[Ffmpeg-devel-irc] ffmpeg-devel.log.20111231

Sun Jan 1 02:05:03 CET 2012

[00:18] <funman> Daemon404: when will we see a ffmpeg port for mmix ?
[00:18] <Daemon404> people use mmix?
[00:19] <Daemon404> actual hardware exists?
[00:23] <funman> don't think so
[01:28] <CIA-101> ffmpeg: 03Mike Melanson 07master * rae51b1a9ba 10ffmpeg/tests/ (fate/h264.mak ref/fate/h264-bsf-mp4toannexb): 
[01:28] <CIA-101> ffmpeg: FATE: Add test for H.264 MP4->annex.B bitstream filter.
[01:28] <CIA-101> ffmpeg: This test uses an existing FATE sample (h264/interlaced_crop.mp4).
[01:28] <CIA-101> ffmpeg: Signed-off-by: Ronald S. Bultje <rsbultje at gmail.com>
[01:28] <CIA-101> ffmpeg: 03Diego Biurrun 07master * rf486fb338e 10ffmpeg/tests/ (4 files in 3 dirs): fate: drop pointless _audio and _video suffixes from xan tests
[01:28] <CIA-101> ffmpeg: 03Diego Biurrun 07master * rd4b63054d9 10ffmpeg/ (11 files in 3 dirs): cosmetics: Drop unnecessary parentheses around return values.
[01:28] <CIA-101> ffmpeg: 03Diego Biurrun 07master * rc7cb1730da 10ffmpeg/tests/ (Makefile fate.mak fate/demux.mak): fate: split off demuxer FATE tests into their own file
[01:28] <CIA-101> ffmpeg: 03Diego Biurrun 07master * r51150cf070 10ffmpeg/tests/ (Makefile fate/voice.mak fate2.mak): fate: split off voice codec FATE tests into their own file
[01:28] <CIA-101> ffmpeg: 03Diego Biurrun 07master * rab2d11f75b 10ffmpeg/tests/ (Makefile fate/audio.mak fate2.mak): fate: split off audio codec FATE tests into their own file
[01:28] <CIA-101> ffmpeg: 03Diego Biurrun 07master * r394f0b18bf 10ffmpeg/tests/ (Makefile fate.mak fate/ea.mak): fate: split off Electronic Arts codec FATE tests into their own file
[01:28] <CIA-101> ffmpeg: 03Diego Biurrun 07master * r9afc025bff 10ffmpeg/tests/fate/ (12 files): fate: whitespace cosmetics
[01:28] <CIA-101> ffmpeg: 03Diego Biurrun 07master * r628637f67b 10ffmpeg/tests/ (Makefile fate.mak fate/video.mak fate2.mak): fate: split off video codec FATE tests into their own file
[01:28] <CIA-101> ffmpeg: 03Michael Niedermayer 07master * r1a6a088f7c 10ffmpeg/: (log message trimmed)
[01:28] <CIA-101> ffmpeg: Merge remote-tracking branch 'qatar/master'
[01:28] <CIA-101> ffmpeg: * qatar/master:
[01:28] <CIA-101> ffmpeg:  fate: whitespace cosmetics
[01:28] <CIA-101> ffmpeg:  fate: split off video codec FATE tests into their own file
[01:28] <CIA-101> ffmpeg:  fate: split off audio codec FATE tests into their own file
[01:28] <CIA-101> ffmpeg:  fate: split off Electronic Arts codec FATE tests into their own file
[01:28] <CIA-101> ffmpeg: 03Diego Biurrun 07master * rf6ba1f2641 10ffmpeg/tests/ (Makefile fate.mak fate/qt.mak fate2.mak): fate: split off QuickTime codec FATE tests into their own file
[01:28] <CIA-101> ffmpeg: 03Diego Biurrun 07master * rcd8d84575b 10ffmpeg/tools/qt-faststart.c: qt-faststart: K&R reformatting; fix comment typos
[01:55] <Matty> Is this right place to discuss, how ffmpeg puts ASS subtitles in Matroska files?
