From: Miguel F. <mi...@ce...> - 2001-11-29 15:51:00
|
i forgot to cc xine-dev, this might be useful for someone else... |
From: Rich W. <rj...@he...> - 2001-11-29 16:14:53
|
On Thursday 29 Nov 2001 3:51 pm, Miguel Freitas wrote: > i forgot to cc xine-dev, this might be useful for someone else... Indeed it was. Just a small quation; below is the trace from my latest=20 attempts to get highlighting to work. These events are sent just at a sce= ne=20 change (jump). As you can see, since there is no frame being shown betwee= n=20 the events, they are both queued and then executed together. The problem = is=20 that the are executed in the reverse order to which they were called (i.e= =2E=20 the hide event is processed before the last show event). Is this a bug in my code or the new OSD stuff? Rich dvdnav_plugin_read_block: Highlight area is (426,145)-(526,350), display = =3D 1 MALLOC1: overlay_event (nil), len=3D4 MALLOC2: overlay_event 0x82dce08, len=3D4 MALLOC1: overlay (nil), len=3D128 MALLOC2: overlay 0x82dce30, len=3D128 BUTTON show=3D1 clut [0 0 0 0] trans [48 100 9 64] left =3D 426 right =3D 526 top =3D 145 bottom =3D 350 dvdnav_plugin_read_block: FIXME: VTS change dvdnav_plugin_read_block: Highlight area is (37156,16505)-(63488,16548),=20 display =3D 0 MALLOC1: overlay_event (nil), len=3D4 MALLOC2: overlay_event 0x82dcf40, len=3D4 MALLOC1: overlay (nil), len=3D128 MALLOC2: overlay 0x82dcf68, len=3D128 BUTTON show=3D0 clut [0 0 0 0] trans [48 100 9 64] left =3D 0 right =3D 0 top =3D 0 bottom =3D 0 dvdnav_plugin_read_block: FIXME: Cell change video_overlay: HIDE MENU NOW 0 MENU BUTTON NOW video_overlay: error: button event received and no overlay allocated. Segmentation fault --=20 Rich You need no longer worry about the future. This time tomorrow you'll be = dead. |
From: Miguel F. <mi...@ce...> - 2001-11-29 16:23:36
|
Rich, i don't know what your debug messages mean, but it seems that you send 2 button events without any menu show event? if sent the show event, have you checked their pts values? i don't know what happens if both have pts = 0, i will check what order they are processed. there might be a bug in this case... Miguel On Thu, 2001-11-29 at 14:14, Rich Wareham wrote: > On Thursday 29 Nov 2001 3:51 pm, Miguel Freitas wrote: > > i forgot to cc xine-dev, this might be useful for someone else... > > Indeed it was. Just a small quation; below is the trace from my latest > attempts to get highlighting to work. These events are sent just at a scene > change (jump). As you can see, since there is no frame being shown between > the events, they are both queued and then executed together. The problem is > that the are executed in the reverse order to which they were called (i.e. > the hide event is processed before the last show event). > > Is this a bug in my code or the new OSD stuff? > > Rich > > dvdnav_plugin_read_block: Highlight area is (426,145)-(526,350), display = 1 > MALLOC1: overlay_event (nil), len=4 > MALLOC2: overlay_event 0x82dce08, len=4 > MALLOC1: overlay (nil), len=128 > MALLOC2: overlay 0x82dce30, len=128 > BUTTON > show=1 > clut [0 0 0 0] > trans [48 100 9 64] > left = 426 right = 526 top = 145 bottom = 350 > dvdnav_plugin_read_block: FIXME: VTS change > dvdnav_plugin_read_block: Highlight area is (37156,16505)-(63488,16548), > display > = 0 > MALLOC1: overlay_event (nil), len=4 > MALLOC2: overlay_event 0x82dcf40, len=4 > MALLOC1: overlay (nil), len=128 > MALLOC2: overlay 0x82dcf68, len=128 > BUTTON > show=0 > clut [0 0 0 0] > trans [48 100 9 64] > left = 0 right = 0 top = 0 bottom = 0 > dvdnav_plugin_read_block: FIXME: Cell change > video_overlay: HIDE MENU NOW 0 > MENU BUTTON NOW > video_overlay: error: button event received and no overlay allocated. > Segmentation fault > > -- > Rich > > You need no longer worry about the future. This time tomorrow you'll be dead. |
