From: IRC <wt...@us...> - 2003-05-20 05:42:30
|
******************************************************************* [03:00] ^iain^ (~ia...@us...) left irc: "POKEY" [03:16] sxpert__ (~sxpert@APh-Aug-105-1-1-129.w217-128.abo.wanadoo.fr) left irc: Read error: 110 (Connection timed out) [03:16] sxpert (~sxpert@APh-Aug-105-1-1-129.w217-128.abo.wanadoo.fr) left irc: Read error: 110 (Connection timed out) [03:16] sxpert (~sxpert@APh-Aug-105-1-6-135.w80-14.abo.wanadoo.fr) joined #gstreamer. [03:16] sxpert__ (~sxpert@APh-Aug-105-1-6-135.w80-14.abo.wanadoo.fr) joined #gstreamer. [04:11] <ds-home> gst_caps_intersect() returning NULL for an empty intersection is, er, buggy [04:21] <Company> lol, true [04:21] <ds-home> ok, tee is sort of working properly now [04:21] <ds-home> I need to spend more time on it [04:22] <Company> we should make that api gboolean gst_caps_intersect (GstCaps *caps1, GstCaps *caps2, GstCaps **result); [04:22] <ds-home> or have a GST_CAPS_EMPTY [04:22] <Company> or that [04:22] <Company> file a bug about it plz so we don't forget [04:23] <ds-home> that way, it can be carried around similar to NAN [04:23] <ds-home> will do [04:24] <Company> GST_CAPS_EMPTY is a great idea [04:28] apoc__ (~ap...@dy...) joined #gstreamer. [04:34] <ds-home> gar [04:34] <ds-home> videotestsrc ! video/raw, format:fourcc=I420 ! xvideosink [04:34] <ds-home> doesn't work [04:34] <ds-home> ???: \ [04:34] <ds-home> unexpected $undefined [04:37] apoc_ (~ap...@dy...) left irc: Read error: 60 (Operation timed out) [04:39] <Company> ds-home: maybe something was escaped wrong [04:39] <ds-home> I enclosed everything in '' [04:40] <Company> ds-home: uhm? [04:40] <Company> ds-home: that's not how it's done [04:40] <Company> if you take exactly that line it and paste it into the shell, it works [04:47] <ds-home> ok, too much gstreamer for one weekend. must work on swfdec now [04:48] walters (wa...@ve...) left irc: Remote closed the connection [05:10] steveb_ (~st...@20...) joined #gstreamer. [05:18] steveb__ (~st...@20...) left irc: Read error: 60 (Operation timed out) [05:38] Company (~Company@pD9E3339B.dip.t-dialin.net) left irc: Remote closed the connection [05:56] Inkwina[m] (~phsi@195.158.106.36) joined #gstreamer. [06:09] Inkwina[m] (~phsi@195.158.106.36) left irc: Remote closed the connection [07:00] ChrisHJW (~chr...@p5...) left irc: Read error: 110 (Connection timed out) [08:21] ChrisHJW (~chr...@p5...) joined #gstreamer. [08:28] harshy (~ha...@dh...) joined #gstreamer. [08:55] ChrisHJW (~chr...@p5...) left irc: Read error: 110 (Connection timed out) [09:30] <ds-home> why did gst-player suddenly stop working on my machine? [10:21] thomasvs_ (~th...@19...) joined #gstreamer. [10:37] thomasvz (~th...@99...) left irc: Read error: 110 (Connection timed out) [10:38] dolphy (~do...@po...) joined #gstreamer. [11:08] ack_ (~ac...@po...) joined #gstreamer. [11:08] Nick change: ack_ -> sack [11:52] vishnu (~joshua@61.11.23.201) joined #gstreamer. [12:08] sack (~ac...@po...) left irc: "new kernel installed , rebooting" [12:25] newt joined #gstreamer. [12:26] _gst_newt_ (~eggdrop@195.162.214.239) left irc: Read error: 104 (Connection reset by peer) [12:27] Nick change: newt -> _gst_newt_ [12:29] Action: sxpert_work compiles Tremor [13:05] gernot (~ge...@a2...) joined #gstreamer. [13:06] matrixise (~stef@213.213.206.2) joined #gstreamer. [13:08] Uraeus (~cschalle@195.18.245.94) joined #gstreamer. [13:08] <Uraeus> ello [13:08] <apoc> yo Uraeus [13:08] <Uraeus> thomasvs_: ping [13:08] <Uraeus> hey apoc [13:09] <apoc> Uraeus : If you can talk with wtay. Can yous ask him if we can add GstProbe for 0.6.2 [13:12] <Uraeus> apoc: what is GstProbe? [13:12] <dolphy> hi Uraeu [13:12] <dolphy> s [13:12] <dolphy> :) [13:13] <apoc> Uraeus : With GstProbe you can get data from pad [13:13] <apoc> yo dolphy [13:14] <apoc> Uraeus : I'm doing a gnome-vfs module based on sound-juicer to rip cd tracks with nautilus [13:14] <apoc> Uraeus : and I need GstProbe for that ;) [13:15] <Uraeus> hey dolphy [13:15] <dolphy> apoc: heya :) [13:15] <Uraeus> apoc: well if it is a new plugin and it works stably there is no reason not to add it, or is gstprobe a core addition? [13:15] <dolphy> the name sounds like a core addition :) [13:15] <apoc> Uraeus : a core addition [13:17] <apoc> We have a release date for 0.6.2 ? next week-end ? [13:18] <dolphy> nope no release date [13:21] <Uraeus> I didn't get to talk to BBB, but he was to work on in yesterday and saturday. And I see lots of commits from him and thomasvs in CVS, so maybe we are getting close [13:22] <vishnu> Uraeus: BBB sent me email saying he was trying for a 0.6.2 today or tomorrow [13:22] <Uraeus> great :) [13:23] <vishnu> Uraeus: yah, i guess, but i'd rather see more work on HEAD [13:24] <dolphy> vishnu: getting 0.6.2 is quite important :) [13:24] <dolphy> vishnu: big fixes inside [13:25] Action: vishnu shrugs [13:25] <vishnu> whatever, i know what fixes i'm waiting for and they're not in 0.6.x [13:26] <Uraeus> vishnu: yeah, well after 0.6.2 is out I think a 0.7.0 release would be good [13:26] <apoc> Uraeus : we [13:27] apoc (~ap...@ww...) left irc: "CGI:IRC 0.4.3 [EOF]" [13:27] <vishnu> i don't even need a release, just changes to CVS ;-) [13:27] <dolphy> vishnu: what kind of fix do you need ? [13:28] <vishnu> bug 113180 [13:28] apoc (~ap...@ww...) joined #gstreamer. [13:35] <vishnu> oops, BBB is telling me to be patient [13:36] Action: vishnu grumbles about being patient [13:36] <md`> hiya apoc [13:43] Nick change: matrixise -> mat_AWAY [13:43] <thomasvs_> pong [13:44] <apoc> hey md' [13:44] <thomasvs_> so [13:44] <thomasvs_> then we need to decide if gstprobe needs to go in now or not [13:46] Company (~as...@rz...) joined #gstreamer. [13:46] <apoc> yo Company [13:46] <Uraeus> hey thomasvs_ [13:48] <Company> hi [13:48] <dolphy> yo Company [13:51] <apoc> not sure GstProbe has stable API ... we should ask wtay [13:51] <apoc> he made an API change 2 months ago ... [13:52] vishnu (~joshua@61.11.23.201) left irc: "Client Exiting" [14:01] <dolphy> ahhg core Head is broken [14:03] <apoc> Can we wait one or two days for the release ? [14:05] <Company> dolphy: what is broken there? [14:05] <Company> (cvs should never be broken...) [14:08] <dolphy> gstclock.c:382: warning: integer constant is too large for "long" type [14:08] <dolphy> max-diff property [14:09] <Company> are you using gcc 3.3 maybe? [14:13] Uraeus (~cschalle@195.18.245.94) left irc: "Client Exiting" [14:13] <dolphy> yup [14:13] <dolphy> sid updated to 3.3 this morning [14:14] <Company> dolphy: glib has issues with that [14:14] <Company> dolphy: i remember wtay and ds talking about that last night [14:15] <dolphy> i see [14:15] Action: dolphy has the feeling he will run into big troubles then :) [14:16] <yippi> what version of gstreamer is about to release? [14:19] <Company> yippi: the next one will be 0.6.2 i think [14:19] <yippi> thx [14:29] <Company> anyone speaking assempler in here? [14:29] <Company> s/assempler/assembler/ [14:30] <thomasvs_> dolphy: maybe check with other gnomers running 3.3 [14:43] gernot (~ge...@a2...) left irc: "Client exiting" [14:52] taaz (~dlehn@66.37.66.32) joined #gstreamer. [14:55] <Company> gah, on my solaris box the inline stuff isn't compiling [14:56] jey (je...