[01:58] <ffmpeguser> Matty: this is the development channel, for ffmpeg usage questions go to this channel #ffmpeg
[01:59] <ffmpeguser> developer*
[02:01] <Matty> Ok, I think ffmpeg produces broken Matroska files containing subtitles that VLC of mplayer do not display.
[02:02] <ffmpeguser> http://ffmpeg.org/trac/ffmpeg/ticket/858, my ticket was closed, what is the problem with the apache license?
[02:02] <ffmpeguser> matty /j #ffmpeg
[02:04] <nevcairiel> ffmpeguser: the apache license is not compatible with (L)GPLv2
[02:04] <CIA-101> ffmpeg: 03Carl Eugen Hoyos 07master * rb46fb615c4 10ffmpeg/libavcodec/tiff.c: tiff.c: Use switch / case instead of if / else where appropriate.
[02:04] <CIA-101> ffmpeg: 03Carl Eugen Hoyos 07master * re767968c6e 10ffmpeg/libavcodec/tiff.c: Simplify tiff decoder by merging two functions.
[02:04] <CIA-101> ffmpeg: 03Carl Eugen Hoyos 07master * r7ee0649f95 10ffmpeg/libavcodec/tiff.c: 
[02:04] <CIA-101> ffmpeg: Fix decoding some 1bpp tiff files, they are palettized.
[02:04] <CIA-101> ffmpeg: Fixes ticket #830.
[02:04] <pasteeater> we support libvo_aacenc and it's APACHE, IIRC.
[02:05] <pasteeater> but nothing is probably stopping anyone implementing apple's ALAC other than interest, time, or manpower
[02:05] <ffmpeguser> nevcairiel: that means that the license is not a problem
[02:06] <nevcairiel> the license is always a problem :)
[02:08] <ffmpeguser> what is incompatible between the 2 licenses?
[02:08] <nevcairiel> ask the FSF, not me
[02:09] <nevcairiel> anyhow, replacing code is never a good option, but if someone wants to submit a wrapper that just uses an external copy of the apple alac library, i'm sure there are no objections
[02:09] <nevcairiel> similar to how the libvo wrapper works
[02:10] <ffmpeguser> in this case it would do good to replace it
[02:10] <iive> isn't alac an apache license ?
[02:10] <ffmpeguser> yes
[02:10] <iive> i think it is gplv3 compatible.
[02:10] <nevcairiel> and APL2.0 is not compatible with (L)GPLv2
[02:11] <nevcairiel> with 3 it is, but that would be a step backwards from a licensing standpoint
[02:11] <ffmpeguser> how do you know?
[02:12] <iive> there is gplv3 code in ffmpeg atm. this includes configure option to enable it.
[02:12] <nevcairiel> indeed, such options would be used to also activate the wrapper around the apple code
[02:12] <nevcairiel> however there still is no reason to replace the perfectly fine code that is already in ffmpeg
[02:13] <iive> also, as I land in the middle of the conversation, I assume you are talking about encoder
[02:14] <pasteeater> ffmpeguser: would you like me to modify the ticket title as a request to add wrapper support?
[02:14] <ffmpeguser> i mainly want the decoder, i have no use for the encoder, but i'm sure that others will use it in the future if you add encoding support
[02:14] <nevcairiel> maybe you should then name one thing that doesn't work in the decoder right now, and then that thing can be fixed
[02:15] <nevcairiel> like you were asked to in the ticket. :)
[02:15] <ffmpeguser> that would involve downloading more things instead of one, ffmpeg only
[02:17] <iive> i thought ffmpeg had the decoder for ages...
