtcp problem you found that day [22:26] <wtay> nope.. [22:26] <Zeenix> wtay: sorry i didnt get the time to copy the latest rtp on vfat today [22:28] <Zeenix> i am getting this error compiling the latest cvs gst: [22:28] <Zeenix> libtool: link: error: cannot link shared libraries into libtool libraries [22:28] <Zeenix> make[3]: *** [libgsttypes.la] Error 1 [22:29] <wtay> where? [22:29] <wtay> oh [22:29] <wtay> make clean? [22:29] <Zeenix> wtay: make [22:29] <wtay> maybe you need a newer libtool (I have 1.4.2)? [22:30] <Zeenix> wtay: website: libtool.sf.net ? [22:31] <wtay> no idea.. [22:33] <Zeenix> wtay: the gsmenc gets its rate in the newcaps( if it has one ) from osssrc & is the rate in gsmenc a user writable property [22:33] <Zeenix> ? [22:34] <wtay> nope [22:35] <Zeenix> i asked 2 questions [22:36] <wtay> yes, to the first question, no to the second [22:40] <Zeenix> wtay: ok, but if i send you the next rtp dir with the same old plugin info stuff, dont change it using GST_PLUGIN_FEATURE stuff [22:40] <wtay> ok [22:41] <omega> ok, I have a windoze filesystem that has been mostly wiped [22:41] <omega> anyone know any programs that can hunt for and attempt to reconstruct files? [22:41] <omega> on a fat32 ;-( [22:42] <Zeenix> omega: i hear these sort of things from my friends, i'll ask them next time i meet them [22:42] <wtay> uhm [22:42] <Zeenix> omega: i may even send you if its a small one [22:44] <Zeenix> wtay: i used the following pipelines: (1) -launch osssrc frequency=8000 format=16 channels=1 ! diksink location=me.wav [22:44] <Zeenix> (2) -launch disksrc ! gsmenc ! rtpsend [22:44] <wtay> ok [22:45] <Zeenix> disksrc location=me.wav i mean [22:45] <wtay> the proble of course is that no caps will be set on rtpsend [22:46] <Zeenix> (3) rtprecv ! gsmdec ! osssink frequency=......... [22:46] <Zeenix> wtay: yes but it if i dont use the gsm in between i hear my sound [22:47] <Zeenix> wtay: using above pipelines i hear crap [22:48] <wtay> uhm [22:51] arik (ar...@sd...) joined #gstreamer. [22:51] <arik> howdy [22:51] <Zeenix> wtay: any idea why ? [22:51] <Zeenix> yo [22:51] <wtay> Zeenix: no idea.. [22:51] <arik> omega: i don't think i understand tunneling correctly [22:52] <ajmitch> hey arik [22:52] <arik> hey aj [22:53] chillywilly (da...@d4...) joined #gstreamer. [22:53] <Zeenix> wtay: & i dont like the idea to check for payload type change on each packet reception *if* rtpsend tells us by sending rtcp packet [22:54] ChiefHighwater (pa...@te...) joined #gstreamer. [22:54] <wtay> Zeenix: rtcp is optional, so.. [22:55] <arik> does anyone understand ssh tunnelling? [22:55] <ajmitch> ahh, not well [22:55] <Zeenix> wtay: optional ? not ATM [22:55] <arik> heh [22:55] <arik> well, prob better then me :-) [22:55] <ajmitch> doubt it ;) [22:55] <arik> do you know how to use it at all? [22:55] <ajmitch> i can't remember if i've even used tunnelling ;) [22:56] <ajmitch> ahh... [22:56] <arik> ah [22:56] <arik> ok [22:56] <arik> maybe not [22:56] <ajmitch> sorry ;) [22:56] <arik> i have directions [22:56] <ajmitch> what's the issue? [22:56] <arik> and i have no idea what i'm supposed to do [22:56] <wtay> Zeenix: it should be optional [22:56] <wtay> Zeenix: especially the type negotiation is APP specific [22:57] Nick change: chillywilly -> danielb [22:58] <Zeenix> wtay: so what changes you are recommending, how could we make it optional ? [22:58] <wtay> Zeenix: I would only use RTCP type negotiation if the parameters for the media type are not defined as a payload in the RFC [23:00] <Zeenix> wtay: s/payload/payload_type ? [23:00] <wtay> yes [23:01] <Zeenix> wtay: payload=media type in the payload( data ), no other parameters of media it repressents [23:01] <Zeenix> s/payload/payloa_type [23:02] <wtay> ? [23:04] Zeenix2 (ze...@13...) joined #gstreamer. [23:04] <Zeenix2> wtay: repeat after: [23:04] <Zeenix2> wtay: payload=media type in the payload( data ), no other parameters of media it repressents [23:04] <Zeenix2> <Zeenix> s/payload/payloa_type [23:04] bstard (Lor...@a2...) joined #gstreamer. [23:04] <wtay> Zeenix2: ? [23:04] <arik> ok [23:04] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-36.brain.net.pk] [23:05] <arik> wtay: do you understand ssh tunnelling? [23:05] <wtay> arik: not really.. [23:05] <arik> heh [23:05] <arik> great :-) [23:05] <Zeenix2> wtay: i mean i got disconnected after the messages i repeated [23:06] <wtay> Zeenix2: I mean, I don't understand what you are saying [23:06] Nick change: Zeenix2 -> Zeenix [23:07] <Zeenix> wtay: the payload_type is a straightly mapped to media types like, 3 for gsm, .... [23:07] <wtay> Zeenix: yes [23:09] <Zeenix> wtay: me too didnt really get what you said about payload_type [23:09] <wtay> Zeenix: the payload type defines the media type, so let's use it [23:13] Zygo (zbl...@24...) left irc: Read error to Zygo[246.chipworks.com]: EOF from client [23:13] <Zeenix> wtay: ok, you are recommending the payload_type check in the pipeline, not by rtcp, right ? [23:14] <wtay> yes [23:14] <Zeenix> wtay: other stuff like rate, frequency will be negotiated by the rtcp ? [23:14] <wtay> yes, if needed [23:14] maYam (ma...@ca...) joined #gstreamer. [23:14] <maYam> hi everyone [23:14] <maYam> can anyone tell me what a 'care package' is? [23:14] tnt-out (tn...@h2...) got netsplit. [23:14] danielb (da...@d4...) got netsplit. [23:14] Zeenix (ze...@13...) got netsplit. [23:14] <wtay> maYam: yo [23:14] tnt-out (tn...@h2...) returned to #gstreamer. [23:14] danielb (da...@d4...) returned to #gstreamer. [23:14] Zeenix (ze...@13...) returned to #gstreamer. [23:15] <arik> um [23:15] <arik> a gift from your family? [23:15] <arik> mostly sent to summer camp? [23:15] <Zeenix> hello maYam [23:15] <maYam> oh right, just a letter or something.. ;) [23:15] <maYam> hi Zeenix [23:15] <arik> um [23:15] <arik> no [23:16] <arik> usually a box of neet stuff [23:16] Shippou (ir...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [23:16] <maYam> a present!! woohoo :) [23:17] <arik> heh [23:17] <Zeenix> wtay: right ? [23:17] <wtay> right [23:17] <Zeenix> wtay: that payload_type checking everytime wont be heavy ? [23:18] <wtay> Zeenix: not at all [23:19] <Zeenix> maYam: tell us the real situation of snooker matches last weekend, i wont believe wtay on that :) [23:20] <maYam> uh.. Zeenix, do i have to? <g> [23:20] <maYam> *blushes* 6-0 a total white-wash [23:21] <wtay> she fought hard, nut I was too good :) [23:21] <maYam> but it was because of the bad cue i was playing with.. ehem [23:21] <wtay> s/nut/but [23:21] <maYam> and the balls i had to hit, weren't round enough [23:22] <wtay> yeah, yeah... [23:22] Action: ajmitch coughs [23:23] <maYam> and wtay injured me severely before we started [23:23] <Zeenix> wtay: which version of mktemp you have ? [23:24] <wtay> Zeenix: no idea.. debianutils 1.15 it says.. [23:29] <Zeenix> wtay: remember i queried you about movie files with .dat ext [23:29] <Zeenix> wtay: the file xxxxxxx.dat say: [23:29] <Zeenix> .dat: RIFF (little-endian) data [23:30] <Zeenix> gst support this format ATM ? [23:31] <wtay> I don't think so [23:32] Uraeus (csc...@c1...) joined #gstreamer. [23:32] <Uraeus> hi dudes [23:32] <arik> hey Uraeus [23:32] <Zeenix> yo [23:33] <Uraeus> arik: hi, gotten anywhere with the GNOME2 gstmediaplayer? [23:33] <arik> no [23:33] <maYam> hey Uraeus [23:33] <arik> no no no no no [23:33] <arik> no [23:33] <wtay> yo [23:34] <arik> eazel-hacking is very very broken on my machine [23:34] <arik> and i haven't emailed darin yet to try to figure out why [23:34] <Uraeus> arik: what distro do you use? [23:34] <arik> redhat [23:35] <Uraeus> arik: you can use Havocs gnomehide RPMS then [23:35] <arik> Uraeus: yeah i know [23:35] <arik> but in the long run [23:35] <ajmitch> hey Uraeus [23:35] <arik> building it makes more sense [23:35] <ajmitch> arik: use v-b-s ;) [23:35] <arik> :-P [23:35] <Uraeus> arik: why spend time building when you could spend it coding :) [23:35] <arik> heh [23:35] <arik> i _like_ eazel-hacking aj :-P [23:36] <ajmitch> arik: i _like_ things that work :P [23:36] <arik> hehe [23:36] <arik> bite me :-) [23:36] <Uraeus> wtay: what is the module name in CVS for the java-bindings again? [23:36] <ajmitch> gst-bind [23:36] <ajmitch> iirc [23:37] <arik> ok aj [23:37] <arik> how does v-b-s work? [23:37] <ajmitch> check it out, RTFM ;) [23:37] <arik> heh [23:38] <ajmitch> cos it has support for various branches & setups [23:39] maYam (ma...@ca...) left irc: Client Exiting [23:41] <wtay> I have to sleep.. cya [23:41] Nick change: wtay -> wtay-zZz [23:51] <omega> anyone know what files or programs I need to touch to change the IP config on an interface on a debian box? [23:51] <arik> omega: hey [23:51] Action: omega is in virus hell [23:51] <arik> ouch [23:51] <arik> on a linux box? [23:51] <Uraeus> omega: I just talked with jrb at redhat, they are staying at a hotel called Howard Johson Oakland, maybe we should stay there too? [23:52] <arik> heh [23:52] <ds> how long are you going to be here? [23:52] <arik> i should go to als [23:52] <omega> Uraeus: too? [23:52] <omega> Uraeus: oh, you mean instead <g> [23:52] <omega> arik: no, windoze [23:52] <arik> omega: fuck it :-) [23:52] <omega> a sircam variant managed to totally trash a fat32 [23:52] <arik> when is als? [23:52] <omega> next week [23:52] <Uraeus> omega: what do you say? [23:52] <arik> heh [23:52] <arik> prob can't go at this point then huh? [23:53] <omega> Uraeus: fine with me, if it's similarly located and priced [23:53] <ajmitch> omega: /etc/network/interfaces [23:53] <omega> arik: yeah, you should be able to get in free still I think [23:53] <Uraeus> omega: I check prices tommorow and make arrangements then [23:53] <omega> linuxsymposium.org [23:53] <arik> omega: hmm [23:53] <omega> Uraeus: ok, ASAP [23:53] <arik> Uraeus: let me know what the hotal price it ok? [23:53] <arik> i may do that [23:53] <omega> arik: and of course, if you come, you've got to be there by the 6th for Battlebots [23:53] <arik> hmm [23:53] <arik> :-) [23:53] Nick change: tnt-out -> tnt [23:54] <omega> ajmitch: ok, um, what's the format? [23:54] <Uraeus> arik: ok, but I guess that if you are comming we could get a 3 person room cheaper than the 2 person price [23:54] <omega> iface eth0 ...... ? [23:54] <ajmitch> omega: text ;) [23:54] <arik> yeah [23:54] <omega> Uraeus: if there is such a thing [23:54] <ajmitch> iface eth0 inet static [23:54] <ajmitch> address 192.168.0.12 [23:54] <ds> you might try looking for hotels in emeryville, which is both a) close to downtown than the airport, and b) a nicer area [23:54] <Uraeus> omega: there where at GUADEC 2 :) [23:54] <arik> omega: what day should i get there / leave [23:54] <omega> Uraeus: heh [23:54] <ajmitch> omega: what else are you wanting to change? [23:54] <omega> arik: see http://www.temple-baptist.com/~omega/als-sched.txt [23:54] <Uraeus> arik: 5th to 12th [23:55] <omega> ajmitch: what about netmask/broadcast/gateway ? [23:55] <arik> hmm [23:55] <Uraeus> ds: well I prefer staying at the same hotel as the rest of the gnomers since it makes hanging out in the evening easier [23:55] <omega> ds: the best western is 8blocks from the conference [23:55] <omega> (11th vs. 3rd or something) [23:55] <omega> same crossstreet [23:55] <omega> Uraeus: yeah [23:56] <arik> omega: ridge run is cool with this? [23:58] <Uraeus> omega: I don't know if you remember this but at GUADEC 2 there was a few people who had managed to book a hotel looong away from the rest of us, and it was a hassle <g> [23:58] <omega> Uraeus: heh, I remember something like that <g> [23:59] Action: Uraeus is now officially enrolled to ALS :) [23:59] <omega> Uraeus: just now? [00:00] --- Wed Oct 31 2001 [00:00] Action: ds wonders if they have adequate bicycle parking [00:02] <Uraeus> omega: yes :) [00:02] <Uraeus> omega: I eithe forgot to enroll earlier or when I remembered the site was giving me a hard time :) [00:04] <Uraeus> omega: I think vishnu wants his mad patch reviewed :) [00:05] <omega> heh, when I get this dealt with, yeah [00:06] <Uraeus> arik: how interested are you in joining me and omega at ALS? should I try to book a three person room tommorow if they have such? [00:07] <arik> Uraeus: very interested, i just need to get a few peices of info [00:07] <arik> i will let you know asap [00:07] <Uraeus> arik: well I am going to sleep now, if you get the info you need tonight please mail me [00:07] <arik> alright [00:07] <arik> i can do that :-) [00:08] <Uraeus> arik: use -> ur...@li... [00:08] <Uraeus> night everyone [00:08] <arik> ok [00:08] Uraeus (csc...@c1...) left irc: Client Exiting [00:16] <arik> alright i need to get offline for a little bit [00:16] <arik> be back later [00:16] arik (ar...@sd...) left #gstreamer. [00:18] <Zeenix> omega: did you get to know that we got rtp to stream today ? [00:18] <omega> cool [00:19] <Zeenix> omega: i got to go now [00:19] <Zeenix> bi [00:20] Zeenix (ze...@13...) left irc: I pay 25 Rs( 0.5$ )/hour [00:31] ChiefHighwater (pa...@te...) left irc: [00:33] Shippou (ir...@ot...) joined #gstreamer. [00:51] Nick change: tnt -> tnt-out [00:55] ajmitch (me...@p5...) left irc: Ping timeout for ajmitch[p55-max3.dun.ihug.co.nz] [01:03] ajmitch (me...@p2...) joined #gstreamer. [01:05] omega (om...@te...) left irc: Ping timeout for omega[temple-baptist.com] [01:08] Action: BBB is away: zzz [01:08] Nick change: BBB -> BBB-zZz [01:11] omega (om...@te...) joined #gstreamer. [01:16] omega (om...@te...) left irc: Ping timeout for omega[temple-baptist.com] [01:24] omega (om...@te...) joined #gstreamer. [01:27] mwc (ma...@ly...) joined #gstreamer. [01:27] ajmitch (me...@p2...) left irc: Ping timeout for ajmitch[p23-max4.dun.ihug.co.nz] [01:27] ajmitch (me...@p1...) joined #gstreamer. [01:33] omega (om...@te...) left irc: Ping timeout for omega[temple-baptist.com] [01:35] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [01:43] omega (om...@te...) joined #gstreamer. [01:43] <omega> bleagh. this server is spontaneously rebooting ;-( [01:45] <ds> Is gstmediaply from HEAD supposed to work? [01:46] ds_ (ds...@c7...) joined #gstreamer. [01:46] <omega> probably not [01:46] <ds> ds_: hi [01:46] ds_ (ds...@c7...) left irc: [BX] Time wasted: 0 days 4 hours 26 minutes 15 seconds 92 milliseconds [01:52] <ds> so what does work? [02:01] Shippou (ir...@ot...) joined #gstreamer. [02:15] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [02:28] omega (om...@te...) left irc: Ping timeout for omega[temple-baptist.com] [02:34] omega (om...@te...) joined #gstreamer. [02:35] Shippou (ir...@ot...) joined #gstreamer. [02:56] <ds> hmmm... it appears the problem is with a52dec [02:59] <vektor> nooooooo |
From: IRC B. <wt...@us...> - 2001-11-01 05:28:08
|
******************************************************************* [03:00] <vektor> Hey ds, know anything about colour spaces and compositing? [03:00] <vektor> Come on man, you're like a smartypants. [03:00] <ds> heh, no [03:00] <vektor> I know you're down with the scene. [03:00] <vektor> Please? [03:01] <vektor> Ah fine, forget it. :) [03:20] omega (om...@te...) left irc: Ping timeout for omega[temple-baptist.com] [03:24] ajmitch (me...@p1...) left irc: Ping timeout for ajmitch[p12-max11.dun.ihug.co.nz] [03:26] ajmitch (me...@p1...) joined #gstreamer. [03:37] danielb (da...@d4...) left irc: Read error to danielb[d41.as12.nwbl0.wi.voyager.net]: EOF from client [03:39] bstard (Lor...@a2...) left irc: Ping timeout for bstard[a213-84-40-242.adsl.xs4all.nl] [05:34] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [06:01] <ds> ack! [06:02] <ds> mpeg2dec from CVS doesn't like gst [06:23] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [06:34] Shippou (ir...@cr...) joined #gstreamer. [06:38] Shippou (ir...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [07:26] <taaz> ds: how so? [07:29] mwc (ma...@ly...) left irc: [x]chat [07:35] <vektor> taaz: you awake? [07:35] steveb (st...@no...) joined #gstreamer. [07:39] <taaz> yeah [07:39] Shippou (ir...@ot...) joined #gstreamer. [07:40] <vektor> k justasec i want to show you something [07:40] <vektor> I'm starting to empart knowledge. [07:41] <vektor> www.dumbterm.net/graphics/dvd/ [07:41] <vektor> It's not done uploading the lawrence pictures though. [07:41] <vektor> Or some of the source. [07:41] <taaz> yo, don't f*ck with the candyman [07:41] <vektor> huh? [07:42] <taaz> scary movie week on tv ya know [07:44] <vektor> ok. [07:44] <vektor> Ok, almost done uploading. [07:45] <vektor> Ok, done. [07:45] <vektor> Check the page. [07:46] Action: vektor sleeps. [07:46] <taaz> so they just changed my cable channel lineup. someone musta thought hard about how to make sure no two consecutive channels are related. [07:48] <vektor> do you like the page? [07:49] <taaz> sure, looks good [07:49] <vektor> yay! [07:50] <taaz> dude, you should look at djangos.com for used dvds too [07:54] <taaz> nap time [07:54] Nick change: taaz -> taazzzz [08:07] <ds> anyone around? [08:07] <taazzzz> zzz [08:07] <ds> I almost believe that... [08:08] <taazzzz> just about to really go zzzz [08:08] <ds> what are people using instead of gstmediaplay? [08:09] shin_bet (sh...@ad...) joined #gstreamer. [08:09] arik (ar...@sd...) joined #gstreamer. [08:09] <arik> ki akk [08:10] <arik> er lo all [08:10] <taazzzz> ds: yell at arik [08:10] <arik> yell at me? [08:10] Action: arik runs away [08:10] <ds> I don't understand his crazy moon language [08:10] <arik> vektor: you around? [08:10] <arik> crazy moon language? [08:10] <taazzzz> ds says gstmediaplay is broke... [08:10] <arik> oh [08:10] <arik> could be [08:10] <taazzzz> sounds like your job ;) [08:10] <arik> heh [08:11] <arik> what's wrong ds? [08:11] <ds> gstmediaplay isn't "broke", more that I broke it. =) [08:11] <arik> oh [08:11] <arik> ok [08:11] Action: taazzzz really goes zzz now [08:11] <arik> what's up ds? [08:12] <ds> I'm not getting sound with gstmediaplay and anything with a52 streams [08:12] <arik> oh [08:12] <arik> ok [08:13] <ds> but I'm having difficulty getting gstreamer-launch to do a52, also [08:13] <arik> ok [08:13] <arik> so it is likely your system then? [08:13] <arik> have you tried reinstalling a52dec? [08:13] <ds> er, I mean, to autoplug a52dec. It will run a52dec if I specify it [08:13] <arik> oh [08:13] <arik> hmm [08:16] <ds> the other problem I've been tracking recently is that mpeg2dec from CVS + gst will segfault on ppc [08:16] <arik> hmm [08:16] <arik> you should ask haddess if he is seeing that too [08:17] <ds> crashed my computer extra-hard by using electric fence with gst... =) [08:18] <arik> ouch [08:18] <arik> :-) [08:24] <ds> oops, there is goes again. [08:24] <arik> i wish i could help [08:24] <ds> my theory of the universe is that it will end by swapping too hard [08:24] <arik> hehe [08:30] <ds> it's rather disturbing that I can kill Linux so easily with a user-space program [08:30] <arik> heh [08:30] <arik> it is isn't it? [08:33] <ds> /dev/hda9: Deleted inode 165706 has zero dtime. FIXED. [08:34] <arik> yep [08:35] <ds> Local time: ... 1962 [08:35] <ds> I love macs... =) [08:35] <arik> hahaha [08:35] <ds> it seems to have an affinity for 1933, though [08:38] <arik> hehe [08:39] <ds> you seem to be laughing a lot. You must be drunker than I [08:39] <arik> heh [08:39] <arik> no, just tired [08:50] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [08:56] Nick change: tnt-out -> tnt [09:20] shin_bet (sh...@ad...) left #gstreamer (Client Exiting). [09:32] <arik> shin bet [09:32] <arik> ok that's weird [09:35] <ajmitch> yep [09:36] <arik> hey aj [09:36] <arik> how goes it? [09:36] <arik> i'm one step closer [09:36] <ajmitch> alright ;) [09:36] <ajmitch> cool [09:36] <arik> to eazel-hacking working [09:36] <arik> hehe [09:36] <ajmitch> oh [09:36] <ajmitch> i thought u had gnome built ;) [09:36] <ajmitch> and were 1 step closer to a working gstplay ;) [09:36] <arik> haha [09:37] <arik> you wish :-P [09:37] <ajmitch> well, yeah ;) [09:37] <arik> but now [09:37] <arik> i'm getting a very weird error [09:37] <arik> bash: cd: foo: No such file or directory [09:37] <ajmitch> probably redhat ;) [09:37] <arik> when i try to run the command that should dir change me [09:37] <arik> haha [09:37] <arik> prounounced [09:37] <arik> ha ha [09:37] <arik> :-) [09:38] <ajmitch> when you get it all working on debian gnu/hurd, then i'll be impressed ;) [09:39] <arik> heh [09:39] <arik> i just want to know how what that error message even mean [09:39] <arik> s [09:40] <ajmitch> hmm [09:40] <ajmitch> something's quite wrong [09:40] <ajmitch> when does this error appear? [09:41] <arik> when i run the gnome2 command [09:41] <arik> which [09:41] <arik> among other things [09:41] <arik> changes the dir to /gnome-source/GNOME2 [09:41] <ajmitch> does it literally have 'foo' in the err message? [09:41] <arik> yep [09:42] <arik> bash: cd: foo: No such file or directory [09:42] <ajmitch> weird [09:42] <arik> ys [09:42] <arik> er yes [09:42] <ajmitch> almost as if a substitution isn't being done, and 'foo' is being used [09:42] <arik> yeah that's what i thought [09:42] <arik> but i can't find it [09:43] <ajmitch> complained on #gnome? [09:43] <arik> better [09:43] <arik> mailed darin [09:43] <ajmitch> ok [09:45] <ajmitch> probably a simple fix ;) [09:45] <arik> prob [09:45] <arik> i'm looking at it now [09:46] <arik> ah! [09:46] <arik> i think i get it [09:46] <arik> it runs this command [09:47] <arik> cd `eazel-build-attributes.sh source_place` [09:47] <arik> and eazel-build-attributes.sh source_place returns [09:47] <arik> foo [09:47] <arik> GNOME2 [09:47] <arik> bar [09:47] <arik> /gnome-source/GNOME2 [09:47] <arik> so it's trying to cd into each of em [09:48] <ajmitch> ok [09:48] <ajmitch> and fails on the cd foo [09:48] <arik> right [09:48] <arik> alright [09:48] <arik> i think i'm on the right tracj [09:48] Action: ajmitch hopes the gnu/hurd distro will start to pick up & get packages compiled for it [09:49] <arik> yes! [09:49] <ajmitch> main problem with hurd - no sound support yet ;) [09:49] <arik> fixed it! [09:49] <ajmitch> excellent! [09:49] <arik> heh [09:49] <arik> you run hurd? [09:49] <arik> alright! [09:49] <arik> i'm building gnome2! [09:49] <ajmitch> from time to time on a spare box [09:49] <arik> ah [09:50] <ajmitch> something interesting to play with ;) [09:50] <arik> now of course i have to wait to check out all of gnome2 [09:50] <ajmitch> hehe [09:50] <ajmitch> modem? [09:50] <arik> then build all of it [09:50] <arik> but at least i can send the patch to darin [09:50] <ajmitch> what speed is your computer? [09:50] <arik> and ask to commit it [09:50] <arik> yeah [09:50] <arik> modem [09:50] <arik> um [09:51] <arik> p3 [09:51] <arik> 700 [09:51] <arik> 256kb cache [09:51] <ajmitch> not bad [09:51] <ajmitch> much better than this heap of sh** [09:52] <arik> nope [09:52] <arik> heh [09:52] <arik> what are you running? [09:52] <ajmitch> 400mhz k6-2 [09:53] <arik> ouch [09:54] <ajmitch> my compiles are done on my 7200rpm drive, and it's got 384mb ram, so it's not a complete sh**heap [09:54] <ajmitch> close enough tho, especially playing video ;) [09:54] <arik> hehe [09:55] <ajmitch> damn i wish gnome builds didnt fail on esound for me [09:55] <arik> patch sent [09:55] <arik> why do they fail? [09:56] <ajmitch> some screwy thing with alsa [09:56] <ajmitch> i just never got round to trying to fix it up :) [09:56] <arik> ah [09:56] <arik> yeah [09:56] <arik> hehe [09:57] <arik> i have to get the sources of [09:57] <arik> autoconf automake libtool gettext pkgconfig gnome-common [09:57] <arik> gnome-xml intltool glib pango atk linc libIDL ORBit2 libxslt gtk+ [09:57] <arik> bonobo-activation gconf libart_lgpl libgnomecanvas libbonobo audiofile esound bonobo-config gnome-vfs libgnome libbonoboui libgnomeui librsvg eel [09:57] <arik> libgnomeprint libzvt [09:57] <arik> i'm up to libtool [09:57] <arik> ooh gettext [09:58] <ajmitch> heh [09:58] <ajmitch> trying to compile esd now, will see how it fails... [09:58] <arik> heh [09:59] Shippou (ir...@ot...) joined #gstreamer. [09:59] <arik> hi [10:00] <arik> once i get gnome2 build [10:00] <ajmitch> hmm, maybe i shoudl just remove these alsa dev files for now [10:01] <arik> i'm building gst under the gnome prefix [10:01] <arik> so i can install it [10:01] <arik> heh [10:01] <arik> maybe you should get a soundcard that doesn't use alsa :-P [10:01] <arik> i wish i could [10:01] <ajmitch> mine doesn;t ;) [10:01] <arik> um [10:02] <arik> why are you using alsa then? [10:02] <arik> pkgconfig! we're smoking now! [10:02] <ajmitch> i'm not ;) [10:02] <ajmitch> the packages are just installed [10:02] <arik> hah [10:02] <ajmitch> i *used* to use alsa [10:02] <arik> ok [10:02] <arik> rmeove the dev files! [10:02] <ajmitch> ok, removed devel packages, we'll see how she goes ;) [10:03] <arik> heh [10:04] <arik> gnome-common nah nah nah nah [10:04] Action: ajmitch has built most of that crap before ;) [10:05] <arik> gnome-xml we smocking now! [10:05] <arik> not smoking [10:05] <arik> smocking [10:05] <ajmitch> this just the checkout? [10:05] <arik> heh [10:05] <arik> me too [10:05] <arik> but not for awhile [10:05] <arik> yep [10:05] <ajmitch> i need a new glib, too :) [10:05] <arik> heh [10:06] <ajmitch> come on.... build.... [10:06] <ajmitch> excellent [10:06] <ajmitch> it works [10:06] <arik> heh [10:06] <arik> there you go [10:07] <ajmitch> cool, now the nightly cron job can work properly ;) [10:07] <arik> woo! [10:07] <arik> what are you cronin? [10:07] <arik> rebuild of cvs? [10:07] <ajmitch> yep [10:07] <arik> man i missed eazel-hacking [10:07] <arik> two commands to build gnome2 [10:07] <ajmitch> will turn on gstreamer builds again too, now that i patched automake [10:07] <ajmitch> heh [10:07] <arik> heh [10:07] <ajmitch> one command for me ;) [10:07] <arik> gnome2 && rebuild-all [10:08] <arik> woo! [10:08] <arik> oh? [10:08] <ajmitch> yeah [10:08] <arik> what is it? [10:08] <ajmitch> just a script with the 2 or 3 things to run for v-b-s ;) [10:08] <arik> oh ok [10:08] <arik> whatever :-) [10:08] <ajmitch> basically it's only 2 commands for v-b-s as well [10:09] <ajmitch> [ajmitch @ ajmitch esound] less /gnome/build-gnome [10:09] <ajmitch> #!/bin/bash [10:09] <ajmitch> PATH=/gnome/head/INSTALL/bin:/opt/gnome/bin:/home/ajmitch/bin:$PATH [10:09] <ajmitch> source /home/ajmitch/bin/gnome-head [10:09] <ajmitch> update-vicious.sh [10:09] <ajmitch> #source /home/ajmitch/bin/gnome-head [10:09] <arik> heh [10:09] <ajmitch> rebuild.sh [10:09] <arik> that is way more complicated looking [10:09] <ajmitch> yeah [10:09] <arik> we all though v-b-s was lame at eazel :-P [10:09] <arik> heh [10:09] <ajmitch> sooo much more complicated [10:09] <ajmitch> ;) [10:09] <arik> we were always trying to get george to switch [10:09] <arik> hahah [10:09] <ajmitch> woohoo, glib is building now [10:10] <arik> heh [10:10] <arik> intltool! [10:10] <ajmitch> i'm gonna do some gnome & gst coding after exams, i'll need this stuff ;) [10:10] <arik> heh [10:10] <arik> i'll believe it when i see it [10:11] <ajmitch> :P [10:11] <arik> glib! [10:11] <arik> heh [10:11] <ajmitch> don't you believe me? ;) [10:11] <arik> we'll see :-) [10:11] <ajmitch> sigh, such little faith ;) [10:11] <arik> i've heard it before [10:11] <arik> heh [10:12] <arik> glib2 baby [10:12] <ajmitch> ok, when shoudl i try & have some code to you by? ;) [10:13] <arik> hehe [10:13] <arik> asap! [10:13] Action: arik cracks the whip [10:13] <ajmitch> :P [10:13] <arik> heh [10:13] <ajmitch> i got another project for the next couple of weeks [10:13] <ajmitch> which i get paid for :P [10:13] <arik> yeah [10:14] <arik> schol [10:14] <arik> er school [10:14] <ajmitch> nope [10:14] <arik> paid? [10:14] <arik> ah [10:14] <ajmitch> stuff done in python [10:14] <arik> joke [10:14] <arik> heh [10:14] <ajmitch> not school stuff [10:14] <arik> ah [10:14] <arik> pango pango [10:15] <arik> pango is sweet [10:15] <ajmitch> hmm [10:15] <ajmitch> had a bit of a problem - quite a few build files were owned by root [10:15] <ajmitch> chown fixes that pretty quick ;) [10:15] <arik> ouch [10:15] <arik> heh [10:15] <arik> v-b-s sucking! [10:15] <arik> jk [10:15] <ajmitch> no, silly me building as root one time ;) [10:16] Action: ajmitch doesn't know why ;) [10:16] <arik> hehe [10:16] <arik> silly silly you [10:17] Action: ajmitch wonders about making a kde gst app ;) [10:17] <ds> you think building as root is bad, try building in 1933 [10:17] <arik> get out now! :-) [10:17] <ajmitch> just to piss off some kde people mainly ;) [10:17] <arik> hehe [10:17] <arik> i am imagining a lot of timestamp error's in your future ds [10:17] <arik> heh [10:17] <arik> i thought kde liked gst? [10:17] <arik> atk! [10:17] <arik> what the heck is atk? [10:18] <ajmitch> umm, some of the more outspken kde multimedia guys *hate* anything starting with g [10:18] <arik> or linc [10:18] <ajmitch> atk is accessibility toolkit [10:18] <ajmitch> linc is to do with ORBit, iirc [10:18] <arik> heh [10:18] <arik> they do hate gnome things [10:18] <arik> but gst i thought they liked [10:18] <arik> there is arts support for gst after all [10:19] <arik> so at least some of the saner kde people [10:19] <arik> are paying attention [10:19] <ajmitch> yes [10:20] <ajmitch> some are rabid haters tho [10:20] <arik> well, there are annoying people everywhere :-) [10:20] Action: ajmitch thinks of Njaard - noatun author, arrogant as hell [10:20] <arik> heh [10:20] <arik> is he? [10:20] <ajmitch> yep [10:21] <arik> heh [10:21] <arik> sounds funny [10:21] <arik> libxslt [10:21] <arik> so noatun boy things he can do everything himself? [10:22] <ajmitch> yep [10:22] <arik> heh [10:22] <arik> well [10:22] <arik> good luck to him [10:23] <ajmitch> i still think noatun is a piece of shite ;) [10:23] <arik> haha [10:23] <arik> i used it once [10:23] <arik> a long long time ago [10:23] <arik> i don't really remember it [10:23] <ajmitch> interesting architecture - all UIs, playlists, effects, etc are plugins [10:23] <ajmitch> so you can have multiple UIs running at once [10:23] <arik> hmm [10:23] <arik> that is interesting [10:23] <arik> weird [10:24] <arik> why is that a good thing? [10:24] <ajmitch> the sort of thing you *could* do for gstmediaplay if you wanted to [10:24] <ajmitch> i dunno ;) [10:24] <ajmitch> makes it easy to switch from winamp skins to a native kde ui, to weird ui plugins [10:24] <arik> oh [10:24] <arik> i'm not making gstmediaplay skinable [10:24] <arik> ever [10:24] <ajmitch> and there are different style playlists, too [10:24] <arik> well [10:25] <arik> hmm [10:25] <arik> that's kinda cool [10:25] <ajmitch> heh [10:25] <arik> it's gonna be theoretically easy to do that with gstmediaplay though [10:25] <ajmitch> good [10:25] <ajmitch> it doesn't need to be [10:25] <arik> cause they whole thing is gonna be a shell embedding bonobo controls [10:25] <ajmitch> someone can write their own damned player with libgstplay ;) [10:25] <ajmitch> k [10:25] <arik> hehe [10:25] <arik> yep [10:25] <arik> the player controls will be a control [10:25] <arik> the playlist will be a control (hopefully haddess') [10:26] <ajmitch> should be nice ;) [10:26] <arik> but the idea that all media apps need to look completely different then the rest of the apps [10:26] <arik> is really weird to me [10:26] <arik> What is Noatun? [10:26] <arik> The ultimate media player for WAV, MP3, OggVorbis, MPEG-1, and DivX. It's for [10:26] <arik> KDE 2. [10:26] <arik> heh [10:26] <arik> the ultimate media player [10:26] <ajmitch> it uses arts for playing stuff [10:26] <arik> right [10:26] <arik> which can use gst [10:27] <arik> heh [10:27] <ajmitch> arts has a hacked up video system tho [10:27] <ajmitch> can arts use gst for much? [10:27] <arik> don't really know [10:27] <arik> hacked up? [10:28] <ajmitch> apparantly [10:28] <arik> hmm [10:28] <ajmitch> dunno much [10:29] <ajmitch> could be ok now [10:29] <arik> heh [10:29] <arik> gtk_ [10:29] <arik> + [10:29] <arik> 17 to go :-) [10:33] <ajmitch> this could take awhile... [10:33] <arik> yes [10:33] <arik> a _long_ while [10:33] <ajmitch> i think i'll turn off other compiles tonight (one starting with K) [10:33] <arik> heh [10:34] <arik> you compile kde? [10:34] <ajmitch> for some weird reason, yeah [10:34] <ajmitch> i haven't used it for months ;) [10:34] <arik> heh [10:34] <arik> months [10:34] <ajmitch> and i only used to use konq [10:34] <ajmitch> now galeon is my friend ;) [10:35] <arik> heh [10:35] <ajmitch> ok, kde build turned off, gstreamer is back on [10:35] <arik> he [10:35] <arik> h [10:36] <ajmitch> gst build was killing my system cos of automake after an update ;) [10:36] <arik> right [10:37] <ajmitch> was bloody annoying to see that most processes had been killed ;) [10:37] <arik> heh [10:37] <arik> anyway i'm off to bed [10:37] <ajmitch> including X [10:37] <ajmitch> night [10:37] <arik> night [10:37] arik (ar...@sd...) left irc: sleeeep [11:15] Nick change: tnt -> tnt-zZz [12:34] ajmitch (me...@p1...) left irc: Ping timeout for ajmitch[p12-max11.dun.ihug.co.nz] [12:37] ajmitch (me...@p1...) joined #gstreamer. [13:02] bstard (Lor...@a2...) joined #gstreamer. [13:43] chillywilly (da...@d1...) joined #gstreamer. [13:44] ajmitch (me...@p1...) left irc: Ping timeout for ajmitch[p12-max11.dun.ihug.co.nz] [13:46] ajmitch (me...@p1...) joined #gstreamer. [13:47] chillywilly (da...@d1...) left irc: Read error to chillywilly[d176.as7.nwbl0.wi.voyager.net]: EOF from client [13:48] danielb (da...@d1...) joined #gstreamer. [13:53] bstard (Lor...@a2...) left irc: Ping timeout for bstard[a213-84-40-242.adsl.xs4all.nl] [13:53] bstard (Lor...@a2...) joined #gstreamer. [13:59] danielb (da...@d1...) left irc: Read error to danielb[d176.as7.nwbl0.wi.voyager.net]: EOF from client [14:06] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [14:18] vektor (reet@HSE-Kitchener-ppp3506142.sympatico.ca) left irc: Ping timeout for vektor[HSE-Kitchener-ppp3506142.sympatico.ca] [14:18] vektor (reet@HSE-Kitchener-ppp3506142.sympatico.ca) joined #gstreamer. [14:48] asmod (stevec@64.5.222.2) joined #gstreamer. [14:48] cinap_lenrek (thefreak@62.153.70.75) joined #gstreamer. [14:48] <cinap_lenrek> hi [14:49] cinap_lenrek (thefreak@62.153.70.75) left irc: User abort with 5 Ctrl-C's [14:51] brainfuck (thefreak@62.153.70.75) joined #gstreamer. [14:52] brainfuck (thefreak@62.153.70.75) left irc: Read error to brainfuck[62.153.70.75]: EOF from client [14:54] brainfuck (thefreak@62.153.70.75) joined #gstreamer. [14:54] <brainfuck> hm [14:55] <brainfuck> can i tell a question? [14:55] Nick change: brainfuck -> brainfuck|bad_english [14:56] <brainfuck|bad_english> hello? [14:57] brainfuck|bad_english (thefreak@62.153.70.75) left #gstreamer. [15:05] Zygo (zbl...@24...) joined #gstreamer. [15:10] Shippou (ir...@ot...) joined #gstreamer. [15:41] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:04] evil_monk (trevo@208.141.162.68) left irc: Ping timeout for evil_monk[208.141.162.68] [16:08] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:13] danielb (da...@d1...) joined #gstreamer. [16:19] danielb (da...@d1...) left irc: Ping timeout for danielb[d10.as29.nwbl0.wi.voyager.net] [16:23] danielb (da...@d8...) joined #gstreamer. [16:29] Action: BBB-zZz is back (gone 15:20:57) [16:29] Nick change: BBB-zZz -> BBB [16:36] danielb (da...@d8...) left irc: Ping timeout for danielb[d88.as7.nwbl0.wi.voyager.net] [16:42] danielb (da...@d1...) joined #gstreamer. [16:51] sienap (sy...@s3...) joined #gstreamer. [16:58] vektor_ (wb...@pe...) joined #gstreamer. [16:58] <vektor_> yo yo yo. [17:01] sienap (sy...@s3...) left irc: Read error to sienap[s340-modem2328.dial.xs4all.nl]: Connection reset by peer [17:10] Nick change: taazzzz -> taaz [17:11] <taaz> ds? [17:18] <vektor_> taaz? [17:18] <vektor_> ah, time for class [17:18] vektor_ (wb...@pe...) left irc: too bad the scene is dead [18:16] Nick change: danielb -> chillywilly [18:22] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [18:24] evil_monk (trevo@208.141.162.68) joined #gstreamer. [18:28] vektor_ (wb...@pe...) joined #gstreamer. [18:28] <vektor_> yo yo [18:33] <steveb> yo yo, yo [18:36] <vektor_> yo yo. [18:48] <taaz> yo [18:49] <taaz> vektor_: your dvd page, you should use png alpha channel rather than just a white background [18:49] <taaz> as if that matters ;) [18:50] <vektor_> taaz: Um. [18:50] <vektor_> taaz: I was making a point, that's all. [18:50] <vektor_> But you're correct. [18:55] wingo (wi...@rd...) joined #gstreamer. [18:55] <wingo> yop [19:02] <vektor_> yo! [19:02] <vektor_> hey so I have a hardcore graphics question. [19:02] <vektor_> Is there someone here who is hardcore? [19:02] <vektor_> I'll just ask. Y'CbCr is a linear transform of gamma corrected R'G'B'. [19:03] <vektor_> Say we have a gamma value of g. [19:03] <vektor_> So, if I want to do a gradient in Y'CbCr and correct for gamma, how do I do it. [19:03] <vektor_> One way is to take my Y'CbCr colour, convert to R'G'B', gamma correct to RGB, interpolate, convert back to R'G'B', then convert back to Y'CbCr. [19:04] <vektor_> However, I'm curious. Can i bypass the R'G'B' stage and just apply the gamma curve to the Y'CbCr components? [19:04] <vektor_> Since they're a linear transform? [19:04] <taaz> sure [19:04] Action: taaz has no clue ;) [19:04] <vektor_> oh. [19:09] Action: wingo is not hardcore :) [19:10] <wingo> intuition says yes [19:10] <wingo> but we all know what assuming does [19:12] <steveb> wingo: makes an arse of me and you? [19:18] <vektor_> haha [19:18] <vektor_> my high school calculus teacher used to say that alot. [19:19] <chillywilly> it is you and me [19:19] <chillywilly> duh [19:20] <chillywilly> ass - u - me [19:20] <chillywilly> :P [19:21] <steveb> i was paraphrasing [19:23] <chillywilly> well just look at the word damnit ;P [19:25] <vektor_> well, going home [19:25] vektor_ (wb...@pe...) left irc: too bad the scene is dead [19:34] <wingo> ok, going to skool. [19:35] wingo (wi...@rd...) left irc: biab. [19:35] <taaz> going.... nowhere [19:41] <chillywilly> going crazy [19:41] <taaz> whoop! [19:42] ChiefHighwater (pa...@te...) joined #gstreamer. [19:43] <chillywilly> chief! [19:44] <ChiefHighwater> Ello [19:44] Action: ChiefHighwater may disappear at any maoment [19:45] <ChiefHighwater> our server is having issues [19:45] <chillywilly> :( [19:45] <ChiefHighwater> yesterday it was a virus on the network, today the proc seems to be flaky [19:46] <chillywilly> I wish my church was more high tech...they don't have a network or anythin [19:46] <ChiefHighwater> we'll see how long the machine stays up [19:46] <chillywilly> winblows? [19:46] <chillywilly> using winblows is like a sin ;) [19:46] <ChiefHighwater> yes [19:46] <ChiefHighwater> a sircam varient [19:47] <ChiefHighwater> chillywilly:isn't there a scripture in Hezekiah about using evil software? [19:47] <chillywilly> haha [19:48] <chillywilly> well I just think Jesus would want me to share my software [19:48] <chillywilly> :) [19:48] <ChiefHighwater> Hezekiah isone of my favorite books, right up there with Second Hesitations [19:52] ChiefHighwater (pa...@te...) left irc: Ping timeout for ChiefHighwater[temple-baptist.com] [19:52] <chillywilly> woops [19:52] <chillywilly> there he goes [20:06] Action: chillywilly is away: soda [20:14] Nick change: tnt-zZz -> tnt [20:14] <tnt> Good morning. [20:16] <ajmitch> morning [20:22] <ds> how do i get gst to run in one process, so debugging is easier? [20:24] thomasvs (th...@21...) joined #gstreamer. [20:25] thomasvs (th...@21...) left irc: [x]chat [20:27] dobey (do...@dr...) joined #gstreamer. [20:28] Action: chillywilly is back (gone 00:22:33) [20:29] <dobey> whee [20:30] <chillywilly> eh? [20:31] <dobey> gnupdate [20:31] <dobey> i think it will make me a happy monkey [20:36] <chillywilly> if you see miguel just give him a swift kick to the head for me ok? [20:37] <dobey> uh [20:37] <dobey> why? [20:37] <chillywilly> many reasons [20:37] <dobey> he's in germany anyway i think [20:37] <dobey> name a few [20:37] <chillywilly> but I'm not getting into it [20:37] <ds> dobey: why not just write deb2rpm? (ducks) [20:37] <dobey> then neither am i [20:37] <dobey> ds: like alien? [20:37] <dobey> because i don't want debs [20:38] <dobey> and rpm is starting to piss me off [20:38] <ds> isn't that just rpm2deb? [20:38] <dobey> because of the building stuff [20:38] wingo (wi...@rd...) joined #gstreamer. [20:38] <dobey> no [20:38] <dobey> alien does *2* [20:38] <tnt> Hey wingo. [20:38] <wingo> hey tnt. [20:50] <dobey> chillywilly: i can't just walk around and give random boots to the head, i need reasons [20:51] <chillywilly> ooooh [20:51] <chillywilly> well, the whole I don't wanna be a GNU project tirade and public dissing of RMS deserves a boot, imho :P [20:51] <dobey> but if you don't have one other than him being miguel or the antitrust thing, then forget it [20:51] <dobey> not really [20:52] <chillywilly> antitrust thing? [20:52] <dobey> you just have to see the issue from the point of view he's coming from [20:52] <dobey> yeah, miguel was in antitrust [20:52] <dobey> for like 5 seconds [21:00] <vektor> huh? [21:00] <vektor> miguel was in antitrust? [21:00] <dobey> yes [21:00] <vektor> haha [21:00] <vektor> that's funny :) [21:00] <dobey> yeah, i know [21:00] <vektor> he's pretty fun. [21:00] <vektor> man, i should get ximian to hire me. [21:01] <dobey> eh [21:01] <vektor> that would be kinda neat. [21:01] <dobey> he's ok [21:01] <dobey> depends on what you do [21:01] <vektor> hey, you work at ximian. [21:01] <vektor> neat. [21:01] <dobey> yes, i do [21:01] <ajmitch> how many ppl work at ximian now? [21:01] <vektor> what's nat doing now? [21:01] <dobey> i don't know, a bunch [21:01] <vektor> nat seems (imho) the smart one. [21:01] <dobey> right now... he's probably in a meeting [21:02] <taaz> ds? [21:03] <vektor> haha. [21:03] <ds> yep [21:03] <taaz> yo, last night you asked about a52 not working in gstmediaplay? [21:03] <vektor> ds: hey ds, you know something about gamma and compositing, right? [21:03] Action: vektor wishes wtay was here. [21:04] <ds> vektor: did you ask me that last night? and I said "no know aqui"? [21:04] <ds> taaz: yes [21:04] <taaz> ds: ok the likely cause of that is the static autoplugger [21:04] <taaz> so mpeg2parse is just choosing to hook audio up to mad [21:05] <taaz> since it has multiple possible outputs (a52, mad, pcm?, whatever) [21:05] <ds> doesn't it try to match mime types or something? [21:06] <taaz> you can use test/mpeg2parse3.c to play vob's if you want since that's hardcoded [21:06] <taaz> ds: it creates the pipeline before looking at the stream. so it doesnt know its a52 vs mpeg audio [21:06] <ds> yes, i've been doing that, roughly, except I'm now tracking down a bug in * [21:07] <ds> * = {libmpeg2dec, gstmpeg2dec, ?} [21:07] <taaz> ds: so the answer is to create the pipeline automagically but i guess that's not done yet and non-trivial [21:07] <taaz> what's the problem with those? [21:07] <ds> segfaults on ppc [21:08] <dobey> ooh [21:08] <dobey> ds has a ppc? [21:08] <ds> gstreamer-launch filesrc=thx.vob ! mpeg2parse video0! queue ! { mpeg2dec ! fakesink } [21:08] <ds> two, actually [21:08] <dobey> ah [21:08] <dobey> which ons? [21:08] <dobey> ones [21:09] <ds> G4 AGP, and a powerbook pismo [21:09] <dobey> ah, cool [21:09] <dobey> i wish this was a pismo [21:10] <ds> taaz: for whatever reason, gstmediaplay doesn't segfault [21:11] <wingo> ds: not smp, is it? [21:12] <ds> wingo: no [21:12] <wingo> it's not supposed to be video_00? /me is ignorant, but i seem to recall seeing that [21:13] <taaz> eh? segfault should be easy to catch [21:13] <taaz> you could run that launch with --gst-mask=-1 at least [21:13] <taaz> or just do $ libtool gdb gstreamer-launch and run it that way [21:13] <ds> yes, it's easy to catch, and even easier to find out that a buffer is getting freed too many times [21:13] <taaz> oh yeah! [21:14] <taaz> so i get those too [21:14] <ds> MALLOC_CHECK_=2 is your friend [21:14] <taaz> actually, i get warnings like buffer_unref called at refcount=0 or something [21:14] <ds> on i386? I don't see this on i386. And not with mpeg2dec-0.2.0. just from cvs [21:18] <taaz> i just use my libmpeg2 debs... i think they have some stuff from cvs but not too much. just enough to make it build PIC i think [21:19] <taaz> i suppose you are using the new altivec stuff.. [21:19] ChiefHighwater (pa...@te...) joined #gstreamer. [21:19] <ds> that's why I want to use CVS [21:22] <taaz> well... figure out which commit to cvs broke it. fun fun fun! [21:23] <ChiefHighwater> taaz:broke what? [21:23] Action: ChiefHighwater is out of the loop [21:23] <wingo> chillywilly: mpeg2dec [21:23] <wingo> not a GP [21:23] <wingo> GStreamer's Problem (tm) [21:23] <wingo> on ppc only [21:23] <wingo> :-) [21:24] <ajmitch> hehe [21:24] <ajmitch> doesn't matter then ;) [21:24] <ds> hmmm... I seem to have less of a handle on the problem than I thought. the -launch line that I gave segfaults, but s/fakesink/xvideosink/ works [21:25] <wingo> strange... [21:25] <wingo> a GP then :-\ [21:28] <ds> unref buffer 0x1009ae0 (1) [21:29] <ds> freeing buffer 0x10090ae0 [21:29] <ds> then aborts because free() is called with an invalid pointer [21:30] Action: taaz whines a little more about the lack of a test suite [21:31] <taaz> if you use a non-altivec mpeg2dec does it work? [21:31] <ds> no [21:32] <ds> I just added an environment variable to libmpeg2dec so I can turn off acceleration [21:32] <taaz> if you just pipe video to fakesink without queue/thread does it work? [21:32] <wingo> maybe it has to do with bufferpools [21:32] <dobey> hrmm [21:32] <taaz> is it always in the same spot? [21:33] <wingo> i don't know if fakesink's bufferpool is sane, or if it has one [21:33] <wingo> i thought it did tho [21:33] <ds> it also has a problem with mpeg2dec ! videoscale ! xvideosink [21:36] <ds> filesrc ! mpeg2parse video_0! fakesink doesn't run [21:36] <tnt> ds: Don't you need to decode it too? [21:37] <ds> no, just dumping the mpeg stream [21:37] <ds> filesrc ! mpeg2parse video_0! queue ! { fakesink } works [21:39] <ds> yes, it seems to always fail in the same spot [21:46] <ds> is it an autoconf bug that causes the failure in make distcheck? [21:47] chillywilly (da...