@je...) joined #gstreamer. [14:57] <Company> ah [14:57] <Company> that stuff requires gcc >= 3.1 [14:57] <jey> Does anyone have info about the status of the RTP plugin in GStreamer? [14:58] <Company> Zeenix would be the one to ask i think [14:59] <jey> Company: Where can I find him/her? [15:00] <Company> jey: in here, but later [15:01] <Company> jey: between 18h and 22h CET would be best i think [15:01] <jey> Company: Thanks. I'll be back. [15:01] jey (je...@je...) left #gstreamer. [15:02] apoc (~ap...@ww...) left irc: "CGI:IRC 0.4.3" [15:06] Vakor (ms...@c1...) left irc: Remote closed the connection [15:07] Vakor (ms...@c1...) joined #gstreamer. [15:18] Company (~as...@rz...) left irc: Remote closed the connection [15:40] ChristianHJW (~Chr...@p5...) joined #gstreamer. [15:58] foser (d0...@22...) joined #gstreamer. [16:07] ChristianHJW (~Chr...@p5...) left irc: Remote closed the connection [16:10] KoRn_w0rk (~ab...@pp...) joined #gstreamer. [16:10] <KoRn_w0rk> hello [16:10] <KoRn_w0rk> I need help [16:10] <KoRn_w0rk> I'm trying to compile gstreamer [16:10] <KoRn_w0rk> works just fine [16:10] <KoRn_w0rk> now I downloaded the tar.gz plugin package [16:10] <KoRn_w0rk> ./configure [16:10] <KoRn_w0rk> checking for gstreamer-0.6 >= 0.6.1... Package gstreamer-0.6 was not found in the pkg-config search path. [16:10] <KoRn_w0rk> Perhaps you should add the directory containing `gstreamer-0.6.pc' [16:10] <KoRn_w0rk> to the PKG_CONFIG_PATH environment variable [16:10] <KoRn_w0rk> No package 'gstreamer-0.6' found [16:10] <KoRn_w0rk> configure: error: no GStreamer found [16:11] <KoRn_w0rk> what's wrong ? it can't find gstreamer, right ? [16:11] <KoRn_w0rk> for gstreamer, I did : ./configure ; make ; make install [16:11] <KoRn_w0rk> I ran ldconfig also [16:12] apoc (~ap...@ww...) joined #gstreamer. [16:27] apoc (~ap...@ww...) left irc: "CGI:IRC 0.4.3" [16:35] <thomasvs_> KoRn_w0rk: export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig [16:35] <thomasvs_> then rerun configure for plugins [16:35] <thomasvs_> KoRn_w0rk: what dist do you run ? [16:37] Company (~Co...@p5...) joined #gstreamer. [16:40] <KoRn_w0rk> redhat 9 [16:41] <KoRn_w0rk> im configuring now [16:43] <thomasvs_> KoRn_w0rk: if you're running redhat, why not install the packages ? [16:43] <KoRn_w0rk> I don't know [16:43] <KoRn_w0rk> I like configuring [16:43] <KoRn_w0rk> ;) [16:43] <KoRn_w0rk> help ! [16:43] <KoRn_w0rk> configure: *** Plug-ins that will not be built : [16:43] <KoRn_w0rk> swfdec [16:43] <thomasvs_> yeah, but you seem to have a lot of problems doing it properly [16:43] <KoRn_w0rk> I need that [16:43] <KoRn_w0rk> how does plugin work in gstreamer ? [16:44] <thomasvs_> do you have swfdec development stuff installed ? [16:44] <KoRn_w0rk> no [16:44] <KoRn_w0rk> I just downloaded 2 files [16:44] <KoRn_w0rk> gstreamer (the core only) [16:44] <KoRn_w0rk> gst-plugins [16:44] <KoRn_w0rk> but isn't swfdec inside the "gst-plugins" ? [16:44] <Company> you need to download all the dependencies for plugins you want [16:45] <KoRn_w0rk> aaaaah [16:45] <KoRn_w0rk> what's "gst-plugins" for then ? [16:45] <KoRn_w0rk> I though "gst-plugins" had all the plugins [16:45] <Company> that's the glue between those libs and our framework [16:46] <KoRn_w0rk> aaah ok. [16:46] <KoRn_w0rk> so everytime I get a plugin, I gotta run gst-plugins again ? [16:46] <KoRn_w0rk> argh : http://sourceforge.net/projects/swfdec/ [16:46] <KoRn_w0rk> 404 [16:47] <Company> they'll be up again in an hour [16:47] <KoRn_w0rk> down ? [16:47] <Company> and yes, if you build from source you have to rebuild everytime you want another of those plugins [16:47] <KoRn_w0rk> aaah yeah : http://www.sourceforge.net/ -- We're Sorry. [16:47] <KoRn_w0rk> The SourceForge.net Website is currently down for maintenance. [16:47] <KoRn_w0rk> We will be back shortly [16:48] <KoRn_w0rk> well... can someone send it 2 me ? ;) [16:48] <Company> yeah, they do routine maintenance more often than you'd like :) [16:48] <KoRn_w0rk> =p [16:48] <KoRn_w0rk> so, while we wait they're back up, tell me [16:49] <KoRn_w0rk> i'm trying to convert an swf file into a bitmap file [16:49] <KoRn_w0rk> I'm installing gstreamer just for that [16:49] <KoRn_w0rk> that should work with that pluggin, right ? [16:49] <Company> uhm [16:49] <Company> shouldn't that be easier with the gimp? [16:49] <KoRn_w0rk> no [16:49] <KoRn_w0rk> I want to automate that [16:50] <KoRn_w0rk> its for an application [16:50] <KoRn_w0rk> I download from internet an swf file (real time traffic map) [16:50] <Company> can't you just use the gdkpixbuf loader then? [16:50] <KoRn_w0rk> but my video player can only read mpeg1, 2 and a few bitmaps [16:51] <Company> ah, you want an animation [16:51] <KoRn_w0rk> so I need to "wget http://file.swf ; convert swf into bitmap ; ftp bitmap to my player ; done [16:51] <KoRn_w0rk> so that needs to be command line stuff [16:51] <KoRn_w0rk> gdkpixbuf ? [16:52] <Company> yeah [16:52] <KoRn_w0rk> im not really a programmer, and I don't know about all these developpment things... [16:52] <KoRn_w0rk> you gotta help me out... [16:52] <Company> that means that every gnome application can load swf files [16:53] <Company> but i don't know if that would help, becuase i don't know if theres an app that does scriptable conversions... [16:53] <KoRn_w0rk> well, my player isn't a linux player. It's a windows closed source player [16:53] <KoRn_w0rk> I came to gstreamer because some guy on a mailing list told me : [16:53] <Company> in any case, gstreamer might be able to do so, i dunno how the still image stuff works for us :) [16:54] <KoRn_w0rk> You can also use GStreamer, which has a swf element. The pipeline "gst-launch filesrc location=/path/to/file.swf ! sfwdec ! colorspace ! pngenc ! filesink location=/path/to/file.png" should create a (PNG) screenshot. [16:54] <Company> try it :) [16:54] <KoRn_w0rk> yeah [16:54] <KoRn_w0rk> but I need to download the thing [16:54] <KoRn_w0rk> lol [16:55] <KoRn_w0rk> server's down [16:55] <Company> that happens [16:55] <KoRn_w0rk> but that sounds right to you ? [16:55] <KoRn_w0rk> the command line ? [16:55] <Company> surf the web for a while :) [16:55] ^iain^ (~ia...@us...) joined #gstreamer. [16:55] <Company> i'm neither into our swf nor into our conversion stuff, but that sounds like it should work [16:55] <KoRn_w0rk> :) [16:56] <KoRn_w0rk> ok ok, at least, I'm on the right way... ;) [16:56] <^iain^> morn' [16:56] <Company> hi ^iain^ [16:56] <KoRn_w0rk> aftern' [16:59] <^iain^> whats been happening in gstreamer world? [17:01] <Company> ^iain^: if you only mean gstreamer stuff: the buffer struct is changed to include a duration and maxsize will be made usable [17:01] <Company> ^iain^: filtered caps have landed in HEAD [17:01] <Company> ^iain^: and if you mean apps using gstreamer: rhythmbox HEAD is the best rhythmbox ever [17:02] <Company> ^iain^: and what is happening in ^iain^ world? [17:03] <Company> s/^iain^ world/^iain^ hacking world/ to be exact :) [17:05] <^iain^> Company; nothing, cos I just got in. although I did make MarlinPositionSpinner work with funny time formats last night [17:10] <Company> so you'Re hacking on marlin currently? [17:10] <Company> and not doing weird dropshadow stuff or so :) [17:11] <^iain^> nah. I need to get marlin working happily [17:12] <Company> i really need to check