From: James Courtier-D. <Ja...@su...> - 2001-11-29 20:46:42
|
Well, if vpts=0; the event will be put at the top of the queue and executed immediately. If two events get sent close together with vpts=0, it should(I had intended it to) do a fifo type function, but there might be a bug. I will play with the latest NAV and SPU code this weekend. Cheers James P.S. Richard, does the latest libdvdnav send NAV packs to the demuxers yet ? > -----Original Message----- > From: xin...@li... > [mailto:xin...@li...]On Behalf Of Miguel > Freitas > Sent: 29 November 2001 16:24 > To: Rich Wareham > Cc: xine-dev > Subject: Re: [Fwd: Re: [xine-devel] reworked spu and osd] > > > Rich, > > i don't know what your debug messages mean, but it seems that you send 2 > button events without any menu show event? > > if sent the show event, have you checked their pts values? i don't know > what happens if both have pts = 0, i will check what order they are > processed. there might be a bug in this case... > > Miguel > > On Thu, 2001-11-29 at 14:14, Rich Wareham wrote: > > On Thursday 29 Nov 2001 3:51 pm, Miguel Freitas wrote: > > > i forgot to cc xine-dev, this might be useful for someone else... > > > > Indeed it was. Just a small quation; below is the trace from my latest > > attempts to get highlighting to work. These events are sent > just at a scene > > change (jump). As you can see, since there is no frame being > shown between > > the events, they are both queued and then executed together. > The problem is > > that the are executed in the reverse order to which they were > called (i.e. > > the hide event is processed before the last show event). > > > > Is this a bug in my code or the new OSD stuff? > > > > Rich > > > > dvdnav_plugin_read_block: Highlight area is > (426,145)-(526,350), display = 1 > > MALLOC1: overlay_event (nil), len=4 > > MALLOC2: overlay_event 0x82dce08, len=4 > > MALLOC1: overlay (nil), len=128 > > MALLOC2: overlay 0x82dce30, len=128 > > BUTTON > > show=1 > > clut [0 0 0 0] > > trans [48 100 9 64] > > left = 426 right = 526 top = 145 bottom = 350 > > dvdnav_plugin_read_block: FIXME: VTS change > > dvdnav_plugin_read_block: Highlight area is > (37156,16505)-(63488,16548), > > display > > = 0 > > MALLOC1: overlay_event (nil), len=4 > > MALLOC2: overlay_event 0x82dcf40, len=4 > > MALLOC1: overlay (nil), len=128 > > MALLOC2: overlay 0x82dcf68, len=128 > > BUTTON > > show=0 > > clut [0 0 0 0] > > trans [48 100 9 64] > > left = 0 right = 0 top = 0 bottom = 0 > > dvdnav_plugin_read_block: FIXME: Cell change > > video_overlay: HIDE MENU NOW 0 > > MENU BUTTON NOW > > video_overlay: error: button event received and no overlay allocated. > > Segmentation fault > > > > -- > > Rich > > > > You need no longer worry about the future. This time tomorrow > you'll be dead. > > > > _______________________________________________ > xine-devel mailing list > xin...@li... > https://lists.sourceforge.net/lists/listinfo/xine-devel |
From: Miguel F. <mi...@ce...> - 2001-11-29 20:52:16
|