[02:17] <nevcairiel> it does
[02:17] <ffmpeguser> yes, i would like that, i'll upload also a file that you can use for testing but on mediafire, because it is over the file size limit but below 10 MB
[02:18] <nevcairiel> not the "official" code mind you, its own implementation, but apparently it works just fine
[02:18] <ffmpeguser> it doesn't decode 6 channel alac streams
[02:19] <ffmpeguser> but it does play it back the stream with -ac 1
[02:19] <ffmpeguser> -it
[02:19] <ffmpeguser> yes
[02:20] <ffmpeguser> in other words the decoding support is incomplete
[02:21] <ffmpeguser> [alac @ 01090AE0] Unsupported channel count: 6
[02:22] <ffmpeguser> [caf @ 00A0B880] skipping CAF chunk: 75756964 (uuid), size 154 [caf @ 00A0B880] skipping CAF chunk: 75756964 (uuid), size 36 [caf @ 00A0B880] skipping CAF chunk: 6F767677 (ovvw), size 13176
[02:22] <ffmpeguser> these also
[02:22] <nevcairiel> thats a demuxer, not related to the audio decoder
[02:23] <nevcairiel> anyway, open issues for the missing features of the existing decoder, and that has a far higher chance of getting fixed
[02:24] <ffmpeguser> the skippings could mean that there's a problem, that's why i mentioned those too
[02:25] <ffmpeguser> alright, i'll open a new ticket
[02:28] <CIA-101> ffmpeg: 03Jean First 07master * r9bfaeee5f3 10ffmpeg/libavcodec/8svx.c: 
[02:28] <CIA-101> ffmpeg: 8svx: remove unused variable in eightsvx_decode_frame
[02:28] <CIA-101> ffmpeg: Signed-off-by: Jean First <jeanfirst at gmail.com>
[02:28] <CIA-101> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[02:28] <CIA-101> ffmpeg: 03Jean First 07master * r3ceb4c907e 10ffmpeg/libavformat/cache.c: 
[02:28] <CIA-101> ffmpeg: cache: remove unused variable in cache_open
[02:28] <CIA-101> ffmpeg: Signed-off-by: Jean First <jeanfirst at gmail.com>
[02:28] <CIA-101> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[02:28] <CIA-101> ffmpeg: 03Jean First 07master * r576ada791a 10ffmpeg/libavformat/cache.c: 
[02:28] <CIA-101> ffmpeg: cache: use av_freep instead of av_free in cache_open
[02:28] <CIA-101> ffmpeg: Signed-off-by: Jean First <jeanfirst at gmail.com>
[02:28] <CIA-101> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[02:28] <CIA-101> ffmpeg: 03Jean First 07master * rb6ffceefb5 10ffmpeg/libavformat/mpegts.c: 
[02:28] <CIA-101> ffmpeg: mpegts: remove unused variable
[02:28] <CIA-101> ffmpeg: Signed-off-by: Jean First <jeanfirst at gmail.com>
[02:28] <CIA-101> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[02:28] <CIA-101> ffmpeg: 03Jean First 07master * r8aed73c355 10ffmpeg/libavformat/movenc.c: 
[02:30] <ffmpeguser> done http://ffmpeg.org/trac/ffmpeg/ticket/859
[02:40] <pasteeater> ffmpeguser: please show an example command of ffmpeg attempting to decode 6 channel alac and include the complete console output
[02:40] <ffmpeguser> ok
[02:53] <ffmpeguser> pasteeater: done
[02:55] <pasteeater> ffmpeguser: thanks. if you edit your post and wrap the command and the output with the code block it will be easier to read.
[02:58] <ffmpeguser> i don't understand
[03:00] <ffmpeguser> text wrapping?
[03:00] <pasteeater> http://ffmpeg.org/trac/ffmpeg/wiki/WikiFormatting#PreformattedText
[03:01] <pasteeater> or the 5th botton from the left above the comment box
[03:01] <pasteeater> button
[03:01] <ffmpeguser> i'll check right now
[03:09] <ffmpeguser> pasteeater: done
[05:14] <buzz_> hmm 27d323577c19af218f8a5ac33364f213b623a023 is causing a problem in xbmc4xbox
[05:15] <buzz_> causes lots of errors like "Truncating packet of size xxxx to 1" . without the patch playback is fine. dont know how to produce this on ffplay, since in this case it is a dvd being played via libdvdnav
[05:20] <CIA-101> ffmpeg: 03Michael Niedermayer 07master * r5500e65342 10ffmpeg/libavformat/flvdec.c: 
[05:20] <CIA-101> ffmpeg: flvdec: Check avpriv_mpeg4audio_get_config() return
[05:20] <CIA-101> ffmpeg: value before using its output.