it out one day [17:12] <Company> so much funky gst software, so little time [17:14] <^iain^> I think marlin is the biggest piece of gst software so far :) [17:14] <^iain^> unless the gst editor is bigger [17:14] <^iain^> but I dunno how big it is [17:17] Action: ^iain^ wonders what mysqlsrc and mysqlsink could be used for [17:18] <^iain^> GST plugins aren't loaded on demand are they? [17:19] <^iain^> they're all loaded at startup [17:23] Nick change: mat_AWAY -> matrixise [17:24] <thomasvs_> I'm pretty sure they're loaded on demand [17:24] <thomasvs_> If they're not, we should fix that :) [17:24] <thomasvs_> it just scans the registry at startup [17:24] <^iain^> well, when I load marlin in gdb it says its linked to all the gstplugins [17:26] <Company> they're loaded on demand [17:27] <Company> gstreamer uses a GstElementFactory that loads its properties from the registry [17:27] <Company> but it only loads the libs when you create an instance of it [17:28] <Company> at least that's how it's supposed to work [17:28] <Company> maybe you don't have a registry and it's rebuilt on startup? [17:29] <^iain^> I do have one [17:29] <^iain^> but if I run marlin, and then run gdb on the process it shows that its loading all the gstplugins [17:29] <Company> not all [17:30] <Company> maybe lots because of the autoplugger [17:30] <^iain^> oh, it might be that [17:30] <Company> try running gdb on marlin before loading a plugin [17:30] <Company> s/plugin/song/ [17:30] <^iain^> yeah, I see what you mean [17:31] <Company> typefinding might load lots of plugins, too [17:35] <^iain^> the sched test in testsuite/refcounting doesn't work on HEAD [17:35] <^iain^> dumps lots of (process:2841): GStreamer-CRITICAL **: file gstelement.c: line 1435 (gst_element_link_pads_filtered): assertion `GST_IS_ELEMENT (src)' failed [17:38] <Company> i'll look at it [17:38] <Company> (i'd be very surprised if a test worked ;)) [17:38] <thomasvs_> Company: all of them work in 0.6 branch [17:39] <thomasvs_> Company: if they don't anymore on 0.7, someone screwed up somewhere :) [17:40] <Company> thomasvs_, ^iain^: that's not really a test but just an error case that wtay dumped there [17:40] <^iain^> it probably shouldn't be run during make check then? [17:40] <^iain^> seeing as it seems to just loop [17:41] <Company> it should check properly for the required elements for example [17:41] <Company> it crashes or hangs if there is no alsasink [17:42] Action: thomasvs_ is doing make distcheck manually based on the Makefile to find why gstbuffer.c gets installed [17:42] <Company> thomasvs_: there is a gstbuffer.c in the tests [17:42] <Company> thomasvs_: maybe that one gets installed? [17:44] <thomasvs_> Company: hm, that might be, but that would be a stupid mistake [17:44] <thomasvs_> and probably mine :) [17:44] <thomasvs_> where is it ? [17:46] Action: thomasvs_ accidentally typed "make distcleanchick" [17:46] <dolphy> lol [17:47] <Company> thomasvs_: find -name \*buffer.c :) [17:48] <Company> ^iain^: could you try again? [17:52] BBB (~rb...@01...) joined #gstreamer. [17:53] <^iain^> * (process:3250): WARNING **: Could not run test, because element type "sinesrc" is not installed. Please retry when it is. Asysuming it works for now... [17:53] <^iain^> okay, works again [17:54] <^iain^> thanks [17:54] <Company> well, i shopuld have checked my spelling, but oh well [17:55] <^iain^> (process:5414): GStreamer-WARNING **: Thread thread0 is destroying itself. Function call will not return! [17:55] <^iain^> thread4 doesn't work either :) [17:55] <^iain^> but the threads stuff was always a bit dodgy [17:59] <Company> ^iain^: thread4 works [17:59] <Company> ^iain^: just not everytime :) [18:02] <^iain^> ah, like I said...a bit dodgy :) [18:02] <Company> yeah [18:03] <Company> because the threading implementors had NO FUCKING CLUE what they were doing [18:04] <Company> and because the queue is still broken... [18:06] <Company> our threading stuff has holes so big you could fly a star destroyer through [18:06] <Company> without even seeing where the hole ends [18:06] <BBB> lol ;) [18:07] <BBB> problem of not having SMP ;) [18:07] <KoRn_w0rk> Company [18:07] <KoRn_w0rk> so now, I'm ./configure ; make ; make install for the swfdec [18:07] <KoRn_w0rk> right ? [18:07] <KoRn_w0rk> do I have to specify an install path ? [18:07] <BBB> no [18:07] <BBB> use rpms ;) [18:07] <KoRn_w0rk> no [18:07] <KoRn_w0rk> ;) [18:09] <dolphy> use debs then :-P [18:09] <KoRn_w0rk> cd /swfdec/ ; ./configure ; make ; make install [18:09] <KoRn_w0rk> is enought [18:09] <KoRn_w0rk> ? [18:09] <Company> yes [18:09] <KoRn_w0rk> now : [18:09] <Company> it'll install to /usr/local then [18:10] <KoRn_w0rk> cd ../gst-plugins-0.6.1 ; ./configure ; make ; make install [18:10] <KoRn_w0rk> ? [18:10] <Company> yes [18:10] <BBB> export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig [18:10] <BBB> :) [18:10] <BBB> or use RPMs [18:10] <Company> true [18:10] <BBB> or debs [18:10] <BBB> or whatever [18:10] <KoRn_w0rk> done that a few hours ago [18:10] <KoRn_w0rk> Company helped me on that [18:10] <KoRn_w0rk> ;) [18:10] <BBB> ok [18:10] ghostar (~ghostar@D-128-95-117-225.dhcp4.washington.edu) joined #gstreamer. [18:10] <KoRn_w0rk> can't gstreamer do the "export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig" by it self ? [18:11] <Company> no [18:11] <KoRn_w0rk> you should make a note then giving the exact command line [18:11] <KoRn_w0rk> for example : [18:11] <KoRn_w0rk> ./configure [18:11] <KoRn_w0rk> ERROR: Try again with "export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig" [18:11] <KoRn_w0rk> :) [18:11] <KoRn_w0rk> now, I know I have to try again with export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig [18:12] <Company> that doesn't work in every case [18:12] <Company> just in your case [18:12] <KoRn_w0rk> configure: *** Plug-ins that will be built : [18:12] <KoRn_w0rk> swfdec [18:12] <KoRn_w0rk> ! [18:12] <KoRn_w0rk> tnx ! [18:12] <Company> but we can expect people building from source to know that :) [18:12] <KoRn_w0rk> just in RH 9 cases then [18:13] <KoRn_w0rk> but you can expect people trying from rpms that screw up the system, and not try anymore... and try from source, but don't know about that [18:13] <KoRn_w0rk> ;))) [18:13] <Company> rpms don't screw up a system normally [18:13] <Company> most of the time --force does that [18:13] <KoRn_w0rk> well... true. But I just had a bad experience with rpms [18:13] <KoRn_w0rk> that's why [18:13] <KoRn_w0rk> apt-get upgrade with redhat9 blows up the system... gotta reinstall [18:14] <Company> use debian [18:14] <KoRn_w0rk> but that's a redhat but [18:14] <KoRn_w0rk> bug [18:14] <KoRn_w0rk> <-- newbi [18:14] <KoRn_w0rk> I tried debian, not so EZ to install [18:14] <Company> yeah, the start is hard [18:16] <^iain^> heck, I've been using linux for 5 years and I can't even install debian [18:17] <Company> hey, don't discourage me [18:17] <KoRn_w0rk> :D [18:17] <Company> if i ever bought an ibook i'd be kinda forced to use debian... [18:19] <KoRn_w0rk> it takes a long time compiling [18:19] <KoRn_w0rk> if I ever want to add a new pluggin [18:19] <dolphy> Company: you can still use Yellowdog [18:19] <KoRn_w0rk> I have to do that "glue" thing again ? [18:19] <dolphy> Company: if you can't bear debian :) [18:19] <dolphy> Company: debianppc is even harder than normal debian [18:19] <thomasvs_> KoRn_w0rk: how does that blow up the system ? [18:20] <KoRn_w0rk> thomasvs_ [18:20] <thomasvs_> KoRn_w0rk: it's fair to expect that users who compile from source know what they're doing [18:20] <KoRn_w0rk> let me show you [18:20] <thomasvs_> KoRn_w0rk: ie, they should know about stuff like PKG_CONFIG_PATH [18:20] <KoRn_w0rk> http://tinyurl.com/9aoi [18:20] <thomasvs_> KoRn_w0rk: you can also read the faq, and discover an easy way to build and run uninstalld gstreamer from cvs [18:22] Uraeus (~csc...