@d1...) left irc: [21:51] Nick change: tnt -> tnt-out [22:01] <dobey> hrmm [22:02] <wingo> ds: i don't completely understand that. it would be a automake thing [22:02] <wingo> the in's are not getting generated [22:02] <wingo> probably due to some strange difference between the set of files in AC_OUTPUT and the set that are included in SUBDIRS and the set of DIST_SUBDIRS (which I don't understand) [22:03] <wingo> but by playing with SUBDIRS i have fixed some of the ones in the past, so that seems to be the genereal strategy for now :-\ [22:13] <wingo> i wonder what version thomasvs has [22:19] Uraeus (csc...@c1...) joined #gstreamer. [22:19] <Uraeus> hi howdy ho [22:19] <dobey> hehe [22:19] <dobey> you rule :-) [22:24] <wingo> hey Uraeus [22:24] <Uraeus> hi wingo [22:25] <ajmitch> hello Uraeus [22:28] <Uraeus> hi ajmitch [22:29] Zeenix (ze...@13...) joined #gstreamer. [22:30] <Zeenix> yo [22:30] <Zeenix> Uraeus: wtay gone till tommorow ? [22:30] <Uraeus> Zeenix: don't know I am not his mother :) [22:31] thomasvs (th...@ad...) joined #gstreamer. [22:31] <thomasvs> hi [22:32] <Zeenix> Uraeus: no, what he said before going, "be back soon" or "gotta sleep now" ? [22:33] <Uraeus> Zeenix: don't think he has been in today [22:33] <Zeenix> Uraeus: ok [22:41] BBB (BB...@uc...) left irc: [x]chat [22:43] <wingo> thomasvs: what version of autoconf do you have? [22:44] BBB (BB...@uc...) joined #gstreamer. [22:44] <wingo> BBB: what did you say you wanted beside your name in the whoswho? [22:44] <BBB> mjpegtools :) [22:44] <BBB> http://mjpeg.sourceforge.net/ [22:45] <wingo> okeydoke, i'll do that this evening :) [22:45] <vektor> BBB: hey dude you there? [22:45] <BBB> yes [22:45] <vektor> BBB: i've been discussing interlacing support in xv on the xpert list. [22:45] <BBB> ok [22:45] <BBB> and? [22:45] <vektor> BBB: thought you might be interested. it's going really well. [22:45] <BBB> oh, good :) [22:45] <vektor> well, they're going to add an xv port attribute that will let us handle interlacing and stills better. [22:45] <vektor> it's coooool. [22:46] <BBB> nice :):):) [22:47] <taaz> so, are pic or shared libXv etc coming out? [22:47] <wingo> you have three heads? ;) [22:47] <wingo> it *is* halloween :) [22:47] <BBB> it is [22:47] <BBB> .... [22:47] <wingo> spooky. [22:48] Action: wingo goes home [22:48] wingo (wi...@rd...) left irc: peace. [22:51] Action: BBB is still doing v4lsrc [22:51] <BBB> and v4lmjpegsrc [22:51] Action: dobey is still looking for a package manager that doesn't suck [22:53] <BBB> apt [22:53] Action: BBB loves apt [22:53] <dobey> no no [22:53] <dobey> a package manager [22:53] <BBB> apt-get install gnome [22:53] <dobey> and one that doesn't suck [22:53] <BBB> dpkg [22:53] <BBB> :) [22:53] <dobey> dpkg sucks [22:53] <BBB> apt rocks [22:53] <dobey> apt doesn't rock [22:54] <dobey> sigh [22:54] <BBB> apt rocks [22:54] <BBB> apt-get install gtk [22:54] <dobey> no [22:54] <dobey> apt sucks [22:54] <BBB> apt-get upgrade [22:54] <BBB> apt-get update [22:54] <dobey> apt does not rock [22:54] <BBB> that's all I need to know [22:54] <BBB> apt rocks [22:54] <dobey> ok [22:54] <dobey> now i'm just going to say stfu [22:54] <ds> apt-get moo [22:54] <dobey> because you aren't saying anything [22:54] <BBB> lol [22:54] <BBB> or else? [22:55] <dobey> or else what? [22:55] Action: BBB is very good in not doing what other people tell him to do [22:55] <dobey> apt is not a fucking package manager [22:55] <Uraeus> ChiefHighwater: you don't happen to have Erik within earshot do you? [22:55] <dobey> anyway [22:55] Action: dobey goes back to reading [22:55] <dobey> and working [22:56] <taaz> what are you looking for? [22:56] <Zeenix> Uraeus: what is the ver. of libtool-libs on your sys ? [22:56] <ChiefHighwater> Uraeus:as a matter of fact, i do [22:56] <dobey> one that does what i want it to do and does require evil insane hacks to do it [22:56] omega (om...@om...) joined #gstreamer. [22:56] <omega> Uraeus: ? [22:56] steveb (st...@no...) left irc: Read error to steveb[node1ee11.a2000.nl]: EOF from client [22:56] <dobey> one that i can help guide it's future [22:56] <ChiefHighwater> there you go Uraeus [22:56] <Uraeus> omega: my flight is KL1140 [22:57] Action: omega goes offline again in a few minutes, working on upgrading server with new disk [22:57] <taaz> dobey: one what? define "package manager" [22:57] <omega> flight into Oakland? [22:57] <dobey> taaz: dpkg is a package manager, rpm is a package manager, gpkg is a package manager [22:57] <Uraeus> omega: into San Francisco [22:57] <Uraeus> omega: is there a separate airport in Oakland? [22:58] <omega> uh, yeah [22:58] <Uraeus> omega: the hotel is HOWARD JOHNSON EXPRESS BAY BRIDGE [22:58] <omega> SF airport is quite a ways from Oakland [22:58] <omega> afaik [22:58] <ajmitch> Uraeus: calm down, no need to shout ;) [22:58] <Uraeus> omega: 423 7TH STREET, BROADWAY AND 7TH [22:58] <Uraeus> ajmitch: I am just cut and pasting :) [22:58] <ajmitch> hehe, i guessed that ;) [22:58] <Uraeus> omega: hmm, guess I need to get a buss or something then [22:59] Zygo (zbl...@24...) left irc: Read error to Zygo[246.chipworks.com]: Connection reset by peer [22:59] <ChiefHighwater> maybe hotel has airport shuttle [22:59] <ChiefHighwater> is likely [22:59] <Uraeus> ChiefHighwater: from SF? [22:59] <ChiefHighwater> possibly, yes [22:59] <ChiefHighwater> SFO is a large airport, many hotels have shuttles to it [22:59] <Uraeus> ok, I will try to call them then to check [22:59] <ChiefHighwater> or they can sheduel one for you [22:59] <ds> Uraeus: exit the terminal, look for "shuttle vans", and take "Bay Porter" to oakland [23:00] <Uraeus> ds: thanks [23:00] <ChiefHighwater> right, unless HoJo will do it for free [23:00] <ds> it will be about $24 [23:00] <omega> ok, so I'll be finding my way to the hotel from the Oakland airport... [23:00] <ds> ChiefHighwater: not likely [23:00] <omega> and I'll get there an hour ahead of you, Uraeus... [23:01] Action: ds should fix his car [23:01] <Uraeus> omega: well the problem is that for me to fly to Oakland would probably be very expensive, when I went to LW in San Jose I took the plane to SF and the train since taking the plane directly to SF cost almost double [23:01] <omega> heh, ok [23:01] <Uraeus> to SJ I mean :) [23:02] <omega> but either I wait there for you (possible) or I need some hotel info from you [23:02] <Uraeus> omega: I just gave it to you above :) [23:02] <Uraeus> omega: booking number at hotel is: 66031416 [23:02] <Uraeus> omega: hotel phone is: Tel 510 451-6316 [23:03] <omega> btw, my cell is (503)539-9852 [23:03] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [23:03] <dobey> is this ALS? [23:03] <omega> all this is going on the als-sched.txt page [23:03] <omega> yeah [23:03] <dobey> ah [23:03] <omega> anything from arik? [23:03] <Uraeus> omega: my cell is +47 95960457 (I am bringing it) [23:04] <omega> good luck getting it to work over here <g> [23:04] <Uraeus> omega: no not yet, but our two person room has the option for an extra bed [23:04] <omega> ok [23:04] <omega> so to get your cell from here its 011 47 ... ? [23:04] chillywilly (da...@d2...) joined #gstreamer. [23:04] <Uraeus> omega: darn, that is right you use that US-only mobile system [23:05] <omega> well, try it, there might be gsm coverage there [23:05] <Uraeus> omega: think so, but dialing plus should be an interernational standard for an international line [23:05] <omega> right, I think that's 011 here [23:05] <Uraeus> ok, international standard except for the US then :) [23:06] <Uraeus> in the rest of the world we actually dial '+' [23:06] <dobey> omega: you might have dial a 1 first [23:07] <Uraeus> ds: so you volunteer to fix you car and be my shuttle buss? <g> [23:07] <omega> Uraeus: remember, this is the US. [23:07] <Uraeus> omega: ah, right THE US [23:07] <Uraeus> :) [23:07] <omega> the great thing about standards is there are so many to choose from [23:07] <dobey> no [23:07] <ds> Uraeus: it's a bit more than "slightly not working". Wish I could. [23:07] <chillywilly> God Bless America [23:08] <chillywilly> :P [23:08] <dobey> the great thing about standards is you can just ignore them [23:08] <omega> dobey: but what's the fun in that? [23:08] <Uraeus> omega: usually correct, but when it comes to mobile phones it is really just the US not following the standard [23:08] <dobey> omega: making your own new "standard" [23:08] <Zeenix> chillywilly: God Bless Freedom [23:08] <dobey> freedom? [23:08] <dobey> where? [23:08] <omega> heh [23:08] <chillywilly> not in the US [23:09] <omega> dobey: that was a week ago, before the "USA PATRIOT Act" [23:09] <dobey> no no [23:09] <dobey> oh wait [23:09] <chillywilly> this is th eubn [23:09] <Uraeus> omega: what aren't you a true PATRIOT who supports the act? :) [23:09] <dobey> i think they have freedom in antarctica [23:09] <omega> Uraeus: a *true* patriot does *NOT* support the act, IMO [23:09] <chillywilly> er, the united corporations of america can we take your order [23:09] <omega> heh [23:10] <omega> oh, wait, now I know why the call it in "Act"..... [23:10] <ChiefHighwater> dobey:not it just sounds like freedom...they are actually freezing, but their teeth chatter to much when the talk [23:10] <omega> s/in/an/ [23:10] <omega> heh [23:10] <Uraeus> well as I often say, who cares about freedom of speech if you are alone [23:10] <omega> you do? [23:10] <dobey> ChiefHighwater: and who is the governing body? and where are the citizens? [23:10] <Zeenix> Uraeus: you would have not said that if you were at my place [23:11] <Uraeus> omega, Zeenix: I do, I mean what is the point of having free speech if there is noone to hear you [23:11] <omega> antarctica is actually more military than anything, afaik, even though it's a bunch of scientists for the most part [23:11] <ChiefHighwater> dobey:Emperor Penguins and Tux is Minister of IT [23:11] <omega> Uraeus: well, there are people to talk to... [23:11] <chillywilly> that's it I am moving to antiartica to start my own country [23:11] <Uraeus> omega: not if you are alone [23:11] <omega> ChiefHighwater: oh, well, in that case, I'll hijack this plane to Oakland all the way to Antarctica. anyone want a ride? [23:11] <dobey> having people to talk to doesn't provide a leader to listen [23:12] <ChiefHighwater> nope [23:12] <Zeenix> Uraeus: i speek the truth when i am all alone [23:12] <ChiefHighwater> omega: you'll run out of gas between la and san diego [23:12] <omega> darn [23:12] <dobey> "in-flight refuel" [23:12] Action: chillywilly is gonna swim for it [23:13] <omega> chillywilly: brings new meaning to your nick... [23:13] <ChiefHighwater> chillywilly:umm, was nice knowin ya [23:13] <chillywilly> us penguins can take it [23:13] <Uraeus> omega: If I understood the travel agency correctly the room costs the same no matter if we are 2 or 3 so if Arik joins us we could save some money [23:13] <omega> ok [23:14] <Uraeus> omega: but I wouldn't trust the agency as they just ripped me of for a 100USD today [23:14] <omega> neat [23:14] <ajmitch> chillywilly: if you want to take a plane down there you'd leave from NZ [23:14] <dobey> you can lie to the hotel anyway [23:14] <chillywilly> ajmitch: oh kewl [23:15] <omega> hide arik in a drawer or something <g> [23:15] <Uraeus> omega: they told me that I could cancel until yesterday, but they 'forgot' to mention that I had to confirm my flight also yesterday at the latest which meant that my airplane ticket increased 100USD in price when I talked to them today [23:15] <omega> neat [23:15] <omega> well, I should go upgrade my server and get it back online..... [23:15] <Zeenix> omega: one problem that i am having is that i need to do bigendian in rtp & it gets me incompatible with osssrc [23:15] <Uraeus> dobey: well since we would need an extra bed for Arik it would be hard to lie, unless omega don't mind having Arik sleep under the same sheets as himself :) [23:16] <omega> Zeenix: you'll probably want to put an endian-swap inside rtpsend/rtprecv [23:16] <dobey> Uraeus: in the us you don't get 3 beds in one room [23:16] <dobey> Uraeus: you get one or two [23:16] <ChiefHighwater> dobey:sure you do, one is just a folding bed [23:16] <Uraeus> dobey: they agency told me that that was an option to have an extra bed/matrass moved in [23:16] <Zeenix> omega: swap ? [23:16] <ajmitch> dobey: that's why someone gets to sleep on the floor ;) [23:16] <ChiefHighwater> two queens and a rolled in full [23:17] <dobey> ajmitch: exactly [23:17] Action: dobey hugs the floor [23:17] <omega> Zeenix: yes, convert big endian to little endian [23:17] <dobey> ChiefHighwater: no [23:17] <ChiefHighwater> dobey:yes [23:17] <ChiefHighwater> have done it several times [23:17] <dobey> and that costs extra [23:17] <dobey> lying to them does not [23:17] <Zeenix> omega: i dont know how to ? [23:17] <ChiefHighwater> usually there is a modest fee for a third bed [23:17] <dobey> hence the floor [23:18] <dobey> or if you get a room with a sofa, it can probably be unfolded out [23:18] <omega> Zeenix: someone else will know, I've gotta go deal with my server [23:18] <Uraeus> dobey: well the extra bed is usually cheap if they charge for it, and sleeping on the floor is not a pleasent thing to do for a week [23:18] <omega> dobey: right, possibly [23:18] <ChiefHighwater> so, the floor and you can steal and lie, or roll in a bed for US$20 [23:18] <omega> l8r all [23:18] <ChiefHighwater> my integrity is worth more than $20 [23:18] <dobey> Uraeus: i've slept on the floor for the past 9 months [23:18] <omega> dobey: that explains things <g> [23:18] <dobey> not really [23:18] Action: Uraeus thinks the same as omega [23:18] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [23:18] <dobey> sleeping on the floor is healthy [23:18] <omega> anyway, bbiab [23:19] omega (om...@om...) left irc: ut 3x [23:19] <dobey> hah [23:19] <dobey> "integrity"? [23:19] <Zeenix> omega: i was to discuss this with wtay today but he isnt arround tonight [23:20] <Uraeus> well I am taking an 'early' night tonight (if you can call 2330 early) so see ya [23:20] Uraeus (csc...@c1...) left #gstreamer (Client Exiting). [23:20] <dobey> heh [23:20] <ds> anyone seen walken recently? [23:20] <dobey> i might watch Suicide Kings tonight [23:20] <dobey> ;-) [23:27] <vektor> ds: he got a girlfriend last i heard ;) [23:27] <taaz> ds: i think some girl is distracting walken from important codec work ;) [23:28] <dobey> lucky bastard [23:29] chillywilly (da...@d2...) left irc: [23:31] chillywilly (da...@d2...) joined #gstreamer. [23:34] Action: BBB is away: zzz [23:34] Nick change: BBB -> BBB-zZz [23:41] thomasvs (th...@ad...) left irc: Client Exiting [23:45] evil_monk (trevo@208.141.162.68) joined #gstreamer. [23:55] asmod (stevec@64.5.222.2) left irc: [x]chat [23:57] ne1 (d...@pc...) joined #gstreamer. [00:00] --- Thu Nov 1 2001 [00:01] ChiefHighwater (pa...@te...) left irc: [00:07] wingo (wi...@rd...) joined #gstreamer. [00:07] <wingo> yop [00:09] <ajmitch> ello [00:11] <wingo> aaa! don't make me think of thursday! ;-) [00:11] <dobey> thursday? [00:12] <dobey> let's go to friday's [00:12] <wingo> bar, yeah, we ran out of beer at home ;) [00:12] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [00:12] <dobey> hah [00:12] <wingo> actually, i better go get more [00:12] Nick change: wingo -> wingo-mission [00:12] <dobey> i am just wondering about this gnupdate thingy [00:12] <wingo-mission> because i'm on a mission :) [00:13] <dobey> haha [00:14] <chillywilly> beer mission impossible [00:15] ne1 (d...@pc...) left irc: Client Exiting [00:16] Zeenix (ze...@13...) left irc: I pay 25 Rs( 0.5$ )/hour [00:16] <dobey> no, that would be me [00:21] dobey (do...@dr...) left #gstreamer (eh). [00:44] chillywilly (da...@d2...) left irc: Read error to chillywilly[d23.as8.nwbl0.wi.voyager.net]: EOF from client [00:46] Nick change: wingo-mission -> wingo [00:46] <wingo> did i hear a naysayer? [00:46] <wingo> mission successful, muahaha [01:04] dobey (do...@cv...) joined #gstreamer. [01:52] ajmitch (me...@p1...) left irc: Ping timeout for ajmitch[p12-max11.dun.ihug.co.nz] [01:53] ajmitch (me...@p5...) joined #gstreamer. [02:00] Nick change: tnt-out -> tnt-halloween [02:03] ajmitch (me...@p5...) left irc: Ping timeout for ajmitch[p51-max10.dun.ihug.co.nz] [02:03] ajmitch_ (me...@p7...) joined #gstreamer. [02:03] <dobey> hey ajmitch_ [02:03] <dobey> so [02:03] Action: dobey has found his package manager [02:04] <ajmitch_> hi [02:04] Nick change: ajmitch_ -> ajmitch [02:04] <ajmitch> oh? [02:04] <dobey> yep [02:04] <dobey> gpkg [02:04] <ajmitch> ok [02:04] <dobey> except i have to create a package format for it [02:04] <dobey> heh |
From: IRC B. <wt...@us...> - 2001-11-02 05:27:52
|
******************************************************************* [03:04] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [03:08] soundmarsc (tho...@so...) joined #gstreamer. [03:09] <soundmarsc> hello [03:09] <wingo> hi [03:09] <vektor> Hello. [03:09] <ajmitch> hi [03:09] <vektor> Hey are you the dude who just posted on Xpert? [03:09] <soundmarsc> hi vektor [03:09] <vektor> Or was that someone else? [03:09] <soundmarsc> I think that was me [03:09] <soundmarsc> let me look [03:09] <vektor> Yeah, um, did you use the name 'vlad' in that email? [03:09] <wingo> let's all sing together, 'it's a small world after all' [03:09] <soundmarsc> yeah.. [03:09] <ajmitch> yay, /me sings along... [03:09] <vektor> Yeah, um, the dude has a full name. [03:09] <vektor> I just didn't remember it. [03:10] <dobey> hrmm [03:10] <soundmarsc> neither did I [03:10] <vektor> I thought I had indicated you should read the list a bit first before posting. :) [03:10] <vektor> So, um, I bet the guy won't know that's intended for him. [03:10] <soundmarsc> oh [03:11] <soundmarsc> well, I thought he would respond to "vlad" [03:11] <soundmarsc> sorry [03:11] <vektor> No, I'm sorry. [03:11] <vektor> I should have been clearer. [03:11] <soundmarsc> I just figured that everyone more or less goes by there codenames here... [03:11] <vektor> anyways, you may want to find out his full name and email him directly. [03:11] <soundmarsc> their [03:11] <vektor> yeah but i just know his name starts with 'vlad' something. i think. [03:12] <vektor> that's not like his 'codename' :)_ [03:12] <vektor> (re:my fault) [03:12] <soundmarsc> ok [03:12] <dobey> heh [03:13] <soundmarsc> singin... [03:13] <soundmarsc> :) [03:13] <wingo> so you're a video fellow then, soundmarsc? [03:13] Action: dobey works on a package file format [03:13] <soundmarsc> sorta... I'm working on it [03:14] <soundmarsc> "fellow" is a bit...formal [03:14] <soundmarsc> cohort maybe [03:14] <soundmarsc> :) [03:15] <wingo> heh [03:15] <wingo> they're equivalent to me [03:15] <wingo> hoy about fella? [03:15] <wingo> er [03:15] <wingo> how even [03:15] <ajmitch> ok, comrade... [03:15] <wingo> :) [03:15] <wingo> well, that would imply that i am a video person, so no ;) [03:16] <vektor> Anyone here care about DVD subtitles? [03:16] <ajmitch> vektor: i can't play dvds, so i don't really care ;) [03:16] <vektor> Hah. [03:17] <ajmitch> plus i dunno how useful subtitles would be for me... [03:17] <soundmarsc> vektor: has he posted this month? [03:17] <wingo> i think taaz cares [03:17] <vektor> soundmarsc: Yes. [03:17] Action: wingo pokes taaz [03:17] <vektor> I know he does, but I updated my webpage and I just wanted opinions on it. [03:17] <soundmarsc> october, for those of you in certain timezones [03:17] <vektor> http://www.dumbterm.net/graphics/dvd/ [03:18] <vektor> soundmarsc: just read all the posts about r128 and Xv and stuff. [03:18] <ajmitch> october? it's november now ;) [03:18] <soundmarsc> I can't find "vlad" on the archive page [03:18] <vektor> maybe I got his name wrong. [03:18] <vektor> See, you didn't even look! ;-) [03:19] Action: vektor bahs. [03:19] <vektor> Vladimir Dergachev [03:19] <vektor> That's his name. [03:19] <vektor> He posts as 'vo...@mi...'. [03:19] <soundmarsc> I did look! [03:19] <vektor> He wrote ati.2, it looks like. [03:20] <vektor> Lots of stuff about the r128 on this list. [03:20] <vektor> That's what you have, right? [03:20] <soundmarsc> oh, he replied to me on the ati-livid list! [03:21] <vektor> Neat. [03:21] Action: soundmarsc is singing again [03:21] <vektor> Did he solve your problem? [03:21] <soundmarsc> no, I was asking the wrong question [03:21] <soundmarsc> I'll email him [03:22] <soundmarsc> at the time, I meant. It was a week or so ago [03:22] <vektor> oh [03:23] <taaz> woof [03:23] <taaz> subtitles? yes, i care [03:23] <vektor> taaz: check the updates to my page and let me know what you think. [03:24] <ajmitch> hey taaz [03:24] <taaz> oh be kay be [03:25] <taaz> i had a pretty crappy radio show this week... happens to be the only one i've recorded ;) [03:25] <dobey> hrmm [03:26] <taaz> stupid mixer... mic shuts off everything else this week... didn't last week. grr. [03:26] <ajmitch> heh [03:26] Shippou (ir...@cr...) joined #gstreamer. [03:30] <taaz> uh, whats the diff between gb pics? [03:31] <vektor> I don't explain that well enough? [03:31] <taaz> well, the images look the same [03:31] <vektor> No, there's a grey outline around the bottom one. [03:31] <vektor> Use xmag and look. [03:31] <dobey> gpkg is so going to rule [03:31] <vektor> If you can't see. [03:31] <taaz> well, my background for the page is gray ;) [03:31] <vektor> ?? [03:31] <vektor> I composited to white. [03:32] <vektor> How did you get a grey background on the image? [03:32] Action: vektor is confued. [03:33] <taaz> i have the whole page background as gray, i see white line on bottom of pics [03:33] <taaz> like below the shadows [03:33] <soundmarsc> who is the guy here that developed gstreamer? [03:33] <vektor> What the hell are you talking about? [03:33] <vektor> Can you take a screenshot? [03:33] <vektor> I mean, like, for each of the subtitles, it should be yellow/black/whatever text on a white background. [03:33] <vektor> Like, in the png, the background is white. [03:33] <vektor> pure white. [03:33] <vektor> for the whole imag. [03:34] <taaz> but like 1 pixel here or there i can't even notice... my freaking glasses bend light in funky ways anyway... i get red/blue fringes on all white/black edges.. ;) [03:34] <vektor> The grey outline around them is pretty damn clear here. [03:34] <vektor> What resolution are you running at? [03:34] <taaz> you want a screenshot? [03:34] <vektor> Please! [03:35] <ajmitch> taaz: then take off your glasses ;) [03:37] <wingo> soundmarsc: omega and wtay are the lead developers, both gone or asleep ;) [03:37] <ajmitch> wingo: omega's probably still trying to fix up a server ;) [03:38] <soundmarsc> darn [03:38] <wingo> heh [03:38] <soundmarsc> it isn't working [03:38] <wingo> the rest of us help though [03:38] <wingo> 0.2.1 or cvs? [03:38] <soundmarsc> I compiled it, but I can't get it to work [03:38] <soundmarsc> 0.2.1 [03:38] <wingo> look at gstreamer.net/wiki/?DichotomousKey [03:38] <soundmarsc> that's why I came here in the first place...heheh [03:39] <wingo> it's my new installation howto [03:39] <vektor> taaz: Like, I'm not using PNG alpha, so the images should all have a white background regardless of your browser's background. [03:39] <wingo> if it doesn't solve your problem it needs to be added to [03:39] <wingo> so let me know how it goes for you [03:39] <ajmitch> wingo: nice name ;) [03:41] <soundmarsc> ok [03:41] <ajmitch> i wonder how arik went with his gnome 2 build... [03:42] <wingo> heh [03:42] <wingo> he failed :) [03:43] <wingo> should have usd v-b-s ;-) [03:43] <soundmarsc> Couldn't create a 'mpeg2dec', no such element or need to run gstreamer-register? [03:43] <soundmarsc> what does that mean? [03:44] <ajmitch> wingo: well v-b-s failed for me too [03:44] <wingo> have you run gstreamer-register? [03:44] <soundmarsc> yep [03:44] <wingo> do you have mpeg2dec? [03:44] <wingo> mpeg2dec.sf.net [03:44] <wingo> iirc [03:44] <vektor> Yes. [03:44] <vektor> And I wish walken were here. [03:44] <soundmarsc> is that a separate library? [03:44] <ajmitch> wingo: arik & i were both trying to compile it last night ;) [03:44] <wingo> sux :-\ [03:44] <vektor> It's been like over a week since we went through the 'let's fix libmpeg2' deal and nothing has happened. [03:45] <wingo> ajmitch: debian is starting to package gnome2, which is nice [03:45] <soundmarsc> vektor: I'm trying to compile that too [03:45] <ajmitch> wingo: good, it failed on building ORBit2 for me ;) [03:45] <vektor> soundmarsc: um, libmpeg2 compiles and works just fine, the only problem is that the API is useless for those of us writing DVD players etc. [03:45] <soundmarsc> oh [03:45] <vektor> soundmarsc: my DVD player (movietime) now distributes the libmpeg2 source with it, along with my patches... Argh... [03:45] <soundmarsc> then I shouldn't waste my time with it? [03:46] <soundmarsc> I want to play divx and dvd [03:46] <soundmarsc> on ppc hardware [03:46] <vektor> soundmarsc: no no, it works great, but if you want to do deinterlacing and stuff you need it to export more data. [03:46] <wingo> taaz: does dvdsrc work? [03:46] <vektor> soundmarsc: but like, xine has a patched copy of libmpeg2 that it distributes along with its source tree also, for example. [03:46] <vektor> soundmarsc: I don't think gstreamer is useful for playing DVDs yet. [03:46] <vektor> soundmarsc: in fact, i'm pretty sure it isn't. [03:47] <soundmarsc> but it plays divx? [03:47] <vektor> soundmarsc: if you have a fast g4, you can try my app (movietime). ;-) [03:47] <soundmarsc> and other mpeg [03:47] <vektor> soundmarsc: gstreamer probably plays them a little. [03:47] <soundmarsc> 450? [03:47] <ajmitch> vektor: umm, i'm sure taaz committed some useful stuff last week for dvd playback? [03:47] <vektor> ajmitch: yeah it's getting there :) [03:47] <ajmitch> at least i hope it's useful ;) [03:47] <soundmarsc> xine works for me except for this video problem I'm having [03:48] <vektor> taaz: dude, your screenshot looks fine. and i can see the grey outline around the bottom pic. [03:48] <vektor> soundmarsc: so shut off Xv. [03:48] <vektor> soundmarsc: no change in app will fix that problem. [03:48] <vektor> soundmarsc: until you fix your buggy driver. [03:49] <soundmarsc> I know, I'm working on that [03:49] <vektor> Well then. nuff said. [03:49] <vektor> You know how to shut off Xv, right? [03:49] <taaz> wingo: dvdsrc should work. gst not smp safe though so i can't play dvds very long (ie, 5 seconds at most it seems) [03:49] <soundmarsc> wingo: that url doesn't work [03:49] <soundmarsc> vektor: actually, I don't [03:50] <vektor> soundmarsc: Ok, well, xine should have a way in its config file or command line to say 'use xshm instead of xv'. However, if you're really stuck, there's some way to shut it off in your XF86Config file. [03:50] <vektor> soundmarsc: But look into the xine docs first, should be trivial. [03:50] <taaz> vektor: doh, sorry, i was being dumb, i see what you mean [03:50] <soundmarsc> ok [03:51] <vektor> soundmarsc: I don't think they have altivec optimizations for the Y'CbCr->R'G'B' conversion though. [03:51] <vektor> taaz: ok. so the web page you think is ok so far though? [03:52] <taaz> it tis good [03:53] <soundmarsc> xine -VXShm works, but the video is choppy as expected [03:54] <vektor> Um. [03:54] <vektor> Shouldn't be tooo choppy if they do it correctly. [03:54] <vektor> But I bet things aren't optimized too much for your ppc. You could look into that. [03:54] <vektor> How choppy is it? [03:54] <vektor> Is it tearing? [03:54] <soundmarsc> yea [03:54] <vektor> Is it just the tearing that's annoying, or something else? [03:54] <vektor> Is it dropping frames? [03:55] <soundmarsc> dropping frames, and only tearing a little [03:56] <soundmarsc> when something else happens on screen [03:56] <soundmarsc> need to recompile with esd support... [03:56] <soundmarsc> It's watchable, I suppose [03:57] <vektor> Don't use esd. [03:57] <wingo> soundmarsc: that url was not accessible? [03:57] <vektor> esd adds an unmeasurable amount of latency to the audio path and will fuck up your attempts to keep audio/video in sync. [03:58] bstard (Lor...@a2...) joined #gstreamer. [03:58] <soundmarsc> wingo: nope [03:58] <soundmarsc> vektor: k [03:58] <soundmarsc> is alsa any good? [03:59] <wingo> um, yes. (re: alsa) [03:59] <vektor> soundmarsc: ALSA is a sound driver. It's good at that. [03:59] <wingo> here it is, directly cut out of my browser (same as before) [03:59] <wingo> http://gstreamer.net/wiki/?DichotomousKey [03:59] <vektor> soundmarsc: It doesn't make up for poor apps though. :) [03:59] <wingo> but your problem was that mpeg2dec was not present at compilation [04:00] <wingo> gstreamer worked fine, you just lacked the libs... probably should add that to the troubleshooter tho [04:00] <ajmitch> when will alsa ever release 1.0? [04:02] <wingo> the pcm api is more or less stable [04:02] <wingo> the sequencer api is changing slightly though [04:02] <vektor> afaik the seq api is useless. [04:02] <vektor> yeah. [04:02] <wingo> the nominal version of all of these betas is 1.0 though [04:02] <wingo> in /proc [04:02] <vektor> In my sequencer I just use /dev/midi*. [04:02] <ajmitch> more importantly - when will hurd get sound support? ;) [04:02] <vektor> Avoid the issue. [04:03] <dobey> more importantly [04:03] <wingo> the sequencer api is fine according to lad... [04:03] <dobey> when will hurd be usable? [04:03] <vektor> I haven't been following lad lately. I should get back into that. [04:03] <soundmarsc> vektor: will alsa solve the latency problems? [04:03] <ajmitch> dobey: it's usable now, for certain tasks [04:03] <wingo> oss by itself would solve them as far as you're concerned [04:03] <vektor> soundmarsc: Dude, I don't think you understand. [04:03] <ajmitch> dobey: runs X, has poor threading still, and falls over under high load [04:04] <soundmarsc> wingo: what is the url for mpeg2dec? [04:04] <dobey> ajmitch: by usable i mean easily compilable, distributable, and does everything linux does [04:04] <ajmitch> but it's improving ;) [04:04] <soundmarsc> vektor: ok, ok, I think I got it [04:04] <ajmitch> dobey: hmm, you can get debian isos [04:04] <wingo> mpeg2dec.sf.net [04:04] <dobey> ajmitch: like i said... [04:04] <vektor> soundmarsc: For Linux there are three kernel driver sets for soundcards. The kernel drivers that ship with the linux source, the commercial OSS drivers, and the ALSA drivers. These are drivers for soundcards in the kernel. Talking to them is as close to the hardware as you can get, so latency is minimal. [04:05] <soundmarsc> wingo: that one doesn't work, I just tried it [04:05] <ajmitch> dobey: but they've only compiled 40% of the thousands of debian packages (a lot are not written to be portable) [04:05] <wingo> soundmarsc: must be a network issue, it works for me (tm) [04:05] <soundmarsc> so the apps should be built to make up for latency? [04:05] <vektor> soundmarsc: For linux there are also various soundservers, namely esd and alsa are the two big ones, these are for people who don't want to talk to the kernel directly, and they want like sound in their window manager, and they want to have their mouse make clicky noises while listening to mp3s (don't like mutual exclusivity of kernel sound drivers). These apps increase latency as the cost of allowing a simple API with sound mixing be [04:06] <dobey> ajmitch: i think you missed my point :-) [04:06] <soundmarsc> wingo: where are you at? [04:06] <ajmitch> dobey: nope, i can see that you won't use or care about hurd until it's better than linux ;) [04:06] <vektor> soundmarsc: so, if your application is a high-quality multimedia app which requires low latency, you need exclusive kernel-level access to the soundcard. no playing mp3s while watching a dvd. this means you must avoid going through other userspace applications like alsa or esd. [04:06] <wingo> nc, usa (east coast) [04:06] <dobey> ajmitch: not better than, but at least equal too [04:06] <vektor> soundmarsc: hope that helps. [04:06] <dobey> ajmitch: sound is very important to me [04:07] <soundmarsc> vektor: thanks [04:07] <wingo> vektor: alsa ain't a sound server... [04:07] <ajmitch> dobey: yeah, it's a pity noone's been working on it [04:07] <dobey> ajmitch: so is usb and various other non-x86 things [04:07] <vektor> wingo: I meant arts. [04:07] <vektor> damn [04:07] <wingo> heh :) [04:07] <vektor> now my statements are all wrong! [04:07] <vektor> :( [04:07] <vektor> i bet i confused him. [04:07] <wingo> lol [04:07] <vektor> soundmarsc: sorry, the two soundservers are esd and arts! [04:07] <vektor> soundmarsc: alsa is not a soundserver!! ;-) [04:07] <vektor> damnit i am a moron. [04:08] <ajmitch> dobey: which is why i'll only be using hurd to play with for quite awhile ;) i like decent hardware support as well [04:08] <vektor> soundmarsc: did you get all that? [04:08] <dobey> ajmitch: and actually, usb is kind of a requirement for me :-) [04:08] <ajmitch> dobey: i can imagine, you've got usb keyboard/mouse? [04:09] <dobey> ajmitch: i've got a powerbook g3 [04:09] <dobey> heh [04:09] <soundmarsc> what is alsa then? [04:09] <vektor> ALSA is a kernel driver set. [04:09] <vektor> aRts is a soundserver. [04:09] <vektor> ALSA is an alternative to the sound drivers shipped with the linux kernel. [04:10] <vektor> it has a different API than the kernel drivers, but it is also compatible with the standard kernel sound API. [04:10] <soundmarsc> what is so special about it? [04:10] <wingo> low latency [04:10] <vektor> Well, nothing really. The API is slightly better than the kernel sound API. [04:10] <vektor> Like, it's cleaner. [04:11] <wingo> support of exotic hardware [04:11] <vektor> wingo: that's bull and you know it. [04:11] <vektor> Yeah, it does support some different hardware. [04:11] <soundmarsc> it doesn't allow multiple clients to one device does it? [04:11] <wingo> sub 3 ms is not possible with oss, yo [04:11] <vektor> soundmarsc: Only if the hardware supports it. [04:11] <ajmitch> soundmarsc: can do, iirc [04:11] <vektor> soundmarsc: Do you require that functionality? [04:11] <vektor> soundmarsc: You know that is nontrivial. [04:11] <soundmarsc> vektor: it would be nice [04:11] <vektor> Ok. Well, for really low latency audio it's impossible. [04:12] <soundmarsc> I don't care to hear my mouse clicks, but other things are good. [04:12] <wingo> soundmarsc: it does on certain devices [04:12] <wingo> setting it up is nontrivial [04:12] <soundmarsc> on ppc hardware? [04:12] <vektor> For DVD playback it can be done, but you need either the kernel help (which I don't like, but ALSA may provide), or you hardware support, or you need a realtime priority userspace sound server that gives you lots of feedback. [04:12] <wingo> sure [04:12] <wingo> it involves hacking undocumented configuration files [04:12] <vektor> soundmarsc: Are you a developer or just a user? [04:13] <wingo> s/just a// ;) [04:13] <soundmarsc> I'm trying to be a developer. I'm a compsci student at the moment [04:13] <soundmarsc> undergrad [04:13] <vektor> soundmarsc: Because I should warn you that multimedia stuff under linux is still a little 'raw'. And especially for ppc support for stuff, I bet you'll have to really know your stuff since you'll be on your own for a while. [04:13] <vektor> soundmarsc: what year? :) [04:13] <wingo> we're all students here :) [04:13] <soundmarsc> sophomore [04:13] <vektor> wtf is sophomore? [04:13] <wingo> 2 [04:13] <wingo> :) [04:13] <vektor> is that second year or third? [04:14] <soundmarsc> sophmore [04:14] <vektor> it's 2, right? [04:14] <soundmarsc> 2nd [04:14] Action: wingo is a fifth year senior :) [04:14] <vektor> ok. [04:14] <vektor> I'm in 4th year. [04:14] <soundmarsc> painfully second [04:14] <vektor> compsci. [04:14] <ajmitch> damn, i'm the youngest here then ;) [04:14] <ajmitch> 1st year compsci [04:14] <vektor> soundmarsc: Like, don't expect lots of glowing and helpful documentation for much of this stuff. [04:14] <vektor> soundmarsc: DVD playback only got really accessable under linux in the last year. [04:15] <vektor> soundmarsc: and all the big advances in dvd playback happened like this summer. [04:15] <vektor> soundmarsc: Xv only got really implemented everywhere in the last year also. [04:15] <soundmarsc> cool [04:15] <vektor> soundmarsc: _alot_ of the multimedia apps for linux are still not very complete. [04:15] <vektor> soundmarsc: and much infrastructure work is still going on. [04:15] <dobey> ajmitch: heh, how old are you? [04:15] <soundmarsc> vektor: I do get that impression [04:15] <ajmitch> 19 [04:16] <soundmarsc> 21 [04:16] <dobey> ah ok [04:16] <dobey> i guess you are the youngest then ;-) [04:16] <vektor> 21 and 2nd year? [04:16] <vektor> seems a bit old :) [04:16] <ajmitch> heh [04:16] <wingo> you vektor? [04:16] <soundmarsc> yep, I stayed out semester after high school [04:16] <vektor> I'm 23 and in 4th year, and that's with a 5-year program and i had an extra year of high school cause i'm in ontario. [04:16] <ajmitch> yay, almost got my summer break ;) [04:17] <wingo> bastard :) [04:17] <ajmitch> almost finished 1st year, thankfully, got one exam left [04:17] <dobey> vektor: yeah, so you are the same as him [04:17] <vektor> dobey: erm, dunno... this is my 5th year really. [04:17] <wingo> 21 / fifth year for myself. dobey, what are you? you're out, no? [04:17] <vektor> i say 4th year but i mean 5th. [04:17] <vektor> since it's a 4-year degree spread over 5 years. [04:17] <dobey> haha, school? what's that? [04:18] Action: dobey is 20 [04:18] <wingo> heh. [04:18] <ajmitch> heh, i got some friends that started uni at 16 [04:18] <vektor> damn them all. [04:18] <vektor> :) [04:18] chillywilly (da...@d9...) joined #gstreamer. [04:18] Action: dobey just skipped the uni [04:18] <vektor> i'm going to fail this assignment :( [04:18] Action: vektor has a formal languages assignment due tomorrow and instead i'm irc'ing. [04:22] <vektor> I need to prove that (a+b+c)* - {a^n b^n c^n : n >= 1} is a CFL. [04:22] <vektor> ugh. [04:22] chillywilly (da...@d9...) got netsplit. [04:22] Shippou (ir...@cr...) got netsplit. [04:22] <ajmitch> heh [04:23] chillywilly (da...@d9...) returned to #gstreamer. [04:23] <ajmitch> fun [04:28] Shippou (ir...@cr...) got lost in the net-split. [04:34] <dobey> ok, i gotta get some sleep [04:34] <dobey> later [04:34] dobey (do...@cv...) left #gstreamer (eh). [04:37] evil_monk (trevo@208.141.162.68) joined #gstreamer. [05:13] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [05:23] Shippou (ir...@ot...) joined #gstreamer. [05:44] arik (ar...@sd...) joined #gstreamer. [05:45] <arik> hi all [05:45] <wingo> hey. [05:45] <ajmitch> hey arik [05:45] <arik> hey aj [05:45] <arik> still building, strange errors [05:45] <ajmitch> mine stuffed up on ORBit2 [05:45] <arik> hmm [05:45] <arik> mine's on libgnomecanvas [05:46] <arik> /gnome/GNOME2/include/gtk-2.0/gtk/gtkstyle.h:33:25: pango/pango.h: No such file or directory [05:46] <ajmitch> you got further than me ;) [05:46] <arik> lot's of errors like tha [05:46] <arik> t [05:46] <arik> hehe [05:46] <ajmitch> idl compiler is giving me weird errors [05:46] <arik> ah [05:47] <arik> i wonder if it's cause pango is now in include/pango-1.0/pango [05:47] <ajmitch> probably [05:47] <ajmitch> go complain in #gnome ;) [05:47] <arik> hehe [05:47] <arik> i should [05:47] <wingo> but does pango-1.0/pango/pango.h exist? [05:48] <arik> i hate going in there [05:48] <arik> yeah it does [05:48] <wingo> sounds like a pkgconfig issue ;) [05:48] <arik> yeah! [05:48] <arik> yeah it prob is [05:49] <arik> i know so little about pkgconfig [05:49] <ajmitch> unless gtkstyle.h is generated i'd say that it's not pkgconfig [05:49] <arik> well [05:49] <arik> the weird thing [05:49] <arik> is that gtk compiles [05:49] <ajmitch> but it could be [05:49] <ajmitch> hehe [05:50] <ajmitch> at least you probably don't have to wait for anoncvs to sync [05:50] <arik> heh [05:50] <arik> no [05:50] <arik> i have gnomecvs access [05:51] <arik> ok [05:51] <arik> that's weird [05:51] <arik> now it compiled [05:51] <ajmitch> weird... [05:52] <arik> ah ok [05:52] <arik> so [05:52] <arik> libgnomecanvas [05:52] <arik> complains about pangoft2.pc [05:52] <arik> not being in the path [05:52] <arik> so i added the pango source dir manually [05:52] <arik> that could be part of the problem [05:52] <ajmitch> yeah [05:55] <ajmitch> why don't they make it nice & easy like KDE? ;) [05:55] <arik> hah [05:57] <arik> so the real question [05:57] <arik> is why doesn't pango install that .pc file [05:58] <ajmitch> hmmmm [05:58] <ajmitch> why indeed? [05:59] <arik> or why libgnomecanvas is linking to it wrong if that's wrong [06:00] omega (om...@om...) joined #gstreamer. [06:00] <arik> hey erik [06:00] <omega> yo [06:00] <omega> so you coming to ALS? [06:00] <arik> i don't think so :-( [06:00] <omega> bleagh, why? [06:00] <arik> rr wasn't pyched at the idea [06:00] <omega> why? [06:00] <arik> cause they feel that we would fall behind [06:01] <omega> um.... fall behind on what? [06:01] <arik> work [06:01] <omega> specifically what? [06:01] <arik> they didn't say i couldn't go [06:01] <arik> my work [06:01] <arik> they just said they thought i might be behind [06:01] <omega> meeting at ALS is the best thing possible, since we can discuss the player stuff [06:01] <arik> yeah but they think the player is done [06:01] <omega> um [06:01] <omega> is it? [06:02] <arik> more or less, the pico one anyway [06:02] <ajmitch> you two not actually get to speak face-to-face? [06:02] <omega> nope [06:02] <ajmitch> sucky [06:02] <omega> that's why I'm hoping he can go to ALS [06:02] <arik> i may just fucking drive to portland [06:02] <omega> heh [06:02] <arik> heh [06:02] <ajmitch> omega: you not been going to boise lately? [06:02] <omega> nope [06:03] <omega> he hasn't gone either [06:03] <arik> nope [06:03] <ajmitch> oh, i see [06:03] <ajmitch> both working from home? ;) [06:03] <arik> yep [06:05] <arik> compling! [06:06] <arik> ah the joy of installing pc files by hand [06:07] <arik> well ok that didn't work [06:07] <arik> why is oft2 not being installed [06:08] <omega> ? [06:08] <arik> er [06:08] <arik> not oft2 [06:08] <arik> ft2 [06:08] <omega> ft@? [06:08] <omega> ft2? [06:08] <arik> pango's freetype2 support [06:08] <omega> oh, no idea, never built it [06:08] <arik> pangoft2.pc and .h and all that [06:08] <arik> ewll [06:08] <arik> er well [06:08] <arik> you can't build gnome2 without it [06:09] <ajmitch> ok, cool, not going to movies tonight... [06:09] <arik> how come? [06:09] <ajmitch> guess that means i should be hacking :) [06:10] <ajmitch> was going to go out with a friend, but am not now ;) [06:10] <arik> hehe [06:10] <arik> ah [06:11] <ds> grrr... my G4 doesn't play sound at 48khz at the wrong speed [06:12] <ds> s/doesn't// [06:14] <arik> argh [06:14] <arik> why doesn't ft2 get installed? [06:15] <arik> um [06:16] <arik> this must have to do with how pkg_config workds [06:16] <ajmitch> gets installed here [06:16] <arik> it gets installed there? [06:16] <arik> wtf! [06:16] <ajmitch> [ajmitch @ ajmitch pkgconfig] ls -la pangoft2.pc [06:16] <ajmitch> -rw-r--r-- 1 ajmitch ajmitch 311 Oct 31 23:27 pangoft2.pc [06:16] <arik> fucking a [06:17] <omega> pkgconfig --list-all [06:17] <arik> hmm [06:18] <arik> ok [06:18] <arik> pkg-config --list-all [06:18] <wingo> hi omega. [06:18] <omega> whatever <g> [06:18] <omega> yo [06:18] <arik> lists pango and pangox [06:18] <arik> but no pangoft2 [06:20] <arik> aj? [06:21] <ajmitch> yeah? [06:21] <arik> do you have pangoft2 when you run pkg-config --list-all? [06:22] <ajmitch> yes [06:22] <arik> well fucking a [06:22] <arik> what's that about [06:22] <ajmitch> missing libs maybe? [06:22] <arik> missing libs where? [06:23] <ajmitch> missing freetype devel libs? [06:23] <arik> hmm [06:23] <arik> then why does pango compile? [06:23] <arik> ooh [06:23] <ajmitch> cos it's optional, i guess [06:23] <arik> i don't have freetype2-devel [06:23] <arik> that could be why [06:23] <ajmitch> yeah [06:23] <ajmitch> could be :) [06:23] <arik> hehe [06:25] Nick change: wingo -> wingo-z [06:25] <wingo-z> zzzz [06:26] Action: arik goes to get freetype2-devel [06:27] Action: chillywilly prefers -dev to -devel [06:27] <chillywilly> :P [06:27] <arik> why? [06:27] <chillywilly> if ya know what I mean [06:27] <arik> heh [06:27] <arik> debian [06:28] <chillywilly> cause debian is the shit [06:28] <ajmitch> hehe [06:28] <arik> hah [06:28] <ajmitch> these damned deblots... ;) [06:28] <arik> religious zealots [06:28] <arik> damned fanatics is what they are [06:28] <arik> debian was my first distro [06:28] <ajmitch> deblot == debian zealot [06:28] <arik> ah [06:28] <ajmitch> how could you switch? [06:28] <chillywilly> freakin' gnu commie hippy bastards [06:28] <arik> um [06:28] <arik> it was a long long time ago [06:29] <arik> well [06:29] <arik> not that long ago [06:29] <ajmitch> heh [06:29] <arik> debian 1.2 or something [06:29] <ajmitch> pretty recent [06:29] <arik> yeah [06:29] <arik> 1998 or 9 or something [06:29] <arik> i was doing litestep before that [06:29] Action: omega babies his laptop so it'll survive another 2 weeks, because it has a shattered frame all around the LCD support [06:29] <ajmitch> heh, i was using mandrake 5.3 in 99 ;) [06:29] <arik> heh [06:29] <ajmitch> omega: you always seem to smash your laptop up [06:29] <arik> 1998 i think [06:29] <arik> maybe 1997 [06:30] <omega> ajmitch: no, the laptop is very poorly designed [06:30] <ajmitch> hmm, i used dragonlinux a little in 98, that was my first distro ;) [06:30] <arik> eh [06:30] <arik> heh [06:31] <chillywilly> No non-free packages installed on obfuscation! rms would be proud. [06:31] <chillywilly> woohhooo [06:31] <ajmitch> heh [06:32] <arik> woo! [06:32] <arik> it's installed! [06:32] <ajmitch> good [06:33] <ajmitch> now you can at least build ;) [06:33] <arik> heh [06:33] <arik> hopefully [06:33] <arik> we shall see [06:33] <ajmitch> i gotta find out why ORBit build is dying [06:33] <arik> yeah [06:33] <arik> you do [06:33] <ajmitch> orbit-interface.idl:15: Error: `TTypeCode' undeclared identifier [06:34] <arik> hmm [06:34] <ajmitch> the thing is, it reads 'TypeCode' in the file ;) [06:34] <arik> missing header? [06:34] <arik> heh [06:34] <ds> how do I get gst to top out below ~100MB of memory usage? [06:34] <ajmitch> ulimit? ;) [06:35] <ds> I want it to still work...