[05:20] <CIA-101> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[05:20] <CIA-101> ffmpeg: 03Michael Niedermayer 07master * r4392b3d11e 10ffmpeg/libavcodec/tiff.c: 
[05:20] <CIA-101> ffmpeg: tiffdec: hotfix for fate failure.
[05:20] <CIA-101> ffmpeg: Only use pal8 as output if the file has a palette or more than 2 colors.
[05:20] <CIA-101> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[05:56] <ffmpeguser> buzz_: the vob files in the DVD
[05:56] <buzz_> unable to reproduce that way.
[05:56] <ffmpeguser> hm
[05:57] <ffmpeguser> you use windows?
[05:57] <buzz_> linux and windows here
[05:59] <buzz_> the error is av_log(0, AV_LOG_ERROR, "Truncating packet of size %d to %"PRId64"\n", size, remaining+1); - and size is say 2345 and remaining is equal to 0
[06:01] <ffmpeguser> you authored the dvd?
[06:01] <buzz_> no
[06:01] <ffmpeguser> who?
[06:01] <buzz_> everything is fine if i revert that ffmpeg commit, but I dont know the ffmpeg code to know what could be going wrong
[06:01] <buzz_> warner bros
[06:02] <ffmpeguser> i'll see the commit
[06:03] <buzz_> basically remaining is always 0 so if i add the logic "remaining>=0" to the if, it should work again. perhaps its due the way xbmc4xbox code uses ffmpeg specifically - perhaps ffmpeg doesnt know how much is remaining in this case ?
[06:04] <buzz_> oh wait. brain glitch. >0 would still match hmm
[06:05] <buzz_> i dont know what the variables are anyway. i just bisected it to find the commit that broke playback
[06:05] <ffmpeguser> it doesn't play back?
[06:06] <buzz_> no.
[06:08] <ffmpeguser> open a ticket in the bug tracker http://ffmpeg.org/trac/ffmpeg/
[06:08] <buzz_> i didnt want to open a ticket until i understand the problem. im assuming remaining must be negative in the case where this works before the commit
[06:09] <ffmpeguser> are you sure that the vob files don't play back?
[06:10] <ffmpeguser> ffplay vobfile
[06:12] <buzz_> they do play back
[06:12] <buzz_> they play fine.
[06:13] <ffmpeguser> you're on windows right now?
[06:13] <buzz_> no
[06:14] <buzz_> well. i have both running. why ?
[06:14] <buzz_> are you an ffmpeg developer, or understand the ffmpeg code etc ?
[06:15] <ffmpeguser> the error you saw, you saw that on your tv if i'm not mistaken, right?
[06:15] <buzz_> i saw it in a log file.
[06:15] <buzz_> why do you ask ?
[06:17] <ffmpeguser> get the latest build for windows from http://ffmpeg.zeranoe.com/builds/
[06:19] <buzz_> im using the ffmpeg dlls in an application. i dont see how getting the latest build for windows will help ? unless im missing something (in which case please explain). the issue is with the latest build. reverting the mentioned commit solves it. im just trying to work out why i need to revert it now (adding some debugging in)
[06:21] <ffmpeguser> xbmc4xbox is for the xbox, that means that the ffmpeg code runs on the xbox
[06:21] <buzz_> yes
[06:22] <buzz_> well seems that remaining is negative - 07:11:13 M: 27897856   ERROR: ffmpeg[184]: size = 2016, remaining = -137248 so ffmpeg doesnt know how much data is left, perhaps due to the way we are decoding. maybe i need an additional ffmpeg flag or something. need to trace this back
[06:23] <ffmpeguser> we'll see if the same thing happens with a dec 26 build which is the latest from zeranoe
[06:23] <ffmpeguser> that error has nothing to do with settings
[06:23] <buzz_> it doesnt happy with commandline. its related to how we are using ffmpeg. 