@wp...) joined #gstreamer. [18:22] <KoRn_w0rk> thomasvs_: yeah... I know... I'm really trying to use a little thing from the application... not the whole gstreamer... that's why I don't want to spend 2 much time on it [18:22] <thomasvs_> KoRn_w0rk: what processor do you run on ? [18:22] <KoRn_w0rk> I got work 2 do [18:22] <KoRn_w0rk> ;) [18:22] <thomasvs_> KoRn_w0rk: which is why you should stick to packages :) [18:22] <KoRn_w0rk> my linux box is on a single PII I think [18:22] <KoRn_w0rk> thomasvs_: maybe your right about the rpms [18:22] <KoRn_w0rk> ;) [18:23] <thomasvs_> KoRn_w0rk: yeah, NTPL doesn't work on less than real i686 I think [18:23] <KoRn_w0rk> but I still kinda like seeing all these compiling lines [18:23] <KoRn_w0rk> NTPL ? what's that ? [18:24] <thomasvs_> the new threading library introduced in rh9 [18:25] <Uraeus> ello [18:26] <^iain^> hey Uraeus [18:26] gort (~jg...@cs...) joined #gstreamer. [18:27] <dolphy> Uraeus: waaaaaaazzzzzzzzzzuuppp :) [18:28] <Uraeus> dolphy: just got back in from the Gym, looking over a new set of Lush icons :) [18:28] Nick change: matrixise -> mat_AWAY [18:28] <dolphy> Uraeus: da themer :) [18:28] <dolphy> Uraeus: i should give you the responsibility of making a nice theme for Getcontrol [18:29] <dolphy> Uraeus: :) [18:29] <Uraeus> dolphy: heh, well I don't really create themes, I just assemble them :) [18:29] <dolphy> Uraeus: exactly what i need :) [18:29] ghostar (~ghostar@D-128-95-117-225.dhcp4.washington.edu) left irc: "Client Exiting" [18:30] <Uraeus> dolphy: and you should use gtk stock icons for GetControl :) [18:30] <taaz> dude, doc fixing! that's so cool! [18:31] md`` (~illuminat@pD9E47891.dip.t-dialin.net) joined #gstreamer. [18:32] <thomasvs_> taaz: I hope you're not making fun of me :) [18:33] md` (~illuminat@pD9533255.dip.t-dialin.net) left irc: Killed (NickServ (Ghost: md``!~illuminat@pD9E47891.dip.t-dialin.net)) [18:33] <dolphy> Uraeus: it would be ugly [18:33] Nick change: md`` -> md` [18:33] <dolphy> Uraeus: getcontrol needs very nice gfx :) [18:36] <taaz> thomasvs_: not at all! up to date and well done docs are a great thing to have. it's one of the reasons i initially started to use gstreamer ;) [18:36] <taaz> thomasvs_: it's just often painful to keep them in good shape [18:38] <thomasvs_> taaz: yeah, same here [18:38] <thomasvs_> taaz: I have yet to do actual cleanups and so on though, I just got them back into shape and working with devhelp [18:38] <thomasvs_> now that I can brwose them easily again, I can fix them properly where needed [18:38] <thomasvs_> dolphy: next on my list is libgstplay docs so you can write them :) [18:39] <Company> i'd prefer dolphy changing libgstplay so it doesn't use 2 videosinks anymore [18:40] ChrJW_log (~cwi...@p5...) left irc: Connection timed out [18:41] <dolphy> Company: the identity trick ? :) [18:42] <Company> dolphy: whatever [18:42] <Company> dolphy: i need that for my embedding stuff [18:50] <KoRn_w0rk> Company [18:50] <KoRn_w0rk> I get an error [18:50] <KoRn_w0rk> gst-launch [18:50] <KoRn_w0rk> -bash: /usr/bin/gst-launch: No such file or directory [18:50] <KoRn_w0rk> now what ? [18:50] <thomasvs_> KoRn_w0rk: you installed the rpms, then ran gst-launch, then removed rpms and installed from source, and your bash doesn't know yet [18:50] <thomasvs_> log into a new terminal and try again [18:51] <KoRn_w0rk> ok [18:51] Nick change: thomasvs_ -> thomasvs [18:51] <thomasvs> I'm curious to see whatever that guy wants to do with mysql [18:52] <Uraeus> thomasvs: I think it is clear, he wants to store the media files in a database [18:52] <^iain^> store files in a mysql server apparently :) [18:52] <Uraeus> thomasvs: we offer that with Oracle too [18:52] <Company> thomasvs: look at the list, he answered [18:52] <Company> thomasvs: text processing via text in blobs [18:56] <thomasvs> have to go, but why would anyone want to store *files* in a *database* ? [18:56] <thomasvs> don't answer, I'll be gone anyway :) [18:57] <^iain^> its a better place than in your pants? [18:57] pb_ (~pb...@pc...) joined #gstreamer. [18:58] <Company> because PHBs think that textfiles store data and data belongs into databases [18:58] <Company> he said his company wants that [18:58] sxpert_ (~sx...@sx...) left irc: Read error: 110 (Connection timed out) [18:59] <dolphy> Company: the last possibility that has been discussed was to put identity elements behind spider and detect if there s video this way moving the videosink to the right bin [18:59] <dolphy> Company: that's a big hack :) [18:59] sxpert_ (~sx...@sx...) joined #gstreamer. [19:00] <Company> dolphy: the current solution isn't much better [19:00] <Company> dolphy: especially because it might require opening 2 xv connections when only one is available in most drivers [19:03] <Company> dolphy: do you have different branches for 0.6 and HEAD yet in the player? [19:04] <KoRn_w0rk> hey guys [19:04] <KoRn_w0rk> me again [19:04] <KoRn_w0rk> it seems that I need more pluggins to have it to work [19:05] <KoRn_w0rk> can I install gstreamer from rpm [19:05] <KoRn_w0rk> and then install the pluggins from source ? [19:05] <KoRn_w0rk> because swfdec doesn't seem to have rpms [19:08] lilo_booter (~charlie@D576AE08.kabel.telenet.be) joined #gstreamer. [19:09] <dolphy> Company: nope and libgstplay is in plugins. [19:09] <dolphy> Company: i agree the actual solution is not that good [19:09] <dolphy> Company: but that's because xvideosink is a bit glummy [19:10] <dolphy> Company: we should be able to have 2 videosink in a same pipeline [19:10] <dolphy> Company: for example a DV acquisition application with video playback and a vuemeter window [19:11] <dolphy> Company: you have 2 videosink [19:11] <dolphy> Company: i could easily make that the player plays a video and in another window make visualisation effects over sound... [19:12] <dolphy> Company: for example looking at a music concert DVD [19:12] <dolphy> Company: i d like to see the video and maybe visualisation through goom too [19:13] <dolphy> other people want to react on that topic ? [19:13] <dolphy> Uraeus ? thomasvs ? [19:13] Nick change: apoc__ -> apoc [19:15] thomasvs (~th...