\ [06:37] Action: ajmitch tries on a clean checkout of ORBit [06:41] <chillywilly> night [06:41] chillywilly (da...@d9...) left irc: Read error to chillywilly[d92.as9.nwbl0.wi.voyager.net]: EOF from client [06:42] <arik> oh god [06:42] <arik> now bonobo is building [06:42] <arik> well, i'll be here awhile :-) [06:45] <ajmitch> haha [06:45] <ajmitch> it can't be that bad ;) [06:46] <ajmitch> bonobo is a nice, compact, clean module ;) [06:46] <arik> hahahahahahahahahaa [06:47] <arik> anyone know why audiofile doesn't exist anymore? [06:48] <ajmitch> dunno, read the mailing list archives [06:48] <ajmitch> i think it's cos the GNOME copy isn't as up-to-date [06:48] <arik> ah [06:49] <ajmitch> it was discussed a couple of days ago [06:49] <ajmitch> been moved to audiofile-historical, iirc [06:49] <arik> yeah it has [06:51] <ajmitch> bonobo built yet? :) [06:51] <arik> yep [06:51] <arik> now i'm on audiofile [06:51] <ajmitch> ok [06:52] <arik> so [06:52] <arik> i've copied my old audiofile [06:52] <arik> and it can build that :-) [06:52] <ajmitch> hmm, weird [06:52] <ajmitch> i already had an old audiofile [06:52] <ajmitch> and audiofile is first in the v-b-s build list ;) [06:52] <ajmitch> dunno if ppl will have breakages.... [06:53] <arik> ah [06:53] <arik> eh [06:53] <arik> he [06:53] <arik> h [06:53] <ajmitch> lol [06:53] <arik> hehe [06:56] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [06:57] <ajmitch> yay [06:57] <arik> built? [06:57] <ajmitch> "rm -rf ORBit2/ && STARTMODULE=ORBIT2 rebuild.sh" works ;) [06:58] <arik> hehe [06:58] <arik> does config.cache not exist anymore or something? [06:58] <ajmitch> why's that? [07:00] <ajmitch> building gnome-xml now [07:00] <ajmitch> mine takes quite a bit longer than yours to compile ;) [07:02] omega (om...@om...) left irc: Ping timeout for omega[omegacs.net] [07:02] omega (om...@om...) joined #gstreamer. [07:06] <arik> i feel like singing an i'm building gnome2 song [07:07] <ajmitch> hehe [07:07] <arik> :-) [07:07] <ajmitch> ah, cool song just started playing in xmms ;) [07:07] <arik> heh [07:07] <arik> xmms doesn't work on my system anymore [07:07] <arik> alsa issues i think [07:07] <ajmitch> ah, ok [07:07] <ajmitch> using an ISA SB16PNP here ;) [07:07] <arik> heh [07:07] <arik> lucky [07:07] <ajmitch> nice compatible card... [07:08] <ajmitch> it's one of the few things i kept from my old computer [07:08] <arik> heh [07:08] <arik> yeah [07:09] <ajmitch> wow, MS has really been slapped on the wrist hard (with a feather) [07:09] <arik> oh? [07:11] <ajmitch> http://linuxtoday.com/news_story.php3?ltsn=2001-11-01-004-20-NW-MS [07:11] <arik> gnome-vfs time [07:13] <ajmitch> or you could read the professional & unbiased opinions of slashdot regulars on it ;) [07:14] <ajmitch> building gconf... [07:14] <arik> hehe [07:15] <ajmitch> dammit this computer is too slow ;) [07:17] <arik> libgnome [07:17] <arik> hehe [07:18] <ajmitch> still gconf... [07:18] <arik> heh [07:18] Action: ajmitch kicks his k6-2 [07:18] <arik> my comp kicks your comp's butt [07:18] <ajmitch> no shit ;) [07:18] <arik> heh [07:19] <ajmitch> the main bottleneck in compiling seems to be the CPU [07:19] <ajmitch> i got enough ram, i got fast enough hard drives.. [07:19] <ajmitch> just an utter POS CPU ;) [07:22] <arik> well [07:22] <arik> libgnome doesn't compile [07:23] <ajmitch> damn [07:23] <arik> make[3]: *** No rule to make target `all'. Stop. [07:23] <arik> make[3]: Leaving directory `/gnome-source/GNOME2/libgnome/help-converters/gnome- [07:23] <arik> vfs-module' [07:23] <ajmitch> gconf just failed here [07:23] <ajmitch> probably cos i shoudl rebuild the earlier modules ;) [07:23] <arik> yep [07:24] <ajmitch> starting again... [07:24] <arik> did libgnome build for you? [07:24] <arik> hmm [07:24] <ajmitch> didn't get that far [07:24] <arik> recent change [07:24] <arik> betting it's an error [07:24] <ajmitch> gconf failed on some gtk+ stuff [07:24] <arik> heh [07:25] <arik> and michael fixed it! [07:25] <arik> woo [07:26] <ajmitch> running my cvs update script to get them all at about the same time ;) [07:26] <ajmitch> otherwise v-b-s will update each module as it goes to build it [07:26] <ajmitch> (which i think can be stupid) [07:27] <arik> hee [07:27] <arik> that can be stupid [07:27] <arik> eazel-hacking gets all of them and then builds [07:28] <ajmitch> yeah [07:28] <ajmitch> which is what i'm doing [07:29] <arik> man, it's so easy to build this, i can't imagine why other people don't port there apps [07:30] <ajmitch> haha [07:31] <ajmitch> hadess has been doing a bit of porting lately [07:31] <arik> oh? [07:31] <arik> where has he been [07:31] <arik> he doesn't come in here anymore [07:32] <ajmitch> he's been in #gnome a fair bit [07:32] <arik> ah [07:33] <ajmitch> i tend to lurk there now [07:33] <arik> i haven't been in there in ages [07:33] <arik> i used to be in there all the time [07:33] <arik> before your time ;-P [07:33] <ajmitch> yeah ;) [07:33] <arik> i have been involved in gnome for far too long [07:33] shin_bet (sh...@ad...) joined #gstreamer. [07:33] <ajmitch> when i was still using KDE 1.93 ;) [07:33] <arik> hehe [07:33] <arik> gnome 0.20.0 [07:34] <arik> that was my first [07:34] <ajmitch> heh [07:34] <ajmitch> i can't remember what came with mandrake 5.3 [07:34] <ajmitch> i still got the cd in my cupboard somewhere... [07:34] <arik> hehe [07:34] <arik> gnome 0.20.0 came with um... redhat 5.1 or something [07:36] <ajmitch> k [07:36] <ajmitch> remember, i'm a newbie ;) [07:36] <arik> hehe [07:39] steveb (st...@no...) joined #gstreamer. [07:40] <ajmitch> hey steveb [07:40] <steveb> hi [07:43] <ajmitch> arik: build finished yet? [07:43] <arik> nope [07:43] <arik> libbonoboui [07:43] <ajmitch> k [07:43] <ajmitch> back near the start for me ;) [07:43] <arik> heh [07:44] <ajmitch> still got another 20 modules to build ;) [07:44] <arik> hah [07:46] <ajmitch> 19 sorry ;) [07:47] <arik> heh [07:47] <ajmitch> ooh, 18... [07:53] <arik> ugh [07:53] <arik> failed on libgnomeui [07:53] <arik> gnome-app-helper.c: In function `help_view_display_callback': [07:53] <arik> gnome-app-helper.c:1303: warning: passing arg 1 of `gnome_help_display' from incompatible pointer type [07:53] <arik> gnome-app-helper.c:1303: warning: passing arg 3 of `gnome_help_display' from incompatible pointer type [07:53] <arik> gnome-app-helper.c:1303: too many arguments to function `gnome_help_display' [07:53] <arik> make[3]: *** [gnome-app-helper.lo] Error 1 [07:54] <ajmitch> damn [07:55] <ajmitch> compiling pango now... [07:55] <arik> heh [07:56] <arik> the funny part of all of this [07:56] <arik> is that i don't get to run anything at the end [07:56] <ajmitch> hehe [07:56] Shippou (ir...@ot...) joined #gstreamer. [07:56] <ajmitch> unless you want to compile the panel & gnome-utils stuff [07:59] <ajmitch> damn, 15 to go still... [07:59] <arik> heh [07:59] <ajmitch> the last few will be the slowest, since they haven't been compiled today ;) [08:03] <shin_bet> shit - libtool: link: error: cannot link shared libraries into libtool libraries - [08:03] <shin_bet> what's this all about? [08:04] <ajmitch> where? [08:04] <shin_bet> gst/types [08:04] <shin_bet> from CVS [08:04] <ajmitch> which libtool ver? [08:04] <shin_bet> 1.3.5 [08:05] <ajmitch> gotta use 1.4 [08:05] <ajmitch> look at REQUIREMENTS for the other stuff you need [08:05] <shin_bet> crap [08:05] <shin_bet> thanks [08:06] <shin_bet> libtool v1.4 or better [08:06] <shin_bet> guess 1.5 is worse :-) [08:06] <arik> librsvg! [08:06] <ajmitch> hehe [08:07] Action: ajmitch doesn't have that module listed [08:07] <shin_bet> god bless rpmfind [08:07] <ajmitch> hmm, wonder why... [08:07] <arik> eel [08:07] <arik> hey aj [08:07] <arik> does v-b-s recompile every time or does it only do it if stuff has changed? [08:08] <ajmitch> arik: umm, it reruns autogen.sh every time [08:09] <arik> heh [08:09] <arik> eww [08:09] <ajmitch> so it might as well recompile it all [08:09] <ajmitch> stupid, eh? [08:09] <arik> yes [08:09] <arik> very very [08:09] <arik> eazel-hacking is better [08:09] <arik> ;-) [08:09] <arik> depend rebuilds [08:09] <arik> they are the way to go [08:10] Action: ajmitch checks out eazel-hacking [08:10] <arik> hehe [08:10] <arik> it is a wonderful thing [08:10] <arik> evil [08:10] <arik> but wonderful [08:11] <ajmitch> you were intern at eazel for a short time? [08:11] <arik> um [08:11] <arik> almost 3 months yeah [08:11] <ajmitch> ok [08:11] <arik> the last 3 months [08:11] <arik> heh [08:11] <arik> i was laid off [08:12] <ajmitch> yeah, i read in your webpage ;) [08:12] <arik> eek [08:12] <arik> scary [08:12] <ajmitch> haha [08:12] <arik> ;-) [08:12] Action: ajmitch has some old websites up on the net somewhere [08:12] <arik> i don't really post anymore [08:12] <arik> at all [08:13] <ajmitch> from when i was using windows ;) [08:13] <arik> hehe [08:15] <arik> libgnomeprint [08:16] <arik> ooh! [08:16] <arik> only one more! [08:16] <ajmitch> damnit ;) [08:17] <arik> hehe [08:17] <arik> ;-P [08:19] <ajmitch> so we'll have a gnome 2 gstmediaplay by morning? ;) [08:20] <arik> bwahahaha [08:20] <arik> matbe [08:20] <arik> er maybe [08:20] <arik> if i can get gstreamer to build [08:20] <ajmitch> shouldn't be hard [08:20] <ajmitch> it's built for me before [08:20] <arik> alright [08:20] <ajmitch> i think they've kept up with glib changes [08:21] <arik> alright [08:22] <arik> oh fucking graet [08:22] <arik> libgnomeprint fails too [08:24] <arik> In file included from gf-fontmap.c:36: [08:24] <arik> /usr/include/libgnome/gnome-util.h:17: syntax error before `char' [08:24] <arik> /usr/include/libgnome/gnome-util.h:29: conflicting types for `G_FILE_TEST_EXISTS' [08:24] <arik> /gnome/GNOME2/include/glib-2.0/glib/gfileutils.h:69: previous declaration of `G_FILE_TEST_EXISTS' [08:24] <arik> /usr/include/libgnome/gnome-util.h:35: conflicting types for `g_file_test' [08:24] <arik> /gnome/GNOME2/include/glib-2.0/glib/gfileutils.h:76: previous declaration of `g_file_test' [08:24] <arik> In file included from gf-fontmap.h:33, [08:24] <arik> from gf-fontmap.c:37: [08:24] <arik> ../libgnomeprint/gp-fontmap.h:18: syntax error before `typedef' [08:24] <arik> wonderful [08:24] <ajmitch> damn [08:24] <ajmitch> API changes still? [08:24] <arik> ooh [08:24] <ajmitch> wait [08:24] <arik> cvs update [08:24] <arik> maybe it fixes it [08:24] <ajmitch> /usr/include [08:24] <arik> oh [08:24] <arik> good point [08:25] Action: arik reruns cg [08:25] <arik> ok it built [08:25] <ajmitch> cool [08:26] <arik> very cool [08:26] <arik> just one more [08:26] <arik> and i'm FUCKING DONE! [08:26] <ajmitch> yay [08:26] <arik> just zvt [08:26] <arik> written by the illustrious michael zucchi [08:26] <ds> coffee or no coffee? [08:27] <arik> aussie extrodinaire [08:27] <arik> hmm? [08:27] <ds> should I make coffee, and thus stay up for N more hours? [08:27] <arik> oh [08:27] <arik> heh [08:28] <ajmitch> aussie? [08:28] <arik> yep [08:28] <arik> ok [08:28] <arik> i'm done! [08:28] <ajmitch> ds: coffee... [08:28] <ajmitch> arik: cool [08:28] Action: ds goes to make coffee [08:28] <arik> well [08:28] <arik> no bets on it working [08:28] <ajmitch> now you can start seeing what needs to be done for porting ;) [08:28] <arik> gtk-demo crashes [08:29] <ajmitch> hehe [08:29] <arik> pango-viewer works [08:29] <arik> ok, now to get gst [08:29] <arik> which takes a year and a day to build [08:29] <ajmitch> well, about 3 hours ;) [08:30] <ajmitch> on a really slow box [08:30] <arik> oh [08:30] <arik> no not that long [08:30] <ajmitch> gtk-demo (pid:5112): GRuntime-WARNING **: invalid cast from `PangoXFont' to `<unknown>' [08:30] Last message repeated 2 time(s). [08:30] <ajmitch> Segmentation fault (core dumped) [08:30] <arik> 10 minutes maybe [08:30] <arik> heh [08:30] <arik> i don't even get that [08:30] <arik> just core dump [08:30] <arik> i think i have something wrong with my system somewhere [08:30] <arik> this happens too often [08:31] Action: ajmitch doesn't have pango-viewer [08:31] <arik> you should [08:31] <ajmitch> ah, wait, it skipped the examples dir i think [08:31] <arik> heh [08:33] <ajmitch> damn, still dies on gconf [08:33] <arik> ouch [08:34] <ajmitch> could be cos of old gtk headers [08:34] Action: ajmitch will delete them [08:34] <arik> heh [08:34] <arik> aclocal: configure.ac: 218: macro `AM_PATH_GLIB' not found in library [08:34] <arik> aclocal: configure.ac: 221: macro `AM_PATH_GTK' not found in library [08:35] <ajmitch> umm [08:35] <arik> not good [08:35] <arik> that's gstreamer [08:35] <ajmitch> this is fscked up, gconf builds before gtk [08:35] <arik> that's not wrong [08:35] <arik> i don't think [08:35] <arik> as long as it's after glib [08:35] <ajmitch> yeah [08:35] <arik> and now [08:35] <ajmitch> but it was including gtk+ headers [08:35] <arik> now i rebuild gnome1 [08:35] <ajmitch> for some sample apps [08:36] <arik> which will take _MUCH_ longer [08:36] <arik> autoconf automake libtool gettext pkgconfig glib gtk+ imlib [08:36] <arik> gtk-engines gnome-xml ORBit audiofile esound gnome-libs gnome-http [08:36] <arik> libglade eazel-tools gnome-common gdk-pixbuf gnome-print intltool [08:36] <arik> oaf gconf gnome-vfs scrollkeeper control-center bonobo ammonite [08:36] <arik> librsvg eel nautilus xpdf libunicode gal gtkhtml bonobo-conf [08:36] <arik> evolution eog gnome-core gob pong grapevine librep rep-gtk sawfish [08:36] <arik> oh ok [08:36] <ajmitch> geez [08:37] <ajmitch> good luck gettng that lot checked out & built ;) [08:37] <arik> yep [08:37] <ds> nice to see that filesrc ! disksink segfaults [08:37] <ajmitch> blame wtay? [08:37] <ds> i was beginning to think there weren't any bugs [08:37] <arik> heh [08:37] <ajmitch> arik: i should rm -rf gconf checkout first [08:38] <arik> gst is pretty nice [08:38] <arik> but there are loads of bugs [08:38] <arik> ;-P [08:38] <ajmitch> hehe [08:38] <arik> right [08:38] <ajmitch> well, time to fix them then ;) [08:38] wingo-z (wi...@rd...) left irc: Ping timeout for wingo-z[rdu25-21-076.nc.rr.com] [08:43] <ajmitch> seems to have worked, building more now [08:43] wingo-z (wi...@rd...) joined #gstreamer. [08:49] <arik> downloading g t k! [08:49] <ajmitch> hehe [08:49] <arik> nah nah nah nah [08:49] <arik> heh [08:49] Action: ajmitch guesses gtk-demo died cos it was a new glib & pango with an old gtk [08:50] <arik> ah yeah [08:50] <arik> mine was dying cause my system is FSCKd [08:51] <ajmitch> forgot that gtk+ hadn't been built then ;) [08:51] <arik> hehe [08:53] <arik> mostly i'm just tired of building gstreamer and not being able to install it [08:54] ajmitch (me...@p7...) left irc: Ping timeout for ajmitch[p7-max11.dun.ihug.co.nz] [08:55] ajmitch (me...@p5...) joined #gstreamer. [08:56] <arik> wb [08:56] <shin_bet> aaahh! just finished second build of gst (1st was ~1 yr ago :-) [08:56] <arik> heh [08:56] <shin_bet> lets see what new goodies [08:57] <shin_bet> (it used to suck big time back then ;-) [08:57] <ajmitch> stupid ISP... [08:58] <ds> now it just sucks time [08:58] <shin_bet> lol [08:59] <arik> downloading imlib, ewwww! [09:00] <ajmitch> ah, imlib, the beautiful imlib ;) [09:00] <arik> hehe [09:01] <arik> gtk-engines [09:01] <shin_bet> Gtk-CRITICAL **: file gtkobject.c: line 726 (gtk_object_set): assertion `object != NULL' failed. [09:01] <shin_bet> shit [09:02] <ajmitch> shin_bet: running what? [09:02] <shin_bet> helloworld :-) [09:02] <shin_bet> anything [09:02] <arik> ouch [09:02] Action: ajmitch is laaggggeeedd [09:04] <ds> shin_bet: do you have mpg123? [09:04] <arik> yeah me too a bit [09:06] Action: ajmitch is checking out all gtk+ 2.0 stuff from cvs [09:06] <arik> hee [09:06] <shin_bet> y [09:06] <arik> er hehe [09:07] omega (om...@om...) left irc: Ping timeout for omega[omegacs.net] [09:07] omega (om...@om...) joined #gstreamer. [09:08] <arik> howdy o [09:16] walken (fo...@c1...) joined #gstreamer. [09:16] <ds> jo! [09:16] <walken> hey :))) [09:16] <walken> whats up ? [09:17] <ds> bug reports [09:17] <walken> about what ? :) [09:17] <ds> among other things, the altivec idct speedup is buggy [09:17] Action: walken cant seem to find any ram that works on his a7v motherboard :/ [09:17] <walken> oh ??? [09:18] <walken> whats up with it [09:18] <arik> gnome-libs!! [09:18] <ds> I'm getting memory corruption [09:18] <ajmitch> hey walken [09:18] <arik> 31 more to download [09:18] <walken> in mpeg2dec ? or using the lib in your own prog [09:18] <ajmitch> arik: hehe [09:18] <ds> in mpeg2dec from CVS [09:19] <walken> damn [09:19] <arik> heh [09:20] <ds> perhaps alignment issues [09:20] Action: BBB-zZz is back (gone 09:46:44) [09:20] Nick change: BBB-zZz -> BBB [09:21] <ds> btw, did I ever tell you about libol? [09:21] <ds> s/^/walken: / [09:22] <walken> whats it? [09:22] <ds> I took the modutils code and made it work with a user-space app, so you can load/relocate .o files at runtime [09:23] <ds> which means you can compile the code non-PIC [09:23] <walken> fun :)) [09:25] <walken> david. do you know when it stopped working ???? [09:25] <arik> course i get a new nautilus and evolution out of it. woooo [09:25] Action: arik spins a twirler [09:26] <arik> can you see the sarcasm? [09:26] <ds> walken: it works most of the time, and I haven't seen it segfault with mpeg2dec. [09:26] <ajmitch> ooh yay [09:26] <ds> I haven't been using CVS, so I don't really know. [09:27] <ds> But I narrowed the problem down to idct-altivec -- disable that, and it always works [09:27] <walken> idct as opposed to mc ? [09:28] <ds> damn. sorry. i misspoke. it's mc, not idct. idct works fine. [09:28] <walken> hmmmm [09:28] <walken> what display ? [09:28] <walken> does it work with "make check" ? [09:29] <walken> maybe its xv that does not provide the right buffer alignment... I didnt test this... [09:29] <ds> I'm using xv [09:29] <ds> ahem, all 0 tests passed [09:29] <walken> you need the mpeg2dec-streams module... [09:30] <walken> extract it in the same top-level-directory as mpeg2dec [09:39] <ds> seems to have passed. otherwise it would output dangerous-looking messages? [09:39] <walken> yes.... [09:39] <walken> it would say it didnt pass [09:39] <walken> I think make would return an error too [09:39] <walken> hmmm [09:39] <walken> OK. so its probably the alignment of xv buffers [09:40] <walken> can you dump frame->base[] after each vo_get_frame ? [09:41] <walken> base[0] should be 16-byte aligned, base[1] and base[2] should be 8-byte-aligned [09:41] <walken> its probably the problem :( [09:43] arik (ar...@sd...) left irc: sleeep [09:44] <walken> RHAAAAAAAAAAAAAAAAAAAAAAAAAA [09:44] <walken> asus wont list sdram that works on their boards [09:44] Action: walken gets annoyed [09:48] <ds> base[0]=0x30429008 [09:48] <walken> ok.... [09:48] <walken> fun :/ [09:48] <walken> hmmmm [09:48] <ds> it's 000 aligned with gstmediaplay [09:50] <walken> hmmmm [09:50] <walken> so the xv module should be hacked to provide the alignment [09:51] <walken> probably by allocating a slightly larger image and then displaying with an offset [09:51] <walken> grrrumpf [09:52] taaz_ (dlehn@66.37.66.32) joined #gstreamer. [09:52] taaz (dlehn@66.37.66.32) left irc: Ping timeout for taaz[66.37.66.32] [09:54] <ds> ok [09:54] <ds> xv seems to give aligned buffers, but most other things do not [09:55] <ds> i'll have to talk to omega and/or wtay about buffer alignment [09:55] <ds> it's kindof important [09:55] <walken> yes [10:04] <shin_bet> I just finished making gst.. but I'm having problmes [10:04] <shin_bet> open("/usr/local/lib/gst/libgstgsm.so.0.0.0", O_RDONLY|O_NONBLOCK|O_LARGEFILE|O_DIRECTORY) = -1 ENOTDIR (Not a directory) [10:04] <shin_bet> ..etc [10:04] <shin_bet> why the O_DIRECTORY? [10:27] shin_bet (sh...@ad...) left irc: Client Exiting [10:28] <ds> all better now [10:28] <ds> unfortunately, glib doesn't seem to have an equivalent of memalign [10:29] <ds> so i hacked it [10:29] <walken> hehe [10:29] <walken> hmmmm [10:40] Action: ds can't seem to handle math with numbers less than 16 [10:40] <ajmitch> heh [10:41] manuel (user422@213.174.92.50) joined #gstreamer. [10:44] <ajmitch> hi [10:57] walken (fo...@c1...) left irc: l8r [11:11] vektor (reet@HSE-Kitchener-ppp3506142.sympatico.ca) left irc: Ping timeout for vektor[HSE-Kitchener-ppp3506142.sympatico.ca] [11:27] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) joined #gstreamer. [11:37] omega (om...@om...) left irc: Ping timeout for omega[omegacs.net] [12:04] <ajmitch> who's awake? [12:11] <ds> i'm in pre-zzz preparation [12:11] <ajmitch> ah, ok [12:11] <ajmitch> was just wanting to find out which plugin is best/works for mpeg4 files? [13:16] BBB (BB...@uc...) got netsplit. [13:16] steveb (st...@no...) got netsplit. [13:16] wingo-z (wi...@rd...) got netsplit. [13:16] taaz_ (dlehn@66.37.66.32) got netsplit. [13:16] manuel (user422@213.174.92.50) got netsplit. [13:17] BBB (BB...@uc...) returned to #gstreamer. [13:17] steveb (st...@no...) returned to #gstreamer. [13:17] wingo-z (wi...@rd...) returned to #gstreamer. [13:17] taaz_ (dlehn@66.37.66.32) returned to #gstreamer. [13:17] manuel (user422@213.174.92.50) returned to #gstreamer. [13:50] chillywilly (da...@us...) joined #gstreamer. [13:52] harobed (harobed@ACAA08BF.ipt.aol.com) joined #gstreamer. [13:55] chillywilly (da...@us...) left irc: Read error to chillywilly[user89.simplexnet.com]: EOF from client [14:22] asmod (stevec@64.5.222.2) joined #gstreamer. [14:53] manuel (user422@213.174.92.50) left irc: Client Exiting [14:55] wingo-z (wi...@rd...) left irc: Read error to wingo-z[rdu25-21-076.nc.rr.com]: EOF from client [14:58] wingo (wi...@rd...) joined #gstreamer. [14:59] <wingo> 'morning, peoples [15:04] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [16:04] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:11] zblaxell (zbl...@24...) joined #gstreamer. [17:07] Nick change: wingo -> wingo-class [17:10] Shippou (ir...@cr...) joined #gstreamer. [17:13] Nick change: taaz_ -> taaz [19:34] Nick change: taaz -> taaz-away [19:34] ChiefHighwater (pa...@te...) joined #gstreamer. [19:48] soundmarsc (tho...@so...) left irc: Client Exiting [20:03] Nick change: wingo-class -> wingo [20:04] <wingo> too... much... indian food... [20:04] Nick change: tnt-halloween -> tnt [20:05] <tnt> Good morning. [20:15] <wingo> 'afternoon, tnt [20:28] chillywilly (da...@d1...) joined #gstreamer. [20:30] steveb (st...@no...) left irc: Ping timeout for steveb[node1ee06.a2000.nl] [20:30] steveb (st...@no...) joined #gstreamer. [20:52] omega (om...@om...) joined #gstreamer. [20:52] <ChiefHighwater> Ello [20:53] <omega> yo [20:58] <chillywilly> yo [21:00] Nick change: taaz-away -> taaz [21:07] <ajmitch> hi [21:18] Zeenix (ze...@13...) joined #gstreamer. [21:19] <ds> jo [21:19] <Zeenix> yo [21:19] <ds> Zeenix: where are you in .pk [21:19] <ds> ? [21:19] <Zeenix> peshawar [21:20] <Zeenix> any sign of wtay today [21:20] <Zeenix> ? [21:21] ChiefHighwater (pa...@te...) left irc: [21:21] <ds> haven't seen him. perhaps he is already flying to ALS? [21:21] <Zeenix> ds: ALS ? [21:22] <omega> ds: only Uraeus and I are going to ALS [21:22] <ds> =) [21:22] <ds> oh yeah [21:23] <ds> omega: am I correct in thinking that there's no current way to get aligned gstbuffers? [21:24] <omega> the buffer pointer can be aligned to whatever you want [21:24] <omega> if you're talking about enforcing alignment when someone else is allocating them, no, not atm [21:25] <ds> ok [21:26] <ajmitch> hi omega [21:26] <omega> yo [21:26] <ds> right now, i see two options -- [21:27] <ds> one is to allocate a buffer, and g_new an area slightly larger, [21:27] <ds> then do a gst_buffer_create_sub() to get another buffer with [21:27] <ds> the correct alignment, then unref the parent buffer [21:27] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-7.brain.net.pk] [21:28] ChiefHighwater (pa...@te...) joined #gstreamer. [21:28] <taaz> doesn't that kill the point of asking another element for the buffer in the first place? [21:28] <ds> the other is to override the default free function [21:28] <ds> taaz: I'm ignoring the problem where you get unaligned buffers [21:30] <omega> ds: where do you need aligned buffers? [21:30] Zeenix (ze...@13...) joined #gstreamer. [21:30] <ds> gstmpeg2dec, on ppc [21:30] <ds> altivec _really_ wants aligned pointers [21:31] <omega> um, but altivec is only used on dct and motion comp data, both of which are allocated by libmpeg2 [21:32] <Zeenix> omega: would you allow to start the rate in oss/gsm/etc from 1000 ? [21:33] <ds> the image data is allocated by the plugin. [21:33] <omega> Zeenix: yes, we discussed this a while ago, if oss allows lower rates, then you can change it to whatever is reasonable [21:33] <omega> ds: which? [21:33] <ds> gstmpeg2dec, if downstream is not using bufferpools [21:34] <Zeenix> omega: ok but we need to set the caps accordingly i mean [21:34] <omega> ds: ok, so you allocate an aligned buffer.... that's not gst's problem at that point [21:35] <omega> if using bufferpools, we need to find a solution of some kind [21:35] <ds> I've noticed that (random plugin) ! osssink goes crazy if (random plugin) and osssink can't agree on a rate [21:35] <ds> omega: bufferpools seem to be ok; don't know why [21:35] <ds> 0x000 aligned. [21:36] <omega> ds: we should really set it up so osssink doesn't work if caps aren't agreed on [21:36] <ds> omega: I'll look into it, since my G4 doesn't work at 48000 [21:36] <omega> bufferpools are only as good as the allocator, which may or may not allocate aligned buffers [21:37] <omega> but your idea of using subbuffers to guarantee alignment is pretty much what we have to do, or a custom free function [21:37] <omega> probably more sane to create a custom free function [21:37] <ds> xvideosink is ok with bufferpools [21:38] <ds> omega: what would be the side effects of changing gst_buffer_destroy to observe buffer->offset? [21:38] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-49.brain.net.pk] [21:38] <omega> it would screw up on all buffers coming from disksrc, for instance [21:39] <omega> offset is supposed to be the absolute source offset [21:41] <ds> but if parent==NULL, then perhaps g_free() could be called with buffer->data - buffer->offset? [21:41] <omega> no, because offset is an absolute origin offset [21:42] <ds> ? not understanding [21:42] <taaz> omega: new NGPT out... what's still lacking in that before gstreamer can try and use it? [21:42] <omega> taaz: lots, it's just the wrong paradigm [21:43] <omega> ds: offset cannot be used for anything else [21:43] <taaz> oh [21:43] <omega> taaz: it's a generic pthreads API [21:43] <omega> which means we have no access to "cothreads" [21:43] <ds> omega: ok, I give up... =) [21:44] <omega> ds: successive buffers from a file have offset of 0, 4096, 8192, etc. [21:44] <omega> that's what offset is for [21:44] <taaz> eh? oh i thought they were going to add that sort of stuff too [21:44] <omega> we can't overload the definition to be a misalignment value [21:45] <omega> taaz: it's maybe possible, but I haven't pursued it because there are other potential problems too [21:45] <ds> omega: I misinterpreted what offset was for, then [21:45] <omega> among other things, it's not clear that it actually *does* user-space threads on top of kernel threads [21:46] chillywilly (da...@d1...) left irc: Read error to chillywilly[d145.as3.nwbl0.wi.voyager.net]: EOF from client [21:46] <omega> ds: it may be that offset as currently defined is irrelevant, in which case it can be redefined [21:46] <omega> but we need to think about that before redefining it [21:47] <ds> do you know if there are glib plans for g_memalign() or similar? [21:50] <omega> no idea [21:50] <omega> suggest it [21:52] CHW_tim (pa...@te...) joined #gstreamer. [21:52] ChiefHighwater (pa...@te...) left irc: [21:52] Nick change: CHW_tim -> ChiefHighwater [22:06] ajmitch (me...@p5...) left irc: Ping timeout for ajmitch[p57-max2.dun.ihug.co.nz] [22:08] ajmitch (me...@p2...) joined #gstreamer. [22:38] chillywilly (da...@d1...) joined #gstreamer. [22:39] Zeenix (ze...@13...) joined #gstreamer. [22:39] <Zeenix> omega: so we were discussing that "rate" ? [22:40] <omega> what about it? if it needs changing, change it [22:41] <Zeenix> omega: yeah i i'll do that in my gst source, but i cant change the original one [22:41] <omega> submit a patch to change the relevant places to the list [22:42] Action: Zeenix doenst know how to make a patch( oops he forgot how to ) [22:43] <wingo> man diff [22:43] <wingo> or makepatch [22:43] <omega> diff -un origfile newfile [22:43] <omega> or -urn origdir newdir for recursive [22:44] <taaz> or 'cvs diff -u' [22:44] <omega> right [22:45] <Zeenix> omega: ok ok :) [22:45] <Zeenix> omega: after calling gst_pad_set_caps(..) on the pad, can we get the actual [22:45] <Zeenix> "rate" set by the peer element by [22:45] <Zeenix> xxx = gst_caps_get_int( gst_pad_get_caps( pad ), "rate" ) ; [22:46] <omega> yes [22:46] <Zeenix> omega: uhm..., i wonder why wtay commented out the last line then [22:47] <omega> no idea [22:47] <Zeenix> omega: do you know where he is these days ? [22:48] <Zeenix> s/he/wtay [22:48] <omega> nope [22:51] <Zeenix> i was trying to get more types into rtpsend/recv today [22:52] ChiefHighwater (pa...@te...) left irc: [22:52] <Zeenix> omega: well i tried playing an mp3 through it, it worked but after some packets it crashed [22:52] <Zeenix> omega: where should we be connecting the mp3parse, rtpsend or rtprecv ? [22:53] <omega> depends on whether rfc2250 requires the rtp packets to contain an integral nu... [truncated message content] |
From: IRC B. <wt...@us...> - 2001-11-03 05:26:20
|
******************************************************************* [03:05] <wingo> nah. just something that was bearing me down :) [03:06] <tnt> What class is it for? [03:07] <wingo> a strange capstone course for this dual-degree program i'm in. the goal is to produce, in a group, a technology assessment of the agricultural impacts of water. [03:07] <wingo> probably 60-70 pages when we're through. [03:07] <wingo> work i can put off till this weekend :-))) [03:07] <tnt> 60-70 pages... sound fun ;-) [03:08] <tnt> What year are you in? [03:10] <wingo> fifth, and last (pretty pleased about that :) [03:10] <dobey> heh [03:10] <tnt> :-) I know how you feel... I graduate a year ago. [03:11] <tnt> I'll go back later... when I forget what homework way like... and get my Masters. [03:12] <wingo> that might be my case as well, working for money would be nice at this point [03:13] <tnt> Money is nice... [03:14] <tnt> ... I actually got laid off a while ago... the place I was working for got hit by the dot-com-crash. [03:14] <wingo> sucks :-\ [03:14] <tnt> ... I'm actually writing a cover letter (for my resume) right now. [03:14] <tnt> Ya, it does suck. [03:15] <wingo> good luck :) [03:15] <tnt> Thanks. [03:23] _gst_newt_ joined #gstreamer. [03:23] wtay-zZz (wi...@ca...) left irc: Ping timeout for wtay-zZz[cable-195-162-215-201.upc.chello.be] [03:23] wtay-zZz (wi...@ca...) joined #gstreamer. [03:24] <wingo> muahaha, /me ports his app to gnome2 [03:48] <dobey> hmm [03:50] <wingo> learning tree{view,model} will be interesting [04:26] <ds> my new audio resampling library rocks! [04:27] <ds> -66 to -70 dB noise level for 48khz to 44.1khz resampling at real-time speeds [04:27] <wingo> yeah? [04:27] chillywilly (da...@d1...) left irc: Read error to chillywilly[d161.as27.nwbl0.wi.voyager.net]: Connection reset by peer [04:27] <wingo> that's nice [04:28] chillywilly (da...@d1...) joined #gstreamer. [04:28] <wingo> it will go into gst, yes? :-) [04:28] <ds> yep [04:28] <wingo> yessssssssss [04:28] <wingo> my precioussssss [04:28] <ds> it needs some massive speedups, though [04:28] <dobey> ugh [04:30] <wingo> um, anyone here familiar with GtkTreeModel? [04:30] <wingo> the merits of various implementations? [04:30] <ds> as a comparison, linear interpolation of samples has a noise level of -35 dB at 2khz for a similar system [04:31] <wingo> interesting [04:31] <wingo> so you were measuring your sinc() algorithm? [04:31] <ds> and the noise is linear with frequency, so at 20 khz, its -15 dB, which is kinda noticable [04:32] <ds> yes, this is with the sinc() algorithm. I also discovered that it is really good for low pass filtering [04:34] <ds> on second thought, the noise for linear interpolation is not linear with frequency. It's more than linear, so you get -0 dB noise at 20 khz. (nasty) [04:35] <dobey> hrmm [04:35] <dobey> anyone in here used AM_PATH_ORBIT () at all? [04:40] <wingo> no, you should use pkg-config [04:41] <dobey> no [04:41] <dobey> wtf would i use pkg-config? [04:41] <dobey> this is gnome1 [04:41] <wingo> there are .m4s for that iirc [04:41] <wingo> er, .pcs [04:41] <wingo> .pc's [04:41] <dobey> one should use ac macros when using ac [04:41] <wingo> the real question is wtf would you use gnome2, imho [04:42] <wingo> PKG_CONFIG_CHECK_MODULES [04:42] <vektor> Woo woo sinc() ! :) [04:42] <wingo> pkg.m4 [04:42] <wingo> er, s/gnome2/gnome2 [04:42] <wingo> aaa! [04:42] <wingo> er, s/gnome2/gnome1 [04:42] <dobey> those ac macros should then use use <package>-config or pkgconfig [04:42] Action: wingo cannot type [04:42] <dobey> uhm [04:42] <dobey> because gnome2 is not fucking stable [04:42] <dobey> that's why [04:42] <dobey> sigh forget [04:42] <wingo> that's cool, encompass isn't either, is it? [04:42] <vektor> ds: tell me about this sinc() algorithm. [04:42] <dobey> fuck you [04:43] <wingo> ooo, testy [04:43] <chillywilly> we don't need no stinkin' stability [04:43] <chillywilly> heh [04:43] Action: chillywilly slowly backs away [04:43] Action: wingo too [04:43] Action: dobey slowly unsheathes his sword [04:44] Action: wingo knows iaido [04:44] Action: dobey knows big ass sword through skull [04:44] <wingo> no offense intended, eh? [04:44] <wingo> sorry. [04:45] arik (ar...@sd...) joined #gstreamer. [04:45] <wingo> yo. [04:45] <arik> hi [04:45] <arik> omega: i liked that comic you sent me btw [04:46] <ds> vektor: sinc(x) = sin(x)/x [04:46] <ds> it's a magic function [04:46] <ds> are you familiar with dirac delta function? [04:48] <vektor> I'm familiar with sinc(). [04:48] <vektor> However. [04:48] <vektor> I want to know how you use it. [04:48] <wingo> i am :-) fellow physics students, unite! [04:48] <vektor> What's your algorithm for approximating it on input? [04:48] <wingo> :-) [04:49] <ds> basically, sinc() has a number of interesting properties: [04:49] <vektor> It's 0 at all integers besides 0 ? [04:50] <vektor> :) [04:50] <vektor> I love sinc() :) [04:50] <ds> sum(sinc(x*n),{n,-inf,+inf}) for all x [04:50] <ds> =1 [04:52] <ds> thus, sum(sinc(x*n) * f(x*n), {n,-inf,+inf}) = f(x*n) [04:52] <vektor> interesting [04:52] <ds> I didn't get it quite right, but that's the idea [04:53] <ds> so within the bounds of how your original data was sampled, [04:53] <ds> you can determine f(x) for any x, whether it is a sampled point or not [04:54] <ds> and by changing the spacing of the sinc()*f() sum, you can also do low pass filtering [04:54] <ds> this is rather important, since downsampling will generally alias unwanted high-frequency noise into lower frequencies [04:55] <ds> so you can resample and low-pass filter in one step [04:55] arik (ar...@sd...) left #gstreamer. [04:55] <vektor> Wow! [04:55] <vektor> ds: Dude, is that well known? [04:55] <vektor> I don't think I've ever seen anyone combine them like that. [04:55] <ds> yes [04:55] <vektor> Wow. [04:55] <vektor> Do you have code? [04:56] <ds> yes [04:56] <vektor> Duuude. [04:56] <vektor> Hey, dude, do you know anything about colour? [04:56] <vektor> I have a math question for you if you do. :) [04:56] <ds> of source, the difficult part is that sin(x)/x is expensive to evaluate. I'm working on that next [04:57] Action: vektor smells a lookup table. [04:57] <vektor> ds: dcc me some source when you have a sec. [04:57] <vektor> i've been wanting to write a resampler for a while. [04:57] <vektor> i'd love you try playing with yours. [04:57] <ds> i'll be putting it into cvs today or tomorrow [04:57] <vektor> reeeet. [04:58] <vektor> I need to idle a bit.... have a midterm tomorrow, going to study a sec. [04:58] <vektor> justasec. [05:07] <dobey> bah [05:07] <dobey> night [05:07] dobey (do...@cv...) left #gstreamer (eh). [05:22] harobed (harobed@ACAA08BF.ipt.aol.com) left irc: Ping timeout for harobed[ACAA08BF.ipt.aol.com] [06:32] chillywilly (da...@d1...) left irc: Ping timeout for chillywilly[d161.as27.nwbl0.wi.voyager.net] [06:32] <tnt> Hey... for everyone running Red Hat 7.x and Ximian GNOME, you might be interested in knowing... [06:32] <tnt> ... that there is a channel to get GNOME 2 now!!! [06:32] <tnt> It's for us developers though. [06:33] <tnt> Wohoo. [06:34] chillywilly (da...@d1...) joined #gstreamer. [06:47] tnt (tn...@h2...) left irc: Client Exiting [06:53] tnt (tn...@h2...) joined #gstreamer. [06:57] <wingo> night all... [06:57] Nick change: wingo -> wingo-zzz [07:09] <tnt> If anyone is installing GNOME 2 from Ximian's Red Carpet, make sure you uninstall pkgconfig first, if you installed it from a tarball. [08:20] steveb (st...@no...) joined #gstreamer. [08:29] Shippou (ir...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [09:22] Nick change: taaz -> taazzzz [09:28] wtay-zZz (wi...@ca...) got netsplit. [09:28] wtay-zZz (wi...@ca...) returned to #gstreamer. [09:36] Shippou (ir...@ot...) joined #gstreamer. [09:36] wingo-zzz (wi...@rd...) got netsplit. [09:36] ajmitch (me...@p2...) got netsplit. [09:37] ajmitch (me...@p2...) returned to #gstreamer. [09:37] wingo-zzz (wi...@rd...) returned to #gstreamer. [09:44] Nick change: tnt -> tnt-zZz [10:03] mwc (ma...@ly...) left irc: [x]chat [11:10] omega (om...@om...) left irc: Ping timeout for omega[omegacs.net] [11:45] harobed (harobed@AC8B37CF.ipt.aol.com) joined #gstreamer. [11:51] harobed (harobed@AC8B37CF.ipt.aol.com) left irc: Ping timeout for harobed[AC8B37CF.ipt.aol.com] [11:53] chillywilly (da...@d1...) left irc: Read error to chillywilly[d174.as7.nwbl0.wi.voyager.net]: EOF from client [11:53] chillywilly (da...@d1...) joined #gstreamer. [12:06] harobed (harobed@AC83DE27.ipt.aol.com) joined #gstreamer. [12:11] chillywilly (da...@d1...) left irc: restarting X [13:02] Action: BBB-zZz is back (gone 13:24:05) [13:02] Nick change: BBB-zZz -> BBB [13:07] Action: BBB is away: buying stuff [13:07] Nick change: BBB -> BBB-away [13:22] chillywilly (da...@d1...) joined #gstreamer. [13:24] chillywilly (da...@d1...) left irc: Read error to chillywilly[d174.as7.nwbl0.wi.voyager.net]: EOF from client [14:46] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [14:48] asmod (stevec@64.5.222.2) joined #gstreamer. [15:42] Zygo (zbl...@24...) joined #gstreamer. [15:44] steveb (st...@no...) left irc: Ping timeout for steveb[node1ee41.a2000.nl] [15:50] harobed (harobed@AC83DE27.ipt.aol.com) left irc: Read error to harobed[AC83DE27.ipt.aol.com]: Connection reset by peer [15:51] thomasvs (thomas@212.100.172.175) joined #gstreamer. [16:23] Action: bstard is away: Weg [16:31] asmod (stevec@64.5.222.2) left irc: [x]chat [16:33] asmod (stevec@64.5.222.2) joined #gstreamer. [16:52] Shippou (ir...@cr...) joined #gstreamer. [17:02] evil_monk (trevo@208.141.162.68) left irc: Ping timeout for evil_monk[208.141.162.68] [17:03] evil_monk (trevo@208.141.162.68) joined #gstreamer. [17:30] Nick change: taazzzz -> taaz [17:34] Nick change: wingo-zzz -> wingo [17:35] <wingo> hi taaz. [17:37] <taaz> yo wingo [17:54] Nick change: taaz -> taaz-away [17:57] harobed (harobed@AC85A024.ipt.aol.com) joined #gstreamer. [17:59] Nick change: wingo -> wingo-away [18:38] steveb (st...@no...) joined #gstreamer. [18:44] Nick change: wingo-away -> wingo [18:44] <wingo> salutations, steveb [18:46] sienap (sy...@s3...) joined #gstreamer. [18:46] <sienap> Hey all [18:46] <sienap> Pretty odd.. parapraxis hasn't been on for a long time.. and also isn't responding to email [18:47] <sienap> anyway... any new cool stuff ? [18:49] <wingo> naw, just old cool stuff :-) [18:49] <wingo> i wonder if he's been committing at all to his sf project... [18:51] <sienap> nmphf [18:51] <sienap> hope he isn't gone or something [18:51] <sienap> Paranormal is born to be coded on :) [18:51] ChiefHighwater (pa...@te...) joined #gstreamer. [18:52] <wingo> hey chief [18:52] <ChiefHighwater> Ello wingo 8-] [18:52] <sienap> heuy [18:52] <wingo> how goes it? /me decided to skip skool today :-) [18:53] <ChiefHighwater> ok, been pretty bust round here last several days cuz of a nasty email virus [18:53] <wingo> yuk [18:53] <wingo> any data loss/ [18:53] <wingo> ? [18:54] <ChiefHighwater> this is the first day this week that I haven't walked in to a crashed computer somewhere in our office [18:54] <ChiefHighwater> yes, we lost one user's hard drive [18:54] <wingo> sucks :-\ [18:54] <ChiefHighwater> omega is going to try to reconstruct the FAT32 by hand, but good luck [18:55] <wingo> thinking about switching to linux with staroffice? ;) [18:55] <ChiefHighwater> not really ;-p [18:55] <ChiefHighwater> but we are going to implement a RAID file server with weekly ofsite backups [18:56] <ChiefHighwater> that will be on a Linux box [18:58] sienap (sy...@s3...) left #gstreamer. [19:00] <ChiefHighwater> software RAID on a rh7.2 box [19:01] <ChiefHighwater> we are also moving to QMail and implementing a mail filter to strip windows executables of off emails [19:03] <wingo> that's good [19:03] <wingo> and qmail is nice [19:25] vektor_ (wb...@pe...) joined #gstreamer. [19:26] Nick change: tnt-zZz -> tnt [19:26] <tnt> Good morning. [19:27] vektor_ (wb...@pe...) left irc: too bad the scene is dead [19:27] <wingo> 'mornin [19:28] <tnt> Hmmm... wtay has been in bed for about 2 days.... [19:41] <wingo> heh [20:10] Nick change: tnt -> tnt-out [20:12] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [20:41] omega (om...@om...) joined #gstreamer. [20:48] <vektor> Hi omega. [20:48] <omega> yo [20:48] <vektor> omega: have you seen my reet dvd tech page? [20:48] <vektor> I'm still updating it though. [20:48] <omega> nope, url? [20:49] <vektor> http://www.dumbterm.net/graphics/dvd/ [20:57] <omega> good, you finally have the Matrix [20:58] <vektor> I've had it forever. [21:00] Nick change: taaz-away -> taaz [21:00] <vektor> taaaz. [21:00] <taaz> veeektor [21:02] <vektor> The movie I saw was 'The Foul King'. [21:03] <taaz> omega: you going to suggest gstreamer for that rta thread on lad? [21:03] <omega> rta? reading [21:05] Nick change: tnt-out -> tnt [21:06] <tnt> Hey... I remember you guys talking about having trouble with threads before, in GStreamer... [21:07] <tnt> ... the ref-ing and unref-ing in GtkObject and GObject are NOT thread safe... could that be the cause of your problems? [21:07] <taaz> feel free to fix it yo [21:07] <tnt> (Maybe early destroying of the Object... or something else screwing up the ref count.) [21:07] <tnt> :-) [21:07] <taaz> please [21:08] <taaz> i kinda stopped working on gst cause i'm too pissed about this untheadsafe crashing randomly stuff [21:08] <taaz> omega: get your smp box working so you can suffer too! [21:08] <omega> timj's response is rather lame [21:08] <tnt> taaz: What file is it in? [21:09] <tnt> schedule.c ? [21:11] <taaz> look in the gst dir ;) [21:11] Nick change: wtay-zZz -> wtay [21:11] <wtay> yawn [21:11] <wtay> yo [21:11] <vektor> wtay! [21:12] <tnt> wtay: How was your (2 day) nap? :-) [21:27] ajmitch (me...@p2...) left irc: Ping timeout for ajmitch[p27-max4.dun.ihug.co.nz] [21:30] ajmitch (me...@p2...) joined #gstreamer. [21:33] Action: bstard is back (gone 05:10:08) [21:40] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) left irc: back in a sec [21:42] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) joined #gstreamer. [21:46] Nick change: wtay -> wtay-tv [21:50] Zeenix (ze...@13...) joined #gstreamer. [21:51] dap (dap@SR3745.resnet.ucsb.edu) joined #gstreamer. [21:54] <wingo> taaz: why don't you build a glib2 gstreamer? [21:54] <wingo> sorry if you've already gone over all this, just thinking out loud [21:57] <Zeenix> are we having H.263 support ATM ? [21:58] <taaz> wingo: does it work with debian glib2? [21:58] <taaz> i havent tried yet... [21:58] <taaz> i just assumed the recent 64bit stuff broke it [22:00] asmod (stevec@64.5.222.2) left irc: [x]chat [22:05] <wingo> no, unfortunately :-\ [22:06] <wingo> the int64 as well as some function renames make it incompatible [22:06] <wingo> Zeenix: you mean in gst? don't think so [22:07] <Zeenix> wingo: yeah [22:08] <Zeenix> wingo: yesterday you told me that 32 bit( int ) dont need byte ordering, then what are g_htonl & g_ntohl for ? [22:08] <wingo> 8 bit int you mean [22:08] <wingo> that's what i said :) [22:09] <wingo> also, i'm not familiar with g_htonl, etc [22:09] <Zeenix> you said only 16 bit ints need that, AFAI remember [22:09] <wingo> oh sorry. i thought you were doing just audio data [22:09] <wingo> where really in most apps just 8 and 16 bits are possibilities [22:10] <wingo> 'byte ordering' implies more than one byte, so [22:10] <Zeenix> yeah [22:10] <wingo> yeah :) [22:12] <Zeenix> wingo: i've done with all but the real data, gst buffers, which rtpsend gets from gsmenc or osssrc [22:13] <wingo> that's cool [22:14] <Zeenix> wingo: any idea ? [22:14] <wingo> so you'll use this for zchat? [22:14] <wingo> ? [22:14] <wingo> about what? [22:14] <Zeenix> wingo: sure i'll be using that for zchat [22:15] dap (dap@SR3745.resnet.ucsb.edu) left irc: Ping timeout for dap[SR3745.resnet.ucsb.edu] [22:15] <Zeenix> wingo: about how to byte-order the real data [22:18] <wingo> dunno, i think there's macros in glib for that [22:19] <wingo> if width=8 or you're doing gsm (which is 8 bits wide, no?) you don't have to worry [22:19] <wingo> otherwise, if you are little endian and playing 16 bit data swap every pair of bytes [22:19] <wingo> iirc :-) [22:20] bstard (Lor...