I introduced a bug reworking the events function... it should be fixed now! :) Miguel On Thu, 2001-11-29 at 18:49, James Courtier-Dutton wrote: > Well, if vpts=0; the event will be put at the top of the queue and executed > immediately. > If two events get sent close together with vpts=0, it should(I had intended > it to) do a fifo type function, but there might be a bug. > I will play with the latest NAV and SPU code this weekend. > > Cheers > James > P.S. Richard, does the latest libdvdnav send NAV packs to the demuxers yet ? > > > > > -----Original Message----- > > From: xin...@li... > > [mailto:xin...@li...]On Behalf Of Miguel > > Freitas > > Sent: 29 November 2001 16:24 > > To: Rich Wareham > > Cc: xine-dev > > Subject: Re: [Fwd: Re: [xine-devel] reworked spu and osd] > > > > > > Rich, > > > > i don't know what your debug messages mean, but it seems that you send 2 > > button events without any menu show event? > > > > if sent the show event, have you checked their pts values? i don't know > > what happens if both have pts = 0, i will check what order they are > > processed. there might be a bug in this case... > > > > Miguel > > > > On Thu, 2001-11-29 at 14:14, Rich Wareham wrote: > > > On Thursday 29 Nov 2001 3:51 pm, Miguel Freitas wrote: > > > > i forgot to cc xine-dev, this might be useful for someone else... > > > > > > Indeed it was. Just a small quation; below is the trace from my latest > > > attempts to get highlighting to work. These events are sent > > just at a scene > > > change (jump). As you can see, since there is no frame being > > shown between > > > the events, they are both queued and then executed together. > > The problem is > > > that the are executed in the reverse order to which they were > > called (i.e. > > > the hide event is processed before the last show event). > > > > > > Is this a bug in my code or the new OSD stuff? > > > > > > Rich > > > > > > dvdnav_plugin_read_block: Highlight area is > > (426,145)-(526,350), display = 1 > > > MALLOC1: overlay_event (nil), len=4 > > > MALLOC2: overlay_event 0x82dce08, len=4 > > > MALLOC1: overlay (nil), len=128 > > > MALLOC2: overlay 0x82dce30, len=128 > > > BUTTON > > > show=1 > > > clut [0 0 0 0] > > > trans [48 100 9 64] > > > left = 426 right = 526 top = 145 bottom = 350 > > > dvdnav_plugin_read_block: FIXME: VTS change > > > dvdnav_plugin_read_block: Highlight area is > > (37156,16505)-(63488,16548), > > > display > > > = 0 > > > MALLOC1: overlay_event (nil), len=4 > > > MALLOC2: overlay_event 0x82dcf40, len=4 > > > MALLOC1: overlay (nil), len=128 > > > MALLOC2: overlay 0x82dcf68, len=128 > > > BUTTON > > > show=0 > > > clut [0 0 0 0] > > > trans [48 100 9 64] > > > left = 0 right = 0 top = 0 bottom = 0 > > > dvdnav_plugin_read_block: FIXME: Cell change > > > video_overlay: HIDE MENU NOW 0 > > > MENU BUTTON NOW > > > video_overlay: error: button event received and no overlay allocated. > > > Segmentation fault > > > > > > -- > > > Rich > > > > > > You need no longer worry about the future. This time tomorrow > > you'll be dead. > > > > > > > > _______________________________________________ > > xine-devel mailing list > > xin...@li... > > https://lists.sourceforge.net/lists/listinfo/xine-devel |
From: Miguel F. <mi...@ce...> - 2001-11-29 16:34:02
|
On Thu, 2001-11-29 at 14:29, Rich Wareham wrote: > Both have a pts of 0. What is hapening is that there is one event sent to > change the CLUT from highlighted to selected (not yet implemented but two > events are sent) and the final one to clear the highlight after the jump. > > It appears in some (not all) cases, these events happen to be queued so that > they both exist in the queue at the same time. In this case, it appears that > you itterate through the queue backwards (maybe). Hum...try again now...a cvs update might be needed... :) btw: i'm on #xine now Miguel |