@19...) left irc: Read error: 110 (Connection timed out) [19:16] <Company> dolphy: 99% of the XFree video drivers don't allow 2 overlays [19:18] <Company> dolphy: try running gst-launch videotestsrc ! xvideosink twice [19:18] <dolphy> Company: then if xv is not available there s no hardware scaling [19:18] <Company> and move the windows next to each other [19:18] <dolphy> Company: is it such a big issue ? [19:18] <BBB> yes [19:19] <Company> gawd, that looks ugly [19:19] Action: Company is gonna start 10 instances of that now :) [19:20] <dolphy> ERROR: /pipeline0/xvideosink0: no image to draw [19:20] <Company> you're using 0.6 ? [19:21] <Company> in that case run gst-launch filesrc location=/path/to/some.mpg ! mpegdemux ! mpeg2dec ! xvideosink twice [19:21] <dolphy> yup 0.6.1 for totem porting [19:23] <Company> the problem is that you really want to just have 1 window, not 2 [19:23] <Company> and my embedding stuff can only embed 1 window [19:23] tjansen (~tjansen@pD9E215F2.dip.t-dialin.net) joined #gstreamer. [19:23] <Company> you could probably hack it to display more than one window, but that is a bit sucky, isn't it? [19:23] <dolphy> that's nice :) [19:24] <Company> :p [19:25] <dolphy> and disabling xv ? [19:25] <dolphy> it does not play with spider [19:25] <dolphy> but will it do the same thing ? [19:26] <yippi> what applications in Gnome 2.2 use gstreamer? [19:26] <^iain^> gnome-sound-recorder and gstreamer-properties [19:27] <yippi> what is gstreamer-properties? [19:27] <yippi> are there no playback tools that are a part of Gnome? [19:27] <yippi> that use gstreaer [19:27] <yippi> gstreamer i mean [19:27] <^iain^> gnome-sound-recorder does playback [19:27] <^iain^> and nautilus-media does as well [19:27] <BBB> and rhythmbox [19:27] <BBB> and gst-player [19:28] <BBB> (but those aren't part of gnome2.2) [19:28] <dolphy> Company: in libgstplay's case the 2 windows are never activated together [19:29] <dolphy> Company: but they can be [19:29] <yippi> is rhythmbox ready for prime time? [19:30] <BBB> it works quite ok, yes [19:30] <Company> dolphy: i haven't really looked at the API, but my code requires embedding before going from NULL->PAUSED [19:31] <apoc> BBB : flacenc need to be backported from HEAD. You prefer a patch or can I commit ? [19:31] <BBB> apoc: make a patch please [19:31] <Company> yippi: it's quite ok, but not "prime time" [19:31] <yippi> when will it be ready, do you think? [19:31] <apoc> BBB : bug report ? [19:31] <BBB> there's some extra constraints for 0_6 commits, so I do prefer to review it [19:31] <BBB> apoc: yeah, please [19:31] <apoc> BBB : ok [19:31] <Company> yippi: 2.6 probably [19:31] gort (~jg...@cs...) left irc: "Client exiting" [19:32] <dolphy> Company: you mean that you want to throw the xid on READY ? [19:32] <apoc> Conpany, BBB : are you agree to add GstProbe to 0.6.2 ? [19:32] <Company> yippi: if the devs get up to snuff on it and it doesn't idle for ages again [19:32] <BBB> what's GstProbe? [19:32] <Company> apoc: no [19:32] <BBB> is it a new struct? then, no please [19:32] <Company> apoc: that'd be a core change [19:33] <apoc> Company: an addition, no a change [19:33] <BBB> apoc: what is it? [19:33] <Company> BBB: pad querying stuff [19:33] <apoc> BBB : with GstProbe you can get buffers from pad [19:33] <Company> apoc: additions are changes [19:35] <BBB> not always [19:35] <BBB> apoc: I need a link with some code or so ;) [19:35] <apoc> I'm writing a gnome-vfs module for sound-juicer and i need gstprobe ;) [19:36] Nick change: Uraeus -> Ura_dinner [19:37] <apoc> BBB : cvs HEAD : gstreamer/gst/gstprobe.[ch] [19:38] <BBB> any reason why it's no a GObject? [19:38] <apoc> BBB : you should ask wtay ;) [19:39] <BBB> wtf does this do? [19:39] <BBB> the header is totally weird ;) [19:40] <dolphy> BBB: Harry Potter emulator :) [19:40] <BBB> right :D [19:40] <BBB> can't make any sense out of the code [19:40] <Company> BBB: it needs an addition to the GstPad struct [19:40] <BBB> then no :) [19:41] <Company> it's untested [19:41] <apoc> BBB : you can't find a example here : gstreamer/tests/probe [19:41] <Company> Freudian slip :p [19:41] <dolphy> apoc: no way to work around without it ? [19:42] thomasvs (~th...@99...) joined #gstreamer. [19:42] <apoc> dolphy: I don't know how to get buffers from pipeline [19:42] <Company> apoc: tee ! fakesink, connect to the handoff signal [19:43] <apoc> Company: handoff ? [19:44] <apoc> Company: do you have an example, please ? [19:45] foser (d0...@22...) left irc: Remote closed the connection [19:45] <Company> apoc: no, but it should be kind of self-explanatory [19:45] <BBB> g_signal_connect (G_OBJECT (fakesink), "handoff", func, ptr); [19:46] <dolphy> that spew buffers ? [19:46] <BBB> yes [19:46] <dolphy> interesting :) [19:46] <apoc> BBB, Company: ok I'll take a look thx for the tips :) [19:47] <dolphy> k time to go back home [19:47] <dolphy> see ya later dudes :) [19:47] <BBB> bye [19:47] dolphy (~do...@po...) left irc: "Network down, IP Packets delivered via UPS" [19:49] <thomasvs> ok, so make distcheck works again [19:49] <thomasvs> is it ok if I branch for release now ? [19:49] <BBB> we don't need to branch for the release [19:49] <BBB> we already have a 0_6 release [19:50] <BBB> and it's a closed branch already [19:50] <thomasvs> ok, is it ok then if I do the prerelease then ? [19:50] <BBB> oh, for gstprops.c:426 - *check++; is invalid C code [19:50] <thomasvs> :) [19:50] <BBB> thomasvs: sure :) [19:50] <BBB> but there's 3 open bugs [19:50] <thomasvs> are we considering them for 0.6.2 ? [19:50] <BBB> s/*check++;/check++;/ is ok [19:50] <BBB> thomasvs: yes [19:50] <thomasvs> which ones specifically ? [19:51] <BBB> at least one of them *must* be fixed imo [19:51] <BBB> http://bugzilla.gnome.org/buglist.cgi?product=GStreamer&bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&target_milestone=0.6.2&email1=&emailtype1=substring&emailassigned_to1=1&email2=&emailtype2=substring&emailreporter2=1&changedin=&chfieldfrom=&chfieldto=Now&chfieldvalue=&short_desc=&short_desc_type=substring&long_desc=&long_desc_type=substring&bug_file_loc=&bug_file_loc_type=substring&status_whiteboard=&status_ [19:51] <BBB> whiteboard_type=substring&keywords=&keywords_type=anywords&op_sys_details=&op_sys_details_type=substring&version_details=&version_details_type=substring&cmdtype=doit&newqueryname=&order=Reuse+same+sort+as+last+time&form_name=query [19:51] <BBB> oops [19:51] <thomasvs> heh, that don't work :) [19:51] <BBB> #111685, #112143 and #111277 [19:51] <BBB> #111277 is a simple request-for-test [19:52] <BBB> I think swfdec is broken on LE machines [19:52] foser (d0...@22...) joined #gstreamer. [19:52] <BBB> I want that tested before release [19:52] <BBB> and 111685 - I want to work on that right after the first prerelease [19:53] <thomasvs> well [19:53] <thomasvs> neither of the three seem resolvable very soon [19:53] <thomasvs> so either we wait for the release until they have clean patches [19:54] <thomasvs> or we don't wait and postpone to 0.6.3 [19:54] ChrisHJW (~chr...