@a2...) left irc: Client Exiting [22:20] <Zeenix> wingo: gsm plays with shorts( 16bit ) [22:21] Nick change: tnt -> tnt-out [22:30] chillywilly (da...@d1...) joined #gstreamer. [22:31] <wingo> yo chillywilly [22:34] <chillywilly> hi [22:38] <Zeenix> i was think if we could install "endianness" as a property in the the osssrc/sink & gsmenc/dec [22:38] <thomasvs> wingo: automake (GNU automake) 1.4-p5 [22:38] <thomasvs> on the build machine [22:39] <chillywilly> man if I catch that fscking cat it will die [22:40] <wingo> hrm. [22:40] <wingo> thomasvs: is that a clean checkout and autogen? [22:40] <thomasvs> wingo: yes, every time [22:40] <thomasvs> it wipes out the previous build and starts from scratch [22:40] <thomasvs> why ? [22:41] <chillywilly> anyone want a stupid cat? [22:41] <wingo> wierd, it seems that the autogen should automake all .in's, then only descend into DIST_SUBDIRS [22:41] <wingo> which cat? [22:41] <wingo> /usr/bin/cat? [22:41] <chillywilly> this stupid cat [22:41] <thomasvs> chillywilly: what does she taste like ? [22:41] <wingo> er, /bin/cat [22:41] <thomasvs> does she taste like chicken ? [22:41] <chillywilly> thomasvs: I dunno, but you can east it if you want to [22:42] <thomasvs> is it a young cat ? I've always wanted to raise a bonsai kitten [22:42] <wingo> hah! [22:42] <chillywilly> it is not too young [22:42] <chillywilly> we have a younger one too, but I like that one [22:42] <thomasvs> oh well... send it over anyway. [22:43] <thomasvs> by overseas mail, I can't afford airmail [22:43] <chillywilly> this one just crapped on the floor in front of me an ran off.....if the box is not completely clean it will not go in it [22:43] <wingo> well, hop to it [22:43] <chillywilly> so you see it must die [22:43] <wingo> clean clean clean [22:43] <chillywilly> not my cat [22:43] <chillywilly> not my responsibility [22:43] <wingo> but your problem [22:43] <chillywilly> yea, well I will have to have a few words with soneone [22:44] <chillywilly> this cat has got to go [22:44] <chillywilly> I have had it [22:45] <Zeenix> i dont know why i get this error now when i try to compile my old cvs gst: [22:45] <Zeenix> . [22:45] <chillywilly> . [22:45] Last message repeated 1 time(s). [22:45] <Zeenix> In file included from ../gst/gstcaps.h:36, [22:45] <Zeenix> from ../gst/gstpad.h:39, [22:45] <Zeenix> from ../gst/gst.h:34, [22:45] <Zeenix> from qtest.c:2: [22:45] <Zeenix> ../gst/gstprops.h:77: parse error before `va_list' [22:45] <Zeenix> In file included from ../gst/gst.h:43, [22:45] <Zeenix> from qtest.c:2: [22:45] <Zeenix> ../gst/gstautoplug.h:68: parse error before `va_list' [22:45] <Zeenix> ../gst/gstautoplug.h:69: parse error before `va_list' [22:46] <Zeenix> make[3]: *** [qtest.o] Error 1 [22:46] <Zeenix> make[2]: [22:46] <Zeenix> its only the tail of it [22:46] <wingo> probably a type you don't have defined [22:46] <wingo> what is on line 77 of gstprops.h? [22:47] <Zeenix> wingo: not on linux box atm [22:47] <chillywilly> slap yourself [22:47] <wingo> you need the first error, that's the important one to fix in this [22:48] <Zeenix> wingo: here is the head of it: [22:48] <Zeenix> In file included from /usr/include/glib-1.2/glib.h:70, [22:48] <Zeenix> from qtest.c:1: [22:48] <Zeenix> In file included from qtest.c:1: [22:48] Zygo (zbl...@24...) left irc: Read error to Zygo[246.chipworks.com]: EOF from client [22:48] Shippou (ir...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [22:48] <Zeenix> In file included from qtest.c:1: [22:49] <Zeenix> heh, i cant get it all copied here, i wonder why [22:51] <Zeenix> In file included from qtest.c:1: [22:51] <Zeenix> ughhh..... [22:52] <wingo> probably a missing semicolon [22:52] <wingo> at the end of one of your header files possibly [22:53] <Zeenix> ok [23:01] <Zeenix> wingo: which ver. of libtool-libs you have ? [23:03] <wingo> 1.4p2 iirc [23:04] <Zeenix> wingo: iirc ? [23:04] <wingo> if i recall correctly [23:09] steveb (st...@no...) left irc: advance! [23:11] <Zeenix> looks like wtay slept with his tv today [23:13] dap (dap@SR3745.resnet.ucsb.edu) joined #gstreamer. [23:19] dobey (do...@dr...) joined #gstreamer. [23:19] Action: dobey kicks wingo in the pants. [23:19] <dobey> ;-) [23:23] <thomasvs> wingo: just reading your comment on galeon ;) [23:23] <dobey> (hence the kick) [23:23] <thomasvs> well, is encompass released yet ? [23:24] <dobey> uhm, yes [23:24] <dobey> allow me to show you something [23:24] <thomasvs> please do :) [23:25] <chillywilly> galeon is the cats meow [23:25] <dobey> not exactly [23:25] <thomasvs> as well as the mutt's nuts [23:26] <dobey> http://primates.ximian.com/~dobey/microsoft.com.png [23:26] <dobey> the only thing galeon has going for it is mozilla, and it's not the greatest thing ever [23:27] <dobey> and yes, that's the same size default font in both windows [23:27] <dobey> both helvetica [23:28] <thomasvs> looks pretty good ;) [23:28] <dobey> and that encompass is before the rewrite that i'm doing right now [23:28] <wingo> heh ;) [23:29] Action: wingo opens the png in galeon ;-) [23:29] <dobey> i wish people would get rid of the ignorance that fills their heads [23:30] <wingo> chillywilly: not your cat though? ;-) [23:34] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-26.brain.net.pk] [23:34] <wingo> that's a neat pair of shots dobey. are there debs for encompass? [23:35] <dobey> wingo: yes, it should be in sid [23:36] Action: wingo installs encompass. [23:37] Action: chillywilly is away: pick up caroline from work [23:39] <dobey> hrmm [23:44] <dobey> brb [23:44] dobey (do...@dr...) left #gstreamer (eh). [23:51] dobey (do...@dr...) joined #gstreamer. [23:55] <dobey> ugh [23:58] <taaz> omega: pbd's reply sounds like a challenge! ;) [23:59] <wingo> indeed :-) [00:00] --- Sat Nov 3 2001 [00:00] <wingo> he might be right though, i was thinking about this a few days ago [00:00] <wingo> the buffer size would have to be teeniny to do anything remotely realtime-style [00:01] <wingo> 128 bytes or so [00:04] Shippou (ir...@ot...) joined #gstreamer. [00:11] <taaz> is that bad? [00:11] <dobey> haha [00:11] <dobey> this article funny [00:11] <dobey> after actually reading the article anyway [00:11] <wingo> the galeon one? yeah. he gives no reasons. really. i do like galeon tho :-) [00:12] <dobey> well, i especially like the binary size comment [00:12] <dobey> trying to indicate that it is "small" [00:12] <wingo> taaz: how would realtime midi be dealt with in a streaming framework? a callback would be called on midi interrupts, no? [00:12] <wingo> we'd have to put it in a queue [00:13] <wingo> and deal with it later [00:13] <taaz> what article? [00:13] <wingo> the latencies pile up, it seems [00:13] <wingo> gnotices.gnome.org, the second one or so about galeon [00:13] <taaz> i really have no idea how midi works ;) [00:13] <dobey> hrmm [00:15] <omega> wingo: huh? [00:15] Action: chillywilly is back (gone 00:38:17) [00:16] <dobey> "Galeon is the bee's knee's. ;) Lee J. ;)" [00:16] <dobey> hahaha [00:16] <dobey> i should put that in my .sig [00:16] <omega> galeon definitely sucks less than pure mozilla [00:17] Nick change: tnt-out -> tnt [00:17] <wingo> omega: do you know much about midi? [00:17] <omega> no, but there's no reason to ever queue anything [00:17] Action: wingo doesn't, for the record [00:17] <wingo> how would you deal with events that can arrive at any time, not just when they're pulled? [00:18] Action: wingo needs to investigate the alsa midi api [00:18] <tnt> I'm working on something like that right now: "event that can arrive at any time". [00:19] <omega> a midisrc would wait for an event, grab it, and push it down the pipeline [00:19] <wingo> immediately? [00:19] <omega> yes [00:19] <wingo> interesting. [00:19] <omega> what on earth would you wait for?? [00:19] <wingo> is there not an overhead to pushing a buffer? [00:19] <tnt> I'm having to extend GObject to has something analygous to "properties"... but for "functions/methods". [00:19] <omega> there's overhead to everything [00:19] <wingo> i thought you'd queue and send every so often [00:19] <taaz> wingo: would you prefer to never push it? ;) [00:19] <wingo> er [00:19] <wingo> heh [00:19] <omega> wingo: no, when you're after lowlatency, you never queue. doesn't matter what kind of data [00:20] <taaz> anyone tried lat recently? [00:20] <wingo> but... while you're waiting (for instance) for some midi sink to process, or your softsynth, [00:20] <wingo> you might get an event [00:20] <omega> taaz: it scales with processor speed, I've verified that [00:20] <wingo> which you would send out the next time you are pulled from [00:20] <wingo> no [00:20] <wingo> ? [00:21] <omega> huh? [00:21] <wingo> ok, let's talk about jack. [00:21] <wingo> jack will call jacksrc periodically. [00:21] <omega> um, how? [00:21] <wingo> what does jacksrc do with that data? gstreamer is pull-driven [00:22] <wingo> omega: to a callback [00:22] <wingo> it has some sort of periodic callback tick [00:22] <omega> wingo: um, you can't call a callback from another process [00:22] <taaz> jack has timer stuff in it? [00:22] <wingo> yes [00:22] <taaz> ouch [00:22] <wingo> somehow by calling jack_client_register_callback, [00:23] <taaz> that's going to suck to implement anywhere [00:23] <wingo> i get callbacks [00:23] <taaz> just use glib stuff ;) [00:23] <wingo> i don't know how the lib communicates with the server, [00:23] <omega> wingo: that callback has to come from somewhere *inside* your app [00:23] <wingo> but i think it's pipe related [00:23] <wingo> duh, but [00:23] <wingo> let me draw: jack server -> jack lib <-> app [00:23] <wingo> lib and app within same address space [00:24] <wingo> i don't know how the lib and server communicate [00:24] <omega> ok, and so you're saying that the jack lib will do Magic and randomly fire a callback into the app? [00:24] <wingo> i get a callback somehow, yes. [00:24] <omega> that's broken [00:24] <wingo> that's cool on a sink, that's what will drive the graph [00:24] <wingo> but on a src it's trickier [00:24] <omega> it is impossible to inteface with the majority of programming models [00:25] Action: omega wonders if the people that dream up these things ever try to *implement* anything with them [00:25] <wingo> he might have a poll(1) interface that's being worked up, but i think was running into problems with that [00:25] Action: wingo whispers 'ardour' [00:25] <wingo> :-) [00:25] <omega> pfff [00:26] <wingo> that's why steveb was talking about jack being a bin rather than a src/sink pair [00:26] <omega> brilliant... [00:26] <wingo> that would suck though [00:27] <taaz> i'm going to, uh, research the competition... time to play with Reason ;) [00:27] <omega> heh [00:27] <omega> let me know when you find the source <g> [00:27] <wingo> ;) [00:28] <taaz> blah.. it's a fun toy, but i can't see how professionals can work in such an environment... i need programatic control of stuff [00:28] <wingo> so... would you have to do funny mutex things on a buffer to integrate the callback model into gst? [00:29] <wingo> in a src element [00:29] <omega> yes [00:29] <omega> nasty bufferig [00:29] <omega> and I thought elliminating that was the whole *point* of jack... [00:29] <wingo> buffering? [00:29] <wingo> if it's just one buffer it might be fine [00:29] <omega> yes, you're going to add latency. [00:30] <wingo> because gst won't get back to processing it but every so often [00:30] <omega> theoretically every iteration, one hopes [00:30] <wingo> right [00:30] <wingo> and theoretically one buffer per iteration, ergo (?) no latency overhead? [00:30] <omega> possibly [00:31] <wingo> i think it's doable and once i get beatbox ported to gnome2 (almost there, libglade issues now) i'm going to play with that again [00:32] <omega> it sound like the jack lib is a poor implementation. I would hope that the protocol itself doesn't force this kind of stuff on an app, and that the lib can just be tossed in favor of a native interface [00:33] <wingo> hmm. unlikely. i'm familiar with the engine, and the callback model is central. [00:33] <omega> what is the communication mechanism? [00:33] <wingo> lemme check. [00:33] <omega> afaik there's only one model that is totally callback based, and that's overuse of signals [00:34] ChiefHighwater (pa...@te...) left irc: [00:34] <dobey> sigh sigh [00:35] <taaz> where's the jack code? [00:35] <wingo> i think it's at http://www.op.net/~pdb/jack-0.2.3.tar.gz [00:35] <wingo> although check the lad archives if that doesn't work [00:35] <wingo> that's one of the issues, [00:35] <wingo> jack needs a mailing list, cvs, and a web page [00:35] <wingo> pbd is reluctant to do all that for some reason [00:36] <taaz> soon sf will be okay for such things... with new mailinst list code ;) [00:36] <wingo> we hope ;) [00:36] <taaz> as long as i can download a mbox file of the whole archive i'm set. mutt is better than web browsing it [00:37] <taaz> dude, Reason is so fun to play with [00:37] <taaz> yeah yeah, i'm on a NT box now ;) [00:37] <wingo> mutt rocks! :) [00:37] <wingo> heh [00:38] <taaz> mutt -almost- rocks... tell me how to save tags like 'D' without a full sync please please? [00:38] <wingo> for imap, you mean? [00:38] <taaz> no for mbox's [00:39] <wingo> oh. i don't actually have any mboxes :-) [00:39] <wingo> i dunno, then [00:39] <wingo> omega: writing on an fd [00:39] <wingo> a socket [00:39] <wingo> and sends the data in shm [00:40] <omega> so there is no reason for it to be callback based [00:40] <wingo> right [00:40] <wingo> ideally we can just poll on the fd then [00:40] <wingo> jack needs cvs. [00:42] <taaz> offer to set it up [00:52] <wingo> i just did, before i saw your note ;) [01:29] <wingo> what's the difference between poll(1) and select(1)? [01:31] <tnt> Isn't select suppose to be better than poll? [01:31] <tnt> I don't have my POSIX book(s) handy, so I'm not sure. [01:31] <ds> depends on what you're doing [01:34] chillywilly (da...@d1...) left irc: Read error to chillywilly[d123.as4.nwbl0.wi.voyager.net]: EOF from client [01:34] <wingo> ds: ? [01:36] <ds> poll() is nicer and more modern [01:36] <ds> you don't need to recreate the FD_SET() every time, since [01:36] <ds> poll doesn't write fields that you probably want to reuse [01:37] <wingo> ah. [01:37] <ds> however, poll() requires copying more data to/from kernel [01:38] <ds> select() is also used to be more portable, but that doesn't matter much anymore [01:39] <wingo> thanks, yo [01:43] Nick change: tnt -> tnt-out [02:04] thomasvs (thomas@212.100.172.175) left irc: Ping timeout for thomasvs[212.100.172.175] [02:04] thomasvs (th...@21...) joined #gstreamer. [02:35] BBB-away (BB...@uc...) left irc: Ping timeout for BBB-away[ucu-104-70.ucu.uu.nl] [02:38] BBB-away (BB...@uc...) joined #gstreamer. |
From: IRC B. <wt...@us...> - 2001-11-04 05:26:12
|
******************************************************************* [03:05] Nick change: dobey -> dobey-out [03:17] Nick change: wtay-tv -> wtay-zZz [03:27] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) left irc: Ping timeout for vektor[HSE-Kitchener-ppp194649.sympatico.ca] [03:28] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) joined #gstreamer. [04:35] Nick change: dobey-out -> dobey [04:42] <dobey> later [04:42] dobey (do...@dr...) left #gstreamer (eh). [05:33] dobey (do...@cv...) joined #gstreamer. [06:23] ChiefHighwater (fl...@su...) joined #gstreamer. [06:51] ChiefHighwater (fl...@su...) left irc: Leaving [06:55] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) left irc: Ping timeout for vektor[HSE-Kitchener-ppp194649.sympatico.ca] [06:55] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) joined #gstreamer. [07:06] Nick change: tnt-out -> tnt [07:07] <tnt> dobey: You there? [07:07] <dobey> yes [07:07] <dobey> sadly enough [07:08] <tnt> :-) [07:08] <tnt> You work for Ximian, correct? [07:08] <dobey> why? [07:08] <dobey> if you are bitching about something, no [07:08] <dobey> i am not in the mood to hear it [07:09] <tnt> It not bitching... I just want to see if you had some info... [07:09] <dobey> nor am i going to get off my depressed lonely ass to fix it [07:09] <tnt> I was just wondering if you knew when the GNOME 2 snapshots (on Red Carpet) get updated. [07:10] <tnt> (Nothing to fix...) [07:11] <dobey> no i don't [07:11] <tnt> OK... thanks anyways. [08:21] Nick change: taaz -> taazzzz [08:27] Nick change: wingo -> wingo-z [08:50] harobed (harobed@AC85A024.ipt.aol.com) left irc: Read error to harobed[AC85A024.ipt.aol.com]: Connection reset by peer [09:08] newt joined #gstreamer. [09:08] _gst_newt_ (eg...@ca...) left irc: Ping timeout for _gst_newt_[cable-195-162-215-201.upc.chello.be] [09:08] Nick change: newt -> _gst_newt_ [09:37] <dobey> bbl [09:37] dobey (do...@cv...) left #gstreamer (eh). [10:24] harobed (harobed@AC95A42B.ipt.aol.com) joined #gstreamer. [10:26] Nick change: tnt -> tnt-zZz [10:56] steveb (st...@no...) joined #gstreamer. [11:19] omega (om...@om...) left irc: Ping timeout for omega[omegacs.net] [11:20] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [12:21] Shippou (ir...@ot...) joined #gstreamer. [13:33] bstard (Lor...@a2...) joined #gstreamer. [14:04] walken (fo...@c1...) joined #gstreamer. [14:05] <walken> hi ? someone ? [14:11] <steveb> hi [14:11] <walken> hi [14:11] <walken> have you ever used ffmpeg ? [14:12] <steveb> um, i might have used the gstreamer plugin [14:12] <steveb> actually, no. i don't have it installed [14:13] <walken> ok [14:36] walken (fo...@c1...) left irc: l8r [15:20] harobed (harobed@AC95A42B.ipt.aol.com) left irc: Ping timeout for harobed[AC95A42B.ipt.aol.com] [15:32] harobed (harobed@ACAD5469.ipt.aol.com) joined #gstreamer. [15:43] chillywilly (da...@d2...) joined #gstreamer. [15:51] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [16:04] harobed (harobed@ACAD5469.ipt.aol.com) left irc: Client Exiting [16:29] chtephan (proxy@pD9547B64.dip.t-dialin.net) joined #gstreamer. [16:30] Action: bstard is away: Weg [16:52] Shippou (ir...@ot...) joined #gstreamer. [16:59] BBB (BB...@bu...) joined #gstreamer. [16:59] <BBB> hmm..... [16:59] Action: BBB should learn to close down irc when he's not at home [17:05] BBB (BB...@bu...) left #gstreamer (Client Exiting). [17:09] ajmitch (me...@p2...) left irc: Ping timeout for ajmitch[p27-max4.dun.ihug.co.nz] [17:11] chtephan (proxy@pD9547B64.dip.t-dialin.net) left irc: Client Exiting [17:13] ajmitch (me...@p2...) joined #gstreamer. [17:26] <vektor> Hi! [17:37] thomasvs_ (th...@ad...) joined #gstreamer. [17:39] Nick change: taazzzz -> taaz [17:40] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [17:57] <steveb> taaz: does ximian have the glib2 debs you need? they have the rpms. [18:04] dobey (do...@cv...) joined #gstreamer. [18:05] <taaz> i can wait for the debs [18:43] Shippou (ir...@ot...) joined #gstreamer. [18:52] dobey (do...@cv...) left irc: Ping timeout for dobey[cv652321-a.brkln1.ma.home.com] [18:52] dobey (do...@cv...) joined #gstreamer. [18:55] dobey (do...@cv...) left irc: Ping timeout for dobey[cv652321-a.brkln1.ma.home.com] [18:55] dobey (do...@cv...) joined #gstreamer. [18:56] Action: chillywilly is away: raking the leaves [18:56] chtephan (proxy@pD9E55C02.dip.t-dialin.net) joined #gstreamer. [18:56] chtephan (proxy@pD9E55C02.dip.t-dialin.net) left irc: Client Exiting [18:58] dobey (do...@cv...) left irc: Ping timeout for dobey[cv652321-a.brkln1.ma.home.com] [19:03] dap (dap@SR3745.resnet.ucsb.edu) left irc: Ping timeout for dap[SR3745.resnet.ucsb.edu] [19:15] ChiefHighwater (fl...@su...) joined #gstreamer. [19:16] hadess (ha...@pc...) joined #gstreamer. [19:16] <hadess> heya boys [19:16] <ChiefHighwater> ello [19:16] <hadess> heya CHW [19:17] <ChiefHighwater> sup hadess 8-] [19:17] <hadess> not much just yet [19:17] dobey (do...@cv...) joined #gstreamer. [19:17] knouf (hu...@t-...) joined #gstreamer. [19:18] <dobey> hrmm [19:18] <dobey> that sucked [19:18] <steveb> hadess: how hard is gnome2 porting? [19:18] <dobey> it's not [19:19] <hadess> steveb: quite easy [19:19] <steveb> i mean is it anything more than fixing compile errors? [19:19] <dobey> yes [19:19] <knouf> hi [19:19] <dobey> well [19:19] <hadess> steveb: hmmm, no, but you might want to use new features in the platform [19:19] thomasvs_ (th...@ad...) left irc: Client Exiting [19:19] <steveb> sure [19:20] <hadess> but to do the basic port, it's just fixing compiling errors ;) [19:21] knouf (hu...@t-...) left irc: Client Exiting [19:47] Action: chillywilly is back (gone 00:51:25) [20:10] Nick change: tnt-zZz -> tnt [20:10] <tnt> Good morning. [20:12] <steveb> hi [20:14] <tnt> Hey steveb. [20:50] wingo-z (wi...@rd...) left irc: Read error to wingo-z[rdu25-21-076.nc.rr.com]: EOF from client [20:50] wingo (lmj...@rd...) joined #gstreamer. [20:50] <wingo> what it is [20:51] <steveb> thats what i want to know [20:51] <wingo> heh [20:51] <wingo> steveb: i've been talking with pbd, and it seems with some hacks jacksrc and jacksink will be possible [20:52] <steveb> is my understanding of it correct? [20:52] <wingo> which is nice, because a jackbin would be nasty [20:52] <wingo> how's that [20:52] <wingo> i think it is, but tell me again [20:52] <steveb> a callback which says "populate this buffer with audio" [20:52] <wingo> yes [20:53] <wingo> but the issue with jacksrc and jacksink is that they would be two clients, connected outside jack's knowledge [20:53] <steveb> so how can we stop gstreamer from replacing buffers and changing buffer boundries? [20:53] <wingo> ? [20:53] <wingo> oh [20:54] <wingo> well, there are a number of api hurdles to go through on the jack side first [20:54] <steveb> jacksrc is given a buffer. jacksink will need that exact same buffer to complete the callback [20:54] <wingo> no, acutally [20:54] <wingo> if jacksrc and jacksink are different clients, that's not an issue [20:54] <steveb> ok [20:54] <wingo> as long as src is called before sink you're cool [20:55] <wingo> then there will be data waiting for you when you pull [20:55] <steveb> hmm, ok [20:55] <wingo> does that make sense? [20:55] <steveb> yeah. so what are the hurdles on the jack side? [20:55] <wingo> letting the engine know that the two schould be called in a specifiend order [20:56] <wingo> as they would be connected outside of jack it's somewhat of a hack [20:56] <steveb> so jack has to call jacksrc callback before jacksink one? [20:57] <wingo> yes, i think, [20:57] <wingo> . [20:57] <steveb> have you looked at LADMEA ;) [20:58] <wingo> heh. not really. i should investigate it though ;O what do you think about it? [20:58] <wingo> i trust my low-latency audio more to jack than to gstreamer though. i think the dictator/democracy analogy was apt [20:59] <steveb> i think both will have their place. LADMEA could make a good generic IPC interface for all gstreamer apps to communicate through [21:01] <wingo> that would be neat [21:01] <tnt> What's the link to LADMEA ? [21:01] <wingo> it would be interesting to see how difficult that would be to do [21:01] <wingo> ladspa.org/ladmea, iirc [21:01] <steveb> yep [21:02] <wingo> so much useless crap in my brain :) [21:02] <wingo> hey steveb, [21:02] <wingo> thought about the plugin ui? [21:02] <steveb> ? [21:02] <wingo> i'm playing with that now in my app [21:02] <wingo> plugin ui lib [21:02] <steveb> ah, we need one :) [21:02] <wingo> check out ambient.2y.net/beatbox/ and click on screenshots to see what i was using [21:03] <wingo> we need to control dparms as well as properties [21:03] <wingo> that was before my gnome2, glib2 shift [21:04] <wingo> i would reckon that compact (optionmenu) and full (big list of properties) interfaces would be nice (selectable [21:04] <wingo> ) [21:04] <steveb> its all good [21:04] <steveb> so does that mean we need to wait for gstreamer to go to gtk2 before we get your widgets ;) [21:05] <wingo> well, it could probably do dparams (whenever i get around to doing that) now, well, no, since we don't have a gtk2 shim :-\ [21:05] <wingo> actually, the big problem with guilaunch porting is gtktree [21:06] <wingo> i tried to do some hacks but the tree has changed quite a bit in gtk2 [21:06] <steveb> surely we can get it working with the deprecated gtk_tree then replace it later? [21:07] <wingo> the plugin ui (i.e., controls for just one plugin) don't (shouldn't) depend on gtktree at all [21:07] <steveb> right [21:08] <steveb> i'm putting dparam ui generation stuff into -guilaunch right now. of course it will eventually end up in libgstui [21:10] <steveb> beatbox looks fun [21:10] <wingo> it will be :) but it has gstreamer issues atm, no sound :-\ [21:11] <wingo> i think i'll steal your dparam stuff, muahaha :) [21:11] <steveb> no sound? [21:14] <wingo> there are dynamic pipeline problems that i have to sort out. by no sound i mean it doesn't work atm ;) [21:16] <steveb> i figured the first sucker to have an app that interactively changes the pipeline would run into problems [21:16] <steveb> s/suck/hack/ [21:16] <wingo> lol [21:16] <wingo> yeah [21:16] <wingo> there are 'issues' ;) [21:17] <wingo> but luckily the gst part is separated from the ui so i can test it even though i haven't completely grokked libglade-2 [21:20] <hadess> libglade-2 is the same as libglade-1, just one change iirc [21:22] <wingo> gnomeapp and gnomedock changed tho [21:22] <wingo> i use gnomedocks and the went to bonobodocks for some unknown reason (do you know why, hadess?) [21:22] <wingo> s/the/they/ [21:23] <wingo> so i am going to break my main window glade file into one per dock, and insert them manually [21:23] <wingo> because gnomeapp has no internal child handler [21:23] <hadess> %s/GnomeDock/BonoboDock/ and add libbonoboui to the list of dependencies [21:23] <wingo> really? i'll try that this evening [21:24] <wingo> and maybe send in a patch for libglade-convert if i get it to work [21:25] <wingo> that's a neat prog btw, combined with glade1 and BUILT_SOURCES i can still use glade1 to edit the ui :-) [21:25] <hadess> yeah, pretty neat [21:25] <wingo> how's gnome-utils porting coming along? [21:26] <hadess> pretty fine, waiting for some fixes in libgnomeui and applets support [21:27] <dobey> heh [21:29] <wingo> it'll be neat, when gnome2 comes out, and gst is maybe up to 0.3, beatbox will be the only gnome mixer application for gst, and that can handle jack streams [21:29] <wingo> i'm pleased ;) [21:29] <hadess> <g> [21:30] <hadess> and i hope there will be a working rhythmbox that would replace xmms in everybody's hearts^Wmachines ;) [21:30] <wingo> i'll take it :) [21:30] <dobey> i do to [21:34] <wingo> it's funny we have (and have had) similar names for the projects, yet they're not connected except by gnome and gst [21:34] <wingo> i'll link to you from my page ;) [21:34] <hadess> soundbox, beatbox, rhythmbox ;) [21:35] <wingo> <insert cheezy joke about thinking outside the box here> [21:35] <dobey> windowbox would be a good name for a WM [21:35] <hadess> window maker is an even better one ;) [21:36] <dobey> why? [21:36] <dobey> *it* doesn't make windows [21:36] <hadess> it's not a box either *slap* [21:36] <dobey> *ding ding* [21:37] <dobey> sigh [22:06] Nick change: dobey -> dobeyshwr [22:10] Zeenix (ze...@13...) joined #gstreamer. [22:16] Action: bstard is back (gone 05:46:39) [22:17] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-9.brain.net.pk] [22:27] Nick change: taaz -> taaz-away [22:34] Nick change: dobeyshwr -> dobey [22:52] dobey (do...@cv...) got netsplit. [22:52] ajmitch (me...@p2...) got netsplit. [22:52] taaz-away (dlehn@66.37.66.32) got netsplit. [22:52] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) got netsplit. [22:52] steveb (st...@no...) got netsplit. [22:52] wingo (lmj...@rd...) got netsplit. [22:52] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) returned to #gstreamer. [22:52] steveb (st...@no...) returned to #gstreamer. [22:52] wingo (lmj...@rd...) returned to #gstreamer. [22:54] taaz-away (dlehn@66.37.66.32) returned to #gstreamer. [22:54] ajmitch (me...@p2...) returned to #gstreamer. [22:54] dobey (do...@cv...) returned to #gstreamer. [23:00] Zeenix (ze...@13...) joined #gstreamer. [23:03] harobed (harobed@AC869DD3.ipt.aol.com) joined #gstreamer. [23:09] Shippou (ir...@ot...) left irc: Write error to Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca], closing link [23:10] <dobey> sigh [23:10] dobey (do...@cv...) left #gstreamer (eh). [23:11] <Zeenix> wingo: online? [23:29] <wingo> yeah [23:31] <Zeenix> wingo: "yeah" means 4 bytes, looks like my ISP has slowed down to 4B/s :) [23:32] <Zeenix> s/s/Hour [23:32] dap (dap@SR3745.resnet.ucsb.edu) joined #gstreamer. [23:34] <wingo> heh [23:36] <Zeenix> wingo: seen wtay around? [23:37] tnt (tn...@h2...) left irc: Client Exiting [23:43] chillywilly (da...@d2...) left irc: Ping timeout for chillywilly[d25.as21.nwbl0.wi.voyager.net] [23:43] tnt (tn...@h2...) joined #gstreamer. [23:44] chillywilly (da...@d9...) joined #gstreamer. [23:44] <wingo> Zeenix: no, not in a while [23:45] <wingo> sorry for the sporadic question-answering, i'm slightly afk atm :) [23:47] <Zeenix> wingo: i have made the latest cvs gst to work on my sys today, but i getting some errors, like 'osssink recieved eos' & stuff [23:47] Nick change: tnt -> tnt-out [23:55] dap (dap@SR3745.resnet.ucsb.edu) left irc: Ping timeout for dap[SR3745.resnet.ucsb.edu] [23:56] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-57.brain.net.pk] [00:00] --- Sun Nov 4 2001 [00:09] Shippou (ir...@ot...) joined #gstreamer. [00:11] dap (dap@SR3745.resnet.ucsb.edu) joined #gstreamer. [00:11] Nick change: hadess -> hds-f00d [00:19] wingo (lmj...@rd...) left irc: beer :) [00:27] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [00:41] steveb (st...@no...) left irc: Read error to steveb[node1ee41.a2000.nl]: EOF from client [00:42] chillywilly (da...@d9...) left irc: Read error to chillywilly[d94.as6.nwbl0.wi.voyager.net]: EOF from client [00:43] chillywilly (da...@d9...) joined #gstreamer. [00:44] chillywilly (da...@d9...) left irc: [00:44] chillywilly (da...@d9...) joined #gstreamer. [00:47] chillywilly (da...@d9...) left irc: Read error to chillywilly[d94.as6.nwbl0.wi.voyager.net]: EOF from client [00:47] chillywilly (da...@d9...) joined #gstreamer. [00:50] chillywilly (da...@d9...) left irc: Read error to chillywilly[d94.as6.nwbl0.wi.voyager.net]: EOF from client [00:50] chillywilly (da...@d9...) joined #gstreamer. [00:51] chillywilly (da...@d9...) left irc: [00:51] chillywilly (da...@d9...) joined #gstreamer. [00:59] ajmitch (me...@p2...) left irc: Ping timeout for ajmitch[p27-max4.dun.ihug.co.nz] [01:14] harobed (harobed@AC869DD3.ipt.aol.com) left irc: Client Exiting [01:17] Nick change: hds-f00d -> hadess [01:19] ajmitch (me...@p2...) joined #gstreamer. [01:30] <ds> ppc is stupid [01:30] <chillywilly> heh [01:30] <ds> it takes 13 instructions to convert from signed short to double [01:30] <ds> because it doesn't have a native instruction [01:31] <chillywilly> it is RISC though [01:32] <hadess> ds: show you clever ppc asm foo ! [01:33] <ds> er? [01:33] <hadess> ds: what kinda stuff you're doing on ppc ? making mpeg2dec faster ? ;) [01:33] <ds> I'm working on audio resampling. [01:34] <hadess> ok [01:34] <ds> the short to double conversion is the most time consuming part right now [01:34] <chillywilly> DSP stuff? [01:35] <ds> yeah [01:35] Action: chillywilly should've taken DSP as an elective...would've been fun [01:35] <hadess> ds: where would that code go ? [01:36] <ds> gstaudioscale [01:36] <ds> and speed [01:36] <ds> which do the same thing, but accidentally [01:37] <hadess> cool [02:00] <ds> i386 is 10 instructions [02:00] <ds> I never realized it was so expensive [02:05] Action: chillywilly is away: dinne [02:05] Action: chillywilly is away: dinner [02:36] Action: chillywilly is back (gone 00:31:17) [02:37] chillywilly (da...@d9...) left irc: Read error to chillywilly[d94.as6.nwbl0.wi.voyager.net]: EOF from client |
From: IRC B. <wt...@us...> - 2001-11-05 05:28:06
|
******************************************************************* [03:12] Nick change: taaz-away -> taaz [03:39] hadess (ha...@pc...) left irc: mooooh! [03:47] ChiefHighwater (fl...@su...) left irc: Leaving [04:03] Nick change: tnt-out -> tnt [04:15] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) left irc: Ping timeout for vektor[HSE-Kitchener-ppp194649.sympatico.ca] [04:15] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) joined #gstreamer. [04:51] tnt (tn...@h2...) left irc: Read error to tnt[h24-83-72-250.vs.shawcable.net]: EOF from client [04:52] tnt (tn...@h2...) joined #gstreamer. [04:57] Nick change: taaz -> taaz-away [05:00] bstard (Lor...@a2...) joined #gstreamer. [06:07] rks (rk...@ho...) joined #gstreamer. [07:11] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [07:20] tnt (tn...@h2...) left irc: Client Exiting [07:47] rks (rk...@ho...) left irc: Client Exiting [08:11] Shippou (ir...@cr...) joined #gstreamer. [08:16] chillywilly (da...@d1...) joined #gstreamer. [08:17] chillywilly (da...@d1...) left irc: Killed (NickServ (This nick is reserved by another user)) [08:18] chillywilly_ (da...@d1...) joined #gstreamer. [08:19] Nick change: chillywilly_ -> chillywilly [08:42] omega (om...@om...) joined #gstreamer. [08:49] <ds> yo [08:49] <omega> yo [09:27] Shippou (ir...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [09:29] <ds> yay! two more milliseconds [09:32] <ds> yay! two more [09:38] chillywilly (da...@d1...) left irc: Read error to chillywilly[d136.as21.nwbl0.wi.voyager.net]: EOF from client [10:07] omega (om...@om...) left irc: sleep [10:15] steveb (st...@no...) joined #gstreamer. [10:27] Shippou (ir...@ot...) joined #gstreamer. [10:41] walken (fo...@c1...) joined #gstreamer. [10:41] <walken> boink [11:29] Nick change: taaz-away -> taaz [11:29] <taaz> yo walken [11:50] <taaz> nap time [11:50] Nick change: taaz -> taazzzz [11:50] <walken> hey [11:50] <taazzzz> oh [11:50] <taazzzz> now you wake up ;) [11:50] <walken> yeah, you woke me up :) [11:50] Nick change: taazzzz -> taaz [11:55] <taaz> hmm.. yeah i think i will go to sleep. g'night [11:55] Nick change: taaz -> taazzzz [13:13] ajmitch (me...@p2...) left irc: Ping timeout for ajmitch[p27-max4.dun.ihug.co.nz] [13:16] harobed (harobed@ACA193A4.ipt.aol.com) joined #gstreamer. [13:23] ajmitch (me...@p2...) joined #gstreamer. [14:04] walken (fo...@c1...) left irc: l8r [14:57] BBB-away (BB...@uc...) got netsplit. [14:57] Shippou (ir...@ot...) got netsplit. [14:57] BBB-away (BB...@uc...) returned to #gstreamer. [14:57] Shippou (ir...@ot...) returned to #gstreamer. [14:57] Topic changed on #gstreamer by ChanServ!s@ChanServ: BeNOW.ca clip: http://gstreamer.net/gstreamer-benow.mp3 [14:57] BBB-away (BB...@uc...) got netsplit. [14:57] Shippou (ir...@ot...) got netsplit. [14:58] BBB-away (BB...@uc...) returned to #gstreamer. [14:58] Shippou (ir...@ot...) returned to #gstreamer. [16:22] ]NEO[ (ne...@fl...) joined #gstreamer. [16:34] dobey (do...@cv...) joined #gstreamer. [17:00] wingo (lmj...@rd...) joined #gstreamer. [17:00] <dobey> hi [17:00] <wingo> hey [17:00] chtephan (proxy@pD9547A52.dip.t-dialin.net) joined #gstreamer. [17:01] <dobey> now to get my gamepad to behave [17:02] <wingo> whip it, into shape, ... [17:02] <dobey> uh [17:03] <wingo> shape it up ... (from the song 'whip it' ;) [17:03] <dobey> i know who devo is :-) [17:03] <wingo> :-) [17:03] <dobey> whee [17:04] <dobey> time to copy the snes9x freeze files over from my windows machine... [17:15] ]NEO[ (ne...@fl...) left irc: Ping timeout for ]NEO[[flat-p08-m090.aruba.it] [17:16] chtephan (proxy@pD9547A52.dip.t-dialin.net) left irc: Client Exiting [18:27] thomasvs (th...@21...) left irc: [x]chat [18:41] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [18:54] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [18:54] <vektor> hey [18:54] <vektor> dude [18:54] <vektor> oh [18:54] <dobey> whee [18:54] <dobey> yo [18:55] <vektor> yo [18:55] <dobey> what's up? [18:56] <vektor> i'm doing a web page demonstrating different deinterlacing algorithms. [18:56] <dobey> oh [18:56] <dobey> fun [19:08] <wingo> hey vekkie [19:08] <vektor> hi [19:09] Uraeus (csc...@c1...) joined #gstreamer. [19:09] <Uraeus> hihi [19:10] <dobey> hi Uraeus [19:11] <Uraeus> hi dobey, Jeremy (in #linuxpower) just promised to put up the interview tommorow morning [19:11] <dobey> sweet [19:16] <dobey> bbiab [19:16] Nick change: dobey -> dobeyshwr [19:19] <wingo> hi Uraeus. what interview is that? [19:19] <Uraeus> wingo: an interview with dobey [19:20] <wingo> really. that's pretty cool [19:42] Shippou (ir...@ot...) joined #gstreamer. [19:50] Nick change: dobeyshwr -> dobey [20:26] tnt (tn...@h2...) joined #gstreamer. [20:27] <dobey> later [20:27] <dobey> going to work [20:27] dobey (do...@cv...) left #gstreamer (eh). [20:28] <wingo> hi tnt [20:29] <tnt> Hey wingo. [20:43] Nick change: taazzzz -> taaz [20:44] <Uraeus> hi taaz [20:44] <taaz> hey [20:54] Zeenix (ze...@13...) joined #gstreamer. [20:54] <Zeenix> hello [20:54] <tnt> hey. [21:02] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-24.brain.net.pk] [21:07] <vektor> Deinterlacing results now up at http://www.dumbterm.net/graphics/dvd/ [21:07] <vektor> I show the images before/after using some algorithms. [21:07] <wingo> cool, i'll check it out [21:07] <vektor> Thanks! [21:07] <vektor> Let me know if you see any bugs/typos. [21:07] <vektor> The new stuff is the bjork results. [21:07] <wingo> okeydoke [21:12] <wingo> i'll have to look when i get back to my desktop, this laptop don't cut it ;) that page is hardcore though [21:24] dobey (do...@dr...) joined #gstreamer. [21:27] Zeenix (ze...@13...) joined #gstreamer. [21:46] bstard (Lor...@a2...) joined #gstreamer. [21:48] sxpert (cha...@r2...) joined #gstreamer. [21:49] Action: Uraeus tries to wake wtay-zZz [21:49] Nick change: wtay-zZz -> wtay [21:49] <wtay> yo [21:50] <Uraeus> impressive what I am able to do these days over IRC :) [21:50] <dobey> whee [21:50] <Zeenix> yo [21:50] <sxpert> dobey: you here too ? [21:51] <dobey> uhm, i'm always here [21:51] <Uraeus> wtay: what are you doing? [21:51] <wtay> Uraeus: I'm writing an XML editor :) [21:51] <Zeenix> wtay: where were you ? [21:52] <wtay> I was here. coding.. [21:53] <Uraeus> wtay: http://idx-getox.idealx.org/ -> XML Editor [21:53] <wtay> Uraeus: xmlspy.com, that's an XML editor :) [21:53] <Zeenix> wtay: hidding from my questions? :) [21:53] <wtay> Zeenix: as always :) [21:54] <dobey> *bszsth* [21:54] Action: sxpert is reading gstreamer's doc and is shitting in his pants as is seems way too easy [21:54] <Zeenix> wtay: a lot of questions for you in the stack [21:55] <wingo> greets, wtay [21:55] <wtay> yo [21:56] <Zeenix> wtay: first of all whats the prob. with the new cvs gst, the disksrc seem to behave stupid [21:57] <Zeenix> wtay: some times it says eos without encountering one [21:57] <Uraeus> wtay: getox can be patched/improved, xmlspy can't :) [21:58] Uraeus (csc...@c1...) left #gstreamer (Client Exiting). [22:00] <sxpert> I come back [22:00] <dobey> hrmm [22:00] sxpert (cha...@r2...) left irc: Read error to sxpert[r205m60.cybercable.tm.fr]: EOF from client [22:00] sxpert_ (sx...@r2...) joined #gstreamer. [22:03] thomasvs (th...@ad...) joined #gstreamer. [22:03] <thomasvs> hi [22:03] Nick change: sxpert_ -> sxpert [22:03] <wtay> Zeenix: no idea [22:03] <thomasvs> anyone around here know how to remove a repository from your cvs using the cvs command ? [22:03] <wtay> Zeenix: use filesrc [22:03] <sxpert> damn, an x-chat instance can't handle more than one server aat a time :( [22:04] <Zeenix> wtay: ok [22:04] <wtay> sxpert: "new server tab" [22:04] Action: sxpert dumb... [22:04] sxpert (sx...@r2...) left #gstreamer (Client Exiting). [22:05] sxpert_ (sx...@r2...) joined #gstreamer. [22:05] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [22:05] <Zeenix> wtay: i think, its not the librtp that is not sending rtcp packets but i think we are doing something wrong with the socket or address [22:05] <sxpert_> back (for good this time) [22:07] <Zeenix> wtay: the select should detect a change on the FD if any data arrives on the socket from anywhere, i tried that, the select doesnt seem to detect any changes [22:08] <wtay> Zeenix: and the RTCP packet is sent? [22:08] <Zeenix> wtay: i think it is [22:09] <sxpert_> the documentation is very nice. but, which example would show how to embed an mpeg1 video player ? [22:09] <Zeenix> wtay: i was sending an integer too by sendto(...) after sending that RTCP packet, but the socket FD in RtpRecv doesnt get it [22:10] <wtay> sxpert_: there are some examples apps that show ho to do that in test/ [22:10] <sxpert_> also gsteditor does NOT work as intended (I get insults about signals) [22:11] Nick change: sxpert_ -> sxpert [22:16] <wtay> sxpert: at startup? thats normal.. [22:16] <sxpert> ah, and then there is not much to do (no gstreamer core objects and so on... [22:17] <wtay> sxpert: click on a thread... [22:18] <sxpert> wtay: on a "bin" object ? [22:18] <wtay> sxpert: a thread object, bin won't work [22:19] <sxpert> wtay: ok, I have a thread. [22:20] <sxpert> then ? [22:20] <wtay> click on the thread to select an element [22:27] steveb (st...@no...) left irc: advance! [22:27] <dobey> hmm [22:28] <sxpert> wtay: what are the pink pads (can't get an elastic thingie from there) ? [22:28] <wtay> sxpert: ? [22:29] <wtay> oh [22:29] <sxpert> for instance on "mpeg1parse" [22:29] <wtay> those are padtemplates, create at runtime [22:29] <wtay> s/create/created [22:29] <wtay> it's not implemented yet :) [22:30] <sxpert> it seems the tool is under "heavy" development ;) [22:30] <wtay> sxpert: heavy bitrot is a better word actually :) [22:30] <Zeenix> sxpert: yes it is [22:31] <sxpert> :( [22:31] <Zeenix> sxpert: i came here to use & ended up writting plugins myself :) [22:31] Action: dobey still needs to "use" [22:31] <Zeenix> sxpert: & now i forgot why i came to gstreamer :) [22:32] <wingo> it's fun though, join the loonie bin [22:33] <wingo> gotta run [22:33] Nick change: wingo -> wingo-out [22:33] <Zeenix> wtay: not getting my queries in the private window? [22:34] <sxpert> all I need atm is to embed a small widget-like mpeg1 player in a gnome app. [22:34] <wtay> uhm no [22:34] <dobey> sxpert: smpeg? [22:35] <wtay> sxpert: or gstplay.. [22:35] <dobey> bonobo-media maybe? [22:35] <sxpert> nah, I checked it out and it does not compile for me :.libs/libsmpeg.so: undefined reference to `operator new[](unsigned)' [22:35] <sxpert> .libs/libsmpeg.so: undefined reference to `operator delete(void*)' [22:36] <sxpert> .libs/libsmpeg.so: undefined reference to `operator delete[](void*)' [22:36] <sxpert> .libs/libsmpeg.so: undefined reference to `operator new(unsigned)' [22:36] <wtay> ah yes.. bonobo-media is nice.. [22:36] <dobey> sxpert: you have all the proper SDL things? [22:36] <sxpert> I compiled and installed the latest and greatest... [22:37] <dobey> ah [22:37] <dobey> that might be the problem then [22:37] <dobey> smpeg 0.4.4 and SDL 1.2.1 work well together [22:38] <sxpert> also, libsmpeg compiles and installs properly (using -k) but the failure is when linking the plaympeg sample [22:38] <sxpert> also, I would like to avoit C++ like anthrax ;) [22:38] <dobey> hrmm [22:38] <dobey> ah well... [22:38] <sxpert> s/avoit/avoid/ [22:38] <dobey> use bonobo-media :-) [22:39] <sxpert> bonobo-media (that uses gstreamer, right?) ? [22:39] <dobey> yep [22:39] <sxpert> ah! [22:40] <wtay> well... [22:40] <wtay> once I get video to cooperate it does.. [22:40] <dobey> yeah, there's always the "well..." part [22:40] <sxpert> trick is, I will need to use gstreamer (console mode) for the rendering engine anyways,,, [22:40] <sxpert> wtay: fast, I need it ;) [22:41] <dobey> sxpert: what are you doing exactly? [22:41] <sxpert> dobey: like I said, an Adobe Premiere ripoff... [22:41] <dobey> oh, right [22:42] Action: sxpert needs all the help he can get with multimedia... ;) [22:47] Zeenix2 (ze...@13...) joined #gstreamer. [22:47] Action: sxpert is fed-up to use premiere on the crappy win2k machine over the network (using vncviewer ;) ) [22:48] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-36.brain.net.pk] [22:48] Nick change: Zeenix2 -> Zeenix [22:50] thomasvs (th...@ad...) left irc: Read error to thomasvs[adsl-64524.turboline.skynet.be]: EOF from client [22:53] <Zeenix> wtay: i was also trying to make rtp plugins capable of new media types like mp3, mpeg video etc [22:53] <sxpert> Zeenix: realaudio/video, mms:// ? [22:53] <wtay> cool [22:54] <Zeenix> sxpert: Real Time Transport Protocal [22:54] <dobey> i wish gnome-vfs could do rtsp: [22:54] <dobey> or something [22:54] <sxpert> implemented as a bonobo-object ;) [22:54] <dobey> heh [22:54] <dobey> well [22:55] <dobey> if bonobo-media can handle rtsp: i don't care about gnome-vfs then :-) [22:55] <Zeenix> wtay: but mp3 plays fine in the start but after a 100 or 200 packets, i start to get these: [22:55] <Zeenix> mpg123: Can't rewind stream by n bits! [22:55] <Zeenix> where n is some integer [22:55] <wtay> Zeenix: yeah, that's packet loss [22:56] <Zeenix> & a big crash after a few of these [22:56] <wtay> that's because the sender doesn't take care of waiting [22:56] <wtay> it just pumps out data and the UDP packets get dropped after a while [22:57] <Zeenix> wtay: queue even doesnt help much in this [22:57] <wtay> Zeenix: nope [22:58] <Zeenix> wtay: why only with mp3, neither oss nor gsm [22:59] <wtay> Zeenix: you'll hear dropouts too with those ones [22:59] <wtay> Zeenix: but they don't care about lost packages that much [22:59] <Zeenix> wtay: mpg123 do care [22:59] <Zeenix> ? [22:59] <wtay> yes [23:00] <Zeenix> wtay: so what should we do? [23:00] <wtay> Zeenix: rtpsend should wait for the timestamp before sending out the packet [23:01] <Zeenix> wtay: timestamp? [23:01] <wtay> Zeenix: each buffer has a timestamp [23:02] <Zeenix> wtay: what do you mean by 'wait for the timestamp' ? [23:03] <wtay> Zeenix: when rtpsend recieves a buffer it should only send the buffer after waiting for the timestamp on that buffer [23:04] <Zeenix> wtay: when does it gets the timestamp & how? [23:04] <wtay> Zeenix: GST_BUFFER_TIMESTAMP gets the timestamp of the buffer [23:05] Action: BBB-away is away: I'm busy [23:05] Action: BBB-away is back (gone 00:00:02) [23:05] Nick change: BBB-away -> BBB [23:05] <Zeenix> wtay: while( GST_BUFFER_TIMESTAMP( buffer ) == NULL ) ; ? [23:07] <wtay> Zeenix: gst_clock_wait (GST_BUFFER_TIMESTAMP (buf)).. [23:07] <wtay> look at xvideosink [23:08] <Zeenix> but we should only be doing it in case of mp3, right ? [23:08] <wtay> Zeenix: no, always [23:09] <Zeenix> & rtpsend should be connected to filesrc having an mp3 file or with mp3parse ? [23:10] <wtay> Zeenix: just filesrc should be enough [23:10] <vektor> wtay: you there? [23:10] <vektor> wtay: check out the deinterlacing algorithm comparison at the bottom of http://www.dumbterm.net/graphics/dvd/ [23:11] <wtay> vektor: I did [23:11] <vektor> oh [23:11] <Zeenix> wtay: & mpg123 with rtprecv ? [23:11] <vektor> and any thoughts? [23:11] <wtay> Zeenix: mp3parse ! mpg123 or use mad [23:11] <wtay> vektor: ffmpeg does fairly well [23:11] <vektor> surprisingly well. [23:11] <vektor> the kernel is so simple too. [23:11] <vektor> filter kernel that is. [23:12] <wtay> nice work though [23:12] <Zeenix> wtay: is the event system on again? [23:13] <wtay> Zeenix: we haven't sorted it out completely [23:13] <Zeenix> wtay: yes but i see 'osssink got eos' like errors in gstmediaplay [23:13] <wtay> Zeenix: yeah.. it somewhat works.. [23:15] <Zeenix> wtay: we are now getting every integer translated to host byte order [23:16] <Zeenix> wtay: but not the buffers themselves ofcourse [23:17] <Zeenix> wtay how to tackle with the buffers ? [23:18] <wtay> what do you mean? [23:18] <Zeenix> wtay: how to translate them network byte order [23:18] <Zeenix> before sending [23:19] <wtay> why do you need that? [23:19] <Zeenix> oh, s/host/network [23:20] <Zeenix> dont they have integers? [23:20] <wtay> ? [23:21] <Zeenix> wtay: arent there integers in the buffers, & dont we need to translate those ints before sending them with rtp [23:21] <Zeenix> s/with rtp/in rtp packets [23:22] <wtay> Zeenix: not really.. [23:23] <Zeenix> wtay: you have a network at your home ? [23:23] <wtay> Zeenix: yes [23:24] <Zeenix> wtay: have you tried the rtp accross that [23:24] <Zeenix> ? [23:24] <wtay> I tried UDP once [23:24] <Zeenix> wtay: yes the only network they worked on :) [23:26] <wtay> it was nice though.. qcamsrc ! mpeg2enc ! udpsink .... udpsrc ! mpeg2dec ! xvideosink [23:26] <Zeenix> wtay: i'll make it good for mp3 tommorow, will you try that accross it ? [23:27] <Zeenix> wtay: i know, it must had been much fun [23:27] <wtay> I have to recompile gstreamer on the other machine then... [23:27] <Zeenix> wtay: maYYam's sys ? [23:28] <wtay> yes.. [23:28] <Zeenix> wtay: i'll talk to her so she could allow you to ... :) [23:30] <Zeenix> wtay: plz update the rate in oss/gsm/etc to start from 1000 instead of 8000 [23:30] <wtay> ok [23:31] <Zeenix> wtay: i think your name should appear in the author's list of RtpRecv before me [23:31] <dobey> later [23:31] dobey (do...@dr...) left #gstreamer (eh). [23:31] <Zeenix> wtay: i really do think that [23:32] <wtay> no way :) [23:33] <Zeenix> wtay: really, since i couldnt contact you these three days, i havent done anything at all [23:34] Action: Zeenix slaps himself really hard [23:37] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-20.brain.net.pk] [23:47] chillywilly (da...@d1...) joined #gstreamer. [23:48] omega (om...@om...) joined #gstreamer. [23:49] <wtay> yo [23:49] <omega> yo [23:49] <vektor> oh [23:50] <vektor> meg [23:50] <vektor> ah [23:50] <vektor> ohmeghah: check www.dumbterm.net/graphics/dvd/ at the bottom for some bjork action [23:50] <vektor> it's bjork like you've never seen her before: deinterlaced! [00:00] --- Mon Nov 5 2001 [00:00] <omega> hrm, some of the shots look really fuzzy. I assume it looks better in motion... [00:07] gtb (syln@Mix-Lyon-106-2-29.abo.wanadoo.fr) joined #gstreamer. [00:08] gtb (syln@Mix-Lyon-106-2-29.abo.wanadoo.fr) left irc: [00:12] Action: BBB is away: zzz [00:12] Nick change: BBB -> BBB-zZz [00:14] <wtay> going to sleep.. [00:14] <wtay> cya [00:14] Nick change: wtay -> wtay-zZz [00:19] bstard (Lor...@a2...) joined #gstreamer. [00:23] dobey (do...@cv...) joined #gstreamer. [00:46] dobey (do...@cv...) left irc: Ping timeout for dobey[cv652321-a.brkln1.ma.home.com] [00:46] dobey-ppc (do...@cv...) joined #gstreamer. [00:46] BBB-zZz (BB...@uc...) left irc: Ping timeout for BBB-zZz[ucu-104-70.ucu.uu.nl] [00:46] BBB-zZz (BB...@uc...) joined #gstreamer. [00:46] Nick change: dobey-ppc -> dobey [00:47] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [00:47] Shippou (ir...@ot...) left irc: Ping timeout for Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca] [01:13] ajmitch__ (me...@p1...) joined #gstreamer. [01:14] ajmitch (me...@p2...) left irc: Ping timeout for ajmitch[p27-max4.dun.ihug.co.nz] [01:14] Nick change: ajmitch__ -> ajmitch [01:16] matju (matju@HSE-Montreal-ppp343437.sympatico.ca) joined #gstreamer. [01:17] Action: matju releases video4jmax 0.3.0 [01:17] <matju> http://artengine.ca/jmax/video4jmax/latest/doc/ [01:18] <dobey> jmax? [01:18] <matju> http://hostname.2y.net/~matju/game-of-life-circuit.png [01:18] <matju> dobey: http://www.ircam.fr/equipes/temps-reel/jmax/en/index.php3 [01:19] omega (om...@om...) left irc: more fixing of computers, yay [01:19] <dobey> oh, so jmax is like BEAST kind of? [01:19] <matju> dobey: what is BEAST ? [01:20] <dobey> a buzz-like audio editor for gnome [01:22] harobed (harobed@ACA193A4.ipt.aol.com) left irc: Client Exiting [01:22] Action: chillywilly is away: dinner [01:23] <matju> dobey: er, i would think so. [01:28] dobey (do...@cv...) left irc: Ping timeout for dobey[cv652321-a.brkln1.ma.home.com] [01:28] dobey (do...@cv...) joined #gstreamer. [01:28] <matju> <matju> dobey: er, i would think so. [01:28] <dobey> yeah [01:28] <dobey> i saw that [01:28] <dobey> :-) [01:40] <tnt> dobey: What's this neon library you need for encompass? [01:41] <dobey> tnt: it's an http transfer library [01:41] <dobey> tnt: http://www.webdav.org/neon/ [01:41] <tnt> dobey: Thanks. [01:48] Shippou (ir...@ot...) joined #gstreamer. [01:50] Action: chillywilly is back (gone 00:28:10) [02:42] matju (matju@HSE-Montreal-ppp343437.sympatico.ca) left #gstreamer. [02:42] <dobey> hrmm |
From: IRC B. <wt...@us...> - 2001-11-07 05:28:02