@p5...) joined #gstreamer. [19:55] <BBB> the first is (I think) a failing change_state in modplug [19:55] <BBB> don't know exactly [19:56] <Company> <BBB> oh, for gstprops.c:426 - *check++; is invalid C code <-- why? [19:57] <thomasvs> BBB: what I mean is, as long as we don't have patches for them it's not worth doing prereleases [19:57] <thomasvs> BBB: unless these are not 0.6.2 blockers [19:58] <thomasvs> the spider one for example, I don't think it's a blocker [19:58] <BBB> Company: it's either check++; or *check++ (which is the same, but gives the value of *check too) and use that value in the expression [19:59] <BBB> e.g. while (*check++ != '\0'); [19:59] <Company> yes [19:59] <BBB> which is an ugly way of finding the end of a string [19:59] <Company> but it's still valid C [19:59] <BBB> ok, ok [19:59] <BBB> but it's unclean [19:59] <Company> it's stupid C, but valid [19:59] <BBB> and unobvious [19:59] <BBB> it should be check++; [19:59] <Company> yeah, i agree there [19:59] <BBB> ok, then we agree ;) [19:59] <Company> you confused me a bit with that [20:00] <BBB> :) [20:00] <BBB> sorry [20:00] <Company> btw: i have no idea about 112143 [20:00] <BBB> oh, that was uraeus' idea: "randomly assign bugs to people" [20:00] <BBB> ;) [20:01] <BBB> reassign to default or so, I'll have a look [20:01] <Company> i just said that, because i'm normally the one to ask about that :) [20:02] <Company> before you start on 0.6.2 prereleases: the dolphy-crash fix with the refcounting is in? [20:02] <Company> and do you want alsa in 0.6? [20:03] <thomasvs> Company: how do I test that fix for dolphy ? [20:03] <Company> and would anybody like to do the backport for me? [20:03] <Company> thomasvs: clicking next in gst-player with a playlist very fast and see if it crashes after 4 or 5 times clicking [20:03] <thomasvs> Company: that's it ? [20:04] <thomasvs> will check that [20:04] <Company> thomasvs: if it crashes the backport is not done yet [20:04] <thomasvs> hm, I should really autoconf/automake-ify ffmpeg [20:04] <Company> thomasvs: it should crash in g_object_unref somewhere [20:05] <Company> thomasvs: and you could try copying HEAD's gstalsa.[ch] into 0.6 and see if that builds [20:05] <Company> that would save me from having to install 0.6 [20:05] <thomasvs> Company: is alsa finished for now, or do you expect more changes later ? [20:05] <thomasvs> Company: heh, use my cvs script :) [20:06] <Company> thomasvs: only bugfixes [20:06] <thomasvs> I did click next [20:06] <thomasvs> then I got [20:06] <thomasvs> ** (lt-gst-player:15091): WARNING **: pad unlink for different groups, implement me [20:06] <thomasvs> and now the ui hangs [20:07] <Company> that doesn't sound like it's in yet :) [20:07] <thomasvs> after a while, things play again [20:07] <thomasvs> might be a timestamping issue [20:07] <BBB> thomasvs: your apt repository is broken... [20:07] <thomasvs> Company: looks like a different issue [20:07] <thomasvs> BBB: how so ? [20:08] <BBB> my installation says that gstreamer-flac depends on flac-1.1.0, but the apt repo carries 1.0.3 [20:08] <BBB> and dvdread-0.9.3 vs. 0.9.4 [20:08] <BBB> hm, maybe it's the other way around, I have 1.1.0 and it requires 1.0.3 [20:08] <thomasvs> [thomas@thocra thomas]$ ldd /usr/lib/gstreamer-0.6/libgstflac.so [20:08] <thomasvs> libFLAC.so.3 => /usr/lib/libFLAC.so.3 (0x4001d000) [20:08] <BBB> don't know exactly [20:09] <thomasvs> BBB: right, so the apt repo is not broken [20:09] <thomasvs> BBB: you have newer libs [20:09] <BBB> hm... [20:09] <BBB> can you upgrade them? [20:09] <thomasvs> in time, ys [20:09] <BBB> ok, thnx :) [20:09] <thomasvs> I'm pretty much done with the autobuilder [20:09] <thomasvs> except for some nagging issues with rpm [20:10] yippi (~br...@be...) left irc: "Client exiting" [20:10] <thomasvs> it's blocking the ui while doing "searching for length". I'lltry with mp3 only [20:11] <thomasvs> ok, so "trying to get length" is pretty much broken over here :) [20:11] <thomasvs> maybe we need to extract a cmdline test app to debug that problem [20:12] <thomasvs> it shouldn't block the ui either for that [20:22] Nick change: mat_AWAY -> matrixise [20:24] ChrHJW_log (~cwi...@p5...) joined #gstreamer. [20:36] matrixise (~stef@213.213.206.2) got netsplit. [20:37] matrixise (~stef@213.213.206.2) returned to #gstreamer. [20:38] ChrHJW_log (~cwi...@p5...) got netsplit. [20:38] Vakor (ms...@c1...) got netsplit. [20:38] KoRn_w0rk (~ab...@pp...) got netsplit. [20:38] camh (~ca...@xi...) got netsplit. [20:38] Titanium (~gi...@po...) got netsplit. [20:38] pb_ (~pb...@pc...) got netsplit. [20:38] Ura_dinner (~csc...@wp...) got netsplit. [20:38] ^iain^ (~ia...@us...) got netsplit. [20:38] taaz (~dlehn@66.37.66.32) got netsplit. [20:38] tsiar ([UnO...@dh...) got netsplit. [20:38] sxpert_ (~sx...@sx...) got netsplit. [20:38] md` (~illuminat@pD9E47891.dip.t-dialin.net) got netsplit. [20:38] sxpert (~sxpert@APh-Aug-105-1-6-135.w80-14.abo.wanadoo.fr) got netsplit. [20:38] lilo (le...@li...eenode) got netsplit. [20:38] matrixise (~stef@213.213.206.2) left irc: "Client exiting" [20:39] ChrHJW_log (~cwi...@p5...) returned to #gstreamer. [20:39] sxpert_ (~sx...@sx...) returned to #gstreamer. [20:39] pb_ (~pb...@pc...) returned to #gstreamer. [20:39] md` (~illuminat@pD9E47891.dip.t-dialin.net) returned to #gstreamer. [20:39] Ura_dinner (~csc...@wp...) returned to #gstreamer. [20:39] ^iain^ (~ia...@us...) returned to #gstreamer. [20:39] KoRn_w0rk (~ab...@pp...) returned to #gstreamer. [20:39] Vakor (ms...@c1...) returned to #gstreamer. [20:39] taaz (~dlehn@66.37.66.32) returned to #gstreamer. [20:39] sxpert (~sxpert@APh-Aug-105-1-6-135.w80-14.abo.wanadoo.fr) returned to #gstreamer. [20:39] lilo (le...@li...eenode) returned to #gstreamer. [20:39] camh (~ca...@xi...) returned to #gstreamer. [20:39] Titanium (~gi...@po...) returned to #gstreamer. [20:39] tsiar ([UnO...@dh...) returned to #gstreamer. [21:03] dolphy (~do...@21...) joined #gstreamer. [21:04] <apoc> dolphy: I keep getting this error with gst-player [21:05] <apoc> dolphy: ** Message: not adding '/home/apoc/15.flac' with mimetype 'application/x-flac' [21:07] <Company> apoc: the playlist is selective about what files to add :) [21:08] <dolphy> apoc: yup playlist [21:08] <dolphy> i ll fix that in totem first [21:08] <apoc> Company: I can't play mod file too same error [21:09] <Company> apoc: the mimetype has to be audio/* or video/* for some stupid reason [21:09] <dolphy> yup i ll check that with hadess [21:10] <Company> and it's heavily un-gstreamerish [21:11] Nick change: wtay-zZz -> wtay [21:11] <wtay> yo [21:11] <^iain^> yo [21:11] <Company> oy [21:12] <apoc> yo wtay [21:12] <Company> wtay: what'S your opinion on http://bugzilla.gnome.org/show_bug.cgi?id=113258 ? [21:13] <apoc> someone using redhat 9 or mandrake 9.1 ? [21:13] <wtay> make clear that a NULL has somewhat different meaning? [21:13] ct (~Chr...@th...) joined #gstreamer. [21:14] <ds> wtay: NULL could have two meanings [21:14] <Company> wtay: gst_caps_intersect (NULL, NULL) means what? [21:14] <wtay> no intersection [21:14] <ct> Company: hey, think i figured out how i'm going to get the gstbuff stuff plugged into xine [21:14] <Company> so i can't connect filesrc to filesink? [21:15] <Company> ct: great [21:15] <wtay> Company: sure, it only fails the connect if one of the caps was not NULL [21:15] <ct> Company: but one question about the linking function... [21:16] <ct> Company: the incoming caps is from the pad requesting the link right? [21:16] <wtay> I have some code and updated plugin with the duration field in gstbuffer [21:17] <BBB> cool :) [21:17] <wtay> + removed automatic OFFSET calculation in subbuffer code [21:17] <wtay> + updated bytestream and filesrc that update the offset [21:17] <BBB> that means that merging two buffers will always copy the buffer? [21:17] <wtay> BBB: er, no [21:18] <BBB> didn't that depend on both having the same parent and using a continuous offset/size? [21:18] <wtay> offset is not used in the subbuffer code to make the subbuffer [21:18] <BBB> (then, it'd recreate a subbuffer of the parent) [21:18] <wtay> BBB: the data pointers + size must be aligned [21:18] <BBB> oh, ok [21:18] <BBB> good enough [21:18] <BBB> :) [21:18] <wtay> correct even :) [21:19] <BBB> :) [21:19] <BBB> which plugins do you have updated? [21:19] <Company> ct: the linking function is supposed to check if it can handle the caps and initialize itself for handling these caps [21:19] <BBB> and something else [21:19] <Company> ct: if it can, it returns GST_PAD_LINK_OK [21:19] <ct> Company: right [21:19] <BBB> in case two elements both use unfixed caps, which values should be used? [21:19] <BBB> the highest or the lowest? [21:19] <ct> Company: but the linking function recieves a pad and a caps [21:20] <ct> Company: the pad that's passed is *the* pad requesting a link, right? [21:20] <wtay> BBB: the plugins decide (one takes initiative) [21:20] <Company> ct: the pad that's passed is the pad that is supposed to accept these caps [21:20] <ct> Company: and the caps that are passed are caps that are associated with the passed pad? [21:20] <wtay> BBB: in most cases the src plugin [21:20] <ct> Company: hmmm [21:21] <ds> anyone have a clue why gst-player (and xvideosink plugging in general) stopped working last night (0.6.1, worked earlier in the evening)? [21:21] <Company> ct: and the caps passed are the caps where the pad should check for acceptanyce [21:21] <ct> Company: *ah* ok so i'm cool then [21:21] <ct> Company: sweet [21:22] <Company> BBB: if no plugin has preferences and the caps are unfixed, just use the first one available [21:22] <Company> BBB: which is the best possible in the ideal case [21:22] <BBB> ok [21:22] <wtay> ds: still works here [21:23] <Company> BBB: alsasrc uses the first one available (which is unsigned 8 bits), because the capslist is generated from alsa headers [21:23] <Company> BBB: but since there is no function for ordering caps I don't care [21:24] <wtay> oh, and I removed the units format [21:26] <BBB> hm... [21:26] <wtay> it's now DEFAULT but I could rename it to UNITS again if that makes more sense [21:26] <BBB> Company: you re-added filtered capsnego to CVS, right? [21:26] <BBB> gst-launch-wise, that is [21:27] <BBB> could you give an example? [21:27] <Company> BBB: yes, HEAD does it, examples are in the gst-launch man page and testsuite/caps/string-conversions [21:28] <BBB> does't exist [21:28] Action: BBB tries manpage [21:28] cschalle_ (~csc...@wp...) joined #gstreamer. [21:28] <Company> testsuite/caps/string-conversions.c [21:29] <BBB> doesn't exist either [21:29] <BBB> ? [21:30] <Company> BBB: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/gstreamer/gstreamer/testsuite/caps/string-conversions.c?rev=1.1&content-type=text/vnd.viewcvs-markup [21:32] md` (~illuminat@pD9E47891.dip.t-dialin.net) left irc: Remote closed the connection [21:33] <BBB> thnx [21:34] <Company> maybe you should do a cvs update? ;) [21:35] <thomasvs> dolphy: while testing gst-player, I did click next [21:35] <thomasvs> then I got: ** (lt-gst-player:15091): WARNING **: pad unlink for different groups, implement me [21:35] <thomasvs> and now the ui hangs [21:36] <thomasvs> dolphy: it always takes a few secs on an ogg to find the length of a track [21:36] <thomasvs> dolphy: any idea why ? [21:36] <BBB> nice stuff :) [21:36] thomasvs (~th...@99...) left irc: Read error: 54 (Connection reset by peer) [21:37] thomasvs (~th...@13...) joined #gstreamer. [21:37] <Company> BBB: thx [21:37] <BBB> I've fixed some v4lsrc threading issues [21:37] <Company> BBB: now you need to extensively test it to find the bugs :) [21:37] <BBB> ... by removing all its threading code [21:37] <Company> hehe [21:38] <BBB> GST_FLAG_SET(element, GST_ELEMENT_THREAD_SUGGESTED); performs equally well [21:38] <Company> like vishnu did with the pipeline parsing - he found loads of bugs in it [21:38] <BBB> I'm still wondering in what kind of moronic state I wrote that code [21:38] <BBB> I'll try :) [21:38] <Company> BBB: that flag does exactly nada, zilch and zero :) [21:38] md` (~illuminat@pD9E47891.dip.t-dialin.net) joined #gstreamer. [21:38] <BBB> Company: I know [21:39] <BBB> Company: but it'll make the application developer (me ;) ) try to put it in a thread with a high priority [21:39] <BBB> so v4lsrc doesn't need any threading code [21:39] <BBB> gst-rec becomes a whole lot more useful then [21:40] <BBB> I've got some ideas written down for the next rewrite [21:40] <BBB> (of gst-rec) [21:40] <Company> yeah, go ahead [21:41] <Company> it sounds as if i need to get some video capture device from somewhere [21:43] <BBB> I'll upload them to CVS for general review :) [21:46] Ura_dinner (~csc...@wp...) left irc: Read error: 110 (Connection timed out) [21:47] <BBB> oh [21:47] <BBB> bw [21:47] <BBB> I've got a problem with a generic VideoSource parentclass [21:47] <BBB> v4l2src should have to decend from it [21:47] <BBB> right? [21:48] <BBB> so v4l2element and v4lelement too [21:48] <BBB> however [21:48] <BBB> v4lmjpegsink and v4l2sink also decend from v4l*element [21:48] <BBB> so it doesn't work! [21:48] <BBB> I need > 1 parent [21:49] <Company> no chance [21:50] <BBB> hm [21:50] <BBB> crap [21:50] Last message repeated 4 time(s). [21:50] <Company> you might do it differently [21:50] <Company> and use V4lPrivate GObjects [21:50] walters (wa...@ve...) joined #gstreamer. [21:54] <thomasvs> crap, I keep getting thrown off :) [21:55] <thomasvs> dolphy: a question [21:55] <thomasvs> <thomasvs> dolphy: while testing gst-player, I did click next [21:55] <thomasvs> <thomasvs> then I got: ** (lt-gst-player:15091): WARNING **: pad unlink for different groups, implement me [21:55] <thomasvs> <thomasvs> and now the ui hangs [21:55] <thomasvs> <thomasvs> dolphy: it always takes a few secs on an ogg to find the length of a track [21:55] <thomasvs> <thomasvs> dolphy: any idea why ? [21:55] <thomasvs> wtay: btw, what does tests/bufspeed do ? is it used for anything ? why does it have it's own gstbuffer code ? [22:01] <dolphy> thomasvs: this warning is something we have for a while :) [22:01] <dolphy> thomasvs: wim told me it has to stay there because someone has to implement it [22:02] <dolphy> thomasvs: for the time to get length i don't know.. [22:02] <Company> dolphy: basically an "i'm lazy and won't do it" ;) [22:02] <dolphy> thomasvs: maybe queries are longer because of a change i don't know about ? [22:03] Marsupilami23 (~Mar...@15...) left irc: "I don't wanna grow up, I'm a Toys R Us Kid. There's a million toys at Toys R Us that I can play with!" [22:03] <apoc> BBB : the patchs attached to the libflac bug, doesn't work with sound-juicer [22:03] <apoc> BBB : I'm looking at this ... [22:04] md` (~illuminat@pD9E47891.dip.t-dialin.net) got netsplit. [22:04] Titanium (~gi...@po...) got netsplit. [22:04] walters (wa...@ve...) got netsplit. [22:04] ^iain^ (~ia...@us...) got netsplit. [22:04] taaz (~dlehn@66.37.66.32) got netsplit. [22:04] tsiar ([UnO...@dh...) got netsplit. [22:04] pb_ (~pb...@pc...) got netsplit. [22:04] sxpert (~sxpert@APh-Aug-105-1-6-135.w80-14.abo.wanadoo.fr) got netsplit. [22:04] sxpert_ (~sx...@sx...) got netsplit. [22:05] walters (wa...@ve...) returned to #gstreamer. [22:05] md` (~illuminat@pD9E47891.dip.t-dialin.net) returned to #gstreamer. [22:05] sxpert_ (~sx...@sx...) returned to #gstreamer. [22:05] pb_ (~pb...