|
******************************************************************* [03:05] gulliver2000 (gul...@bi...) joined #gstreamer. [03:07] <ds> ok, my bad. Glibc now has LinuxThreads, not UofF [03:07] <ds> I've used POSIX.4, but not in Linux [03:07] <tnt> I think all of Linux's pthread stuff is implemented over top of LinuxThreads. [03:08] gulliver2000 (gul...@bi...) left #gstreamer (Client Exiting). [03:08] <tnt> What did you use POSIX.4 on? [03:08] <tnt> (Are there any free OSes that implement POSIX.4?) [03:09] <ds> RTAI, mainly [03:47] Shippou (ir...@ot...) left irc: Client Exiting [03:48] Shippou (zbl...@cr...) joined #gstreamer. [03:52] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [03:58] Shippou (zbl...@cr...) joined #gstreamer. [04:22] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [05:08] Shippou (zbl...@cr...) left irc: Terminated with extreme prejudice - dircproxy 1.0.0 [05:08] Shippou (zbl...@cr...) joined #gstreamer. [05:12] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [05:12] Shippou (zbl...@cr...) joined #gstreamer. [05:29] Action: dap is away: I'm busy [05:29] Action: dap is back (gone 00:00:02) [05:48] chillywilly (da...@d6...) left irc: Read error to chillywilly[d60.as15.nwbl0.wi.voyager.net]: Connection reset by peer [05:52] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [05:52] Shippou (zbl...@cr...) joined #gstreamer. [06:09] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [06:09] Shippou (zbl...@cr...) joined #gstreamer. [07:03] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [07:03] Shippou (zbl...@cr...) joined #gstreamer. [07:24] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: Connection reset by peer [07:24] Shippou (zbl...@cr...) joined #gstreamer. [07:25] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [07:25] Shippou (zbl...@cr...) joined #gstreamer. [08:03] steveb (st...@no...) joined #gstreamer. [08:03] <tnt> Hey steveb. [08:04] <steveb> hi [08:05] <ajmitch> hi steveb [08:28] Nick change: taaz -> taazzzz [08:42] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [08:43] Shippou (zbl...@cr...) joined #gstreamer. [08:47] walken (fo...@c1...) joined #gstreamer. [08:47] walken (fo...@c1...) left irc: l8r [09:06] walken (fo...@c1...) joined #gstreamer. [09:37] mattias (ma...@ga...) joined #gstreamer. [09:39] <mattias> Hi! I've got a problem building gstreamer.. any one awake to give a helping hand? :) [09:39] <ajmitch> i suppose... :) [09:39] <tnt> (I am too.) [09:39] <mattias> mkay I've fallen onto a syntax bug in the configure script that builds gstreamer [09:40] <ajmitch> oh? [09:40] <mattias> it has to do with pkg-config I think.. [09:40] <ajmitch> what is it? [09:40] <tnt> What does it say? [09:40] <mattias> ./configure: line 7922: syntax error near unexpected token `PKG_CHECK_MODULES(GLIB2,' [09:40] <mattias> ./configure: line 7922: ` PKG_CHECK_MODULES(GLIB2, glib-2.0 gobject-2.0 gthread-2.0 gmodule-2.0, HAVE_GLIB2=yes,HAVE_GLIB2=no)' [09:40] <ajmitch> heh [09:40] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: Connection reset by peer [09:40] Shippou (zbl...@cr...) joined #gstreamer. [09:40] <mattias> I can't find where it is defined.. [09:40] <ajmitch> first, check that you have the packages needed as listed in REQUIREMENTS [09:41] <mattias> this might have to do with the fact that I'm running a Debian system, which is lagging behind Redhat etc in the GNOME-department [09:41] <ajmitch> debian sid? [09:41] <mattias> yup unstable [09:41] <tnt> No... I had this problem too. [09:41] <tnt> You just need to upgrade a whole bunch of stuff. [09:41] <ajmitch> well, mine's a couple of weeks behind on updates, and i can build all of the gnome libs, so it's not too bad :) [09:41] <mattias> hehe ok [09:41] <tnt> Check the REQUIMENTS file, like aj said. [09:42] <mattias> yep will do that again! thnx :) [09:42] <ajmitch> dpkg -l pkg-config [09:42] <tnt> Like pkgconfig needs to be 0.8.0... libtool needs ti be ... ummm... check the REQUIREMENTS file. [09:42] <ajmitch> libtool has to be 1.4 [09:42] <mattias> ajmitch: what sources do you have in sources.list ? [09:43] <ajmitch> deb http://http.us.debian.org/debian unstable main contrib [09:43] <ajmitch> deb http://non-us.debian.org/debian-non-US unstable/non-US main contrib [09:43] <tnt> Eventually the errors go away... after you do a whole bunck o' upgrading. [09:43] <ajmitch> tnt: i've not had this problem [09:43] <ajmitch> tnt: and i've been doing nightly builds for a looong time now [09:43] <mattias> hmm I don't have pkg-config installed as a package simply because it's not on any of my Debian-mirrors [09:43] <ajmitch> mattias: there's your problem :) [09:43] <mattias> so I did a manual install minutes ago [09:43] <mattias> :) [09:44] <tnt> I had this problem when I first installed from the CVS... before the last merge happened. [09:44] <ajmitch> it's not including pkg-config.m4 [09:44] <mattias> ah! [09:44] <tnt> omega walked me through it... [09:44] <ajmitch> pkg-config.m4 resolves these symbols... [09:44] <mattias> should I add it to gstreamer.m4 ? [09:44] <ajmitch> no! [09:44] <tnt> ... the record is going to be in the IRC logs... if anyone wants to search it :-) [09:45] <mattias> so how do I add it to the process? [09:45] <ajmitch> mattias: the pkg-config lines resolve to the configure flags, etc for glib and all that [09:45] <ajmitch> i bet pkgconfig is installed in /usr/local... [09:45] <mattias> yup I got that much [09:45] <mattias> yes it is [09:45] <ajmitch> that's not a good place for it to be :) [09:45] <mattias> :/ [09:45] <ajmitch> aclocal don't like it :) [09:45] <ajmitch> export ACLOCAL_FLAGS="-I/usr/local/share/aclocal" [09:46] <ajmitch> and try to rebuild ;) [09:51] <mattias> I'm waiting for the autogen.sh script.. so I might as well ask if anyone knows what libglib-versions use the GLib 2.0 model? [09:51] <mattias> btw there should be a space after the "-I..." above :) [09:51] <mattias> aclocal complained [09:52] <mattias> hmm the configure script still fails with the same error [09:52] <ajmitch> oh, ok [09:53] <ajmitch> it's been awhile since i tried playing with that stuff [09:53] <ajmitch> hmm, k [09:53] <ajmitch> why does your mirror not have pkgconfig? [09:53] <ajmitch> apt-cache search pkg-config [09:54] <mattias> good question.. I use ftp.acc.umu.se [09:54] <ajmitch> hmm [09:55] <ajmitch> that mustn't have all the debian packages, because pkg-config is being used quite a bit now in newer stuff [09:55] <mattias> yeah I'd think so too.. [09:55] <mattias> Get:1 http://ftp.acc.umu.se unstable/main pkg-config 0.8.0-1.1 [44.6kB] [09:55] <mattias> doh! :) [09:55] <ajmitch> hehe [09:56] <ajmitch> now you should remove pkg-config from /usr/local to avoid errors [09:56] <mattias> done [09:56] <ajmitch> now it should hopefully work [09:56] <mattias> now I'm getting somewhere ! [09:57] <ajmitch> cool [09:57] <sxpert> it seems that the gstreamer samples are _really_ prone to core-dumping [09:59] <ajmitch> oh, the barely maintained tests & examples? ;) [09:59] <mattias> what deb-package provides libHermes ? [09:59] <ajmitch> [ajmitch @ ajmitch gstreamer-anon] dpkg -l |grep hermes [09:59] <ajmitch> ii hermes1 1.3.2-2.1 The Hermes pixel-format library [09:59] <ajmitch> ii hermes1-dev 1.3.2-2.1 Development libraries for the Hermes pixel-f [09:59] <mattias> hermes1 [09:59] <ajmitch> stupid naming, shoudl follow debian policy... [09:59] <mattias> I should think more.. :) [10:03] <sxpert> I got that library, but it core dumps on me too [10:03] <ajmitch> sxpert: sorry, what coredumps? [10:05] <sxpert> apparently, hermes core sumps when it is called by the mpeg2parse3.c sample when giving an mpeg1 stream to it. [10:06] <tnt> Are there any examples yet at using a filter with two inputs ??? [10:07] <ajmitch> hmm... [10:08] <sxpert> so I removed the element, and the video played, but no sound could be heard [10:46] thomasvs (th...@21...) joined #gstreamer. [10:47] <ajmitch> hi thomasvs [10:48] <thomasvs> hi there [10:48] <thomasvs> anything up ? [10:49] <ajmitch> not much [11:01] ajmitch (me...@p1...) left irc: Ping timeout for ajmitch[p19-max2.dun.ihug.co.nz] [11:02] ajmitch (me...@p5...) joined #gstreamer. [11:17] sxpert (sx...@r2...) left irc: Client Exiting [11:20] mattias (ma...@ga...) got netsplit. [11:20] mattias (ma...@ga...) returned to #gstreamer. [12:06] hadess (ha...@de...) joined #gstreamer. [12:06] <hadess> heya [12:12] <ajmitch> hey hadess [12:13] <hadess> hi ajmitch [12:19] <ajmitch> haven't seen you in here for awhile hadess [12:20] <hadess> yeah, been busy [12:29] walken (fo...@c1...) left irc: l8r [12:29] Nick change: tnt -> tnt-zZz [12:36] bstard (Lor...@a2...) joined #gstreamer. [12:51] sxpert (sx...@r2...) joined #gstreamer. [12:51] <sxpert> hello [12:52] <sxpert> while compiling today's CVS : [12:52] <sxpert> make[4]: Entering directory `/home/sxpert/devel/gstreamer/gst/types' [12:52] <sxpert> /bin/sh ../../libtool --mode=link gcc -I../.. -I../../include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -I/usr/include/glib-1.2 -I/usr/lib/glib/include -D_REENTRANT -I/usr/include/gtk-1.2 -I/usr/include/glib-1.2 -I/usr/lib/glib/include -I/usr/X11R6/include -I/usr/include/gnome-xml -O6 -Wall ../../gst/libgst.la -o libgsttypes.la -rpath /usr/lib/gst -version-info 0:0:0 gsttypes.lo -L/usr/lib -rdynamic -lgmodule -lgthread -lglib -lpthread [12:52] <sxpert> -ldl -L/usr/lib -L/usr/X11R6/lib -lgtk -lgdk -rdynamic -lgmodule -lglib -ldl -lXi -lXext -lX11 -lm -L/usr/lib -lxml -lz [12:52] <sxpert> libtool: link: error: cannot link shared libraries into libtool libraries [12:53] <ajmitch> libtool --version [12:53] <ajmitch> you have 1.4 or higher? [12:54] <sxpert> 1.3.5, i'm afraid :( [12:54] <ajmitch> get 1.4 [12:54] <ajmitch> it's listed in REQUIREMENTS [12:55] Action: sxpert can't read I guess... [12:57] <sxpert> ajmitch: btw, is Hermes *really* needed ? [12:58] <ajmitch> nope, just for the colorspace plugin [12:58] <ajmitch> colorspace is used if you don't have Xv, afaik [12:59] <sxpert> gstelement.c: In function `gst_element_class_init': [12:59] <sxpert> gstelement.c:147: warning: assignment from incompatible pointer type [12:59] <sxpert> now that's a bug. [12:59] <hadess> it's better to have hermes installed still [13:00] <ajmitch> hadess: yeah [13:00] <sxpert> ajmitch: same error with libtool 1.4.2 installed [13:00] <hadess> sxpert: re-run the autogen [13:01] <sxpert> configure.ac:1152: LIBS was set with `+=' and is now set with `=' ;( [13:01] <ajmitch> you're using automake 1.5, i guess [13:02] <sxpert> ajmitch: yeah (it said so in the README ;) [13:03] <hadess> that's just a warning, not an error [13:05] <sxpert> GCC 3.0.2 seems to be *very* anal about the code... ;) [13:07] <sxpert> also, shouldn't the makefile uses -march=i686 ? [13:07] <hadess> CFLAGS="-march=i686" ./autogen.sh [13:08] <sxpert> this should be discovered on configure ;) [13:08] Action: sxpert runs [13:08] <ajmitch> why shoudl the makefile use that? not everyone has an i686 cpu ;) [13:08] <hadess> especially not me ;) [13:08] <ajmitch> hadess: exactly :) [13:08] <sxpert> hadess: you have -march=alpha ? [13:09] <hadess> ppc [13:09] <sxpert> ach! a mac guy ! [13:09] Action: ajmitch has a k6-2, which doesn't do i686 afaik [13:10] Action: sxpert has a thunderbird 1.4G... ;) [13:11] Action: sxpert was grossly overpaid last year [13:12] Action: ajmitch is a student [13:13] Action: sxpert will be an IBM'er next year [13:14] Action: ajmitch will be a student next year, and the year after.... [13:14] <ds> HAVE_CPU_S390? [13:14] <sxpert> ds: soon ;) [13:14] <hadess> mehehe [13:15] <sxpert> ds: it's all Power 3 anyways nowadays ;) [13:16] <sxpert> [sxpert@r205m60 test]$ ./mpeg2parse3 ~/*.mpg [13:16] <sxpert> have 2 args [13:16] <sxpert> INFO (15097:-1) Initializing GStreamer Core Library [13:16] <sxpert> INFO (15097:-1) CPU features: (c1c7f9ff) MMX 3DNOW MMXEXT [13:16] <sxpert> ** WARNING **: gstplugin: registry needs rebuild: run gstreamer-register [13:16] <sxpert> lt-mpeg2parse3: relocation error: /usr/lib/gst/libmpeg2dec.so: undefined symbol: gst_plugin_add_factory [13:17] <ajmitch> ** WARNING **: gstplugin: registry needs rebuild: run gstreamer-register [13:17] <ajmitch> you did run gstreamer-register, no? ;) [13:17] <sxpert> gstreamer-register: relocation error: /usr/lib/gst/libmpeg2dec.so: undefined symbol: gst_plugin_add_factory [13:18] <ajmitch> looks like a plugin wasn't rebuilt properly [13:19] <ajmitch> ah... [13:19] <ajmitch> it's using installed plugins [13:19] <ajmitch> not the ones that have just been built, i guess [13:23] BBB (BB...@uc...) left irc: [x]chat [13:27] <ajmitch> Checking for patched automake... not found. [13:27] <ajmitch> Detected automake version 1.4 (or near) without patch. [13:27] <ajmitch> Your version of automake needs a patch applied in order to operate correctly. [13:27] <ajmitch> something tells me that autogen.sh doesn't test for a patched automake very well [13:29] ds (ds...@c7...) left irc: Ping timeout for ds[c799921-a.pinol1.sfba.home.com] [13:31] <ajmitch> ah, i know why, bad $PATH (cruft from gnome 2 libs build) [13:36] arik (ar...@sd...) joined #gstreamer. [13:36] <arik> morning [13:36] <arik> well [13:36] <arik> actually [13:36] <arik> late late late night here [13:36] <ajmitch> morning arik [13:36] <ajmitch> hehe [13:36] <arik> heh [13:36] ds (ds...@c7...) joined #gstreamer. [13:36] <ajmitch> sounds like me last night [13:36] Action: arik is playing around with gtk+2 [13:36] <arik> no gtk apps run on my system atm [13:36] <ajmitch> got to bed about 6:30am, got up about 7:30 :) [13:36] <arik> 2.0 apps that is [13:36] <arik> hehe [13:36] <ajmitch> ooh, why not? [13:36] <ajmitch> not even the demos? [13:36] <arik> nope [13:37] <arik> not even the demos [13:37] <arik> i have no idea [13:37] <ajmitch> suck [13:37] <arik> yep [13:37] <ajmitch> works for me :) [13:37] <arik> i'm updating and recompling gtk now [13:37] <arik> :-P [13:37] <arik> that's my answer to that [13:37] <ajmitch> hehe [13:37] <ajmitch> i'm rebuilding gstreamer right now [13:37] <arik> hah [13:37] <arik> suck [13:37] <ajmitch> yeah, slooow [13:37] <ajmitch> dunno why i'm doing it [13:38] <ajmitch> cron will rebuild it in a couple of hours [13:38] <arik> i'm starting a new test gst app i think [13:38] <ajmitch> cool [13:38] <arik> mix tape app [13:38] <ajmitch> mix tape app? [13:38] <arik> yep [13:38] <arik> it's gonna be an app to help you make mix tapes [13:38] <ajmitch> heh, ok [13:38] <arik> eh [13:38] <arik> i want one [13:39] <arik> i figure why not use it as a test bed for some gst stuff [13:39] <ajmitch> i'm building an app to help record from tape -> ogg || mp3 [13:39] <arik> ooh [13:39] <arik> that's actually usefull [13:39] <ajmitch> yeah [13:39] <arik> mine is just for my own fiendish purposes [13:39] <ajmitch> so is mine :) [13:39] <arik> heh [13:39] <arik> hadess: are you around? [13:39] <ajmitch> got about 100 tapes from a seminar or two that i'm gonna put on a cd [13:39] <hadess> si sen~or [13:39] <ajmitch> heh [13:40] <arik> hadess: wanna talk about rb? [13:40] <arik> :-) [13:40] <arik> specifically [13:40] <ajmitch> playlist... [13:40] <arik> are you gonna make the playlist a control? [13:40] <arik> exactly [13:40] <ajmitch> hehe [13:40] <hadess> hold on [13:40] <arik> cause you really really should [13:40] <arik> sure [13:40] <ajmitch> arik: you gotta grovel! [13:40] <arik> heh [13:40] <arik> fuck that :-) [13:40] <ajmitch> haha [13:41] <arik> i'll write my own damn playlist if i have to! :-P [13:41] <arik> but i like hadess' [13:41] <hadess> i don't have any clue on how to do a playlist control... [13:41] <ajmitch> arik: here's your chance, you could make a control ;) [13:41] <arik> hadess, just make it into a bonobo control and then embed it in rb [13:41] <hadess> what methods should it have ? etc. [13:41] <arik> ajmitch: hehe [13:41] <arik> hadess: yeah, it's not simple [13:42] <arik> but it would be really really usefull [13:42] <arik> cause otherwise it's gonna be super complicated for me to use it in gstplay [13:42] <arik> how is rb coming btw? i haven't build it in awhile [13:42] <hadess> yeah... [13:42] <arik> (cause i don't like installing gst) [13:42] <hadess> i haven't worked on it for quite a while... [13:42] <arik> oh [13:43] <arik> busy? [13:43] <arik> or bored :-) [13:43] <ajmitch> busy porting stuff to gnome 2? [13:43] <arik> heh [13:43] <hadess> at one point waiting for a fix in gst, but i think it still isn't there [13:43] <arik> fix for what? [13:45] <hadess> destroying a thread from within one of its callbacks [13:45] <arik> hmm [13:45] <arik> i have no idea :-) [13:45] <arik> i'm not using threads yet [13:45] <hadess> ie. destroying the play thread when reaching eos [13:45] <arik> i should [13:45] <arik> right [13:46] <arik> eos still isn't working quite right anyway [13:46] <hadess> do you want to talk about the methods the control should have ? i always had troubles using only the property bags for communication between the container and the component [13:47] <arik> yeah sure [13:47] <arik> my bonobo knowledge is not huge (about to get bigger though *grin*) [13:48] <ajmitch> it'll have to get bigger ;) [13:48] <arik> yes [13:48] <arik> yes it will [13:48] <arik> ok, well what are you thinking control wise hadess? [13:48] <arik> we should eventually talk about this on gst-devel [13:48] <arik> cause i know somepeople feel very strongly about it [13:49] <ajmitch> yep [13:49] <arik> i'm not one of them :-) i just want something i can use [13:49] <ajmitch> hehe [13:49] <arik> :-) [13:49] <arik> in general i don't like long arguments about stuff like this [13:50] <ajmitch> i dunno if i would use a playlist or not - i'm not looking to make a player at the moment [13:50] <arik> heh [13:50] <arik> rb and gstplay are really the only major apps that need it [13:50] <arik> but might as well make them use the same one [13:50] <ajmitch> easier just to get it done rather than endlessly discuss details? [13:50] <arik> yep [13:50] <hadess> the playlist is the major part of rhythmbox' code [13:51] <arik> yep [13:51] <hadess> and i'm still working on it [13:51] <arik> i know :-) [13:51] <arik> i've read every line [13:51] <hadess> hehe [13:51] <arik> i've even incorporated it into gstplay [13:51] <arik> at least twice [13:51] <arik> i understand the code pretty well [13:52] <ajmitch> but you want it to be shared.... [13:52] <arik> yep [13:52] <hadess> hmmm.... [13:54] <hadess> should the playlist be the part that's clever about the types ? [13:54] <ajmitch> night all [13:54] <arik> you mean have the gst code in the player or do you mean the media type? [13:54] <arik> night aj [13:55] <hadess> ie. should you choose to add just a file, or add a SongInfo ? [13:55] <hadess> yeah [13:55] Shippou (zbl...@cr...) left irc: Getting off stoned server - dircproxy 1.0.0 [13:55] <arik> hmm [13:55] <hadess> night aj [13:55] <arik> i think you [13:55] Shippou (zbl...@cr...) joined #gstreamer. [13:55] <arik> should just add a file [13:55] <arik> types should be in the playlsit [13:55] <arik> hmm [13:55] <arik> i think [13:56] <hadess> ok... [13:56] <arik> you could have the app fill in the SongInfo [13:56] <arik> but [13:56] <arik> that seems like more work then needed [13:56] <arik> besides, that's how the code works now :-) [13:56] <hadess> yeah, we could then add more types to the playlist [13:56] <hadess> that would be pretty cool :) [13:56] <arik> hehe [13:56] <arik> upgrade the playlist and the app supportes the type? [13:57] <arik> that would be neet [13:57] <hadess> yeah [13:57] <arik> yeah i agree with you [13:57] <hadess> and gstmediaplay and rb would be very integrated, same playlist, same look... [13:57] <arik> i think [13:57] <arik> yeah [13:57] <arik> make it very easy to use from a user perspective [13:58] <arik> i think [13:58] <hadess> we have to think of 'what methods it gives to the application' and 'what facilities it gives to the user' [13:58] <arik> right [13:58] <arik> the app adds the file [13:59] <arik> and asks for the next song to play and whatnot [13:59] <hadess> 1) add items (as SongInfo's) [13:59] <hadess> 2) remove item(s) (selected) [13:59] <arik> as songinfo;'s? [13:59] <arik> i thought we said not as songinfo's [13:59] <arik> or do you mean internally? [13:59] <hadess> oh, yeah, sorry, *dang* [13:59] <arik> heh [13:59] <arik> 1) add items as uri's [13:59] <arik> 2) remove item(s) [13:59] <hadess> gnome-vfs :) [13:59] <arik> yes :-) [14:00] <hadess> get a SongInfo from the current selection/currently played [14:01] <arik> 3) provides uri's to play [14:01] <arik> yeah [14:01] <arik> so you can use the info [14:01] Action: arik is very lagged atm [14:01] <arik> ah ok [14:01] <arik> to the user it needs to manage the library [14:01] Action: hadess adds to list [14:01] <arik> i like the way windows media handles genre's [14:02] <arik> like you can play all the files you have of a genre [14:02] <arik> we want to be able to view the library by information fields [14:02] <hadess> that's the big part [14:03] <hadess> the sorting/searching/filtering [14:03] <arik> yeah [14:03] <arik> exactly [14:03] <arik> the playlist needs to support video [14:03] <arik> file types that is [14:03] <hadess> i'll need to extend SongInfo to add more info ;) [14:03] <arik> yes :-) [14:03] <arik> or make a VideoInfo [14:03] <arik> :-P [14:04] <hadess> no, they should be the same type [14:04] <arik> or make it FileInfo or something [14:04] <arik> yeah they should [14:04] <hadess> MediaInfo or whatever [14:04] <arik> yeah [14:04] <hadess> hmm [14:04] <arik> they should be the same type [14:04] <hadess> then, we need to be able to toggle stuff like repeat/shuffle [14:05] <arik> yep [14:06] <hadess> then it needs to get the next element to play [14:07] <arik> and you need to be able to construct a view based on criteria [14:07] <arik> like in one part of my app i might need a playlist control that only shows the current active temp playlist and only the song names [14:07] <arik> and in another part i want to see the whole library view [14:07] <arik> yeah the next and previous elements [14:07] <hadess> oh previous, yeah :) [14:07] <arik> :-) [14:08] <arik> i can help you code this, i'm not expecting you do pull this all out :-) [14:08] <arik> er to [14:08] <hadess> that's very interesting, because when this is done, i can actually start working on making the app just *work* [14:08] <arik> yes :-) [14:09] <arik> which will be nice as xmms doesn't run on my system :-) [14:09] <hadess> about the whole library vs. playlist [14:10] <arik> yes? [14:10] <hadess> we should then have a 'add playlist' 'add radio listing' 'create library' [14:10] <arik> hmm [14:10] <hadess> the playlist would keep track of them [14:10] <arik> ok [14:10] <arik> so then the app makes a temp playlist? [14:10] <arik> and just doesn't save it? [14:10] <arik> or the playlist just doesn't save it? [14:10] <hadess> and the app would say: "how many items ? what's their names ? what type ?' [14:11] <arik> ok [14:11] <hadess> you would just ask the playlist to save this and that playlist to this and that files [14:11] <arik> oh ok [14:11] <arik> that sounds good [14:11] <arik> so the playlist manages the library and i just add files and get info [14:12] <hadess> exactly [14:12] <arik> right [14:12] <arik> makes sense to me :-) [14:13] <arik> i'm sure it won't be this simple when we do it :-0 [14:13] <hadess> it should *quite* straight forward for the most part actually [14:13] <arik> i hope you're right ;_) [14:14] <hadess> oh, we also need to be able to create filters [14:14] <arik> yep [14:14] <arik> file masks [14:14] <hadess> vfolders dude :) [14:14] <arik> oh my god [14:14] <arik> :-) [14:14] <arik> that is how we do the sorting [14:14] <arik> with vfolders [14:14] <arik> damn [14:14] <arik> a "Rock" vfolder for example [14:14] <hadess> for example [14:15] <arik> that would be very powerful [14:15] <hadess> but we also need a quick way of browsing through [14:15] <arik> right [14:15] <hadess> like setting a "search" item [14:15] <arik> well in windows media there is an All Audio folder and an All Video folder [14:15] <arik> search would be good [14:16] <hadess> yeah, that's just a vfolder [14:16] <arik> yep [14:16] <hadess> selected on type [14:16] <arik> but also have search [14:16] <hadess> like in evo [14:16] <arik> that makes loads of sense [14:16] <arik> right [14:16] <arik> the vfolder stuff is alot of code though [14:16] <arik> just saying [14:16] <hadess> it can be very very simple, only one criteria at first [14:16] <arik> right [14:16] <arik> and we can copy it from evo [14:17] <arik> but it's still a big task [14:17] <hadess> yeah [14:17] <hadess> but i think it's easier this way [14:17] <arik> very big reward at the end though [14:17] <arik> yeah i think so too [14:17] <arik> and more usable [14:18] <hadess> lemme print what i have so far [14:18] <arik> ok [14:18] <hadess> flood [14:18] <hadess> - add playlist items (radio/playlist/audio cd/vfolders) [14:18] <hadess> - add items to library and/or playlist(as URIs, gnome-vfs) [14:18] <hadess> - remove item(s) (selection) [14:18] <hadess> - get currently selected items (GList of MediaInfos) [14:18] <hadess> - get currently playing item (MediaInfo) [14:18] <hadess> - get next/previous element to play [14:18] <hadess> - set a search string (view filters on this) [14:18] <hadess> - toggle Shuffle, Repeat (one/all) [14:18] <hadess> - save playlist items [14:19] <arik> how about get a GList of MediaInfo's of certain criteria? [14:19] <hadess> hmm, why would you need that ? the playlist control does the display [14:19] <arik> hmm [14:19] <arik> that's true [14:19] <arik> i guess it depends on how versatile our display is :-) [14:20] <hadess> hehe [14:20] <hadess> i need to go for food, bbiab [14:20] <arik> for now i think you have a good basic list though [14:20] <arik> ok [14:20] Nick change: hadess -> hds-food [14:55] <arik> great [14:55] <arik> all gtk apps still crash for me [14:55] <arik> on gtk2 [14:59] sienap (sy...@s3...) joined #gstreamer. [14:59] <sienap> Goodday [14:59] <arik> gday [15:01] Nick change: hds-food -> hadess [15:01] <arik> wb [15:01] <hadess> ta [15:01] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [15:02] <hadess> so, what do you think of this ? [15:02] Shippou (zbl...@cr...) joined #gstreamer. [15:02] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [15:02] <hadess> the main problem with a control is that you can only have one view [15:03] <arik> that is a problem [15:03] <arik> why is that true? [15:03] Shippou (zbl...@cr...) joined #gstreamer. [15:03] <sienap> Hey hadess! .. how is porting stuff goign ? [15:03] <hadess> a control is just an out-of-process (still) clever widget [15:03] <arik> right, but you can embed multiple controls [15:03] <hadess> sienap: stalling right now, waiting for the panel to be fixored [15:04] <hadess> arik: true, but the playlists would be different (not the library though) [15:04] <arik> hadess: hmm [15:06] BBB (BB...@uc...) joined #gstreamer. [15:06] <sienap> hadess hehe ;) [15:06] <arik> hadess: what do you suggest? [15:07] <sienap> hej BBB [15:07] <sienap> goddamn i want vesa 1.2 support for FB [15:07] <sienap> problem with vesa 1.2 is that you can't linear map it.. you have the pageflip for every 64k page. [15:07] <hadess> arik: well, i don't have any problems with only one visible playlist ;) [15:07] <arik> hadess: heh, i don't have any problems with only one _visibile_ playlist i don't think [15:07] <hadess> arik: cool, so we agree :) [15:08] <arik> hadess: yep i think so :-) [15:08] <BBB> hello sienap [15:08] <BBB> :) [15:08] Action: BBB wonders how Xine works [15:08] <BBB> I just set up my DXR3 [15:08] <BBB> now how do I get Xine to use it? [15:11] <sienap> HAhaha.. the george lebl candidate anouncement is cool :) [15:11] <hadess> of course [15:13] <sienap> hadess i saw you're electable as well ? :) [15:13] <arik> ugh [15:13] <arik> dangereous stuff hadess [15:13] <arik> you might be elected [15:13] <arik> and then where would you be? [15:13] <sienap> hehe ;) [15:13] <sienap> yeah what then! [15:15] <hadess> "<hadess>: step forward or I will be on the Board" <- topic on #gnome [15:15] <arik> heh [15:15] <sienap> ghe ghe [15:18] <sienap> This program runs on Linux S.O and its license is GPL, we hope that this be useful to the investigators community. [15:18] <sienap> The Linux S.O [15:18] <sienap> dat je het maar weet [15:18] <sienap> :) [15:18] <arik> sienap, where are you? [15:19] <sienap> ehm.. sorry wrong channel ;) [15:19] <sienap> i am from holland [15:19] <arik> hehe [15:19] <arik> hej [15:19] <sienap> so so :) [15:19] <arik> heh [15:20] <arik> that's pretty much the entent of my knowledge :-P [15:21] <sienap> now use that for your coding and it is allright ;) [15:21] <hadess> and you know what ? it also works in danish, swedish and finnish :) [15:21] <arik> i know [15:21] <arik> it's so useful :-) [15:21] <sienap> =] [15:21] <arik> that and um kuhl [15:21] <arik> and rov [15:21] <hadess> "/hej/, pretty blonde" [15:21] <arik> hehe [15:22] <sienap> Nah saying hi to a nice girl in dutch is more like "Goeiendag geile sloerie.. zuig mijn lul" [15:22] <hadess> gibberish [15:30] <arik> heh [15:30] <thomasvs> sienap: some people understand what you're saying you know ;) [15:31] <sienap> thomas that is a detail :) [15:31] <arik> heh [15:31] <arik> thomasvs, where are you? [15:32] <thomasvs> i'd translate but I'm sure some sitewide bot would kick me off for foul language ;) [15:32] <thomasvs> arik: physically ? belgium. [15:32] <arik> yes i know belgium :-) where in belgium? [15:32] <thomasvs> atm, brussels, but I live in Gent [15:32] <arik> ah ok [15:32] <arik> i was in brussels um [15:32] <arik> over new years [15:32] <sienap> ooh you can ask the same in belgium as well [15:32] <sienap> guess you'll get the same response as well [15:32] <sienap> :) [15:33] <sienap> nah.. ask.. say :) [15:33] <thomasvs> sienap: 99 out of 100 times I'll get slapped, 1 out of 100 I'll get really lucky ;) [15:33] <thomasvs> those are good odds actually, maybe I should consider it [15:33] <thomasvs> arik: partying, you mean ? [15:33] <arik> brussels is nice [15:33] <arik> um [15:33] <arik> visiting yeah [15:33] <arik> for a week [15:33] <sienap> thomasvs 99 slaps / 1 sucky sucky a day.. he.. good pay off :) [15:39] wingo (wi...@rd...) joined #gstreamer. [15:51] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [15:52] Shippou (zbl...@cr...) joined #gstreamer. [16:04] <mattias> how do I build gstreamer without avifile support ? it seems broken for me.. [16:05] Action: mattias is using current cvs [16:05] <mattias> flood [16:05] <mattias> gstwinenc.cc: In function `void gst_winenc_chain(GstPad *, GstBuffer *)': [16:05] <mattias> gstwinenc.cc:280: `BitmapInfo' undeclared (first use this function) [16:05] <mattias> gstwinenc.cc:280: (Each undeclared identifier is reported only once [16:05] <mattias> gstwinenc.cc:280: for each function it appears in.) [16:05] <mattias> gstwinenc.cc:280: parse error before `)' [16:05] <mattias> gstwinenc.cc:280: invalid use of undefined type `class CImage' [16:05] <mattias> /usr/local/include/avifile/videoencoder.h:12: forward declaration of `class CImage' [16:05] <mattias> gstwinenc.cc:289: invalid use of undefined type `class CImage' [16:05] <mattias> /usr/local/include/avifile/videoencoder.h:12: forward declaration of `class CImage' [16:05] <arik> um yeah [16:05] <arik> you need to include a file [16:05] <arik> the one in which BitmapInfo is declared [16:05] <mattias> uhum [16:05] <arik> i can't remember which [16:06] <arik> it's an error [16:06] <thomasvs> if you don't need it, just remove the directory from Makefile.am [16:06] <mattias> ok so I should add #include <whatever.h> in gstwinenc.cc ? [16:06] <mattias> thomasvs: ok will consider it [16:06] <arik> yeah [16:06] <arik> just include it [16:07] Action: mattias greps for `BitmapInfo' [16:12] Action: mattias added "#include <image.h>" to gstwinenc.cc [16:12] <mattias> oh sorry that should be "avifile/image.h" [16:12] <arik> yep [16:12] <mattias> seems to work [16:12] <arik> should compile [16:12] <arik> yeah [16:12] <arik> need to tell wtay about that [16:16] sienap (sy...@s3...) left irc: User abort with 5 Ctrl-C's [16:17] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [16:22] <mattias> I'm using the current cvs version of libmpeg2 and there is no libmpeg2dec compiled, only libmpeg2 [16:23] <mattias> Am I missing something? Cause gstreamer wants to link against libmpeg2dec [16:23] <arik> no idea [16:23] <arik> my brain has turned off [16:23] <mattias> :) [16:23] <arik> :-) [16:25] <mattias> time to check the REQUIREMENTS file again I guess [16:25] <arik> heh [16:28] <sxpert> gstreamer-register: relocation error: /usr/lib/gst/libmpeg2dec.so: undefined symbol: gst_plugin_add_factory [16:28] <sxpert> what is that ? [16:29] arik (ar...@sd...) left #gstreamer. [16:31] <mattias> uhm yeah.. [16:31] <mattias> what is that? :) [16:34] <mattias> the configure script should have failed when it couldn't find the library.. I hope I don't have to disable this one [16:35] <sxpert> mattias: that's what I get when I run gstreamer-register [16:37] <mattias> this is strange, kind of like a chicken-or-egg problem.. /usr/lib/gst/libmpeg2dec.so is built with gstreamer? [16:38] <mattias> I had the impression that libmpeg2dec belongs to an external package [16:39] <mattias> http://sourceforge.net/projects/libmpeg2/ [16:39] <sxpert> mattias: I have one in /usr/lib and one in /usr/lib/gst [16:40] <mattias> can you find out which package the "/usr/lib/libmpeg2dec.so" belongs to? I assume it's part of a package in the Linux distro you are using ? [16:41] <mattias> "dpkg -S /usr/lib/libmpeg2dec..." for Debian [16:41] <sxpert> mattias: I compiled it by hand, still have the source somewhere [16:41] <mattias> hehe ok [16:42] <mattias> if you find it, let me know! [16:43] <sxpert> mattias: found the directory [16:44] <mattias> ok! unfortunately I have to leave work but I will be here tomorrow, just mention it in the channel :) [16:44] <sxpert> ok. [16:45] <mattias> what version, an url to where I can find it would be great [16:45] Nick change: mattias -> mattias|na [16:45] <sxpert> found it on their web site, version is 0.2.0 (latest I guess) [16:46] <sxpert> link by freshmeat [16:46] <sxpert> http://libmpeg2.sourceforge.net/ [17:35] Action: hadess started writing nifty .idl file for the playlist component [17:37] <sxpert> hadess: found something that people should be careful with (I got bitten): [17:38] <sxpert> mpeg2dec should be configured with --enable-shared (which is disabled by default :( ) [17:38] <hadess> oh, ok [17:39] Nick change: taazzzz -> taaz [17:39] <taaz> it's disabled for a reason... PIC code slows it down a lot [17:39] <sxpert> taaz: PIC ? [17:40] <sxpert> hadess: will the bonobo-streamer-player-control be ready for consumption soon ? [17:41] <hadess> that's not what i'm working on [17:41] <sxpert> hadess: ah :( [17:41] <sxpert> hehe [17:41] <sxpert> ah, the playlist component (should read slower sometimes, i stopped at ".idl" ;) [17:41] <hadess> i work on a playlist control, for use in rhythmbox and gstmediaplay [17:42] <hadess> and probably other apps as well ;) [17:42] <sxpert> taaz: what do you mean, PIC code ? [17:42] <sxpert> I am currently working on a control to store the bunch of clips (the thing on the right in imovie, with the rows of pictures)) [17:43] <hadess> coolio [17:44] <sxpert> I did all the GList-based stuff to handle the clip info, but now I don't know where to start to create a bonobo control ;) [17:44] <sxpert> that should be a ;( then... [17:44] <hadess> http://www-106.ibm.com/developerworks/library/co-bnbo3.html [17:44] <hadess> and djb's page, lemme look the url [17:44] <hadess> http://linux.denhaag.org/bonobo_controls/ [17:44] <taaz> PIC=position independent code, it's what shared librarys use. on x86 it requires a register be used up for it, which ends up slowing down libmpeg2 7% or something vs a static lib [17:45] <taaz> (roughly... look it up for more info) [17:46] <sxpert> taaz: ah, ok. (I supposed that on i686, athlons and others, it mostly does MMX stuf (and 3Dnow and whatever) so the slowdown is not noticeable), but I may be all wrong ;) [17:46] <sxpert> taaz: I have an athlon 1400 so I don't really care about a 7% slowdown ;) [17:47] <hadess> and i have an endian that's bigger than yours [17:48] <sxpert> hadess: you are wrong, you have an *-endian [17:48] <taaz> just test it yourself [17:48] <taaz> it takes like 5 minutes [17:48] <sxpert> (PPC can take it both ways, no pun intended) [17:48] sienap (sy...@s3...) joined #gstreamer. [17:48] <taaz> MMX doesn't solve the problem [17:48] <hadess> just find me *one* ppc system that's running on little endian [17:49] <sxpert> hadess: well, a mac running virtual-pc will run on little-endian most of the time the virtual pc task is scheduled [17:50] <hadess> hmm, i didn't know this was possible like that... [17:50] <sxpert> hadess: a connectix engineer told me so. (the PPC can switch endianness real-time (just an instruction) and back (another instruction) [17:51] <hadess> oh, cool [17:51] <Shippou> I'm told it's not quite that simple [17:51] <sxpert> IBM engineers are the best ;) [17:51] <Shippou> Stuff almost works [17:51] <Shippou> but some of the instructions are either not as fast or not supported in both endian modes [17:52] <Shippou> it might work for all the instructions you can use from user-space though [17:52] <sxpert> depends if you have an IBM or a Motorola processor (the IBM has it for every single instruction. in fact, it's the bus handler doing all the work, and it's invisible for the core) [17:53] <Shippou> ah [17:53] <Shippou> it's model-specific! [17:53] Action: Shippou whimpers [17:53] <sxpert> Shippou: jut like altivec ;) [17:54] <sxpert> it's model-specific. [17:54] <hadess> only motorola does altivec [17:54] <sxpert> BTW, this is the reason the Nintendo-game-cube has an IBM processor, because of the endianness thing... (they need to eventually emulate the older systems which were little-endians [17:55] <sxpert> hadess: right [17:55] <hadess> coolio [17:55] <hadess> that's neat shit indeed [17:58] <sienap> hell yeah [18:03] <sxpert> waiting impatiently for ST: Enterprise to show up on IRC ... [18:11] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) got netsplit. [18:11] sxpert (sx...@r2...) got netsplit. [18:11] BBB (BB...@uc...) got netsplit. [18:11] wingo (wi...@rd...) got netsplit. [18:11] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) returned to #gstreamer. [18:11] sxpert (sx...@r2...) returned to #gstreamer. [18:11] BBB (BB...@uc...) returned to #gstreamer. [18:11] wingo (wi...@rd...) returned to #gstreamer. [18:11] Topic changed on #gstreamer by ChanServ!s@ChanServ: BeNOW.ca clip: http://gstreamer.net/gstreamer-benow.mp3 [18:12] Shippou (zbl...@cr...) left irc: Getting off stoned server - dircproxy 1.0.0 [18:12] Shippou (zbl...@cr...) joined #gstreamer. [18:13] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [18:13] Shippou (zbl...@cr...) joined #gstreamer. [18:15] <sxpert> waiting impatiently for ST: Enterprise to show up on IRC ... [18:15] Shippou (zbl...@cr...) left irc: Getting off stoned server - dircproxy 1.0.0 [18:15] Shippou (zbl...