@pc...) returned to #gstreamer. [22:05] ^iain^ (~ia...@us...) returned to #gstreamer. [22:05] taaz (~dlehn@66.37.66.32) returned to #gstreamer. [22:05] sxpert (~sxpert@APh-Aug-105-1-6-135.w80-14.abo.wanadoo.fr) returned to #gstreamer. [22:05] Titanium (~gi...@po...) returned to #gstreamer. [22:05] tsiar ([UnO...@dh...) returned to #gstreamer. [22:06] Titanium (~gi...@po...) got netsplit. [22:06] md` (~illuminat@pD9E47891.dip.t-dialin.net) got netsplit. [22:06] sxpert (~sxpert@APh-Aug-105-1-6-135.w80-14.abo.wanadoo.fr) got netsplit. [22:06] sxpert_ (~sx...@sx...) got netsplit. [22:06] ^iain^ (~ia...@us...) got netsplit. [22:06] tsiar ([UnO...@dh...) got netsplit. [22:06] taaz (~dlehn@66.37.66.32) got netsplit. [22:06] pb_ (~pb...@pc...) got netsplit. [22:06] walters (wa...@ve...) got netsplit. [22:07] walters (wa...@ve...) returned to #gstreamer. [22:07] md` (~illuminat@pD9E47891.dip.t-dialin.net) returned to #gstreamer. [22:07] sxpert_ (~sx...@sx...) returned to #gstreamer. [22:07] pb_ (~pb...@pc...) returned to #gstreamer. [22:07] ^iain^ (~ia...@us...) returned to #gstreamer. [22:07] taaz (~dlehn@66.37.66.32) returned to #gstreamer. [22:07] sxpert (~sxpert@APh-Aug-105-1-6-135.w80-14.abo.wanadoo.fr) returned to #gstreamer. [22:07] Titanium (~gi...@po...) returned to #gstreamer. [22:07] tsiar ([UnO...@dh...) returned to #gstreamer. [22:12] <thomasvs> dolphy: any idea why the ui blocks while querying for length ? [22:13] ct (~Chr...@th...) left irc: "Client exiting" [22:14] <dolphy> thomasvs: because the query to element is done in the main thread [22:14] <dolphy> thomasvs: i think [22:17] <thomasvs> dolphy: can't the ui be done in a separate thread then ? [22:17] <Company> thomasvs: it is already [22:17] <Company> thomasvs: just this one thing cannot be done without adding _another_ thread [22:18] <dolphy> thomasvs: most of the stuff is in a separate thread already [22:18] <dolphy> thomasvs: but signal emiting and element querying is in main thread [22:19] <Company> which seems right somehow (apart from the fact that the query isn't handed over) [22:19] <thomasvs> hm, so basically I'm screwed ? [22:20] <dolphy> i don't know why this query takes that much time now [22:20] <dolphy> i thought it was on mutexes [22:20] <dolphy> but it's not [22:20] cschalle_ (~csc...@wp...) left irc: "Client exiting" [22:23] <thomasvs> does it take a long time for you too then ? [22:23] <thomasvs> if (play->length_nanos == 0LL){ [22:23] <thomasvs> /* try to get the length up to 16 times */ [22:23] <thomasvs> play->get_length_attempt = 16; [22:23] <thomasvs> play->timeout_add_func (200, (GSourceFunc) gst_play_get_length_callback, play); [22:23] <thomasvs> maybe this might be it ? :) [22:24] <thomasvs> it seems kind of backwards that a simple operation like get-length should take this long and be handles async [22:27] <BBB> apoc: I've had a quick look, the patch seems fine to me :) [22:27] <thomasvs> dolphy: well, I wanted to test if the unref problems are solved, but I can't even press next quickly enough this time :) [22:30] <apoc> BBB : I have the last version with a metadata property this time ;) [22:32] <BBB> flacenc doesn't have a metadata property? [22:32] <BBB> hm... [22:32] <BBB> rrrrrrrright [22:32] <BBB> :) [22:33] tjansen_ (~tjansen@pD9E213C8.dip.t-dialin.net) joined #gstreamer. [22:40] <ds> what was the flac change in configure.ac about? [22:40] Nick change: ds -> ds-work [22:41] <apoc> ds-work: flacenc switched from FLAC__stream_encoder to FLAC__seekable_stream_encoder [22:41] <ds-work> what version? [22:42] <apoc> libflac version ? [22:42] <ds-work> yes [22:42] <apoc> 1.1 here [22:43] <ds-work> so they changed the ABI again. how nice. not [22:44] <^iain^> FLAC RULES! [22:46] <apoc> FLAC__seekable_stream_encoder was added in 1.0.4 [22:46] <wtay> thomasvs: bufspeed is an old experiment.. you can probably remove the code.. [22:46] <ds-work> apoc: but the old symbol was removed [22:47] <apoc> ds-work: no, but with FLAC__seekable_stream_encoder, the encoder update the STREAMINFO [22:48] <apoc> ds-work: with last flacenc, i get no errors with metaflac [22:50] <BBB> mp4 (mpeg audio v.4) is gonna be lossless too [22:50] tjansen (~tjansen@pD9E215F2.dip.t-dialin.net) left irc: Read error: 110 (Connection timed out) [22:50] <BBB> probably just like mp3 vs. ogg [22:50] <BBB> let's hope flac settles first [22:51] <apoc> BBB : yes [22:53] <taaz> what's bufspeed? is that as elite as lat.c? ;) [22:54] <dolphy> thomasvs: the unref problems are solved for sure :) [22:59] <thomasvs> wtay: ok, will do [22:59] <thomasvs> wtay: so that memchunk stuff isn't going to be kept somehow ? [22:59] <wtay> thomasvs: it's obsolete now [22:59] <thomasvs> ok, removing it then [23:04] <ds-work> so is flac-1.1.0 supposed to be ABI and API stable now? [23:06] <ds-work> Company: could you test the new sparc asm? [23:06] <thomasvs> ds-work: check their homepage for info on how they change version numbers [23:06] <thomasvs> I believe the minor gets changed on file format changes or something [23:08] <Company> ds-work: one moment [23:09] <ds-work> thomasvs: I find no information about API. Just stream format, which is rather independent [23:11] Nick change: wtay -> wtay-away [23:13] <Company> ds-work: gstatomic_impl.h: In function `gst_atomic_int_dec_and_test': [23:13] <Company> gstatomic_impl.h:305: parse error before `[' [23:17] tjansen_ (~tjansen@pD9E213C8.dip.t-dialin.net) left #gstreamer. [23:27] tjansen (~tjansen@pD9E213C8.dip.t-dialin.net) joined #gstreamer. [23:27] thomasvs_ (~th...@13...) joined #gstreamer. [23:27] thomasvs (~th...@13...) left irc: Read error: 104 (Connection reset by peer) [23:36] <apoc> night all [23:36] Nick change: apoc -> apoc-zZz [23:43] <ds-work> Company: sorry, only fixed one function [23:46] <ds-work> Company: try it now [00:00] --- Tue May 20 2003 [00:07] <thomasvs_> ds-work: how optional is mad for swfdec ? [00:07] <thomasvs_> ds-work: and am I free to commit spec changes for it ? a lot of buildreqs are missing [00:07] <ds-work> thomasvs_: in any sane configuration, mad is required [00:07] <ds-work> and please fix as necessary [00:08] <thomasvs_> ok [00:08] <thomasvs_> *** attempt to put segment in horiz list twice [00:08] <ds-work> libart bug [00:08] <thomasvs_> are those important warnings ? [00:08] <thomasvs_> ok [00:08] <ds-work> nothing I can do about it [00:09] <ds-work> I should probably submit a patch, though :) [00:09] <ds-work> er, to make it shut up [00:16] <dolphy> ok fullscreen and hide/show controls is fixed in gst-totem [00:16] <dolphy> finally :) [00:17] <dolphy> i now have to fix scale ratio handling [00:18] <thomasvs_> dolphy: does totem work against 0.6 or only 0.7 ? [00:18] <dolphy> 0.6.1 [00:18] <thomasvs_> ok [00:18] Nick change: thomasvs_ -> thomasvs [00:18] Action: thomasvs starts building [00:18] <dolphy> and it will work like a charm with 0.6.2 [00:21] <thomasvs> dolphy: will it also have length query problems ? [00:22] <ds-work> building swfdec or gstreamer-0.6.2? [00:22] <thomasvs> ds-work: swfdec is built, rpms and all, will be in the repo tomorrow [00:22] <thomasvs> on to totem [00:22] <dolphy> thomasvs: should :) [00:22] <ds-work> cool [00:23] BBB (~rb...@01...) left irc: "Client exiting" [00:40] taaz (~dlehn@66.37.66.32) left irc: Read error: 110 (C... [truncated message content] |