@cr...) joined #gstreamer. [18:23] <sxpert> apparently, Hermes has problems : [18:23] <sxpert> Program received signal SIGSEGV, Segmentation fault. [18:23] <sxpert> [Switching to Thread 1024 (LWP 19843)] [18:23] <sxpert> 0x40f29c13 in _Hermes_X86_CPU () from /usr/lib/libSDL-1.2.so.0 [18:23] <sxpert> (gdb) bt [18:23] <sxpert> #0 0x40f29c13 in _Hermes_X86_CPU () from /usr/lib/libSDL-1.2.so.0 [18:23] <sxpert> #1 0x410af044 in plugin_desc () from /usr/lib/gst/libcolorspace.so [18:23] <sxpert> #2 0x410d4127 in Hermes_Init () at Hermes.c:116 [18:23] <sxpert> #3 0x410a3e05 in plugin_init () from /usr/lib/gst/libcolorspace.so [18:23] <sxpert> #4 0x4005dff5 in gst_plugin_register_func () from /usr/lib/libgst.so.0 [18:23] <sxpert> #5 0x4005e224 in gst_plugin_load_plugin () from /usr/lib/libgst.so.0 [18:23] <sxpert> #6 0x4005df2c in gst_plugin_load_absolute () from /usr/lib/libgst.so.0 [18:23] <sxpert> #7 0x4005da3a in gst_plugin_load_recurse () from /usr/lib/libgst.so.0 [18:23] <sxpert> #8 0x4005d955 in gst_plugin_load_recurse () from /usr/lib/libgst.so.0 [18:23] <sxpert> #9 0x4005dae4 in gst_plugin_load_all () from /usr/lib/libgst.so.0 [18:23] <sxpert> #10 0x4005d6e4 in _gst_plugin_initialize () from /usr/lib/libgst.so.0 [18:23] <sxpert> #11 0x4004d2ee in gst_init () from /usr/lib/libgst.so.0 [18:23] <sxpert> #12 0x08048f02 in main () at eval.c:41 [18:23] <sxpert> #13 0x403b5627 in __libc_start_main (main=0x8048e90 <main>, argc=1, [18:24] <sxpert> ubp_av=0xbffff9d4, init=0x8048958 <_init>, fini=0x8049110 <_fini>, [18:24] <sxpert> rtld_fini=0x4000dcd4 <_dl_fini>, stack_end=0xbffff9cc) [18:24] <sxpert> at ../sysdeps/generic/libc-start.c:129 [18:36] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [18:37] Shippou (zbl...@cr...) joined #gstreamer. [18:37] <hadess> 0x40f29c13 in _Hermes_X86_CPU () from /usr/lib/libSDL-1.2.so.0 <- wtf is that ? [18:38] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [18:39] Shippou (zbl...@cr...) joined #gstreamer. [18:57] <hadess> hadess.net/files/stuff/Gnome_Rhythmbox_Playlist.idl <- can you guys tell arik to look at that ? check it as well please [18:57] <hadess> need to go [18:57] <hadess> cya [18:57] hadess (ha...@de...) left #gstreamer (Client Exiting). [19:06] evil_monk (trevo@208.141.162.68) joined #gstreamer. [19:07] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) got netsplit. [19:07] sxpert (sx...@r2...) got netsplit. [19:07] BBB (BB...@uc...) got netsplit. [19:07] wingo (wi...@rd...) got netsplit. [19:08] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) returned to #gstreamer. [19:08] sxpert (sx...@r2...) returned to #gstreamer. [19:08] BBB (BB...@uc...) returned to #gstreamer. [19:08] wingo (wi...@rd...) returned to #gstreamer. [19:09] BBB (BB...@uc...) left irc: Ping timeout for BBB[ucu-104-70.ucu.uu.nl] [19:16] BBB (BB...@uc...) joined #gstreamer. [19:36] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [19:38] evil_monk (trevo@208.141.162.68) joined #gstreamer. [19:54] Shippou (zbl...@cr...) left irc: Getting off stoned server - dircproxy 1.0.0 [19:55] Shippou (zbl...@cr...) joined #gstreamer. [19:56] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [19:56] Shippou (zbl...@cr...) joined #gstreamer. [19:59] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [19:59] Shippou (zbl...@cr...) joined #gstreamer. [20:00] Shippou (zbl...@cr...) left irc: Getting off stoned server - dircproxy 1.0.0 [20:00] Shippou (zbl...@cr...) joined #gstreamer. [20:01] CrunchyBear (na...@53...) joined #gstreamer. [20:09] Shippou (zbl...@cr...) left irc: Getting off stoned server - dircproxy 1.0.0 [20:09] Shippou (zbl...@cr...) joined #gstreamer. [20:17] CrunchyBear (na...@53...) left #gstreamer. [20:23] ChiefHighwater (pa...@te...) joined #gstreamer. [20:24] Nick change: wingo -> wingo-klass [20:29] <steveb> ds: there? [20:37] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [20:37] Shippou (zbl...@cr...) joined #gstreamer. [20:38] Nick change: tnt-zZz -> tnt [20:38] <tnt> Good morning. [20:40] <ChiefHighwater> Ello [20:40] <tnt> Hello ChiefHighwater. [20:47] <evil_monk> hmm, is ganso even still being developed? [20:47] Shippou (zbl...@cr...) left irc: Ping timeout for Shippou[cr934547-a.flfrd1.on.wave.home.com] [20:48] Shippou (zbl...@cr...) joined #gstreamer. [20:56] Shippou (zbl...@cr...) left irc: Getting off stoned server - dircproxy 1.0.0 [20:56] Shippou (zbl...@cr...) joined #gstreamer. [20:58] <sxpert> evil_monk: not since last february :( [21:01] <sxpert> <sxpert> apparently, Hermes has problems : [21:01] <sxpert> <sxpert> Program received signal SIGSEGV, Segmentation fault. [21:01] <sxpert> <sxpert> [Switching to Thread 1024 (LWP 19843)] [21:01] <sxpert> <sxpert> 0x40f29c13 in _Hermes_X86_CPU () from /usr/lib/libSDL-1.2.so.0 [21:01] <sxpert> <sxpert> (gdb) bt [21:01] <sxpert> <sxpert> #0 0x40f29c13 in _Hermes_X86_CPU () from /usr/lib/libSDL-1.2.so.0 [21:01] <sxpert> <sxpert> #1 0x410af044 in plugin_desc () from /usr/lib/gst/libcolorspace.so [21:01] <sxpert> <sxpert> #2 0x410d4127 in Hermes_Init () at Hermes.c:116 [21:01] <sxpert> <sxpert> #3 0x410a3e05 in plugin_init () from /usr/lib/gst/libcolorspace.so [21:01] <sxpert> <sxpert> #4 0x4005dff5 in gst_plugin_register_func () from /usr/lib/libgst.so.0 [21:01] <sxpert> <sxpert> #5 0x4005e224 in gst_plugin_load_plugin () from /usr/lib/libgst.so.0 [21:01] <sxpert> <sxpert> #6 0x4005df2c in gst_plugin_load_absolute () from /usr/lib/libgst.so.0 [21:01] <sxpert> <sxpert> #7 0x4005da3a in gst_plugin_load_recurse () from /usr/lib/libgst.so.0 [21:01] <sxpert> <sxpert> #8 0x4005d955 in gst_plugin_load_recurse () from /usr/lib/libgst.so.0 [21:01] <sxpert> <sxpert> #9 0x4005dae4 in gst_plugin_load_all () from /usr/lib/libgst.so.0 [21:01] <sxpert> <sxpert> #10 0x4005d6e4 in _gst_plugin_initialize () from /usr/lib/libgst.so.0 [21:01] <sxpert> <sxpert> #11 0x4004d2ee in gst_init () from /usr/lib/libgst.so.0 [21:01] <sxpert> <sxpert> #12 0x08048f02 in main () at eval.c:41 [21:01] <sxpert> <sxpert> #13 0x403b5627 in __libc_start_main (main=0x8048e90 <main>, argc=1, [21:01] <sxpert> <sxpert> ubp_av=0xbffff9d4, init=0x8048958 <_init>, fini=0x8049110 <_fini>, [21:01] <sxpert> <sxpert> rtld_fini=0x4000dcd4 <_dl_fini>, stack_end=0xbffff9cc) [21:02] <sxpert> <sxpert> at ../sysdeps/generic/libc-start.c:129 [21:02] <sxpert> any ideas ? [21:17] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [21:34] sienap (sy...@s3...) left irc: User abort with 5 Ctrl-C's [21:54] Action: tnt is away: I'm busy [21:58] steveb (st...@no...) left irc: advance! [22:02] wingo-klass (wi...@rd...) left irc: going home [22:10] Zeenix (ze...@13...) joined #gstreamer. [22:15] Shippou (zbl...@cr...) left irc: Getting off stoned server - dircproxy 1.0.0 [22:16] Shippou (zbl...@cr...) joined #gstreamer. [22:24] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [22:24] Shippou (zbl...@cr...) joined #gstreamer. [22:26] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: EOF from client [22:26] Shippou (zbl...@cr...) joined #gstreamer. [22:36] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: Connection reset by peer [22:36] Shippou (zbl...@cr...) joined #gstreamer. [22:55] Action: ds boing! [22:56] evil_monk (trevo@208.141.162.68) joined #gstreamer. [23:08] chillywilly (da...@d1...) joined #gstreamer. [23:11] <Zeenix> anyone arround ? [23:11] <chillywilly> nope [23:12] <sxpert> yes [23:13] <Zeenix> sxpert: i needed someone to answer my questions :) [23:14] <sxpert> Zeenix: me too ;) [23:14] <Zeenix> sxpert: lets send police to search out wtay for us :) [23:14] <sxpert> gstreamer-register crashes with #0 0x40f29c13 in _Hermes_X86_CPU () from /usr/lib/libSDL-1.2.so.0 [23:15] <sxpert> that's whan backtracking with gdb ;( [23:15] Nick change: wtay-zZz -> wtay [23:15] <wtay> yo [23:15] <Zeenix> yo yo [23:15] <wtay> sxpert: SDL is borken [23:16] <Zeenix> wtay: i made the RTCP work [23:16] <wtay> sxpert: it includes it's own version of Hermes [23:16] <wtay> s/it's/its [23:16] <wtay> Zeenix: cool [23:17] <Zeenix> wtay: its doing fine, but dont know why the i am not getting the correct payload_type on rtp packet( not the rtcp packet ) [23:17] <Zeenix> s/on/in [23:18] <Zeenix> wtay: i just sent you the plugins [23:18] <wtay> ok [23:19] <Zeenix> the prob. was that we didnt realize that the newcaps is called before the state_change [23:20] <wtay> oh [23:20] <Zeenix> wtay: could the newcaps be called after the state_change ? [23:20] <wtay> it should be I think... [23:20] <wtay> actually, you cannot rely on that i guess.. [23:22] harobed (harobed@ACAD586D.ipt.aol.com) joined #gstreamer. [23:22] <Zeenix> so we need a boolean, so in the newcaps that we could know that the socket is initialized before we send the rtcp packet [23:23] <wtay> yes [23:23] <Zeenix> or we assign a 0 to the socket when its not initialized [23:24] <Zeenix> & this will serve as our boolean [23:24] <ds> wtay: you may have noticed libs/resample. what do you want me to do with the Makefile.am changes? [23:25] <wtay> ds: what's wrong? [23:25] <ds> I just wanted to coordinate [23:25] <ds> get an ok [23:25] <wtay> what's the problem then? [23:26] Action: tnt is back (gone 01:32:31) [23:26] <ds> nothing, I guess. But I'd like to add libs/resample to configure.base and libs/Makefile.am, and then fix audioscale to use the lib [23:26] <wtay> oh, ok [23:26] <Zeenix> wtay: did you tried those mp3 piplelines i requested you to ? [23:27] <wtay> Zeenix: nope, as I said I'm doing other stuff, non gst related [23:27] <Zeenix> wtay: office work? [23:27] <wtay> ds: did you know that I removed the audio resample libs from libs/ a few months ago? :) [23:27] <sxpert> wtay: should I downgrade ? [23:27] <wtay> Zeenix: naah, writing an app I always wanted to write [23:28] <ds> wtay: er? [23:28] <Zeenix> wtay: ahh, video editting.... [23:28] <wtay> sxpert: ii libsdl1.2 1.2.2-3 Simple DirectMedia Layer [23:28] <wtay> Zeenix: nope, xml editor [23:28] <wtay> ds: yeah :) [23:29] <taaz> people shouldn't have to write raw xml [23:29] <wtay> ds: omega is not very fond of the libs/ stuff, but I think of it as libs that should become separate libs in the end.. [23:30] <ds> wtay: that is sort of what I was planning to do. [23:30] <wtay> ds: yup, perfect [23:33] Action: sxpert found some subreptiously RPM installed obsolete libSDL [23:34] <sxpert> whatever happened to that spanish-written really cool XML editor ? [23:35] <wtay> ? [23:35] <ds> wtay: wingo wanted to use it for a slightly different plugin, speed [23:35] <Zeenix> wtay: whats the handoff signal in disksrc/filesrc for ? [23:36] <wtay> Zeenix: if the app wants to be notified when data is pushed [23:39] <Zeenix> wtay: dont have time to run that mp3 pipeline atm ? [23:39] <wtay> Zeenix: nope, gotta finish this first [23:39] <Zeenix> wtay: on single machine i mean [23:41] <wtay> hrrm, ok.. [23:42] <wtay> what should I do? [23:42] <Zeenix> --launch disksrc location=xxx.mp3 ! rtpsend payload_type=mp3 [23:42] <wtay> oh, I need to port your plugin too... [23:43] <ds> should .../libxing/config.h.in be in CVS? autoheader keeps regenerating it for me. [23:43] <Zeenix> --launch rtprecv ! mp3parse ! mpg123 ! osssink [23:43] <Zeenix> wtay: port? [23:44] <wtay> ds: it can go (so can the libxing dir)... [23:45] <wtay> Zeenix: my librtp is different, it doesn't compile here [23:45] <Zeenix> wtay: librtp-0.99-3cl [23:46] <Zeenix> wtay: i have the rpm named: librtp-0.99-3cl.rpm [23:46] <wtay> I have cvs rtp [23:47] <Zeenix> cant you find the one i am having, i got it from rpmfind.net [23:47] <wtay> Zeenix: debian? [23:47] <Zeenix> wtay: yeah, aint there any debianfind.net [23:47] <Zeenix> ? [23:52] <Zeenix> wtay: is there any rtcp_compound_add_bye() in the cvs librtp( rtcp-packet.h ) [23:53] <wtay> rtp.h:44: conflicting types for `PAYLOAD_GSM' [23:53] <wtay> /usr/local/include/rtp/rtp-audio.h:32: previous declaration of `PAYLOAD_GSM' [23:53] <wtay> rtp.h:45: conflicting types for `PAYLOAD_L16_MONO' [23:53] <wtay> /usr/local/include/rtp/rtp-audio.h:33: previous declaration of `PAYLOAD_L16_MONO' [23:53] <wtay> .... [23:53] bstard (Lor...@a2...) joined #gstreamer. [23:54] <Zeenix> wtay: what i commented out that #include <rtp-audio.h) [23:55] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [23:57] Shippou (zbl...@cr...) left irc: Getting off stoned server - dircproxy 1.0.0 [23:57] Shippou (zbl...@cr...) joined #gstreamer. [23:57] evil_monk (trevo@208.141.162.68) joined #gstreamer. [23:57] <Zeenix> wtay: ? [23:58] Action: BBB is away: zzz [23:58] <wtay> Zeenix: refuses to compile.. fixing.. [23:58] Nick change: BBB -> BBB-zZz [23:59] <Zeenix> wtay: but what happend, got that librtp-0.99-3 [23:59] <Zeenix> ? [23:59] <wtay> Zeenix: nope, I'm porting to cvs rtp [23:59] <wtay> it compiled now.. [00:00] --- Wed Nov 7 2001 [00:00] Action: Zeenix declares people, not using rpm, criminals :) [00:01] harobed (harobed@ACAD586D.ipt.aol.com) left irc: Client Exiting [00:01] <sxpert> I reinstalled it all, but Hermes still crashes. [00:01] <wtay> Zeenix: it works.. I hear something [00:01] <sxpert> :( [00:02] <wtay> sxpert: remove SDLvideosink or colorspace [00:02] <sxpert> That I just did :) [00:03] <Zeenix> sxpert: it crashes registering colorspace plugin ? [00:03] <wtay> Zeenix: trick is now to add mp3parse before rtpsend so the timestamps are calculated before sending stuff [00:04] <wtay> Zeenix: that's when mp3parse actually does timestamps... [00:04] <sxpert> Zeenix: yes :( [00:04] <Zeenix> wtay: ok, but did it worked, or got the core dumped [00:05] <sxpert> Zeenix: no, it code-dumped :( [00:05] <wtay> Zeenix: it works fine [00:05] <wtay> Zeenix: I don't use mpg123 though, I'm more a madman myself :) [00:06] <wtay> Zeenix: sounds stops after a while because of packet loss [00:06] <sxpert> #0 0x40f29c13 in _Hermes_X86_CPU () from /usr/lib/libSDL-1.2.so.0 [00:06] <sxpert> #1 0x410af044 in ?? () [00:06] <sxpert> #2 0x410d4127 in ?? () [00:06] <sxpert> #3 0x410a3e05 in ?? () [00:06] <sxpert> #4 0x4005dff5 in gst_plugin_register_func () from /usr/lib/libgst.so.0 [00:06] <sxpert> #5 0x4005e224 in gst_plugin_load_plugin () from /usr/lib/libgst.so.0 [00:06] <sxpert> #6 0x4005df2c in gst_plugin_load_absolute () from /usr/lib/libgst.so.0 [00:06] <sxpert> #7 0x4005da3a in gst_plugin_load_recurse () from /usr/lib/libgst.so.0 [00:06] <sxpert> #8 0x4005d955 in gst_plugin_load_recurse () from /usr/lib/libgst.so.0 [00:06] <sxpert> #9 0x4005dae4 in gst_plugin_load_all () from /usr/lib/libgst.so.0 [00:06] <sxpert> #10 0x4005d6e4 in _gst_plugin_initialize () from /usr/lib/libgst.so.0 [00:06] <sxpert> #11 0x4004d2ee in gst_init () from /usr/lib/libgst.so.0 [00:06] <sxpert> #12 0x08048f02 in main () at eval.c:41 [00:06] <sxpert> #13 0x403b5627 in __libc_start_main (main=0x8048e90 <main>, argc=1, [00:07] <sxpert> ubp_av=0xbffff974, init=0x8048958 <_init>, fini=0x8049110 <_fini>, [00:07] <sxpert> rtld_fini=0x4000dcd4 <_dl_fini>, stack_end=0xbffff96c) [00:07] <sxpert> at ../sysdeps/generic/libc-start.c:129 [00:07] <Zeenix> wtay: use that now & see if you get "mpg123: Can't rewind stream by 8 bits!" like errors [00:07] <sxpert> here is the backtrace [00:07] <wtay> sxpert: you removed colorspace? [00:07] <wtay> Zeenix: I have those too [00:07] <sxpert> now, without /usr/lib/gst/libcolorspace* it worked [00:07] <Zeenix> sxpert: make uninstall in the plugins/filter/colorspace [00:07] <wtay> sxpert: right [00:08] <sxpert> I jsut did, and it worked [00:08] <wtay> sxpert: nothing that can be done about that unfortunatly [00:08] <Zeenix> sxpert: i am having this error since a long time ago, so i always remove the colorspace [00:09] <Zeenix> wtay: i am a bit confused in max_level & level prop. in the queue element [00:10] <sxpert> so I also have to remove the color element in the /test/mpeg2parse3 sample [00:10] Shippou (zbl...@cr...) left irc: Read error to Shippou[cr934547-a.flfrd1.on.wave.home.com]: Connection reset by peer [00:11] Shippou (zbl...@cr...) joined #gstreamer. [00:11] <wtay> Zeenix: max_level is the maximum number of queued buffers, level is the current amount of queued buffers [00:11] <ajmitch> sxpert: are you installing a CVS version, or installing from a tarball? [00:11] <sxpert> ajmitch: CVS. [00:11] <ajmitch> hmmm... [00:11] <sxpert> it was doing the same thing with the tarball by the way [00:14] <Zeenix> wtay: connecting the mp3parse before rtpsend would mean to also send the props on the caps set by mp3parse, oh another headache ? [00:14] <Zeenix> s/send/rtcp_send [00:15] <wtay> Zeenix: it should work out of the box with your current code [00:15] <ajmitch> sxpert: hmm, afaik, it's not best to install a CVS version (ask wtay about it) [00:17] <Zeenix> wtay: how? dont need to communicate the bit_rate & stuff to rtprecv? [00:18] <wtay> Zeenix: no, it outputs audio/mp3, no bitrate at all [00:20] <Zeenix> wtay: you mean that mp3parse se... [truncated message content] |
From: IRC B. <wt...@us...> - 2001-11-08 05:27:11
|
******************************************************************* [03:05] Zygo (zbl...@cr...) left irc: Read error to Zygo[cr934547-a.flfrd1.on.wave.home.com]: Connection reset by peer [03:05] Zygo (zbl...@cr...) joined #gstreamer. [03:07] taaz (dlehn@66.37.66.32) joined #gstreamer. [03:16] chillywilly (da...@d1...) left irc: booting kernel 2.4.14 [03:17] Zygo (zbl...@cr...) left irc: Read error to Zygo[cr934547-a.flfrd1.on.wave.home.com]: Connection reset by peer [03:20] <taaz> ds: getting funky with the resampler eh? is your new code pretty good? [03:21] <taaz> it's gotta work at like +/-8% or so to make turntable emulation easy ;) [03:21] <ds> is 10x real-time with near perfect resampling good? [03:21] <taaz> how much can you slow it down before it sounds like crap? [03:22] <taaz> (i'm not at home or i'd try...) [03:22] <taaz> err.. slow the audio down, not the performance ;) [03:22] <ds> 48khz->22050 sounds ok [03:23] <ds> I'm more interested in rate resampling, not time. Although they are identical, of course. [03:23] <taaz> eh.. is resampling the same as scaling? [03:23] <taaz> my brain is fried right now it seems ;) [03:24] <ds> yeah, it's the same. [03:24] <ds> wingo is working on the speed plugin, which is essentially the same [03:24] <taaz> i'll have to make a nifty turntable demo app sometime ;) [03:24] <ds> but takes x*44.1k samples and turns it into 44.1k [03:25] <taaz> well... they probably all use the same data munging code, but timestamps are probably different for different uses [03:25] <ds> right. and caps negotiation. So the underlying code can be the same, but the plugins are different [03:25] <taaz> whee! reuse! [03:26] <ds> btw, do you know how to force a video stream into rgb? I want to test videoscale with RGB [03:26] <taaz> huh? [03:27] <ds> wait, you're not a video guy, are you? [03:27] <taaz> do with have a converter? [03:27] <taaz> like mpeg2dec will dump YUV (aka Y'CrCb ;)) [03:27] <ds> ... ! colorspace ! videoscale ! xvideosink [03:28] <taaz> colorspace... didnt know about that one [03:28] <ds> yes, I know xvideosink understands YUV, but I want to force RGB so I can test videoscale [03:28] <ds> Maybe I should write a force_caps plugin [03:29] <taaz> i really need to get glib2.0 stuff running and see if it's thread safe or not [03:30] <taaz> i'm still in shock that i'm the only person getting hit by gstreamer having been totally un threadsafe while based on gtk [03:30] ChiefHighwater (pa...@te...) left irc: Ping timeout for ChiefHighwater[temple-baptist.com] [03:31] <ds> you have SMP? [03:31] <taaz> yeah [03:31] <ds> my SMP machine is too slow to do anything useful with gstreamer [03:31] <ds> (except audio resampling, now...) [03:31] <taaz> SMP+threads+gtksignals == lockup [03:32] <taaz> yeah.. maybe i'll work on that l8r t0n1gh7 [03:33] <taaz> i -really- need to work on python bindings [03:33] <taaz> i'm a python junkie now... [03:33] <ds> i've started to really like combinations of C and gst_parse_launch() [03:34] <ds> although python would replace the C quite easily [03:34] chillywilly (da...@d1...) joined #gstreamer. [03:35] <taaz> skool stuff taking 110% of my time lately :( i need to fit gstreamer into my project somehow... [03:35] chillywilly (da...@d1...) left irc: Killed (NickServ (This nick is reserved by another user)) [03:35] Action: taaz ponders [03:35] chillywilly_ (da...@d1...) joined #gstreamer. [03:37] chillywilly_ (da...@d1...) left irc: Read error to chillywilly_[d17.as21.nwbl0.wi.voyager.net]: EOF from client [03:37] chillywilly (da...@d1...) joined #gstreamer. [03:37] <taaz> i want to try actually doing the data processing in python too ;) i get laughed at for suggesting that for some reason [03:37] <taaz> NumPy stuff is pretty darn cool though [03:38] <taaz> it's just a question of how much overhead there is jumping in/out of python vm [03:39] <ds> rule of optimization #1: don't do it very often [03:40] <taaz> i prefer not to speculate on the overhead till i've tried it [03:41] <ds> rule of optimization #2: don't divide [04:08] Shippou (zbl...@ot...) joined #gstreamer. [04:19] <vektor> Hello. [04:20] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [04:20] <tnt> Hey vektor. [05:17] mwc (ma...@ly...) left irc: Read error to mwc[lychee.ntu.edu.au]: EOF from client [06:34] Action: ds boing! [06:34] <ajmitch> ello [06:34] <ds> what's new? [06:35] <ajmitch> not much, irc seems dead at the moment [06:35] <ajmitch> i'm meant to be studying so have no time for oding [06:35] <ajmitch> c/oding/coding/ [06:35] <ajmitch> bah [06:35] <vektor> i wish i had more time for oding. [06:35] <ajmitch> too hot here [06:35] <vektor> i love to ode. [06:36] <ajmitch> can't think (or type) [06:36] <ajmitch> wassup vektor? [06:36] <vektor> bzy [06:37] <ajmitch> :) [06:38] <ds> not hot here. My house seems to defy the laws of thermodynamics by being continuously colder than the surroundings [06:38] <ajmitch> heh [06:39] <ajmitch> my room tends to be a demonstration of the greenhouse effect [06:39] <ds> 'course it's pushing winter here, too [06:39] <ajmitch> getting on towards summer here [06:39] <ajmitch> good thing i'm only living here for another 3 days [06:42] <ajmitch> ah well, bbl ;) [07:35] wingo (wi...@rd...) joined #gstreamer. [07:35] <ds> yo [07:35] <wingo> hey [07:35] <ds> so I ci'd libs/resample [07:35] <wingo> i've seen your work on the sampler-doober [07:35] <wingo> yeah [07:35] <ds> it works with both mono and stereo s16 [07:35] <wingo> would it be hard to add float? [07:36] <ds> float should be easy to add [07:36] <wingo> cool [07:36] <ds> it converts to float first [07:36] <wingo> i'll play with it when i get some time [07:36] <ds> cool [07:37] <wingo> nice work tho. thanks for libifying it [07:38] <ds> no problem. [07:40] <wingo> gnight [07:40] Nick change: wingo -> wingo-zzz [07:58] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) got netsplit. [07:58] tnt (tn...@h2...) got netsplit. [07:58] ds (ds...@c7...) got netsplit. [07:58] taaz (dlehn@66.37.66.32) got netsplit. [07:58] sxpert (sx...@r2...) got netsplit. [07:59] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) returned to #gstreamer. [07:59] sxpert (sx...@r2...) returned to #gstreamer. [07:59] tnt (tn...@h2...) returned to #gstreamer. [07:59] ds (ds...@c7...) returned to #gstreamer. [07:59] taaz (dlehn@66.37.66.32) returned to #gstreamer. [08:46] sxpert (sx...@r2...) left irc: Ping timeout for sxpert[r205m60.cybercable.tm.fr] [08:46] sxpert (sx...@r2...) joined #gstreamer. [09:12] Nick change: taaz -> taazzzz [09:28] chillywilly (da...@d1...) left irc: [11:00] arik (ar...@sd...) joined #gstreamer. [11:03] arik (ar...@sd...) left irc: Ping timeout for arik[sdn-ar-001waseatP050.dialsprint.net] [11:03] <ajmitch> hey arik [11:03] <ajmitch> bah [11:28] tnt (tn...@h2...) left irc: Ping timeout for tnt[h24-83-72-250.vs.shawcable.net] [11:31] hadess (ha...@de...) joined #gstreamer. [11:31] <hadess> 'lo [11:32] arik (ar...@sd...) joined #gstreamer. [11:32] <hadess> arik ! [11:32] <arik> hi [11:32] <arik> hey hadess [11:32] <arik> how goes it? [11:32] Action: arik is building gnome2 from scratch... again [11:32] <arik> *sigh* [11:32] <hadess> http://hadess.net/files/stuff/Gnome_Rhythmbox_Playlist.idl <- tell me what you think of that [11:32] <arik> ok [11:32] <arik> one sec [11:32] <arik> btw [11:33] <arik> i really think someone should write a daemon [11:33] <arik> that polls the x server every second [11:33] <arik> and if it can't talk to it kills the process [11:33] <ajmitch> hey hadess, arik [11:33] <hadess> arik: huh ? [11:33] <arik> well [11:33] <arik> it would stop those xlockups from crashing your system [11:33] <arik> as just happened to me [11:34] Action: arik is reading the .idl [11:34] <ajmitch> x locks up? [11:34] <ajmitch> heh, i only have lockups when the room is *way* too hot [11:35] <arik> hmm [11:35] <arik> x locks up on me periodically [11:35] <arik> only reason my comp has ever crashed [11:35] <ajmitch> ok [11:35] <arik> hadess, just so i'm clear, explain the SourceInfo stuff to me? [11:35] <arik> i think i understand it mind [11:35] <arik> oh ok [11:35] <arik> nm [11:35] <ajmitch> mine has been locking up a few times in the last month or so (gettng into summer weather) [11:35] <arik> i get it [11:35] <arik> ajmitch: my comp does run hot [11:35] <hadess> hehe, ok [11:36] <hadess> it's a playlist/library/audio cd/radio listing [11:36] <arik> right [11:36] <arik> i got it [11:36] <arik> maybe we should have an add_media_to_playlist and an add_media_to_library [11:36] <arik> that's how windows media works [11:36] <arik> so that you can have temp playlists [11:37] <arik> where the file isn't added to the library [11:37] <hadess> add_media_to_playlist adds it directly to the library [11:37] <arik> hmm [11:37] <hadess> add_media_to_playlist where "playlist" is the library adds the media if it's not already there [11:37] <arik> ok [11:37] <arik> in the .idl you have add_media_to_library [11:38] <arik> and i think we need something to add to the library [11:38] <arik> and something to add to new temp playlist [11:39] <arik> what does set_next_repeat_type do? [11:39] <hadess> it's a toggle button [11:40] <hadess> so .... it toggles the value ;) [11:40] <arik> what is it for? [11:40] <arik> ok [11:40] <arik> that's strange [11:40] <arik> but ok :-) [11:40] <arik> well [11:40] <hadess> no, that's so that you don't have to care about the value of it, just say "next type" [11:40] <arik> oh ok [11:40] <arik> that makes sene [11:40] <arik> alright [11:41] <hadess> instead of next repeat type, i should say toggle... [11:41] <arik> other then the add_media_to_playlist / library debate i agree with this file [11:41] <arik> yeah [11:41] <arik> that would be better [11:42] <hadess> ha, no, not for the repeat [11:42] <hadess> it's a 3-state toggle, and it's easier that way [11:42] <arik> right [11:42] <arik> ok [11:42] <hadess> the shuffle is a plain on/off toggle [11:42] <arik> so here is my thing about adding to the library / playlist [11:42] <arik> you want to be able to add a file to the library [11:43] <arik> but you also want to be able to simply create a temp playlist view that doesn't get saved (or maybe just saves state) and doesn't add files to the library [11:43] <arik> i kinda understand what you are saying [11:44] <hadess> add_media_to_playlist(..., boolean add_to_library) ? [11:44] <arik> yeah that would work [11:44] <arik> a bit hacky [11:44] <hadess> i'd rather have a different call though [11:44] <arik> but it would work [11:44] <arik> yeah me too [11:44] <arik> how about [11:44] <arik> add_media_to_playlist and [11:44] <arik> um [11:44] <arik> add_media_to_playlist_and_library or something? [11:44] <hadess> hmmm [11:45] <ajmitch> long names... [11:45] <arik> yeah [11:45] <hadess> add_media_to_playlist_only [11:45] <arik> ok [11:45] <arik> that's a bit less clear [11:45] <arik> but ok [11:45] <ajmitch> you need to learn how to cut them down, start by removing '_' and all vowels ;) [11:45] <arik> haha [11:45] <arik> i like long names [11:45] <arik> just use meta-/ in emacs :_P [11:46] <ajmitch> oh, and then ROT13 them.. ;) [11:46] <arik> hehe [11:46] <arik> that's much more clear aj [11:46] <ajmitch> who cares about clear? [11:46] <hadess> *slap* [11:46] <arik> hehe [11:47] Action: ajmitch should write a program to do that to all variable names ;) [11:47] <arik> hehe [11:47] <arik> that would be funny [11:47] <hadess> arik: updated [11:48] <arik> ok [11:48] <arik> ok [11:48] <arik> this looks good for now [11:48] <arik> i'm sure we will change things later on [11:48] <arik> when we realize how broken it all is :-P [11:48] <hadess> hehe [11:49] <ajmitch> sure [11:49] <hadess> oh, we also need a "summary" [11:49] <ajmitch> that's the whole point of design, work out what to break in the next revision ;) [11:49] <arik> hadess, should we make a new gnome module for this or leave it in the rb module? [11:49] <arik> summary? [11:49] <arik> ajmitch: yep, and since only me and hadess are using it for now we can break ALL WE WANT! :-) [11:49] <hadess> like it would appear in a status dock [11:49] <ajmitch> damn, there are so many gnome modules already... [11:50] <arik> hadess: hmm [11:50] Action: ajmitch might decide to use it tho [11:50] <arik> ajmitch: well we can leave it in rb's module too [11:50] <ajmitch> arik: doesn't worry me [11:50] <arik> hehe [11:50] <hadess> arik: like "323 Files, 32343 Megs, 787876:00 Play Time" [11:50] <ajmitch> i have rb checked out [11:50] <arik> hadess: right right, makes sense [11:50] <arik> we should add that [11:51] <hadess> 3 different calls for that ? [11:51] <arik> hmm [11:51] <arik> i suppose [11:51] <arik> seems a bit much [11:51] <arik> how about returning a struct? [11:51] <hadess> or one that does everything into a string ? [11:51] <arik> with one call [11:51] <hadess> yeah, ok [11:51] <ajmitch> i'm gonna need far more than a simple playlist thing for my app, i think i'll use a DB for storing info (gotta have decent descriptions, etc of each tape i store) [11:51] <hadess> pffff [11:52] <arik> ajmitch: is this the tape -> mp3 thing? [11:52] <ajmitch> and i'm not planning to use my app for much playback [11:52] <ajmitch> yep [11:52] <ajmitch> finally getting time to work on it [11:52] <arik> this is no mere playlist aj [11:52] <arik> this is the worlds greatest playlist :-P [11:52] <ajmitch> arik: that's good [11:52] <arik> hehe [11:53] <hadess> TapeOff ! [11:53] <ajmitch> hadess: you remember it? [11:53] <ajmitch> hadess: i sorta have the motivation to do it now ;) [11:53] <ajmitch> (and a bit more programming skill) [11:53] <hadess> which else ;) [11:53] <arik> heh [11:54] <ajmitch> arik: hadess thought up the name for me ;) [11:54] <arik> haha [11:54] <arik> tapeOff [11:54] <arik> not bad [11:54] <arik> i am horrible at coming up with names [11:54] Action: ajmitch has some stuff in cvs here [11:54] <hadess> arik: hmm, actually i will add the summary bits to the source info [11:54] <ajmitch> i tend to use my own cvs server for my apps :) [11:54] <arik> hadess: ok, that makes some sense [11:54] <arik> sort of [11:55] <arik> mabe [11:55] <arik> er maybe [11:56] <hadess> brb fag [11:56] <arik> k [11:56] <arik> ajmitch: that means something very different here then he meant it [11:56] <arik> oh btw [11:56] <arik> according to earthlink i have dsl now [11:56] <arik> of course they haven't bothered to ship me my modem yet [11:56] <ajmitch> haha [11:57] <arik> so i guess i'm supposed to just enjoy the idea of having it [11:57] <arik> there order status page says [11:57] <ajmitch> arik: what do you mean, means something different? [11:57] <arik> "ready to be shipped" [11:57] <arik> oh [11:57] <arik> brb fag is not a nice thing to say here :-) [11:57] <arik> it doesn't mean going for a smoke [11:57] <ajmitch> heh [11:58] <ajmitch> i understand what you mean ;) [11:58] <arik> :-) [11:58] Action: arik could sing a building gnome2 from scratch song [11:59] <arik> i'm on orbit2 [11:59] <ajmitch> gtk+ build is failing at the moment, some BS about install pkg-config files [11:59] <arik> yeah [11:59] <arik> i had that problem [11:59] <arik> it's pango [11:59] <ajmitch> oh? [11:59] <arik> as far as i remember anyway [11:59] <arik> i guess we'll see when i get there [12:00] <ajmitch> really doesn't seem to be pango's fault here, will see what i can do tho [12:00] <arik> hm [12:00] <arik> i could be remembering wrong [12:00] <arik> i will find out in a few minutes i suspect :-) [12:01] <ajmitch> cp: `gdk-x11-2.0.pc' and `gdk-2.0.pc' are the same file [12:01] Last message repeated 1 time(s). [12:01] <ajmitch> that's what i get, cos one is symlinked to the other [12:01] <arik> ah ok [12:01] <arik> that's something new [12:01] <ajmitch> i just gotta screw around with it and rm the files ;) [12:01] <arik> well once it works tell me how you did it :-) [12:02] <ajmitch> cd /gnome/head/INSTALL/lib/pkgconfig && rm gdk* && gtk* [12:03] <ajmitch> oops, missed an rm [12:03] <arik> heh [12:03] Action: ajmitch didn't cut & paste that, btw [12:03] <arik> hehe [12:03] <arik> ok [12:03] <arik> i'm on libxslt [12:04] <ajmitch> cool [12:05] <arik> it's a good thing i've played around with rb as much as i have [12:05] <arik> it will make working on this with hadess much easier [12:06] <ajmitch> yeah [12:06] Action: ajmitch is unsure about whether to use the playlist or not [12:06] <arik> heh [12:06] <hadess> arik: does using gal sound good to you ? [12:06] <arik> etable? [12:06] <hadess> yeah [12:06] <arik> yeah [12:07] <arik> i think so [12:07] <hadess> michael already kinda ported it to gnome2 [12:07] <ajmitch> cool [12:07] <arik> well we could use jonathans new tree [12:07] <hadess> so it would be an easy path from gnome1 to gnome2 [12:07] <arik> but i don't know much about it [12:07] <ajmitch> what's the docs like for gnome 2 stuff? [12:07] <arik> almost non existant afaik aj [12:07] <arik> ok i am on gtk+ [12:07] <ajmitch> removing those bad symlinks worked fine btw [12:08] <hadess> i quite like e-table, and gtktreeview sounds like it's much the same, but e-table exists for both platforms... [12:08] <arik> ah ok [12:08] <arik> although gnome1 is gonna go away pretty fast [12:08] <arik> but still [12:08] <arik> why bother changing now [12:08] <arik> etable is fine [12:08] <hadess> cool [12:11] <ajmitch> ok, 2 days of study, 1 exam on saturday, then 3 months off uni [12:11] <arik> hehe [12:12] <ajmitch> i should find time for tapeoff (especially as my flatmate will work on it as well) [12:16] <arik> so how do you want to start hadess? [12:16] <arik> i have time tonight to work on this [12:16] <arik> heh [12:17] <hadess> i'll be starting in a minute adding a couple of files in CVS [12:17] <arik> ok [12:17] <arik> i'll be starting by checking them out then :-) [12:17] <hadess> i'm taking care of some other stuff right, be with it soon ;) [12:17] <arik> sure [12:17] <arik> need to start using your changelog :-P [12:17] <ajmitch> ah well, i only have anoncvs access, it'll be a few hours before i can look at the stuff you add ;) [12:17] <arik> hehe [12:17] <arik> write something [12:17] <arik> anything [12:17] <arik> and get cvs acces [12:17] <arik> s [12:17] <ajmitch> hehe [12:17] <ajmitch> why bother? ;) [12:18] <arik> hehe [12:18] <arik> good point [12:18] <arik> i only got mine when i started working at helix code [12:18] <ajmitch> i'm more into stuff that i'd store on my cvs server [12:18] <arik> hmm [12:18] <ajmitch> a bit of web-based stuff [12:18] <arik> ok [12:19] <ajmitch> and some other stuff, which i don't really care about where it goes [12:24] <arik> woo! [12:24] <arik> gtk compiled perfectly [12:25] <ajmitch> same here [12:25] <arik> yeah [12:25] <arik> but mine did it without the removing of symlinks [12:25] <ajmitch> it just had 2 bad symlinks which it choked on in installation [12:25] <arik> ;-P [12:25] <arik> oh [12:25] <arik> hmm [12:25] <arik> it might still do that then :-) [12:26] <ajmitch> i dunno why, just some cruft from the nightly build i guess (someone changed it) [12:26] <arik> ok [12:26] <arik> well i rm -rf'd the whole thing [12:26] <arik> so maybe i will be alright [12:26] <arik> yep [12:26] <arik> on to bonobo-activation [12:26] <ajmitch> hehe [12:26] <ajmitch> that's one way to do it ;) [12:27] <arik> yep :-) [12:27] <arik> i'm hoping i'll get working system this time :_) [12:27] <arik> ah to dream the impossible dream [12:27] <ajmitch> works for me ;) [12:27] <arik> hehe [12:27] <arik> :-P [12:27] Action: ajmitch has gtk-demo up on the screen [12:28] <arik> i hate you [12:28] <arik> really i do [12:28] <arik> :-) [12:28] <ajmitch> where does one enable AA text tho? ;) [12:28] <arik> not sure [12:30] <hadess> export GDK_USE_XFT='yes yes yes' [12:30] <arik> hehe [12:30] <ajmitch> heh [12:30] <arik> do you have to do three yes's? [12:31] <hadess> no, it just needs to be defined [12:31] <arik> hehe [12:31] <arik> ok [12:31] <ajmitch> yep, that worked [12:32] <ajmitch> using a fugly font tho ;) [12:33] <arik> hehe [12:33] <arik> i will try it if i ever get gtk-demo working [12:33] <arik> i'm on libglade [12:33] <arik> ARGH! [12:33] <arik> still doesn't work [12:33] <arik> gtk-demo (pid:20059): GRuntime-CRITICAL **: gtype.c:1824:g_type_register_static(): initialization assertion failed, use g_type_init() prior to this function [12:33] Last message repeated 1 time(s). [12:33] <arik> gtk-demo (pid:20059): GRuntime-WARNING **: invalid cast from (NULL) pointer to `<invalid>' [12:35] <ajmitch> weird [12:35] <arik> very [12:36] <arik> and very frustrating [12:44] <arik> the weird part is that i can compile gtk2 programs [12:44] <arik> i just can't run them [12:44] <ajmitch> hmm [12:44] <ajmitch> something is quite broken [12:44] <arik> yep [12:44] <arik> and i have no idea what [12:44] <ajmitch> complained in #gtk+ ? [12:44] <arik> no [12:44] <arik> i should [12:55] _gst_newt_ joined #gstreamer. [12:55] <arik> could be [12:55] <ajmitch> you don't tend to get internal error on code syntax errors [12:55] <arik> yeah it is [12:55] <arik> course that doesn't help me :-) [12:55] <ajmitch> recompiling audiofile gives the same error? [12:55] <arik> yep [12:56] <ajmitch> hmm, not good [12:56] <arik> no :-) [12:56] <arik> i am trying again [12:57] <arik> failed again [12:57] <arik> damnit [12:57] <ajmitch> damn [12:57] <arik> i could just not compile it [12:58] <ajmitch> it's not in cvs anymore, so v-b-s doesn't build it [12:58] <arik> ah [12:58] <arik> eazel-hacking fetches the tarball now [12:58] <ajmitch> ok [12:58] <arik> removing the test from the makefile worked [12:58] <arik> so now i move on :-) [12:59] <arik> not that any of this matters [12:59] <arik> if gtk doesn't work it's pretty useless [12:59] Action: ajmitch might get enough money over summer for a *decent* computer ;) [12:59] <arik> heh [12:59] <arik> good luck [12:59] <ajmitch> one that can compile this stuff at a decent speed [12:59] <arik> hehe [12:59] <arik> yeah [12:59] <arik> 4am [12:59] <ajmitch> but i'll probably use this computer here for compiling stuff nightly [12:59] <ajmitch> 4am? [13:00] <ajmitch> 1am here [13:00] <arik> it's 4am here [13:00] <ajmitch> 21 hours difference :) [13:00] <arik> yes :-) [13:00] <ajmitch> might as well be 3 hrs [13:00] <arik> yep [13:00] <arik> effectivly 3 [13:01] <ajmitch> what sorta system you recommend, athlon xp? [13:02] <arik> or p4 [13:02] <ajmitch> flatmate will probably be doing a fair bit of graphics work (eg raytracing) on it [13:02] <arik> neet [13:03] <ajmitch> of course he's gonna get a new box too, so it'll go a bit faster ;) [13:03] <arik> hehe [13:03] <ajmitch> anyway, bedtime, gotta be up by 8am ;) [13:04] <ajmitch> night ;) [13:04] <arik> alright [13:04] <arik> night [13:04] <arik> :-) [13:08] <hadess> arik: do you know how i "declare" a integer in a .idl ? [13:08] <arik> mm [13:08] <arik> i'm not sure [13:08] <arik> i haven't played around with idl stuff that much [13:09] <arik> what are you trying to do? [13:09] <hadess> ok, lemme just commit that [13:09] <arik> ok [13:10] <arik> looks like my gtk2 build has failed again [13:10] <hadess> arik: ok, committed [13:10] <arik> ok [13:10] <arik> let me just update [13:11] <arik> hmm [13:11] <arik> integererface Playlist { [13:11] <arik> ? [13:11] <hadess> oops ;) [13:11] <arik> hehe [13:11] <arik> also [13:11] <arik> you don't have an #endif [13:11] <arik> :-) [13:13] <hadess> fixed [13:13] <hadess> rhythmbox-playlist.idl:62: Error: `integer' undeclared identifier [13:13] <arik> hmm [13:14] <arik> i thought that was how you did it [13:15] <hadess> ha, it's short or long, not integer [13:15] <arik> ah ok :-) [13:21] <arik> we'll be done in no time :-P [13:21] <hadess> it compiles now :) [13:21] <arik> woo! [13:21] <hadess> no GList though... need to find about these ones [13:21] <arik> that will be more complicated [13:22] <hadess> yeah [13:22] <arik> there are docs though [13:22] <arik> is it an any or something? [13:22] Action: arik is building gnome1 so he can build rb [13:22] <hadess> hmm, yeah, any might be alright [13:23] <arik> :_) [13:23] <hadess> yeah, workie [13:24] <hadess> ok, so the .idl is generated alright [13:24] <arik> woo! [13:24] <arik> ok [13:24] <arik> nice [13:24] <arik> now to make an actual control :-) [13:24] <hadess> hehe, i'll try to add some code that does something [13:24] <arik> ok [13:24] <arik> :-) [13:25] <arik> i'm gonna be working on building gnome [13:25] <arik> for the moment [13:46] <hadess> hmmm... i think the best is too start making the playlist as a widget, then turning it into a component... [13:46] <arik> hmm [13:47] <arik> you're prob right [13:47] <hadess> otherwise it will be an oaf mess... [13:47] <arik> yeah [13:47] <arik> let's do it as a widget first [13:52] <arik> it will be easier to do anyway [13:54] <hadess> hmm, actually not exactly :/ [13:54] <arik> oh [13:54] <arik> great :-) [13:54] <arik> what's the problems? [13:54] <hadess> the interface [13:54] <hadess> the types and stuff like that for the interface [13:54] <arik> right [13:55] <arik> what's the problem? [13:55] Action: arik is a bit confused to be honet [13:55] <arik> honest [14:02] <arik> hadess? [14:02] <hadess> si ? [14:02] <arik> what is the problem exactly? [14:02] <hadess> i'm doing a component [14:02] <arik> ok [14:03] <hadess> "lemme hack" ;) [14:03] <arik> hehe [14:03] <arik> ok [14:03] <arik> let me know when i should look at it [14:03] <hadess> yeah, sure [14:05] <hadess> i'm off for lunch [14:05] Nick change: hadess -> hds-food [14:05] <arik> alright [14:05] <hds-food> bbiab [14:05] <arik> i'll be here most likely [14:05] <hds-food> ok [14:14] sienap (sy...@s3...) joined #gstreamer. [14:14] <sienap> http://linuxvideo.org/oms/ ... oms is death ? [14:14] <arik> oms is death? [14:15] <sienap> nah .. unmaintained.. as the site claism [14:15] <arik> ah dead [14:15] <arik> not death :-) [14:15] <sienap> dead that is blah.. :) [14:15] <arik> hehe [14:23] <sienap> <http://www.nluug.nl/events/nj01/fotos/hubbard.jpg> [14:23] <sienap> Last modified: September 7, 2001 [14:23] <sienap> [[Go to: Is Linux een alternatief voor Windows?] ] [[Go to: Rapid Application Development with GNOME and Python]] [[Go to: Index]] [14:23] <sienap> ehm.. [14:23] <sienap> BLAH [14:24] <sienap> Redael .. what should i think of .. ? :) [14:24] <arik> i remember that site [14:24] <sienap> This ehm.. IQ / EM something workbench.. [14:24] <sienap> Emotional Intelligence Workbench to be more exact [14:24] <sienap> ;) [14:24] <arik> heh [14:51] arik (ar...@sd...) left irc: SANDWHICHS! [15:21] arik (ar...@sd...) joined #gstreamer. [15:21] <arik> dum de dum [15:26] Nick change: hds-food -> hadess [15:26] <hadess> lalala [15:26] <hadess> ;) [15:26] <arik> wb hadess [15:26] <arik> :-) [15:26] <arik> i'm still building [15:26] <arik> heh [15:26] <sienap> hey hadess! [15:26] <hadess> hey sienap [15:26] <sienap> hadess any coding lately ? :) [15:27] <arik> heh yes [15:27] <sienap> what ? :) [15:27] <hadess> sienap: yeah, working on the playlist component [15:27] <arik> woo! [15:27] <sienap> for rhythmbox i assume ? :) [15:27] <sienap> hell yeaah! :) [15:27] <arik> and for gstplay [15:27] <sienap> what ? :) [15:27] <arik> yep [15:29] <hadess> sienap has gone def because of too much wanking [15:29] <arik> heh [15:29] <arik> ok [15:30] <sienap> hadess ..gghe.. :) [15:30] <sienap> anyway... ehm who did code WHAT for gstplay ? :) [15:30] <arik> i coded gstplay [15:31] <sienap> btw did someone hear from parapraxis lately ? ... i did mail him but no response yet.. [15:31] <sienap> arik and gstplay being a cool playing toy ? :) [15:31] <arik> it will get better :-) [15:32] <sienap> how far it is for now ? [15:32] <arik> um [15:32] <arik> the backend is getting quite good [15:32] <arik> the frontend needs major work [15:33] <sienap> an backend / frontend seperation ? :) [15:33] <sienap> do you have any screenshots [15:33] <arik> yep [15:33] <sienap> ? [15:33] <arik> um [15:33] <arik> no [15:33] <sienap> make them [15:34] <sienap> >:) [15:34] <arik> heh [15:34] <sienap> *uses his puppy eyes looking cute trick on arik* [15:34] <arik> hehe [15:34] <arik> once the new frontend is in cvs [15:34] <arik> i can take ss's :-) [15:34] <arik> right now it's nothing interesting to look at [15:35] <sienap> now the final quesiton.. gstplay will be an replacement of gstmediaplay ? [15:49] <arik> gstplay is gstmediaplay [15:50] <sienap> oi [15:51] Action: arik is compiling gst [15:57] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:03] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [16:03] Action: arik is installing gst [16:04] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:09] <mattias|na> Do any of you know which version of mpeg2dec is compatible with gst? I saw this thing in the changelogs of mpeg2dec: [16:09] <mattias|na> "renamed libmpeg2dec to libcpuaccel" [16:09] <arik> no idea [16:10] <mattias|na> which explains why the current cvs version of gst wants to link to libmpeg2dec [16:10] <mattias|na> am I right? [16:10] Nick change: mattias|na -> mattias [16:12] <mattias> gonna try and compile an older version of mpeg2dec [16:12] Action: arik is installing ogg [16:12] <mattias> and see if it build libmpeg2dec [16:15] evil_monk (trevo@208.141.162.68) left irc: Ping timeout for evil_monk[208.141.162.68] [16:15] <sxpert> mattias: you probably have an old version laying around (same thing with SDL by the way) [16:15] <sxpert> mattias: however, colorspace seems hopeless [16:16] <sxpert> mattias: I got it to work yesterday with mpeg2dec CVS head ;) [16:17] <hadess> arik: i'm getting there... [16:17] <arik> hadess: oh? [16:17] <arik> wow [16:17] Action: arik is building gal [16:18] <hadess> $ ls [16:18] <hadess> 541k rhythmbox-playlist [16:18] <mattias> sxpert: sorry to say I have no old files laying around [16:18] <arik> hehe [16:18] <arik> wow [16:18] <arik> it will prob get bigger [16:18] <arik> :-) [16:18] <mattias> but what do you mean, "colorspace seems hopeless" [16:18] <mattias> ? [16:18] <hadess> it builds, we just need to implement everything now ;) [16:18] <arik> hehe [16:18] <arik> the fun part! [16:19] <arik> as soon as i have this building i'll check it out [16:19] <hadess> yeah, this stuff is quite a mess :) [16:19] <arik> we should make a todo list and divide it up or something [16:19] <arik> 541k just to build is insane [16:20] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:21] <hadess> arik: 41k stripped... [16:21] <arik> hehe [16:21] <arik> wow [16:21] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [16:21] <hadess> there's nothing but the skels/stubs and other bonobo foo in there [16:21] <arik> uep [16:21] <arik> er yep [16:22] <hadess> i need to do the .oaf file now [16:22] <sienap> 541k..... .. . :) [16:22] <sienap> damn i wish parapraxis came back online [16:23] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:24] <sienap> goodday [16:25] sienap (sy...@s3...) left irc: User abort with 5 Ctrl-C's [16:27] mattias (ma...@ga...) got netsplit. [16:27] mattias (ma...@ga...) returned to #gstreamer. [16:27] <mattias> wb [16:29] <mattias> Apparently the mpeg2dec plugin of gst wants the old LinuxVideo version of mpeg2dec, not the current cvs [16:29] <arik> ARGH! [16:29] <arik> ok [16:29] <arik> rb core dumps on startup for me [16:30] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) got netsplit. [16:30] ds (ds...@c7...) got netsplit. [16:30] taazzzz (dlehn@66.37.66.32) got netsplit. [16:30] hadess (ha...@de...) got netsplit. [16:30] BBB-zZz (BB...@uc...) got netsplit. [16:31] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) returned to #gstreamer. [16:31] ds (ds...@c7...) returned to #gstreamer. [16:31] taazzzz (dlehn@66.37.66.32) returned to #gstreamer. [16:31] BBB-zZz (BB...@uc...) returned to #gstreamer. [16:32] hadess (ha...@de...) returned to #gstreamer. [16:32] <hadess> hmm, crappy network :/ [16:34] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [16:34] <arik> wb hadess [16:34] <arik> hehe [16:35] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:35] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [16:36] <hadess> ok, doing the .oaf file [16:38] <arik> ok [16:38] <arik> i'm still trying to get rb to run [16:38] <hadess> where does it crash ? [16:38] <arik> it core dumps on startip [16:38] <arik> so does gstmediaplay [16:38] sienap (sy...@s3...) joined #gstreamer. [16:39] <arik> i think my gstreamer is b0rked [16:39] <hadess> that might be gst... [16:39] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:39] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [16:46] Action: arik recompiles gst [16:47] Action: mattias is doing make install of gst [16:47] Action: mattias holds breath [16:47] <arik> hehe [16:47] <arik> i've had some problems with my gst for awhile [16:48] <sxpert> mattias: wierd, it worked for me yesterday ;) [16:50] Shippou (zbl...@ot...) left irc: Terminated with extreme prejudice - dircproxy 1.0.0 [16:50] Shippou (zbl...@ot...) joined #gstreamer. [16:53] <mattias> sxpert: that's weird because the only libs built by current cvs are: libcpuaccel and libmpeg2 [16:53] <mattias> but the old mpeg2dec-livid builds these: libmpeg2, libmpeg2dec and libvo [16:53] <sxpert> yeah, and it compiled and searched for mm_acces() in libcpuaccel properly [16:54] <sxpert> s/mm_acces()/mm_accel() [16:54] <mattias> hmm [17:08] Action: hadess pokes arik [17:08] <hadess> arik: so ? [17:08] <arik> yes? [17:08] <arik> still compiling [17:08] <arik> gst takes forever and aday [17:08] <hadess> hehe [17:09] <arik> :-) [17:09] <arik> and prob won't work when i'm done anyway [17:09] <arik> ugh [17:09] evil_monk (trevo@208.141.162.68) joined #gstreamer. [17:12] <arik> i don't actually think it's gst's fault though [17:12] <arik> cause other gst based programs work [17:12] <arik> at least in the past [17:12] <hadess> hmmm [17:12] <arik> but we shall see [17:12] <hadess> yeah [17:12] <arik> how are you coming along? [17:12] <hadess> rebuilding gst [17:12] <arik> heh [17:12] <arik> same boat :-) [17:13] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) got netsplit. [17:13] ds (ds...@c7...) got netsplit. [17:13] taazzzz (dlehn@66.37.66.32) got netsplit. [17:13] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) returned to #gstreamer. [17:13] ds (ds...@c7...) returned to #gstreamer. [17:13] taazzzz (dlehn@66.37.66.32) returned to #gstreamer. [17:14] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [17:17] <arik> still fucking core dumps [17:17] <hadess> :/ [17:18] <hadess> what gcc you're using ? [17:18] <arik> hmm [17:18] <hadess> if you say 3.0 i kick you [17:18] <arik> hehe [17:18] <arik> 2.96 [17:18] <hadess> RH mangled ? [17:18] <arik> yeah prob [17:19] <arik> is that the prob? [17:19] <hadess> i have 2.95.4 here, and it works fine [17:19] <arik> hmm [17:19] <sienap> hadess is there some rhythmbox code online yet ? [17:19] <hadess> gcc 3 problably works better [17:19] <arik> maybe i should downgrade [17:19] <hadess> sienap: gnome cvs [17:19] <arik> or upgrade [17:19] <arik> :-) [17:19] <hadess> heheh [17:19] <arik> i think i'm getting rh 7.2 on thursday [17:20] <sienap> hadess.. ooh let's check some out [17:20] <sienap> under what module name ? [17:20] dobey (do...@dr...) joined #gstreamer. [17:20] <sienap> dobey! [17:20] <arik> you think it's a gcc problem? [17:21] <dobey> sienap: ! [17:23] <arik> ok [17:23] <arik> i'm downloading a new 99 [17:23] <dobey> what's up? [17:23] <dobey> brb [17:23] <arik> er 96 [17:23] <arik> 96-98 instead of -54 [17:23] <arik> maybe that will fix it [17:26] <arik> maybe 2.96-54 had some problem :-P [17:26] <arik> who knows [17:26] wingo (wi...@rd...) joined #gstreamer. [17:26] <hadess> hey wingo [17:26] wingo-zzz (wi...@rd...) left irc: Read error to wingo-zzz[rdu25-21-076.nc.rr.com]: EOF from client [17:26] <wingo> heya [17:33] <arik> course now i have to recomplie gst [17:33] <arik> again [17:33] <arik> yay! [17:41] mattias (ma...@ga...) got netsplit. [17:41] bstard (Lor...@a2...) got netsplit. [17:41] <hadess> hehe [17:43] mattias (ma...@ga...) returned to #gstreamer. [17:47] bstard (Lor...@a2...) got lost in the net-split. [17:47] <wingo> this is *neat*: http://www.cs.brown.edu/software/xmx/ [17:47] Action: arik rebuids gst [17:47] <arik> again [17:47] <arik> it prob won't work [17:47] <arik> again [17:47] <arik> but i do now have a new gcc [17:47] <arik> although still 2.96 [17:50] <arik> but a new rpm version [17:50] <dobey> hah [17:50] <dobey> whee [17:50] <dobey> rpm fun [17:51] <arik> oh yeah [17:51] <arik> rpm [17:51] <arik> 's suck [17:51] <arik> they suck the same as debs [17:51] <arik> but they suck [17:51] <dobey> no [17:51] <dobey> they suck differently [17:51] <arik> yes [17:51] <arik> but equally [17:51] <dobey> nah, dpkg sucks more [17:51] <arik> dpkg is pretty sucky [17:51] <arik> but spec.in files are pretty sucky too [17:51] <dobey> yep [17:51] <dobey> spec.in files aren't rpm [17:52] <arik> true [17:52] <arik> but they make making rpm's suck [17:52] <dobey> no [17:52] <arik> spec files suck [17:52] <dobey> well, they make rpm -tb suck [17:52] <arik> spec.in files suck [17:52] <arik> rpm fucks [17:52] <arik> er sucks [17:52] <arik> excuse me :-) [17:53] <arik> spec files do make -tb suck [17:53] <arik> which is making rpms [17:53] evil_monk (trevo@208.141.162.68) joined #gstreamer. [17:53] <dobey> no [17:53] <dobey> never mind [17:53] <arik> um [17:53] <arik> i know what i'm talking [17:53] <arik> abot [17:53] <arik> i really do [17:53] <dobey> riiiight [17:53] <arik> i used to make helix gnome [17:53] <arik> like all of it [17:53] <arik> all the rpms anyway [17:54] <arik> for all the platforms [17:54] <arik> all i did was make rpms [17:54] <dobey> all i'm saying is [17:54] <dobey> spec files don't suck [17:55] <dobey> rpmbuild does [17:55] <arik> hmm [17:55] <dobey> i have read the code [17:55] <arik> spec files suck too, imho [17:55] <arik> rpmbuild is nasty [17:55] <arik> all of the rpm source code is nasty [17:55] <arik> perl ported to c will prob always be nasty but rpm is particularly nasty [17:55] <dobey> spec files suck because most people don't know how to make them correctly [17:55] <arik> heh [17:55] <arik> spec files also suck cause many programs have sucky build systems [17:56] <arik> and because they just suck [17:56] <arik> :-) [17:56] <arik> but that's just my opinion [17:56] Action: dobey reiterates his previous statement [17:56] <arik> and i'm bitter :-) [17:56] Action: dobey goes to get a snickers [17:56] <hadess> spec files suck because rpm is not nazi enough [17:56] <arik> hehe [17:57] petert (pe...@h0...) joined #gstreamer. [17:57] <arik> hi peter [17:57] <arik> haven't seen you in awhile [17:57] <petert> hi [17:57] <hadess> look at the debian build system, and you see it's very hard to make crappy debs [17:57] <arik> it is possible though hadess [17:57] <arik> look at old mc debs [17:57] <arik> ewww [17:57] <hadess> hehe [17:57] <arik> :-) [17:57] <dobey> hadess: nah, it's pretty easy to make crappy debs [17:57] <dobey> hadess: i could do it in about 10 seconds :-) [17:58] <arik> peter, how are you doing? [17:58] <hadess> dobey: peter's debs are crap... because he doesn't care about the lintian ouput [17:58] <arik> hehe [17:58] <petert> arik: not bad [17:58] <arik> glad to hear it ;-) [17:58] <hadess> dobey: if you do the debs properly, lintian is run on them, and it tells you all the problems with it [17:58] <petert> hadess: we're accepting bug reports [17:58] <arik> heh [17:58] Action: arik steps away from the discussion [17:58] <hadess> ooops ;) [17:59] <hadess> petert: fakeroot lintian usermode*deb | mail pe...@xi... [17:59] <dobey> hehe [17:59] <arik> heh [17:59] <petert> oh, god. usermode. [17:59] <dobey> haha [18:00] <hadess> i must say i didn't look at the other ones [18:00] <hadess> but this one deserves to die ;) [18:00] <arik> hadess, does audioscale compile for you? [18:00] <dobey> hadess: usermode probably will die [18:00] <hadess> arik: everything compiled ok [18:00] <hadess> dobey: i hope so [18:00] <petert> hadess: erm, there is nothing serious in the lintian report there [18:00] <hadess> petert: huh ? [18:00] <petert> hadess: i'm using the potato lintian, though [18:00] <arik> hadess: hmm, i'm getting: libs/resample/resample.h: No such file or directory [18:00] <hadess> petert: ha, potato lintian... [18:01] <dobey> hadess: it's good enough for potato :-) [18:01] <petert> well, no actually. it isn't. [18:01] <arik> hadess, if you have that file can you tell me what it's part of? [18:01] <petert> but it's all i have [18:01] <dobey> yeah [18:01] <dobey> heh [18:02] <hadess> E: usermode: binary-without-manpage consolehelper [18:02] <hadess> E: usermode: binary-without-manpage userhelper [18:02] <hadess> E: usermode: binary-without-manpage userinfo [18:02] <hadess> E: usermode: binary-without-manpage usermount [18:02] <hadess> E: usermode: binary-without-manpage userpasswd [18:02] <hadess> E: usermode: FSSTND-dir-in-usr usr/man/ [18:02] thomasvs (th...@21...) left irc: [x]chat [18:02] <petert> right [18:02] <hadess> plus, no proper builddeps [18:02] <petert> i don't consider missing links to the undocumented man page to be serious issues [18:02] <hadess> i agree [18:02] <petert> ok, yeah. build dependencies are broken across ximian gnome [18:02] <arik> hadess: ? [18:03] <hadess> find -name resample.h [18:03] <hadess> ./libs/resample/resample.h [18:03] <arik> right [18:03] <arik> but what package is it part of? [18:03] <hadess> arik: update your cvs properly ;) [18:03] <hadess> arik: it's in gst [18:03] <arik> heh [18:03] <arik> ah ok [18:03] <arik> i think i did [18:03] <arik> hmm [18:04] <petert> hrm, and that /usr/man is just being picked up because usermode installed empty man directories [18:04] <hadess> petert: build-deps is the biggest grief i have against your debs [18:04] <petert> hadess: yeah, that's definitely a valid gripe [18:04] <arik> hmm [18:04] <arik> i wonder if cvsu isn't doing -dP [18:06] <arik> anyway [18:06] <arik> building again [18:06] <arik> ugh [18:06] <arik> i really should be working hadess [18:06] <arik> but this is more fun :-) [18:07] <arik> which is sad [18:07] <hadess> lol [18:07] <arik> :-) [18:09] <arik> hadess: earthlink claims i have dsl now, but they haven't bothered to send me the modem yet so :-) [18:09] <hadess> ha [18:10] <arik> it makes it harder to use :-) [18:10] <hadess> yeah ;) [18:12] steveb (st...@no...) joined #gstreamer. [18:12] <arik> howdy [18:12] <steveb> hi [18:12] Matty_Boy (Ma...@os...) joined #gstreamer. [18:13] <hadess> heya steveb [18:14] <Matty_Boy> Hi. I am a gstreamer newbie. What would be the best OS for developing with gstreamer. Mandrake or Redhat? [18:14] <hadess> they're the same OS [18:14] <sienap> agrees on that statement [18:14] <sienap> they are indeed ;) [18:15] <Matty_Boy> OK. distribution. [18:15] <hadess> debian ;) [18:15] <sienap> hehe.. hadess is lying! :) [18:15] <sienap> matty i do not think it makes big difference :) [18:15] <arik> hehe [18:15] <arik> redhat! [18:15] <arik> solaris! [18:15] <arik> multics! [18:15] <hadess> most of the packages used by gst are in debian [18:15] <mattias> Debian rulz d0od! :D [18:15] <dobey> uhm [18:16] <sienap> so. [18:16] <Matty_Boy> I was just having a problem getting all the include files in the right place with Mandrake, and wondered if Redhat was setup better initially for compiling against gstreamer? [18:16] <dobey> install redhat or buy a mac and use yellowdog [18:16] <hadess> "mandrake is bad" (tm) [18:16] <arik> hehe [18:16] <dobey> heh [18:16] <arik> mandrake is french and therefore bad :-P [18:16] <sienap> rofl :) [18:16] <dobey> mandrake is very bad for development [18:16] <sienap> .. looks at hadess.. :) [18:17] Action: mattias has touched down, a succesful gst build is functional :) [18:17] Action: hadess hits arik with a michael jackson CD: "Who's bad ?" [18:17] <arik> hahaha [18:17] Action: arik sings billie jean! [18:21] <sxpert> mattias: whick package was the culprit ? [18:21] <sxpert> s/whick/which/ (well that's whicked anyways ;) [18:22] <hadess> arik: dude, do you have the current playlist working in gstplay ? [18:23] <arik> yeah i do [18:23] <hadess> arik: i wanted to know how you handled having 2 different media types in succession [18:23] <arik> hadess: example? [18:23] <arik> oh [18:23] <arik> you mean pipeline wise? [18:23] <hadess> oh, but you told me eos didn't work [18:23] <arik> yeah it doesn't really [18:23] <hadess> nm [18:24] <arik> but my pipeline would support it [18:24] <hadess> can you explain me how you do it ? [18:24] <arik> i null out the pipeline and rebuild it :-) [18:24] <arik> not the best way [18:24] <arik> but prob the only way for now [18:24] Action: hadess is crashing a machine remotely [18:25] <sienap> yay [18:25] <sienap> :) [18:25] <arik> ooh it built [18:25] <hadess> arik: the problem is that at eos, the pipeline would be null'ed from one of its callbacks [18:25] <arik> now to install [18:25] <hadess> arik: and that makes rb crash... [18:25] <arik> hadess: cause of the threads? [18:25] <hadess> nod [18:25] <arik> hmm [18:25] <hadess> wtay said he'd fix it, but it requires some pretty complicated changes iirc [18:25] <arik> you should mention that to omega [18:25] <arik> ah [18:28] <mattias> sxpert: it was mpeg2dec, I had to use the old LinuxVideo build (just checked out mpeg2dec-livid from cvs at sourceforge) [18:31] <arik> ARGH!!!!! [18:31] <arik> still crashes [18:31] <sienap> hehe [18:31] <arik> damnit [18:32] <arik> i give up for now [18:32] <arik> i will try again later [18:32] <arik> see ya all later [18:33] <arik> and hadess i will check out whatever you have in a few hours [18:33] arik (ar...@sd...) left irc: sleeeeeeep [18:33] <hadess> arik: ok [18:33] <dobey> hah [18:33] <hadess> petert: so what are you in for ? :) [18:56] Nick change: dobey -> dobey-eat [19:01] hadess (ha...@de...) left irc: faster ! [19:21] petert (pe...@h0...) left irc: Read error to petert[h00601d1f5850.ne.mediaone.net]: EOF from client [19:24] sienap (sy...@s3...) left irc: User abort with 5 Ctrl-C's [19:34] taazzzz (dlehn@66.37.66.32) left irc: Ping timeout for taazzzz[66.37.66.32] [19:34] taazzzz (dlehn@66.37.66.32) joined #gstreamer. [19:37] wingo (wi...@rd...) left irc: yog [19:37] taazzzz (dlehn@66.37.66.32) left irc: Ping timeout for taazzzz[66.37.66.32] [19:39] taazzzz (dlehn@66.37.66.32) joined #gstreamer. [19:41] evil_monk (trevo@208.141.162.68) left irc: Ping timeout for evil_monk[208.141.162.68] [19:44] evil_monk (trevo@208.141.162.68) joined #gstreamer. [20:20] Nick change: dobey-eat -> dobey [21:29] <dobey> hrmm [21:29] <dobey> bbl [21:29] dobey (do...@dr...) left #gstreamer (eh). [21:30] ChiefHighwater (pa...@te...) joined #gstreamer. [21:45] Zeenix (ze...@13...) joined #gstreamer. [21:49] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-55.brain.net.pk] [22:17] Nick change: taazzzz -> taaz [22:25] Zeenix (ze...@13...) joined #gstreamer. [22:29] steveb (st...@no...) left irc: Ping timeout for steveb[node1ee33.a2000.nl] [22:31] <Zeenix> sxpert: online? [22:32] <sxpert> yep [22:32] <Zeenix> sxpert: got the answers to all the questions [22:32] <Zeenix> ? [22:33] <sxpert> well, I got rid of the colorspace thingie, but that pisses me off, as I can't use some of the (un)maintained examples [22:33] <sxpert> sorry, I'm also trying to get ST: Enterprise ;) [22:34] <Zeenix> sxpert: me too :_ [22:34] <Zeenix> sxpert: ST ? [22:34] <sxpert> channel #enterprise on efnet ;) [23:03] <ds> sxpert: I think people have started using gstreamer-launch instead of the examples [23:04] <sxpert> ds : ah, ok .... [23:09] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-39.brain.net.pk] [23:13] Zeenix (ze...@13...) joined #gstreamer. [23:13] <Zeenix> plz repeat last message for me, if any [23:13] <Zeenix> s/message/messages [23:14] <ds> <sxpert> channel #enterprise on efnet ;) [23:14] <ds> <ds> sxpert: I think people have started using gstreamer-launch instead of the [23:14] <ds> examples [23:14] <ds> <sxpert> ds : ah, ok .... [23:15] <ds> that's it [23:15] <ds> kinda quiet [23:15] <sxpert> man ST: Enterprise Rocks ;) [23:15] <Zeenix> ds: didnt get my question ? [23:16] <ds> no [23:16] <Zeenix> ds: whats the gui-launch, a seperate app ? [23:16] <ds> sxpert: it still is below my television-watching-threshhold =) [23:17] <ds> Zeenix: try it. graphical front end to INFO and DEBUG debugging, apparently [23:23] hadess (ha...@pc...) joined #gstreamer. [23:23] <hadess> heya [23:24] <Zeenix> yo [23:24] <hadess> heya Zeenix [23:25] Matty_Boy (Ma...@os...) left irc: Read error to Matty_Boy[os125.orthoii.com]: EOF from client [23:26] <Zeenix> any one having star virgo or pisces ? [23:28] <hadess> nope [23:29] sienap_ (sy...@s3...) joined #gstreamer. [23:29] <sienap_> hey guys.. [23:29] Nick change: sienap_ -> drunknp [23:29] <drunknp> ehm [23:29] Nick change: drunknp -> drunknap [23:29] <Zeenix> hadess: yours? [23:29] <hadess> sagitarius [23:30] <Zeenix> hadess: yeh, me too [23:30] <hadess> hehe [23:31] <Zeenix> hadess: i've started to believe in it as when it talks of "me", it really does [23:32] <hadess> that's uber bullshit though [23:33] <Zeenix> hadess: go on, you are proving yourself a sagittarian( outspoken ) :) [23:33] <hadess> hehe [23:37] <ChiefHighwater> most of that is just speaking shrouded generalities that are true of most people [23:37] <hadess> indeed [23:38] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-52.brain.net.pk] [23:38] <ChiefHighwater> quess he didn't want to talk about it (or his bill was getting high again) [23:40] <hadess> whatever [23:40] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [23:43] <drunknap> ooh [23:43] <drunknap> i forgot this console [23:43] <drunknap> hi all [23:44] Zeenix (ze...@13...) joined #gstreamer. [23:44] evil_monk (trevo@208.141.162.68) joined #gstreamer. [23:45] <drunknap> hm [23:45] <drunknap> beh [23:45] <drunknap> :) [23:46] <Zeenix> drunknap: really drunk or kidding? [23:47] <drunknap> ejm [23:47] <drunknap> nag let's say [23:47] <drunknap> i am not stable anymore :) [23:49] <drunknap> 'sg zeenix hi [23:49] <drunknap> eh [23:50] drunknap (sy...@s3...) left #gstreamer. [00:00] --- Thu Nov 8 2001 [00:35] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-53.brain.net.pk] [00:36] bstard (Lor...@a2...) joined #gstreamer. [00:38] ChiefHighwater (pa...@te...) left irc: [00:43] drunknap (sy...@s3...) joined #gstreamer. [00:43] <drunknap> o my [00:49] Nick change: taaz -> taaz-away [00:49] chillywilly (da...@d4...) joined #gstreamer. [00:49] <drunknap> hey [00:50] <chillywilly> ello [00:50] <drunknap> ello [01:00] sxpert (sx...@r2...) left irc: Client Exiting [01:02] drunknap (sy...@s3...) left irc: drunknap has no reason [01:03] hadess (ha...@pc...) left irc: mooooh! [01:29] dobey (dobey@24.38.230.22) joined #gstreamer. [02:00] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: Connection reset by peer [02:11] ds (ds...@c7...) left irc: reboot 2.4.14 [02:56] CrunchyBear (na...@53...) joined #gstreamer. |
From: IRC B. <wt...@us...> - 2001-11-09 05:26:19
|
******************************************************************* [03:24] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [03:24] Nick change: taaz-away -> taaz [03:32] CrunchyBear (na...@53...) left irc: [03:40] ds (ds...@c7...) joined #gstreamer. [04:34] taaz (dlehn@66.37.66.32) left irc: Ping timeout for taaz[66.37.66.32] [04:34] dobey (dobey@24.38.230.22) left #gstreamer (eh). [04:34] taaz (dlehn@66.37.66.32) joined #gstreamer. [05:15] omega-als (omega@64.71.176.6) joined #gstreamer. [05:15] <ds> yo! [05:16] <omega-als> yo [05:16] <omega-als> so ds, where are you, again? [05:16] <ds> apparently, about 8 miles north of your [05:17] <ds> s/r$// [05:17] <omega-als> ok, but you're not coming to als? [05:17] <ds> tomorrow, probably [05:17] <ds> if i wake up [05:17] <omega-als> ok, cool. how do I ID you? [05:17] <ds> I have _really_ bleached hair that needs to be cut [05:18] <ds> I'll be hanging around the Debian booth, probably [05:18] <ds> and my number is 510-367-5884 [05:18] dap (dap@SR3745.resnet.ucsb.edu) left irc: Read error to dap[SR3745.resnet.ucsb.edu]: Connection reset by peer [05:18] <omega-als> wireless is pretty touchy around here... [05:20] <omega-als> adding #, sec [05:20] <taaz> omega-als: jeff might be there tomorrow... or friday or saturday ;) [05:20] <taaz> you remember frizzy hair jeff? ;) [05:21] <taaz> i hear your doing a gstreamer bof? [05:21] <ds> omega-als: btw, that's a cell phone [05:24] dap (dap@SR3745.resnet.ucsb.edu) joined #gstreamer. [05:27] Nick change: ds -> ds-coffee [05:30] vektor (reet@HSE-Kitchener-ppp194649.sympatico.ca) left irc: too bad the scene is dead [05:32] <ds-coffee> oh yeah. how do I find you? [05:33] <taaz> he'll be at the gstreamer booth... <g> [05:34] <ds-coffee> there;s a booth? [05:34] <taaz> maybe next year... [05:37] <omega-als> back, jrb and hp hijacked my laptop <g> [05:38] <omega-als> um, I'll be, um... [05:38] <omega-als> (503)539-9852 [05:38] <omega-als> taaz: yeah, remember jeff, and am doing a bof in ~25hrs [05:38] <omega-als> no idea what we'll do, but it depends heavily on who's there [05:39] <taaz> discuss gtk signals in multithreaded apps on SMP systems ;) [05:39] <omega-als> yeah, well, timj isn't here [05:39] <omega-als> but owen is [05:39] <omega-als> in fact, if I want to go try the food at the bar, I could bug him right now [05:39] <taaz> heh... if its fixed in gobject it's all good... else i'll be upset [05:39] <omega-als> taaz: test it [05:40] <omega-als> ds-coffee: oooh, you've been busy [05:40] <taaz> i'm dreadfully busy last week or so... demo of stuff next week [05:42] ds-coffee (ds...@c7...) left irc: Ping timeout for ds-coffee[c799921-a.pinol1.sfba.home.com] [06:09] omega-als (omega@64.71.176.6) left irc: food..... [06:59] walken (fo...@c1...) joined #gstreamer. [06:59] <walken> hi ? [06:59] <ajmitch> hello ? [06:59] <walken> :) [06:59] <walken> just wanted to say hi [07:01] <ajmitch> hehe, ok ;) [07:38] arik (ar...@sd...) joined #gstreamer. [07:38] <arik> hi [07:38] <taaz> yoyoyo [07:38] <arik> yo [07:38] <ajmitch> hey hey arik [07:38] <walken> hi :) [07:38] <taaz> so are you actually working on gstmediaplay? [07:38] <arik> hey aj and walken :-) [07:38] <arik> heh [07:38] <arik> i was all day yesterday [07:38] <walken> hehe [07:38] <taaz> you plan to put that in cvs? [07:39] <ajmitch> arik: how's it coming along? [07:39] Action: taaz wants new code to play with ;) [07:39] <arik> i was working with hadess on the playlist component taaz [07:39] <arik> the fruits of that are in cvs [07:39] <arik> in the rb cvs tree [07:39] <ajmitch> cool , i should cvs up then ;) [07:39] <arik> ajmitch: alright, gst doesn't compile correctly atm for me but... [07:40] <arik> heh [07:40] <arik> it's pretty basic at this point [07:40] <ajmitch> arik: gst doesn't? [07:40] <ajmitch> hmm [07:40] <arik> ajmitch: every app core dumps on startup [07:40] <arik> ajmitch: it could be a gcc problem i don't know [07:41] <ajmitch> doesn't compile for me, but that's cos is apt-get installed libavifile0-dev [07:41] <arik> heh [07:42] <ajmitch> some weird stuff with that, probably just not the right versions or something [07:43] <arik> hmm [07:43] <arik> yeah [07:43] <arik> i'm wondering if gcc-2.96 rh version is broken [07:43] <arik> with gst [07:43] <ajmitch> dunno [07:43] <ajmitch> what does omega use? [07:44] <arik> no idea [07:44] <arik> and i haven't seen him in days [07:44] Action: arik is gonna do a test [07:45] <taaz> omega is at als, was here a while ago [07:46] <ajmitch> ah, ok [07:46] <arik> ah ok [07:47] <ajmitch> heh [07:48] <arik> hmm [07:48] <arik> @GST_PKG_DEPS@ is not being replaced in gstreamer-uninstalled.pc [07:50] Action: ajmitch shrugs :) [07:51] Action: arik recheckes out ALL OF GST [07:51] <arik> *sigh* [07:51] <ajmitch> that's not that much [07:51] <ajmitch> why you doing that? [07:51] <arik> i wonder when my dsl modem is gonna get here [07:51] <arik> um [07:51] <ajmitch> you just trying to clean out the cvs tree? [07:51] <arik> long story [07:51] <arik> heh [07:51] <arik> yeah [07:52] <ajmitch> i like the KDE makefiles, they have a cvs-clean target [07:52] <ajmitch> removes anything not in CVS/Entries [07:52] <arik> interesting [07:52] <ajmitch> should put that in gst's Makefile ;) [07:52] <arik> how does that work? [07:53] <arik> heh [07:53] <ajmitch> paste in /query ? [07:53] <arik> sure [07:57] ChiefHighwater (pa...@te...) joined #gstreamer. [07:57] <ajmitch> hey chief [07:57] <ChiefHighwater> Ello [07:57] <arik> chief [07:57] <ChiefHighwater> 8-] [07:57] <ajmitch> what's up? [07:57] <ChiefHighwater> lookin for Omega, he's at ALS [07:58] <ajmitch> hehe, ok [07:58] <arik> yeah [07:58] <ChiefHighwater> hopin he would log in [07:58] <ajmitch> he did earlier :) [07:58] <ChiefHighwater> rats, and I misssed him [07:59] <ajmitch> was here 2 hrs ago [07:59] Action: arik checks out gst and waits... and waits... and waits... [07:59] <arik> i'm looking forward to the plugins becoming modules [07:59] <ajmitch> [17:38:16] <omega-als> um, I'll be, um... [07:59] <ajmitch> [17:38:22] <omega-als> (503)539-9852 [07:59] <arik> cause honestly i don't need most of them [07:59] <ajmitch> i think that's his phone number [07:59] <ChiefHighwater> yup [07:59] <ChiefHighwater> cell [08:00] <arik> heh [08:00] <arik> i should call him and ask him what gcc he used :-P [08:00] <arik> uses [08:01] <ajmitch> LOL [08:01] <ajmitch> [17:26:43] <jamesh> how's ALS? [08:01] <ajmitch> [17:26:45] <jrb> jamesh: I'm not really here [08:01] <ajmitch> [17:26:51] <jrb> jamesh: I'm on a stolen laptop [08:01] <ajmitch> [17:26:59] <jrb> jamesh: I think I need to go now [08:01] <ajmitch> [17:27:01] <jamesh> the same one hp was on? [08:01] <ajmitch> [17:27:02] <jody_> jrb: not really on irc or not really @ als ? [08:01] <ajmitch> [17:27:12] <jrb> jamesh: I was hp. [08:01] <ajmitch> [17:27:22] <jrb> jody_: we're all at ALS [08:01] <ajmitch> [17:27:43] * jamesh wonders who omega might be :) [08:01] <ajmitch> [17:27:48] <jody_> jrb: we all of you are there, who is doing any work ? [08:01] <ChiefHighwater> he put 2.96 on the machine here [08:01] <ajmitch> [17:27:57] <jrb> jamesh: dunno but he has a big knife [08:01] <ajmitch> [17:28:05] <jrb> jamesh: (actually -- it's the gstreamer guy :-) [08:01] <arik> hahha [08:01] <arik> the gstreamer guy [08:02] <arik> als would have been loads of fun [08:02] <ajmitch> yeah [08:02] <arik> but i just need to work [08:02] <ChiefHighwater> well need to get him to change his nick to "The GStreamer Guy" [08:02] <ajmitch> hehe [08:02] <ajmitch> cos hp logged on as: [08:02] <ajmitch> [17:23:01] --> hp (omega@64.71.176.6) has joined #gnome [08:02] <ChiefHighwater> I think he needs a theme sone and a costume to 8-] [08:02] <ajmitch> hehe [08:03] <arik> hehe [08:03] <ChiefHighwater> I got it! A slug costume 8-] [08:03] <arik> hehe [08:03] <ajmitch> brb, gotta use phone [08:03] <ChiefHighwater> Themesong: Hello World [08:04] <arik> hehee [08:11] <arik> does it go int main (int argc, char **argv) { printf "Hello World!"; return 0; }? [08:26] harobed (harobed@194.51.105.35) joined #gstreamer. [08:26] <arik> brb [08:26] arik (ar...@sd...) left #gstreamer. [08:26] arik (ar...@sd...) joined #gstreamer. [08:26] <arik> back [08:29] <ajmitch> wb [08:29] Action: ajmitch is back too ;) [08:29] <arik> thank you kindly [08:30] <arik> need to run for a minute [08:30] <arik> (returning videos) [08:30] <arik> be back in a bit [08:30] arik (ar...@sd...) left #gstreamer. [08:33] chillywilly (da...@d4...) left irc: [08:38] steveb (st...@no...) joined #gstreamer. [08:40] <ajmitch> hi steveb [08:40] <ajmitch> when do you come to dunedin? [09:02] arik (ar...@sd...) joined #gstreamer. [09:03] <arik> yo [09:05] <ajmitch> yo arik [09:15] Action: arik compiles gst... again :-) [09:15] <ajmitch> hehe [09:15] <ajmitch> you love it ;) [09:15] <arik> hahahahahaha [09:17] Action: ajmitch wonders if he has the right rb source, can't find the latest stuff that's been checked in [09:18] Action: ajmitch uses cvs2cl to see what has been added ;) [09:18] <arik> what's that? [09:18] <ajmitch> ah yep, it's not the gnome cvs version [09:18] <ajmitch> ah, cv2cl grabs all the cvs log messages & produces a changelog from them [09:19] <arik> ah ok [09:19] <arik> neet [09:19] <ajmitch> yup [09:19] Action: ajmitch cvs co's rb [09:19] <arik> heh [09:22] Action: arik builds avifile [09:35] Nick change: taaz -> taazzzz [09:36] <ajmitch> night taazzzz :) [09:39] <arik> heh [09:40] <ajmitch> it's so quiet in here tonight [09:40] <arik> yep [09:40] <arik> no omega [09:40] <arik> no hadess [09:40] <arik> and i'm working :-) [09:41] <ajmitch> heh [09:42] Shippou (zbl...@ot...) left irc: Read error to Shippou[ottawa-hs-64-26-169-108.s-ip.magma.ca]: Connection reset by peer [09:42] Shippou (zbl...@ot...) joined #gstreamer. [09:54] Action: arik wonders why noone has sent wtay a patch for gstwinenc.cc including <avifile/image.h> and decidedes to do so himself [09:56] walken (fo...@c1...) left irc: ircII EPIC4-1.0.1 -- Are we there yet? [09:58] Action: arik sent patch [09:59] Action: ajmitch didn't even bother to look at the gstwinec.cc code to see why it failed to compile ;) [10:00] <arik> hehe [10:00] <arik> ugh [10:00] <arik> now i play with icecast [10:01] <arik> cause that plugin doesn't work either [10:01] <arik> yay [10:01] <arik> the plugins SHOULD REALLY have readme's [10:01] <ajmitch> heh [10:01] <ajmitch> yeah, i see the avi plugin fix was real easy (thx ;) [10:01] <arik> no problem :-) [10:02] <ajmitch> so you want to be the README person, eh? ;) [10:02] <ajmitch> i remember that was discussed quite awhile ago [10:03] <arik> hahaha [10:03] <arik> no [10:03] <arik> i don't want to be [10:03] <arik> i want the plugin authors to be [10:03] <ajmitch> i remember trying to write READMEs for a plugin either the 0.1.1 or 0.2.0 release [10:03] <arik> hehe [10:04] Action: arik updates his libshout [10:04] Shippou (zbl...@ot...) left irc: Getting off stoned server - dircproxy 1.0.0 [10:05] <arik> woo [10:05] <arik> compiles [10:06] <arik> i want a clean build [10:06] Shippou (zbl...@ot...) joined #gstreamer. [10:06] <arik> no not building a plugin [10:06] <arik> or anything like that [10:06] <ajmitch> sorry? [10:06] <arik> what? [10:06] <arik> i'm just saying i'm actually fixing whatever build problems i'm having [10:06] <arik> rather then working around them [10:06] <ajmitch> ah, ok [10:07] <arik> so i have a clean build [10:07] <arik> hopefully it will make it work :-) [10:07] <ajmitch> it helps :) [10:07] <arik> yes [10:07] <arik> :-) [10:08] <ajmitch> i usually always have a clean build [10:08] <arik> you suck :-P [10:08] <ajmitch> i use debian ;) [10:09] <arik> heh [10:09] <arik> i would prob like debian if i knew exactly how it worked [10:09] <ajmitch> easier to keep stuff up to date, i find [10:09] <arik> but i'm to used to redhat [10:09] <ajmitch> yeah [10:09] <ajmitch> i was used to mandrake [10:09] <ajmitch> then i switched ;) [10:09] <arik> i don't like mandrake :-) [10:09] <arik> i hate all the distro's actually [10:09] <ajmitch> hehe, why not [10:09] <arik> they all suck in there own ways :-) [10:09] <arik> well [10:09] <ajmitch> mandrake's QA seems to suck [10:10] <arik> having to package helix gnome for all of them made me hate them all [10:10] <ajmitch> and their GUI tools are hideous [10:10] <arik> yes [10:10] <ajmitch> ah, yes [10:10] <ajmitch> dobey has those sorta feelings about distros [10:10] <arik> heh [10:10] <ajmitch> you are him are just the same ;) [10:10] <arik> i know [10:10] <arik> no [10:10] <arik> we are not :-) [10:10] <ajmitch> s/are/and/ [10:10] <ajmitch> hehe [10:10] <arik> he was arguing with me last night in fact [10:10] <ajmitch> he argues with a few people :) [10:11] <arik> heh [10:11] <arik> some young whippersnapper! he has no idea what it was like back in the day :-) [10:11] <arik> seriously though [10:11] <arik> it totally sucked when i was there [10:11] <arik> i was the only person maintaining all the rpm based distro's [10:11] <arik> with no build tools [10:12] <arik> not fun [10:12] Action: ajmitch read your discussion ;) [10:12] <ajmitch> ouch [10:12] <ajmitch> how the heck did you manage? [10:12] <arik> people would be like "why are the rpm's not as uptodate as the debs?" [10:12] <arik> and i'm like "cause peter does one fucking distro" [10:12] <arik> i do 9 [10:12] <arik> to be honest [10:12] <ajmitch> "arik's not at work yet..." [10:12] <arik> i'm not sure how i manged :-) [10:12] <arik> hehe [10:12] <arik> arik's not at work yet... [10:12] <arik> heh [10:13] <arik> i worked endless hours [10:13] <ajmitch> or did you never leave? ;) [10:13] <arik> hehe [10:13] <arik> not much [10:13] <arik> it got old [10:13] <ajmitch> yeah [10:13] <arik> plus i had almost 0 building experience before hand [10:13] <arik> learned alot [10:13] <ajmitch> maintaining rpms would not be fun ;) [10:13] <arik> the hard way [10:13] <arik> no [10:13] <arik> no it is not :-) [10:13] <ajmitch> yeah, i can imagine you'd need to learn a fair bit ;) [10:13] <arik> heh [10:13] <arik> yes [10:14] <arik> and you get a healthy (and not so healty) hatred for all sorts of things :_) [10:14] <ajmitch> :) [10:14] <arik> heh [10:14] <ajmitch> you hate gstreamer yet? ;) [10:14] <arik> hehe [10:14] <arik> not quite yet :_) [10:14] Action: ajmitch hates exams ;) [10:14] <arik> i'm working on it :-P [10:14] <arik> yeah [10:15] <ajmitch> i really shouldn't be on irc, but oh well.... ;) [10:15] <arik> hehe [10:15] <arik> STUDY! [10:15] <arik> oh well [10:15] <arik> i should be working [10:15] <ajmitch> yeah [10:15] <ajmitch> study is sooo much fun [10:15] <arik> instead of building gst [10:15] <arik> hehe [10:16] <arik> either way [10:16] <arik> core dumping on startupo [10:16] <arik> before loading _anything) [10:17] <ajmitch> damn, that's not good [10:17] <arik> is usually a sign of compiler problems [10:17] <arik> i would think [10:17] <ajmitch> yeah, since noone else seems to get these problems [10:17] <ajmitch> and you're having probs wil gtk+ too [10:17] <arik> right [10:17] <arik> yep [10:17] <arik> i updated my compiler a bit [10:17] <arik> from 2.96-54 to -98 [10:17] <arik> will see if it helps [10:18] <ajmitch> ok [10:18] <arik> prob not [10:18] <ajmitch> it shouldn't affect it too much, i think the main bugs were c++ related [10:18] <arik> hmm [10:18] <arik> i didn't update my g++ [10:18] <arik> i should [10:18] <ajmitch> ok [10:18] <arik> but that wouldn't affect gtj [10:19] <ajmitch> you should grab any other updates like libc6/glibc [10:19] <ajmitch> brb [10:19] <arik> hmm [10:20] <arik> should do [10:23] Action: arik downloads g++ [10:25] <arik> woo [10:25] <arik> done compiling [10:25] <arik> now to install [10:25] <arik> then to watch not work :-) [10:26] <ajmitch> back... [10:26] <arik> wb [10:26] <ajmitch> ok, gst is still building... [10:27] <ajmitch> almost finished (on the editor now) [10:27] <arik> yeah [10:27] <arik> that's near the ned [10:27] <arik> end [10:28] <ajmitch> yep, does it come before or after tests & examples? [10:28] <arik> after [10:28] <ajmitch> hmm, k, i've disabled building the tests, etc [10:28] <arik> hehe [10:28] <ajmitch> done [10:28] <arik> good idea [10:29] <ajmitch> --disable-tests disable building test apps [10:29] <ajmitch> --disable-examples disable building examples [10:29] <arik> ah [10:29] <arik> nice [10:29] <ajmitch> useful for my slow system ;) [10:29] <arik> i should do that [10:29] <arik> yes [10:29] <arik> usefull for anyone [10:29] <ajmitch> was real easy to add to configure.base too [10:29] <arik> must remove a third of the time [10:29] <ajmitch> yeah [10:29] <ajmitch> i got real sick of building them all the time [10:29] ChiefHighwater (pa...@te...) left irc: [10:30] <arik> yep [10:30] <ajmitch> about the only useful patch i committed :) [10:30] <arik> hehe [10:30] <arik> woo [10:30] <arik> done installing [10:30] <ajmitch> excelletn [10:30] <arik> still core dump [10:30] <arik> s [10:30] <ajmitch> crap [10:31] <arik> what's weird is that [10:31] <ajmitch> gdb shows nothing? [10:31] <arik> with prefix=/usr it doesn't core dump [10:31] <arik> and with prefix=/gnome/GNOME1 [10:31] <arik> it does [10:31] <ajmitch> linking in different libraries [10:31] <ajmitch> possibly cos you compiled your own gtk+ 1.2 ? [10:31] <arik> yeah [10:31] <arik> could be [10:31] Action: ajmitch stabs wildly in the dark... [10:32] <arik> hehe [10:32] <arik> i can't get gdb to work with gst [10:32] <ajmitch> works for me [10:32] <arik> hmm [10:32] <arik> works now for me [10:33] <ajmitch> [ajmitch @ ajmitch tools] gdb .libs/gstreamer-register [10:33] <arik> #1 0x4001a048 in gdk_init_check () at eval.c:41 [10:33] <arik> #2 0x403ef351 in gtk_init_check (argc=0xbffff650, argv=0xbffff654) [10:33] <arik> at gtkmain.c:247 [10:33] <arik> #3 0x403ef9d1 in gtk_init (argc=0xbffff650, argv=0xbffff654) at gtkmain.c:471 [10:33] <arik> #4 0x401570f6 in gst_init () from /gnome/GNOME1/lib/libgst.so.0 [10:33] <arik> #5 0x8048b07 in main () at eval.c:41 [10:33] <arik> #6 0x405804b7 in __libc_start_main (main=0x8048af0 <main>, argc=1, [10:33] <ajmitch> yeah [10:33] <arik> ubp_av=0xbffff6b4, init=0x80488a4 <_init>, fini=0x8048c20 <_fini>, [10:33] <arik> rtld_fini=0x4000dbb4 <_dl_fini>, stack_end=0xbffff6ac) [10:33] <ajmitch> in gtk+ as i thought [10:33] <arik> at ../sysdeps/generic/libc-start.c:129 [10:33] <arik> ooh [10:33] <arik> it is a gtk problem [10:33] <arik> hmm [10:33] <arik> well what now [10:33] <arik> fucking a [10:33] <ajmitch> woohoo, wild stab wasn't too far off [10:33] <arik> hehe [10:33] <arik> but what to do now [10:34] <arik> interesting [10:34] <ajmitch> what is eval.c:41 ? [10:34] <arik> i've built gtk1.2 and gtk2 and both fail [10:34] <arik> hmm [10:34] Action: arik goes to look [10:35] <arik> not sure where eval.c is [10:35] <ajmitch> what are you trying to run? [10:36] <arik> gstmediaplay [10:36] <ajmitch> k [10:36] <arik> hmm [10:37] <arik> i can't find eval.c [10:38] <ajmitch> woohoo, massive failure on gstmediaplay [10:38] <arik> hehe [10:38] <arik> i can find gtktypebuiltins_eval.c [10:38] <ajmitch> hmm [10:39] <ajmitch> ok, fixed big bad stuff with re-running gstreamer-register (after rm'ing bad xmms plugin) [10:39] <arik> hmm [10:39] <ajmitch> [ajmitch @ ajmitch gstplay] ./gstmediaplay /home/ajmitch/DAVE\ DOBBYN\ -\ LOYAL.MP3 [10:39] <ajmitch> INFO ( 2748:-1) Initializing GStreamer Core Library [10:39] <ajmitch> INFO ( 2748:-1) CPU features: (808029bf) MMX 3DNOW [10:39] <ajmitch> Gtk-WARNING **: invalid cast from (NULL) pointer to `GtkCList' [10:39] <ajmitch> Gtk-CRITICAL **: file gtkclist.c: line 2622 (gtk_clist_append): assertion `clist != NULL' failed. [10:39] <ajmitch> still get these errors (as if they matter) [10:39] <ajmitch> it still works, and i'm listening to dave dobbyn's tunes now :) [10:40] <arik> oh yeah [10:40] <arik> yep [10:40] <arik> i know what those are [10:40] <arik> don't worry bout em [10:41] <ajmitch> listening to "Loyal", it's quite a good song :) [10:41] <arik> #0 0x40a5932c in ?? () [10:41] <arik> #1 0x4001a048 in gdk_init_check () at eval.c:41 [10:41] <arik> #2 0x406d0351 in gtk_init_check (argc=0xbffff4c8, argv=0xbffff4cc) [10:41] <arik> at gtkmain.c:247 [10:41] <arik> #3 0x406d09d1 in gtk_init (argc=0xbffff4c8, argv=0xbffff4cc) at gtkmain.c:471 [10:41] <arik> #4 0x404236f9 in gnome_add_gtk_arg_callback (con=0x8067ae8, [10:41] <arik> reason=1084592940, opt=0xbffff4c8, arg=0x0, data=0x0) at gnome-init.c:139 [10:41] <arik> #5 0x405dc40f in invokeCallbacksPOST () from /usr/lib/libpopt.so.0 [10:41] <arik> #6 0x405dc3e6 in invokeCallbacksPOST () from /usr/lib/libpopt.so.0 [10:41] <arik> #7 0x405dd37c in poptGetNextOpt () from /usr/lib/libpopt.so.0 [10:41] <arik> #8 0x407e244d in gnomelib_parse_args (argc=1, argv=0xbffff6b4, popt_flags=0) [10:41] <arik> at gnome-popt.c:86 [10:41] <arik> #9 0x40423a26 in gnome_init_with_popt_table (app_id=0x805701b "rhythmbox", [10:41] <arik> app_version=0x8057025 "0.1.0", argc=1, argv=0xbffff6b4, options=0x0, [10:41] <arik> flags=0, return_ctx=0x0) at gnome-init.c:457 [10:41] <arik> #10 0x40423aa3 in gnome_init (app_id=0x805701b "rhythmbox", [10:41] <ajmitch> always fails in the same place? [10:41] <arik> app_version=0x8057025 "0.1.0", argc=1, argv=0xbffff6b4) at gnome-init.c:495 [10:41] <arik> #11 0x804e3a6 in main (argc=1, argv=0xbffff6b4) at main.c:41 [10:41] <arik> #12 0x408914b7 in __libc_start_main (main=0x804e370 <main>, argc=1, [10:41] <arik> ubp_av=0xbffff6b4, init=0x804d3c8 <_init>, fini=0x8056bd0 <_fini>, [10:41] <arik> rtld_fini=0x4000dbb4 <_dl_fini>, stack_end=0xbffff6ac) [10:41] <arik> at ../sysdeps/generic/libc-start.c:129 [10:41] <arik> that's for rb [10:42] <arik> yep [10:42] <arik> in eval.c [10:42] <arik> which doesn't exist as far as i can tell [10:42] <ajmitch> weird [10:42] <ajmitch> you'd think it's in the gdk part of the code [10:43] <arik> yeah [10:43] <arik> hmm [10:43] <arik> but i can't find any eval.c [10:44] Action: arik is installing another gtk+ [10:44] Action: ajmitch searches for eval.c on gnome bonsai [10:44] <arik> thanks [10:47] <ajmitch> http://cvs.gnome.org/bonsai/cvsquery.cgi?treeid=default&module=gtk%2B&branch=gtk-1-2&branchtype=match&dir=&file=eval.c&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=all&mindate=&maxdate=&cvsroot=%2Fcvs%2Fgnome [10:47] <ajmitch> doesn't show up in gtk+ :( [10:47] <arik> yep [10:47] <ajmitch> searching whole repository now [10:48] <arik> ok [10:48] <arik> doubt you will find it [10:48] <arik> seems more like a libc thing [10:49] <ajmitch> yeah [10:49] <ajmitch> this is puzzling [10:49] <arik> yes [10:50] <arik> stripchart/eval.c [10:50] <arik> heh [10:50] <ajmitch> or gnumeric [10:50] <arik> heh [10:50] <ajmitch> or dr-genius... [10:51] <arik> hehe [10:51] <arik> wtf is eval.c [10:52] Action: arik goes to ask in gtk+ [11:19] <arik> /usr/lib/libgdkxft.so.0 => /usr/lib/libgdkxft.so.0 (0x40018000) [11:19] <arik> libgtk-1.2.so.0 => /gnome/GNOME1/lib/libgtk-1.2.so.0 (0x4001d000) [11:19] <arik> libgdk-1.2.so.0 => /gnome/GNOME1/lib/libgdk-1.2.so.0 (0x4015e000) [11:19] <arik> libgmodule-1.2.so.0 => /gnome/GNOME1/lib/libgmodule-1.2.so.0 (0x40193000) [11:19] <ajmitch> umm... [11:19] <arik> heh [11:19] <arik> wrong window [11:19] <arik> ;-) [11:19] <ajmitch> i noticed ;) [11:24] arik (ar...@sd...) left irc: leaving [11:27] ds-coffee (ds...@c7...) joined #gstreamer. [11:28] Nick change: ds-coffee -> ds [11:30] arik (ar...@sd...) joined #gstreamer. [11:30] <arik> woo! [11:31] <arik> brb [11:31] arik (ar...@sd...) left #gstreamer. [11:31] arik (ar...@sd...) joined #gstreamer. [11:31] <arik> woohoo! [11:31] <ajmitch> yay! [11:31] <arik> hehe [11:31] <arik> thanks :-) [11:31] <ajmitch> no probs [11:31] <ajmitch> lupus helped more, i think ;) [11:32] <arik> yes [11:32] <arik> :-) [11:32] <arik> the new gtk looks sweet [11:32] <ajmitch> yeah [11:32] <ajmitch> fonts still look ugly tho ;) [11:32] <arik> hehe [11:32] <arik> they do [11:32] <arik> scrollbars do to [11:32] <ajmitch> yep [11:32] <ajmitch> what 1.2 theme do you use? [11:33] <arik> um [11:33] <arik> cruz [11:33] <arik> er crux [11:33] <ajmitch> same [11:33] <ajmitch> it's quite good [11:34] <arik> yep [11:34] <arik> arlo is very very good [11:34] <arik> shame eazel went under [11:34] <arik> he was gonna do alot more [11:34] <ajmitch> yeah [11:34] <ajmitch> such as? [11:35] <arik> um [11:35] <arik> mozilla theme [11:35] <ajmitch> ok [11:35] <arik> he and i were gonna do a new gtk file open [11:36] <ajmitch> that would have been great [11:36] <ajmitch> many people think the current one is ugly ;) [11:36] <ajmitch> it works [11:37] <arik> yeah [11:37] <arik> it would have been really neet [11:37] <arik> based on user testing [11:37] <arik> and pretty [11:37] <arik> but alas [11:37] <arik> gtktreeview looks nice [11:38] <ajmitch> heh, found maps of dunedin on the net, but can't search by street name [11:38] <ajmitch> wanna see where i live? :) [11:38] <arik> heh [11:38] <arik> sure [11:39] <arik> woo! [11:39] <arik> my hello world works! [11:39] <arik> now i can actually write a new gstplay [11:40] <ajmitch> http://www.atoz-nz.com/Dunedin/map03.html [11:40] <arik> ok [11:40] <ajmitch> that's the university campus [11:40] <arik> ok [11:40] <ajmitch> i'm in building 29 :) [11:40] <arik> i see [11:41] <ajmitch> (middle right) [11:41] <arik> ah [11:41] <arik> ok [11:41] <ajmitch> nice & close to it all [11:41] <ajmitch> next year i won't be so close [11:42] <arik> ok [11:42] <arik> that sucks [11:42] <ajmitch> not really [11:42] <ajmitch> will be flatting, and get DSL ;) [11:42] <ajmitch> http://www.atoz-nz.com/Dunedin/map01.html#Map2 [11:43] <ajmitch> i'll be near that red dot in the lower right corner [11:43] <arik> hehe [11:43] <arik> ok [11:43] <ajmitch> it's not far from campus :) [11:45] <arik> good :-) [11:45] <arik> ok [11:45] <arik> now i need to compile gst for gnome2 [11:45] <ajmitch> cool, it's not hard : [11:45] <ajmitch> ;) [11:45] <arik> hehe [11:45] <arik> should be pretty easy [11:45] <ajmitch> --enable-glib2 [11:45] <arik> oh [11:45] <arik> ok [11:45] <ajmitch> is all you need i think (and maybe --prefix) [11:45] <arik> yeah [11:46] <arik> woo [11:46] <ajmitch> of course the stuff like gstmediaplay, gsteditor, will fail to build [11:46] <arik> yep [11:46] <arik> course [11:46] <ajmitch> they should be disabled in the makefile [11:46] <arik> shouldn't even attempt to build em [11:46] <ajmitch> dunno if they are [11:46] <arik> hopefully [11:47] <ajmitch> hopefully not hard to fix up [11:47] <arik> pretty big school [11:47] <ajmitch> yeah about 17000 students [11:47] <arik> wow [11:47] <arik> my school has um [11:47] <arik> 600 [11:47] <arik> something like that [11:47] <ajmitch> plus the students at otago polytechnic, and the college of education, would be about 23000 in total [11:47] <arik> wow [11:48] <ajmitch> guess what the city population is :) [11:48] <arik> um [11:48] <arik> 1billion people? [11:48] <ajmitch> hah [11:48] <ajmitch> funny [11:49] <ajmitch> i dunno of any city with 1billion ppl ;) [11:49] <arik> nope [11:49] <arik> none [11:49] <arik> hmm [11:49] <arik> aclocal: configure.ac: 218: macro `AM_PATH_GLIB' not found in library [11:49] <arik> aclocal: configure.ac: 221: macro `AM_PATH_GTK' not found in library [11:49] <ajmitch> hmm [11:50] <arik> wonderful [11:50] <ajmitch> what's your --prefix ? [11:50] <arik> /gnome/GNOME2 [11:51] <ajmitch> tri export ACLOCAL_FLAGS="-I /gnome/GNOME2/share/aclocal/" [11:51] <ajmitch> s/tri/try/ [11:51] <arik> hmm [11:51] <ajmitch> dunno if that works [11:51] <arik> first i'm gonna check it out again from cvs [11:53] Action: ds things users suck [11:54] <ajmitch> i'll try a build here [11:54] <ds> thinks, too [11:54] <arik> ok [11:54] <ds> user to ds: I have this problem [11:54] <ds> ds to user: send me the output of X [11:54] <ds> ... three days [11:55] <ds> user to list: I have this problem [11:55] <arik> hehe [11:55] <arik> arik to earthlink: send my my damn dsl modem [11:55] <arik> earthlink to arik: no [11:55] <arik> arik to earthlink: you suck [11:56] <ajmitch> aclocal: configure.ac: 226: macro `AM_PATH_GLIB' not found in library [11:56] <ajmitch> aclocal: configure.ac: 229: macro `AM_PATH_GTK' not found in library [11:56] <ajmitch> nope, i get the same ;) [11:56] Action: ajmitch tries to remember the fix for this [11:56] <arik> heh [11:56] <arik> i hope you remember [11:57] <ajmitch> i've built it fine before :) [11:57] <arik> heh [11:58] <ajmitch> aclocal accepts the following options: [11:58] <ajmitch> --acdir=DIR [11:58] <ajmitch> Look for the macro files in DIR instead of the [11:58] <ajmitch> installation directory. This is typically used for [11:58] <ajmitch> debugging. [11:58] <ajmitch> -I DIR Add the directory DIR to the list of directories [11:58] <ajmitch> searched for .m4 files. [11:59] <ajmitch> doesn't mention environment variables [11:59] <arik> wonderful [11:59] <ajmitch> nah, ACLOCAL_FLAGS is substituted in by autogen.sh [12:00] <ajmitch> should be :) [12:00] <arik> hehe [12:02] <ajmitch> trying again... [12:03] <ajmitch> wow, it's definitely checking that dir [12:04] <ajmitch> i get 1001 'duplicated macro' warnings [12:04] <arik> hmm [12:04] <arik> wonderful [12:04] <arik> i remember those [12:05] <ajmitch> but... [12:05] <ajmitch> it shouldn't be checking for those macros [12:06] <ajmitch> shoudl only check for those if using glib & gtk+ 1.2 [12:06] <arik> hmm [12:06] <ajmitch> read configure.base, you'll see what i mean [12:07] <arik> i'm still waiting for the checkout to finish [12:08] <ajmitch> ok [12:09] <ajmitch> you don't mind me pasting a few (10 or so) lines, right? ;) [12:09] <arik> hehe [12:09] <arik> go right ahead [12:09] <ajmitch> hmm, not here, it's a bit more than 10... ;) [12:10] <arik> heh [12:18] harobed (harobed@194.51.105.35) got netsplit. [12:18] mattias (ma...@ga...) got netsplit. [12:18] <arik> i'm still checking out [12:20] <ajmitch> heh [12:22] <ajmitch> it's rather interesting [12:22] <arik> hmm? [12:22] <ajmitch> if i use the environment for v-b-s, it fails [12:22] <arik> ooh! it's checkedout [12:22] <ajmitch> if i use my normal environment, it works [12:23] <arik> ooh [12:23] <arik> what was that command again? [12:23] <ajmitch> PKG_CONFIG_PATH="/gnome/head/INSTALL/lib/pkgconfig" ./autogen.sh --prefix=/gnome/head/INSTALL --enable-glib2 --disable-docs-build --disable-tests --disable-examples [12:23] <ajmitch> for me :) [12:23] <arik> i'm gonna try it [12:24] mattias (ma...@ga...) got lost in the net-split. [12:24] harobed (harobed@194.51.105.35) got lost in the net-split. [12:24] <ajmitch> cool [12:24] <ajmitch> it should work for you :) [12:25] mattias (ma...@ga...) joined #gstreamer. [12:25] harobed (harobed@194.51.105.35) joined #gstreamer. [12:25] <arik> woo [12:25] <arik> nope [12:25] <arik> not working so far [12:25] <ajmitch> what fails? [12:25] <ajmitch> same stuff? [12:25] <ajmitch> paste me your 'export' results [12:25] <arik> /gnome/GNOME2/lib/pkgconfig [12:26] <ajmitch> not just that [12:26] <arik> not sure what you want [12:26] <ajmitch> the whole env if you want [12:26] <ajmitch> mine fails in one case but not the other, i dunno why [12:26] <arik> PWD=/gnome-source/GNOME2/gstreamer [12:26] <arik> GNOME_INSTALL=/gnome [12:26] <arik> MOZILLA_CVSROOT=:pserver:ano...@cv...:/cvsroot [12:26] <arik> GNOME_CVSROOT=:pserver:ar...@cv...:/cvs/gnome [12:27] <arik> WINDOWID=31457535 [12:27] <arik> PAGER=less [12:27] <arik> EAZEL_USER_DIR=/home/arik/.eazel [12:27] <arik> HOSTNAME=localhost [12:27] <arik> HISTFILESIZE=100000 [12:27] <arik> LD_LIBRARY_PATH=/gnome-source/MOZILLA/mozilla/dist/bin:/gnome/GNOME2/lib:/gnome/GNOME1/lib [12:27] <arik> EAZEL_CVS_COMMAND= cvs -z3 [12:27] <arik> EAZEL_SERVICES=1 [12:27] <arik> QTDIR=/usr/lib/qt-2.2.0 [12:27] <arik> EAZEL_HACKING_MODULES_DIR=/etc/eazel/modules [12:27] <arik> LESSOPEN=|/usr/bin/lesspipe.sh %s [12:27] <arik> NAUTILUS_DEBUG=1 [12:27] <arik> MOZILLA_FIVE_HOME=/gnome-source/MOZILLA/mozilla/dist/bin [12:27] <arik> KDEDIR=/usr [12:27] <arik> EAZEL_BUILD_MODE=GNOME2 [12:27] <arik> USER=arik [12:27] <arik> LS_COLORS=no=00:fi=00:di=01;34:ln=01;36:pi=40;33:so=01;35:bd=40;33;01:cd=40;33;01:or=01;05;37;41:mi=01;05;37;41:ex=01;32:*.cmd=01;32:*.exe=01;32:*.com=01;32:*.btm=01;32:*.bat=01;32:*.sh=01;32:*.csh=01;32:*.tar=01;31:*.tgz=01;31:*.arj=01;31:*.taz=01;31:*.lzh=01;31:*.zip=01;31:*.z=01;31:*.Z=01;31:*.gz=01;31:*.bz2=01;31:*.bz=01;31:*.tz=01;31:*.rpm=01;31:*.cpio=01;31:*.jpg=01;35:*.gif=01;35:*.bmp=01;35:*.xbm=01;35:*.xpm=01;35:*.png=01;35:*.tif [12:27] <arik> GNOMAD_SKIP_INTRO_MESSAGE=1 [12:27] <arik> MACHTYPE=i386-redhat-linux-gnu [12:27] <arik> OMF_DIR=/gnome/GNOME2/share/omf:/usr/share/omf [12:27] <arik> GNOME_SOURCE=/gnome-source [12:27] <arik> MAIL=/var/spool/mail/arik [12:27] <arik> INPUTRC=/etc/inputrc [12:28] <arik> CVS_RSH=ssh [12:28] <arik> OLDPWD=/gnome-source/GNOME2 [12:28] <arik> BASH_ENV=/home/arik/.bashrc [12:28] <arik> LANG=en_US [12:28] <arik> EAZEL_HACKING_DATA_DIR=/etc/eazel [12:28] <arik> EAZEL_USER_PROFILE=/home/arik/.eazel/eazel_arik_profile [12:28] <arik> COLORTERM=gnome-terminal [12:28] <arik> DISPLAY=:0 [12:28] <arik> FIGNORE=.o:/home/arik:.elc [12:28] <arik> LOGNAME=arik [12:28] <arik> SHLVL=4 [12:28] <arik> GNOME_DISABLE_CRASH_DIALOG=1 [12:28] <arik> SESSION_MANAGER=local/localhost:/tmp/.ICE-unix/6430 [12:28] <arik> SHELL=/bin/bash [12:28] <arik> USERNAME= [12:28] <arik> CVSEDITOR=emacs -nw [12:28] <arik> PKG_CONFIG_PATH=/gnome/GNOME2/lib/pkgconfig [12:28] <arik> HOSTTYPE=i386 [12:28] <arik> CVSROOT=:pserver:ar...@cv...:/cvs/gnome [12:28] <arik> OSTYPE=linux-gnu [12:28] <arik> HISTSIZE=100000 [12:28] <arik> HOME=/home/arik [12:28] <arik> TERM=xterm [12:28] <arik> PATH=/usr/bin/eazel-hacking-hacks:/gnome/GNOME2/bin:/gnome/GNOME1/bin:/usr/local/bin:/bin:/usr/bin:/usr/X11R6/bin:/home/arik/bin [12:28] <arik> OAF_DEBUG_OUTPUT=1 [12:29] <arik> BONOBO_ACTIVATION_DEBUG_OUTPUT=1 [12:29] <arik> _=/usr/bin/env [12:29] <arik> wow [12:29] <arik> that's a flood :-) [12:29] <ajmitch> hehe [12:29] <arik> PKG_CONFIG_PATH="/gnome/GNOME2/lib/pkgconfig" ./autogen.sh --prefix=/gnome/GNOME2 --enable-glib2 --disable-docs-build --disable-tests --disable-examples [12:29] <arik> that's the command i'm running [12:29] <arik> :-) [12:29] <ajmitch> it's only happening cos pkg-config is failing to pickup those packages [12:29] <arik> ok [12:30] <ajmitch> but pkg-config --libs glib-2.0 works fine [12:30] <arik> yep [12:30] <arik> -L/gnome/GNOME2/lib -lglib-1.3 [12:31] <ajmitch> fsck it, what is going on? [12:31] <arik> no idea :-) [12:31] <arik> that's your job [12:31] <ajmitch> export PATH=/usr/bin:$PATH [12:31] <ajmitch> then it works [12:32] <arik> hmm? [12:32] Action: ajmitch shrugs [12:32] <ajmitch> if i use the pacakged pkg-config, it goes, if i use the gnome-installed pkg-config, it fails [12:32] <arik> oh [12:32] <arik> weird [12:32] <ajmitch> quite [12:32] <ajmitch> does it work for you now? [12:33] <arik> if i export the path? [12:33] <ajmitch> yeah [12:33] <arik> um [12:33] <ajmitch> if you put /usr/bin first [12:33] <arik> yeah it looks like it does [12:34] <ajmitch> weird stuff, this... [12:34] <arik> it si [12:34] <arik> it is [12:35] <arik> it is compiling [12:35] <ajmitch> cool [12:36] hadess (ha...@de...) joined #gstreamer. [12:36] <hadess> heya [12:36] <ajmitch> hey hadess [12:37] <hadess> heya aj [12:37] <arik> heya haddess [12:37] <arik> i got rb running! [12:37] <hadess> 'lo arik [12:37] <hadess> wd :) [12:37] <ajmitch> arik & i are just having loads of fun with glib 2 stuff & building gstreamer :) [12:37] <hadess> hehe, pretty straight forward [12:37] <arik> hehe [12:37] <ajmitch> if only it were [12:38] <ajmitch> pkg-config is acting weird [12:38] <arik> rb doesn't actually do anything though [12:38] <hadess> arik: dnd a song in the library [12:38] <arik> oh [12:38] <arik> dnd [12:38] <hadess> arik: then press play [12:38] <ajmitch> :) [12:38] <arik> i don't run any app to dnd it from ;-) [12:38] <hadess> yeah, i didn't have time to implement anything else [12:38] <arik> but i will try [12:38] <ajmitch> what do i need for rb to build? lotsa cvs libraries? [12:38] <hadess> no [12:39] <ajmitch> that's good :) [12:39] <arik> mostly just gal [12:39] <hadess> i need to change the REQUIREMENTS though [12:39] <hadess> and libcdaudio [12:39] <arik> ah yeah [12:40] <arik> hmm [12:40] <arik> didn't quite work [12:40] <hadess> what did it do ? [12:40] <arik> i just don't hear anything [12:40] <arik> it says it's playing [12:40] <hadess> you use oss ? [12:40] <arik> no [12:40] <arik> alsa [12:41] <arik> but with oss faking [12:41] <hadess> weird, should work... [12:41] <hadess> but well, you get the point [12:41] <arik> yep [12:41] <arik> :-) [12:42] Action: arik is compiling gst for glib [12:42] <arik> 2 [12:49] <arik> wow [12:49] <arik> gst compiles WAY faster with that stuff aj [12:50] <ajmitch> yeah [12:50] <arik> thatnks [12:52] <ajmitch> no probs ;) [12:55] <arik> hmm [12:55] <arik> after installing gst [12:55] <ajmitch> oh? [12:55] <arik> and running pkg-config --list-all [12:55] <arik> i get [12:55] <arik> Package libxml was not found in the pkg-config search path. [12:55] <arik> Perhaps you should add the directory containing `libxml.pc' [12:55] <arik> to the PKG_CONFIG_PATH environment variable [12:55] <arik> Package 'libxml', required by 'GStreamer', not found [12:55] <ajmitch> umm [12:55] <hadess> upgrade your libxml [12:55] <ajmitch> could be because it's using /usr/bin/pkg-config [12:55] <arik> ah yeah! [12:55] <arik> that's right [12:56] <arik> no [12:56] <arik> i reset my path [12:56] <ajmitch> hmm, ok [12:56] <ajmitch> ls -l /gnome/GNOME2/lib/pkgconfig/gst* [12:56] <arik> i have libxml2 [12:56] <ajmitch> this is sooo slow [12:56] <arik> i have gstreamer.pc [12:57] <ajmitch> which pkg-config [12:57] <hadess> you need libxml1 iirc [12:57] <arik> ah [12:57] <hadess> locate libxml.pc [12:57] <hadess> /usr/lib/pkgconfig/libxml.pc [12:57] <arik> hmm [12:58] <hadess> dpkg -S `locate libxml.pc` [12:58] <hadess> libxml1: /usr/lib/pkgconfig/libxml.pc [12:58] <arik> ok [12:58] Last message repeated 1 time(s). [12:58] <arik> what module is it [12:58] <arik> ? [12:58] <ajmitch> hadess: the poor person doesn't use debian ;) [12:58] <hadess> dunno [12:58] <hadess> ajmitch: what a mistake ;) [12:58] <arik> hmm [12:58] <ajmitch> hehe [12:59] <hadess> btw, you're still working on the packages ? [12:59] <arik> i have libxml [12:59] <arik> hmm? [12:59] <arik> libxml2 anyway [13:00] <hadess> you need libxml1, capish ? [13:00] <arik> hehe [13:00] <arik> i get it :-) [13:00] <arik> but i wonder why eazel-hacking doens't grab it [13:00] Action: hadess hits arik with a DV [13:00] <arik> hmm [13:00] <arik> it's in the same dam module [13:00] <arik> it's in the same damn module [13:00] <arik> gnome-xml [13:00] <arik> well great [13:01] <hadess> use cvs to see which branch it is, and modify the modules.sh file to take it into account [13:02] <arik> i need both :-) i'll just copy it to a gnome-xml1 dir [13:04] <arik> ok [13:04] <arik> works [13:05] <arik> woo [13:05] <arik> gstreamer for gnome2! [13:05] <ajmitch> cool [13:05] <ajmitch> now could you be kind & donate some money o me for a faster computer? ;) [13:06] <arik> hehe [13:06] Action: ajmitch is poor student ;) [13:06] Action: ajmitch senses an overwhelming lack of pity [13:06] <arik> heh [13:06] <arik> yes :-) [13:07] <ajmitch> curses ;) [13:07] <arik> how does one build a gnome2 app? [13:07] <arik> pkg-config --cflags --libs what? [13:08] <ajmitch> ask hadess, i'm going to bed ;) [13:08] <arik> cause i'm getting weird errors [13:08] <arik> hehe [13:08] <arik> night aj [13:08] <ajmitch> early start in the morning [13:08] <arik> /tmp/ccPalYXr.o(.text+0x2c): undefined reference to `gnome_init' [13:08] <arik> /gnome/GNOME2/lib/libbonoboui-2.so: undefined reference to `Bonobo_Control_setWindowId' [13:08] <arik> collect2: ld returned 1 exit status [13:08] <ajmitch> cy'll l8r ;) [13:08] <arik> usr/bin/ld: warning: libbonobo-activation.so.2, needed by /gnome/GNOME2/lib/libgnomeui-2.so, may conflict with libbonobo-activation.so.3 [13:08] <arik> later [13:08] <hadess> PKG_CHECK_MODULES(GNOME_UTILS, libgnome-2.0 >= $LIBGNOME_REQUIRED libgnomeui-2.0 >= $LIBGNOMEUI_REQUIRED) [13:08] <hadess> AC_SUBST(GNOME_UTILS_CFLAGS) [13:08] <hadess> AC_SUBST(GNOME_UTILS_LIBS) [13:09] <hadess> stuff like that [13:09] <arik> right [13:09] <arik> i know [13:09] <arik> i am using [13:09] <arik> both of those [13:09] <arik> oh well [13:09] Action: arik goes to look for docs [13:12] <arik> ugh [13:12] <arik> food time [13:12] <arik> back later :-) [13:12] arik (ar...@sd...) left irc: leaving [14:33] Nick change: hadess -> hds-food [15:28] Nick change: hds-food -> hadess [15:52] evil_monk (trevo@208.141.162.68) joined #gstreamer. [16:04] Shippou (zbl...@ot...) left irc: I shouldn't really be here - dircproxy 1.0.0 [16:04] Zygo (zbl...@ot...) joined #gstreamer. [16:10] Zygo (zbl...@ot...) left irc: Read error to Zygo[ottawa-hs-64-26-169-108.s-ip.magma.ca]: EOF from client [16:11] Zygo (zbl...@ot...) joined #gstreamer. [16:27] MM (co...@39...) joined #gstreamer. [16:27] MM (co...@39...) left irc: Killed (NickServ (This nick is reserved by another user)) [16:28] _MM_ (co...@39...) joined #gstreamer. [16:42] Nick change: mattias -> mattias|na [16:50] Action: BBB-zZz is back (gone 40:52:18) [16:50] Nick change: BBB-zZz -> BBB [17:57] harobed (harobed@194.51.105.35) left irc: Client Exiting [19:03] hadess (ha...@de...) left irc: Client Exiting [19:13] _MM_ (co...@39...) left #gstreamer. [19:14] Nick change: Zygo -> Shippouh [19:14] Nick change: Shippouh -> Shippou [19:18] wingo (wi...@rd...) joined #gstreamer. [19:26] CrunchyBear (na...@53...) joined #gstreamer. [19:38] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [20:04] CrunchyBear (na...@53...) left irc: Ping timeout for CrunchyBear[5399-3-p5.ras.network-i.net] [20:19] nash (yar@203.128.13.31) joined #gstreamer. [20:26] ChiefHighwater (pa...@te...) joined #gstreamer. [20:35] evil_monk (trevo@208.141.162.68) joined #gstreamer. [20:48] _gst_newt_ joined #gstreamer. [20:59] omega (omega@64.71.176.6) joined #gstreamer. [20:59] Nick change: omega -> omega-als [21:01] <taazzzz> yo als boy [21:01] <omega-als> yo [21:01] Nick change: taazzzz -> taaz [21:02] <taaz> having fun there? [21:04] <ChiefHighwater> ello 8-] [21:15] Zeenix (ze...@13...) joined #gstreamer. [21:17] <Zeenix> hello all [21:17] <ChiefHighwater> Ello [21:21] <Zeenix> is there an online system like gstreamer-inspect ? [21:29] <omega-als> nope, but we could do something statically easily [21:29] <omega-als> ..enough [21:29] <omega-als> I can try it out in a little while [21:34] omega-als (omega@64.71.176.6) left irc: Ping timeout for omega-als[64.71.176.6] [21:47] omega-als (om...@pu...) joined #gstreamer. [21:52] sienap (sy...@s3...) joined #gstreamer. [21:52] <sienap> Hey.. all :) [21:52] <ChiefHighwater> ello [21:52] <sienap> NLUUG autumn conference 2001 has ben way cool! [21:52] <sienap> met jorn (from galeon) and dirk jan c binnema and some other people [21:52] <sienap> had a cool time today [21:55] <BBB> sienap: you're a member there? [21:55] <BBB> cool :) [21:55] <sienap> no i am not ;) but that doesn't make any difference [21:55] <sienap> the najaars conferencie was just way awesome! :) [21:55] <BBB> :) [21:55] Action: BBB is NLLGG member [21:55] <sienap> we really had a good time.. and good speakings as well! [21:55] <sienap> he ok :) [21:56] <sienap> Christian Egli had a cool presentation [21:56] <sienap> about rad development with gnome / glade / phython [21:56] Action: BBB will go follow the nluug as well in the future [21:56] <sienap> anyway his presentation was GREAT! :) [21:56] <BBB> nice :) [21:56] <BBB> sounds all pretty interesting :) [21:57] <BBB> really a development meeting, or also for "simple" unix users? [21:57] <sienap> yeah it was cool [21:57] <sienap> nah for sure not aimed at simple users [21:57] <BBB> hmm...... [21:57] <sienap> the topic was "desktop on unix" [21:57] <BBB> sounds good :D [21:57] <sienap> but there were some good technical presentations [21:57] <BBB> desktop on unix is also the main topic on the NLLGG mailinglists now :) [21:58] <sienap> and Jordan Hubbard did a talk about Mac os X.. really intresting as well [21:58] <sienap> he is a good speaker [21:58] <BBB> mac os X is actually quite interesting [21:58] <sienap> and jeroen blaten did the key word ( actually it is baten.. anyway he blaats ) :) [21:58] <sienap> but i was up to checking some news site [21:58] <BBB> lol [21:58] <sienap> so going to do that first ;) [21:58] <BBB> go ahead ;) [22:06] wingo (wi...@rd...) left irc: going home, yo [22:06] <Zeenix> i need mp3parse to timestamp the outgoing buffers, any idea how to ? [22:06] omega-als (om...@pu...) left irc: Ping timeout for omega-als[public-nat.als01.linuxshowcase.org] [22:21] omega-als (omega@64.71.176.6) joined #gstreamer. [22:22] <ds> yo [22:22] Action: omega-als throws something at wtay [22:22] <omega-als> ds: so, you're not here... [22:22] <Zeenix> yo [22:22] <Zeenix> his spirit is [22:23] <ds> I typically wake up at 1pm. today is no different =) [22:23] <omega-als> heh [22:23] Action: Zeenix picks that thing & throws at wtay again [22:24] taaz (dlehn@66.37.66.32) left irc: Ping timeout for taaz[66.37.66.32] [22:24] <Zeenix> omega-als: the rtcp is also working now [22:24] <omega-als> cool [22:24] taaz (dlehn@66.37.66.32) joined #gstreamer. [22:27] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-19.brain.net.pk] [22:27] taaz (dlehn@66.37.66.32) left irc: Ping timeout for taaz[66.37.66.32] [22:30] omega-als (omega@64.71.176.6) left irc: [x]chat [22:30] omega_ (omega@64.71.176.6) joined #gstreamer. [22:32] bstard (Lor...@a2...) joined #gstreamer. [22:41] taaz (dlehn@66.37.66.32) joined #gstreamer. [22:43] Zeenix (ze...@13...) joined #gstreamer. [22:46] <Zeenix> omega_: you wanted to tell me abuot timstamping the buffers in mp3parse [22:46] <omega_> huh? [22:47] <Zeenix> omega_: infact it was my question you said you'll answer after... [22:47] <omega_> um, what was the question? [22:47] <Zeenix> omega_: how to timestamp the buffers? [22:48] <omega_> um, you set the timestamp. GST_BUFFER_TIMESTAMP(buf) = ... [22:49] <Zeenix> omega_: that simple? [22:49] <omega_> yes [22:50] <ChiefHighwater> GStreamer:making the hard things simple, and the impossible possible 8-] [22:51] <Zeenix> omega_: i dont know why but the disksrc/filesrc sucks, on my PC atleast [22:51] <omega_> in what way? [22:51] <Zeenix> omega_: core-dumps [22:51] <Zeenix> err [22:51] <Zeenix> omega_: seg-faults [22:51] <Zeenix> after some reading [22:52] <Zeenix> omega_: -launch disksrc location=me.wav ! gsmenc ! rtpsend payload_type=gsm [22:52] <omega_> Zeenix: have you confirmed that the segfault is in disksrc? [22:53] <Zeenix> omega_: there are some other errors that seems from disksrc/filesrc [22:53] <Zeenix> omega_: getting error in gstmediaplay that 'oss got eos' [22:53] <Zeenix> s/oss/osssink [22:56] <Zeenix> omega_: -launch osssrc ! gsmenc ! rtpsend runs infinitelly [22:56] <omega_> Zeenix: have you isolated it to disksrc?? [22:56] ChiefHighwater (pa...@te...) left irc: Ping timeout for ChiefHighwater[temple-baptist.com] [22:57] <Zeenix> omega_: ? [22:58] <omega_> you said it segfaults. where? [22:58] <Zeenix> omega_: infact i cant run gst-launch in gdb, gdb treats the element name as files & say it cant find them [22:59] <Zeenix> s/cant/dont know how to [22:59] <omega_> libtool gdb ./gstreamer-launch [22:59] <omega_> r <pipeline goes here> [22:59] Nick change: wtay-zZz -> wtay [22:59] <wtay> yo [22:59] <omega_> yo [22:59] <Zeenix> yo yo [22:59] Nick change: omega_ -> omega_als [23:00] ChiefHighwater (pa...@te...) joined #gstreamer. [23:00] <wtay> interesting stuff? [23:00] Nick change: Zeenix -> Zeenix-away [23:00] <omega_als> wtay: sorta, in a UML talk atm [23:00] Nick change: Zeenix-away -> Zeenix [23:00] <omega_als> but I'm not so interested in it, so I'm just dealing with my mailserver ;-( [23:01] <wtay> doh [23:01] <Zeenix> wtay: completed your App [23:01] <Zeenix> ? [23:01] <omega_als> s/sendmail/qmail/ has issues, and the machine is cooking itself [23:01] <wtay> Zeenix: no, give a few more months.. [23:03] <Zeenix> wtay: do you want me to do that timestamp thing in mp3parse, omega made it too simple for me :) [23:03] <omega_als> timestamping in mp3parse is tricky, I have something in progress that I need to finish [23:03] <wtay> yeah.. [23:04] <wtay> does it work with VBR? [23:04] <omega_als> mine? doesn't care [23:04] <omega_als> but will be seekable, etc. [23:04] <omega_als> now that bytestream exists, it'll be a lot easier to rewrite [23:05] nash (yar@203.128.13.31) left #gstreamer. [23:06] steveb (st...@no...) left irc: boodle boodle boodle [23:09] ds (ds...@c7...) left irc: reboot [23:09] <Zeenix> wtay: what would the pipelines look like for mpeg videos ? [23:09] <Zeenix> wtay: using rtp of course [23:09] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [23:10] <wtay> Zeenix: audio/video or both? [23:11] <Zeenix> wtay: video, we are having seperate payload type for video & audio [23:12] omega_als (omega@64.71.176.6) left irc: [x]chat [23:12] <wtay> Zeenix: I mean, do you want to keep the muxed mpeg or a demuxed version? [23:12] omega_als (omega@64.71.176.6) joined #gstreamer. [23:14] <Zeenix> wtay: muxed, demuxed, dont know exactly about them [23:15] <wtay> Zeenix: mpeg comes as a raw video stream (demuxed) or as audio/video in one (muxed) [23:15] <Zeenix> wtay: talking of raw video of course [23:15] <omega_als> ok, dvdplay is still segfaulting on my machine [23:15] <omega_als> anyone have dvdplay work? [23:16] <wtay> Zeenix: for mpeg over rtp I guess you want to send PES packets over RTP [23:16] <omega_als> fault is in mpeg2dec, but I get an assert failure in buf_unref first [23:17] <Zeenix> wtay: video source ! mp2parse ! rtpsend ; rtprecv ! mpeg2dec ! videosink ? [23:18] <omega_als> wtay: what was the env var to force a stack trace on assert? [23:18] <wtay> oh, is there any? [23:18] evil_monk (trevo@208.141.162.68) joined #gstreamer. [23:18] <wtay> Zeenix: maybe [23:18] <omega_als> afaik [23:18] <wtay> use the source, luke [23:18] <wtay> er omega_als :) [23:18] <omega_als> yeah, right <g> [23:25] bstard (Lor...@a2...) left irc: Read error to bstard[a213-84-40-242.adsl.xs4all.nl]: EOF from client [23:26] <Zeenix> wtay: i'll be writting the sequencing algorithem today or tommorrow with your quidelines in mind [23:26] <wtay> cool [23:26] <omega_als> wtay: can you run dvdplay? [23:27] <wtay> nope, it needs a DVD it seems.. [23:28] <omega_als> I don't think it's a locking race because it happens at the same place every time afaict, even with DEBUG on [23:28] <omega_als> don't you have one? [23:29] <wtay> omega_als: not in this PC, And I can't touch the other PC [23:29] <omega_als> bah [23:29] Action: omega_als throws something at taaz [23:30] sienap (sy...@s3...) left irc: pff mclight shut up [23:35] <taaz> ouch [23:36] <taaz> huh what? where? who? [23:36] <omega_als> does dvdplay work for you? [23:36] <taaz> no [23:36] <taaz> or not really [23:36] <omega_als> what fails? [23:36] <taaz> a little bit maybe [23:36] <taaz> hmm... i think it just locks up somehow [23:37] <taaz> maybe some warnings about buffer unrefs at refcount=0 [23:37] <omega_als> right, I'm trying to track that, and it's not making any sense [23:37] <taaz> quite possibly bugs in there [23:37] <omega_als> but I know it's not a locking race, cause it's totally predictable [23:38] <taaz> that's good [23:38] <Zeenix> wtay: what pipeline you used to test your udp plugin between your & maYYam's PC [23:38] <omega_als> taaz: no ideas? [23:38] <taaz> i tried to track that down a little bit... kind of hard to wade through tons of debug refcount msgs though [23:38] <wtay> Zeenix: same one as the one tested locally (except for the ip address) [23:39] <taaz> um... ideas on where the problem is or how to find it? [23:39] <omega_als> ok, now the talk I wanted to hear is up... [23:39] <omega_als> taaz: both [23:39] <omega_als> taaz: I may have to go get that real trace stuff going, though the debug stuff can be used as far as grepping for the buf*'s [23:40] <taaz> yeah... i was trying that [23:40] <taaz> i think the messages i got were insufficient to track it down though [23:40] <taaz> what problems are you having? [23:40] <omega_als> same thing [23:40] <taaz> ok, i'm not at home now... hard to test ;) [23:41] <taaz> does it print the obj id (address) on those messages? [23:41] <omega_als> yes, and I'll commit a change that prints the pre-operation refcount [23:41] <omega_als> committed [23:42] <taaz> well... you can put on mask=-1 and search for stuff on that obj addr [23:42] <omega_als> right, and it didn't show me much ;-( [23:42] <omega_als> but I need to analyze it better [23:42] <omega_als> after this talk <g> [23:42] <taaz> i think that's about where i was last time i worked on it [23:42] <taaz> been distracted by other stuff recently [23:42] <omega_als> like trying the threadsafety stuff with glib2 maybe? [23:43] <taaz> and skool stuff [23:43] <taaz> mostly skool [23:43] <taaz> does mpeg2parse3 work for you? [23:43] <taaz> cause dvdplay.c is just about the exaxt same code [23:43] <taaz> exact too [23:44] <omega_als> plays fine [23:47] omega_als (omega@64.71.176.6) left irc: Ping timeout for omega_als[64.71.176.6] [23:47] <wtay> ok, I have to sleep, cya [23:47] omega (omega@64.71.176.6) joined #gstreamer. [23:47] Nick change: wtay -> wtay-zZz [23:48] <omega> stupid xv [23:48] Nick change: omega -> omega_als [23:51] <taaz> this is the sort of thing that would be easier to fix if you knew where the buffer came from [23:53] omega_als (omega@64.71.176.6) left irc: Ping timeout for omega_als[64.71.176.6] [23:55] Zeenix (ze...@13...) left #gstreamer. [23:55] omega_als (omega@64.71.176.6) joined #gstreamer. [23:56] <omega_als> blah [23:56] <taaz> definately need to do some cleanup in dvdsrc.c [23:56] <omega_als> eta? [23:59] <omega_als> hrm, this talk isn't that good... [23:59] <taaz> eta? uh... next week [23:59] <omega_als> ok, well, I'm trying to get DVD playback to work now, because at 10pm is the BOF [23:59] <taaz> i was just looking for buffer errors [23:59] <omega_als> I want to be able to demo something cool [00:00] --- Fri Nov 9 2001 [00:00] <taaz> that's 7 hours... plenty of time ;) [00:00] <omega_als> right, if we can figure out what's wrong [00:00] <taaz> well a start is to associate the broken buffers with where they were created [00:01] <taaz> buffer_new() starts with refcount=1 right? [00:01] <omega_als> right, which is rather hard without some kind of mechanism for determining calling function [00:01] <omega_als> which we need a stack-trace reader for [00:01] <omega_als> yes [00:02] <taaz> there are some failure paths that dont unref the buffer but that's probably not an issue here [00:02] <omega_als> where? [00:02] <taaz> in _loop where it returns out in the middle [00:02] <omega_als> of what?> [00:02] <taaz> dvdsrc [00:02] <omega_als> ok [00:03] <omega_als> shoudln't cause this though [00:03] <taaz> so _new() _push(buf) _unref(buf) is the proper loop sequence right? [00:03] <omega_als> no [00:03] <omega_als> most definitely not [00:03] <omega_als> _new, _push [00:04] <taaz> well... then remove that _unref in dvdsrc _loop after the _push [00:04] <taaz> that's goofy... why is the policy not to unref anything you create? [00:04] <omega_als> the base reference travels [00:04] <omega_als> else you're reffing and unreffing it all over the place for no good reason [00:05] <taaz> 1 good reason, people like me didn't know this ;) [00:05] <omega_als> ooooh, ther ewe go [00:05] <taaz> or something.. [00:05] <taaz> working? [00:05] <omega_als> yup [00:05] <taaz> is it totally out of sync? ;) [00:06] <omega_als> didn't look like it, but was lurching [00:06] <omega_als> dma is on, bleck [00:06] <omega_als> yes, out of sync [00:07] <omega_als> ok, I'm heading up to the gnome hacking room, dunno if it has 'net connect [00:07] <taaz> so... like [00:07] <taaz> if you want to fix that [00:07] <taaz> you may need to add timestamps to a52dec [00:07] <omega_als> um, yeah [00:07] <taaz> via bytestream [00:07] <omega_als> ok, I'm out of here [00:07] <taaz> which is why i keep asking abut that ;) [00:08] omega_als (omega@64.71.176.6) left irc: heading up to gnome hackers room [00:25] mattias|na (ma...@ga...) got netsplit. [00:27] mattias|na (ma...@ga...) returned to #gstreamer. [00:33] chillywilly (da...@d1...) joined #gstreamer. [00:47] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [00:54] omega_als (om...@pu...) joined #gstreamer. [01:02] evil_monk (trevo@208.141.162.68) joined #gstreamer. [01:05] dobey (do...@cv...) joined #gstreamer. [01:15] sienap (sy...@s3...) joined #gstreamer. [01:15] <sienap> good evening [01:15] <sienap> night to be more exact. [01:15] <dobey> heh [01:15] <sienap> <- tired :) [01:15] <sienap> eoc [01:15] sienap (sy...@s3...) left #gstreamer. [01:15] <dobey> right... [01:26] ChiefHighwater (pa...@te...) left irc: [01:41] Athos (al...@ic...) joined #gstreamer. [01:46] omega_als (om...@pu...) left irc: Ping timeout for omega_als[public-nat.als01.linuxshowcase.org] [01:47] <dobey> hrmm [01:57] Action: BBB is away: zzz [01:57] Nick change: BBB -> BBB-zZz [02:02] Athos (al...@ic...) left #gstreamer. [02:15] omega (omega@64.71.175.7) joined #gstreamer. [02:16] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [02:18] evil_monk (trevo@208.141.162.68) joined #gstreamer. [02:31] omega (omega@64.71.175.7) left irc: food [02:33] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: EOF from client [02:40] evil_monk (trevo@208.141.162.68) joined #gstreamer. |
From: IRC B. <wt...@us...> - 2001-11-10 05:27:24
|
******************************************************************* [03:04] ShrimpX (marius@131.252.244.168) joined #gstreamer. [03:04] evil_monk (trevo@208.141.162.68) left irc: Read error to evil_monk[208.141.162.68]: Connection reset by peer [03:10] evil_monk (trevo@208.141.162.68) joined #gstreamer. [03:24] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [03:30] evil_monk (trevo@208.141.162.68) joined #gstreamer. [03:32] wingo (wi...@rd...) joined #gstreamer. [03:32] <wingo> hey folks [03:44] <dobey> night [03:44] dobey (do...@cv...) left #gstreamer (eh). [04:03] lucio (lu...@pp...) joined #gstreamer. [04:13] lucio (lu...@pp...) left irc: Ping timeout for lucio[ppp-62-11-6-44.dialup.tiscali.it] [04:14] mattias|na (ma...@ga...) got netsplit. [04:14] ShrimpX (marius@131.252.244.168) got netsplit. [04:14] dap (dap@SR3745.resnet.ucsb.edu) got netsplit. [04:14] chillywilly (da...@d1...) got netsplit. [04:14] BBB-zZz (BB...@uc...) got netsplit. [04:14] taaz (dlehn@66.37.66.32) got netsplit. [04:14] wingo (wi...@rd...) got netsplit. [04:14] evil_monk (trevo@208.141.162.68) got netsplit. [04:14] ajmitch (me...@p5...) got netsplit. [04:14] vini (vi...@bo...) got netsplit. [04:14] vini (vi...@bo...) returned to #gstreamer. [04:14] ajmitch (me...@p5...) returned to #gstreamer. [04:14] BBB-zZz (BB...@uc...) returned to #gstreamer. [04:14] dap (dap@SR3745.resnet.ucsb.edu) returned to #gstreamer. [04:14] taaz (dlehn@66.37.66.32) returned to #gstreamer. [04:14] mattias|na (ma...@ga...) returned to #gstreamer. [04:14] chillywilly (da...@d1...) returned to #gstreamer. [04:14] ShrimpX (marius@131.252.244.168) returned to #gstreamer. [04:14] evil_monk (trevo@208.141.162.68) returned to #gstreamer. [04:14] wingo (wi...@rd...) returned to #gstreamer. [06:17] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [06:48] chillywilly (da...@d1...) left irc: Read error to chillywilly[d116.as21.nwbl0.wi.voyager.net]: EOF from client [07:03] ShrimpX (marius@131.252.244.168) left irc: Remember the... the... uhh..... [08:15] steveb (st...@no...) joined #gstreamer. [08:17] <ajmitch> hi steveb [08:19] <steveb> hi [08:20] <ajmitch> when you gonna be in dunedin? [08:23] <steveb> the friday before xmas [08:23] <steveb> then we'll be in Ohau on Saturday [08:23] <ajmitch> fun [08:23] <ajmitch> why ohau? ;) [08:24] <steveb> my family has a house there [08:24] <ajmitch> ok [08:25] <ajmitch> i suppose, i'll be in balclutha then [08:25] <steveb> its hardly wanaka, but thats the point [08:25] arik (ar...@sd...) joined #gstreamer. [08:25] <arik> hmm [08:25] <arik> i wonder why gstreamer-register is still trying to use /etc even though my prefix is /gnome/GNOME1 [08:26] <steveb> probably [08:26] <ajmitch> het arik [08:26] <ajmitch> hey [08:26] <ajmitch> :) [08:26] <arik> hey aj [08:26] <arik> :-) [08:26] <arik> oh i know why [08:26] <ajmitch> arik: sysconfdir? [08:27] <arik> yep :-) [08:27] Action: arik reruns configure [08:27] <ajmitch> it should be $PREFIX/etc, from what configure --help says [08:28] <arik> yep [08:28] <arik> i know [08:29] <ajmitch> ok, bbl, gotta run off & study (then sit a fun exam tomorrow) [08:29] <ajmitch> by tomorrow night, i'l be done for the year [08:29] <arik> hehe [08:29] <arik> woo! [08:29] <arik> have fun [08:29] <arik> see you tomorrow [08:30] <arik> wtf? [08:30] <arik> still putting it in /etc [08:35] <taaz> that stuff is hard hard to get done right [08:35] <arik> yep [08:35] <arik> it really is [08:36] <arik> bbiab [08:36] arik (ar...@sd...) left irc: leaving [09:43] Nick change: taaz -> taazzzz [09:46] mattias|na (ma...@ga...) left irc: Client Exiting [10:01] mattias (ma...@ga...) joined #gstreamer. [11:29] harobed (harobed@194.51.105.35) joined #gstreamer. [11:31] ajmitch (me...@p5...) left irc: Ping timeout for ajmitch[p54-max1.dun.ihug.co.nz] [11:32] ajmitch (me...@p4...) joined #gstreamer. [11:59] harobed_ (harobed@194.51.105.35) joined #gstreamer. [11:59] hadess (ha...@de...) joined #gstreamer. [11:59] <hadess> hallo [11:59] <ajmitch> hi [11:59] harobed (harobed@194.51.105.35) left irc: Read error to harobed[194.51.105.35]: Connection reset by peer [11:59] <hadess> heya aj [12:14] ShrimpX (marius@131.252.244.168) joined #gstreamer. [12:31] _gst_newt_ joined #gstreamer. [12:31] wtay-zZz (wi...@ca...) left irc: Ping timeout for wtay-zZz[cable-195-162-215-201.upc.chello.be] [12:34] wtay-zZz (wi...@ca...) joined #gstreamer. [12:52] ajmitch (me...@p4...) left irc: Ping timeout for ajmitch[p42-max7.dun.ihug.co.nz] [12:53] ajmitch (me...@p4...) joined #gstreamer. [13:41] Action: BBB-zZz is away: weekend [13:41] Nick change: BBB-zZz -> BBB-away [13:43] Nick change: hadess -> hds-food [13:52] vini (vi...@bo...) left irc: Read error to vini[bones.kulnet.kuleuven.ac.be]: EOF from client [14:37] Nick change: hds-food -> hadess [15:51] Shippou (zbl...@ot...) got netsplit. [15:51] mattias (ma...@ga...) got netsplit. [15:51] steveb (st...@no...) got netsplit. [15:51] dap (dap@SR3745.resnet.ucsb.edu) got netsplit. [15:51] ShrimpX (marius@131.252.244.168) got netsplit. [15:51] BBB-away (BB...@uc...) got netsplit. [15:51] hadess (ha...@de...) got netsplit. [15:51] wtay-zZz (wi...@ca...) got netsplit. [15:51] ajmitch (me...@p4...) got netsplit. [15:57] ShrimpX (marius@131.252.244.168) got lost in the net-split. [15:57] hadess (ha...@de...) got lost in the net-split. [15:57] mattias (ma...@ga...) got lost in the net-split. [15:57] steveb (st...@no...) got lost in the net-split. [15:57] Shippou (zbl...@ot...) got lost in the net-split. [15:57] dap (dap@SR3745.resnet.ucsb.edu) got lost in the net-split. [15:57] BBB-away (BB...@uc...) got lost in the net-split. [15:57] wtay-zZz (wi...@ca...) got lost in the net-split. [15:57] ajmitch (me...@p4...) got lost in the net-split. [16:00] _gst_newt_ joined #gstreamer. [16:18] Nick change: mattias -> mattias|na [17:07] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [17:08] evil_monk (trevo@208.141.162.68) joined #gstreamer. [17:57] evil_monk (trevo@208.141.162.68) left irc: Client Exiting [17:58] evil_monk (trevo@208.141.162.68) joined #gstreamer. [18:00] ShrimpX_ (marius@Photon.pdx.netroedge.com) joined #gstreamer. [18:45] harobed (harobed@AC8A041D.ipt.aol.com) joined #gstreamer. [18:45] bstard (Lor...@a2...) got netsplit. [18:49] bstard (Lor...@a2...) returned to #gstreamer. [18:52] ChiefHighwater (pa...@te...) joined #gstreamer. [18:58] hadess (ha...@de...) left irc: Client Exiting [19:06] omega (omega@64.71.176.6) joined #gstreamer. [19:22] thomasvs (th...@21...) joined #gstreamer. [19:22] <thomasvs> ello [19:22] Nick change: omega -> omega_als [19:22] <omega_als> yo [19:22] <dobey> YO! [19:22] <dobey> is uraeus there? [19:22] harobed (harobed@AC8A041D.ipt.aol.com) left irc: Read error to harobed[AC8A041D.ipt.aol.com]: Connection reset by peer [19:22] <omega_als> yup [19:23] <dobey> tell him the interview isn't up [19:24] <omega_als> ok, sec, we're at the tail end of a talk [19:24] <dobey> ok [19:25] <omega_als> dobey: ok, [19:26] <omega_als> dobey: I will look into it [19:26] <dobey> yeah, ok [19:26] <omega_als> (was Uraeus) [19:26] <dobey> i'm trying to bug jeremy too [19:45] harobed (harobed@ACAF8134.ipt.aol.com) joined #gstreamer. [20:27] Action: bstard is away: Weg [20:37] harobed (harobed@ACAF8134.ipt.aol.com) left irc: Read error to harobed[ACAF8134.ipt.aol.com]: Connection reset by peer [20:40] omega_als (omega@64.71.176.6) left irc: Ping timeout for omega_als[64.71.176.6] [21:30] Nick change: taazzzz -> taaz [21:48] <taaz> jackit.sf.net soon to be live [21:52] <steveb> take your pants off, and jackit [21:52] <steveb> (Blink 182) [22:05] nash (ya...@13...) joined #gstreamer. [22:19] ChiefHighwater (pa...@te...) left irc: [22:22] bstard (Lor...@a2...) left irc: Client Exiting [22:40] steveb (st...@no...) left irc: Read error to steveb[node1ee7b.a2000.nl]: EOF from client [22:54] Zeenix (ze...@13...) joined #gstreamer. [22:55] Action: Zeenix announces rtp plugin with gsm, oss, mp3 & mpeg videos support [22:57] <taaz> cool [22:58] <Zeenix> taaz: i was busy today setting the mpeg video support [22:59] <Zeenix> taaz: & i succeeded atlast, i tested it with some small mpeg videos i had [23:03] <taaz> dvd over rtp here we come! ;) [23:03] <taaz> how hard is it to add support for a new format? [23:05] <Zeenix> taaz: a day's work :) [23:06] <Zeenix> taaz: i'll do that for you, if you really need it, but not free of cost :) [23:07] <taaz> it would be nice if it were somewhat generic [23:07] <taaz> so we could just send any packetized data [23:08] <taaz> of course i have no idea how rtp works... [23:08] <Zeenix> taaz: the payload_type( media type ) is attached with each packet [23:09] <Zeenix> taaz: its in the RFCs [23:09] <taaz> yeah... i just havent looked into it [23:10] <Zeenix> taaz: i dont know anything at all about DVDs unfortunatelly [23:10] <taaz> you know people are going to want to use the profiles for ogg/vorbis, mpeg2, a52, etc etc [23:12] <Zeenix> taaz: seen wtay today ? [23:13] <taaz> nope [23:22] ds (ds...@c7...) joined #gstreamer. [23:42] dobey (do...@dr...) left #gstreamer (eh). [23:44] omega_als (om...@pu...) joined #gstreamer. [23:44] <Zeenix> yo [23:45] <Zeenix> omega_als: now we are streaming mpeg videos through rtp [23:45] <omega_als> cool [23:48] <taaz> how did the bof go? [23:49] <omega_als> pretty good [23:49] <omega_als> so, I need to write up a summary to the list.... [23:49] <ds> um... except for the projector [23:49] <omega_als> er, yeah [23:50] <ds> omega's laptop is too leet [23:50] Action: omega_als needs to call walken again [23:50] <omega_als> no, X is too 'smart' [23:53] <ajmitch> heh, sounds fun [00:00] --- Sat Nov 10 2001 [00:06] chillywilly (da...@d1...) joined #gstreamer. [00:07] chillywilly (da...@d1...) left irc: Killed (NickServ (This nick is reserved by another user)) [00:08] chillywilly_ (da...@d1...) joined #gstreamer. [00:09] chillywilly_ (da...@d1...) left irc: Read error to chillywilly_[d164.as14.nwbl0.wi.voyager.net]: EOF from client [00:09] chillywilly (da...@d1...) joined #gstreamer. [00:34] gnome (gn...@fl...) joined #gstreamer. [00:34] <gnome> hi dudes [00:34] Nick change: gnome -> Uraeus_als [00:34] <ds> clever disguise, that [00:34] <Uraeus_als> yes, I paid a fortune for that :) [00:35] <Uraeus_als> ds: i forgot to try your phone last night :( [00:35] <Zeenix> yo [00:35] <Uraeus_als> hi zeenix [00:36] <Uraeus_als> omega_als: I have a happy hacking t-shirt for you [00:39] <omega_als> cool [00:39] Action: omega_als waves to Uraeus [00:39] <omega_als> Uraeus_als: look up [00:40] <omega_als> what, you want me in the booth?? [00:40] <Uraeus_als> yes [00:40] <omega_als> bah <g> [00:40] <Uraeus_als> it is boring here by myself [00:40] <omega_als> pff, ok, sec [00:44] omega_als (om...@pu...) left irc: Ping timeout for omega_als[public-nat.als01.linuxshowcase.org] [00:45] omega_als (om...@pu...) joined #gstreamer. [00:45] <omega_als> ok, here. happy? <g> [00:45] <Zeenix> wb [00:47] <ds> omega_als: what was the web page for MAS? [00:55] <omega_als> DNE [00:56] <omega_als> well, shiman.org/mas/ maybe [00:56] <omega_als> s/org/com/ ? [00:56] <omega_als> .com [01:00] <ds> thanks [01:07] Zeenix (ze...@13...) left irc: Ping timeout for Zeenix[13-47.brain.net.pk] [01:08] omega_als (om...@pu...) left irc: Ping timeout for omega_als[public-nat.als01.linuxshowcase.org] [01:08] omega_als (om...@pu...) joined #gstreamer. [01:12] omega_ (om...@fl...) joined #gstreamer. [01:13] omega_als (om...@pu...) left irc: Ping timeout for omega_als[public-nat.als01.linuxshowcase.org] [01:13] Nick change: omega_ -> omega_als [01:18] Uraeus_als (gn...@fl...) left #gstreamer (Client Exiting). [01:19] nash (ya...@13...) left irc: Ping timeout for nash[13-12.brain.net.pk] [01:24] Zeenix (ze...@13...) joined #gstreamer. [01:27] <Zeenix> omega_als: Uraeus says you have been telling people about my "great work", which great work do you think i ever did? [01:31] <omega_als> rtp [01:32] <Zeenix> is robla there too? [01:32] <omega_als> not afaict [01:33] <Zeenix> he wanted 2 things in it: mp3 & some video type i dont remember the name of [01:34] <Zeenix> but gst doesnt have plugin for that video type ATM AFAIK [01:35] <Zeenix> & mp3 is working clean if that mp3parse timestamps the buffers [01:38] <Zeenix> a bundle of thanx for promoting me [01:39] <Zeenix> need to go now, bi [01:39] Zeenix (ze...@13...) left irc: I pay 25 Rs( 0.5$ )/hour [01:42] Action: chillywilly is away: bank [01:46] omega_als (om...@fl...) left irc: attempting a rh72 upgrade |