[06:24] <buzz_> doesnt matter. i dont this you can help.
[06:24] <buzz_> s/happy/happen
[06:25] <ffmpeguser> if they have modified the ffmpeg code extensively then you should open a ticket in their bug tracker in they have one
[06:25] <buzz_> well "remaining" variable starts at 0 and goes progressively negative with each call to ffio_limit
[06:25] <buzz_> ffmpeguser, that would be opening a bug for myself then
[06:26] <buzz_> i am developing xbmc4xbox
[06:27] <buzz_> we havnt modified the ffmpeg code extensively.. slightly. but we are obviously doing something in a way this change doesnt like. ill continue debugging. thanks anyway
[06:27] <ffmpeguser> maybe it is something obvious
[06:40] <buzz_> hmm s->maxsize is always 0 in this case, so then it does int64_t remaining= s->maxsize - avio_tell(s); which sets remaining to a negative number (maxsize = 0, avio_tell = 1769495 for example)
[06:56] <CIA-101> ffmpeg: 03Michael Niedermayer 07master * r91ab557b68 10ffmpeg/libavcodec/aacdec.c: 
[06:56] <CIA-101> ffmpeg: Revert "aacdec: Avoid unneeded reinit due to ADTS headers"
[06:56] <CIA-101> ffmpeg: This has been fixed differently and its revert avoids error messages
[06:56] <CIA-101> ffmpeg: when decoding xx.flv. This also reduces the difference to qatar.
[06:56] <CIA-101> ffmpeg: This reverts commit 5a2b3f3a52070232a68582a502fb60a9226ca86a.
[06:56] <CIA-101> ffmpeg: Reviewed-by: Derek Buitenhuis <derek.buitenhuis at gmail.com>
[06:56] <CIA-101> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[07:21] <buzz_> michaelni / any other devs, would appreciate any feedback regarding http://www.ffmpeg.org/trac/ffmpeg/ticket/861 (middle of night in eurozone i realise now). 
[07:26] <michaelni> buzz_, ill add aworkaround tomorrow but why does your avio_size() return 0 ?
[07:26] <michaelni> that would be filesize = 0
[07:27] <buzz_> i assume the way the data is passed to ffmpeg. via libdvdnav. im not all that familiar with the code in this part of xbmc4xbox
[07:30] <michaelni> workaround that ill probably add tomorrow: s->maxsize= newsize - !newsize;
[07:31] <michaelni> that is if it still makes sense after sleeping
[07:32] <buzz_> wish i understood some of this code better - this is another patch i had to revert for the dvd playback to work - http://sourceforge.net/apps/trac/xbmc4xbox/browser/trunk/xbmc/cores/dvdplayer/Codecs/ffmpeg/patches/ours/0001-Revert-Fix-url_feof-for-file-appendings.patch again i assume its related to the way the data is sent to ffmpeg, so eof doesnt work right ?
[07:33] <buzz_> xbmc are on an older ffmpeg, so they will bump into these issues when they upgrade, and no doubt understand it better since they wrote this code. im just maintaining the old xbox version of the player
[07:33] <buzz_> but we have a newer ffmpeg.
[07:33] <buzz_> anyway. have a nice sleep :)
[07:48] <buzz_> yeh i guess we wont know the file size, since we are processing dvd video blocks. the seek call doesnt do anything but say seeking is possible, and seeking is handled via dvdnav
[07:57] Action: buzz_ gets confused/lost by the code
[11:46] <CIA-101> ffmpeg: 03Stefano Sabatini 07master * r549b53e8cb 10ffmpeg/libavfilter/avfilter.h: 
[11:46] <CIA-101> ffmpeg: lavfi: remove redundant definition of avfilter_copy_frame_props() in avfilter.h
[11:46] <CIA-101> ffmpeg: Rationale: avfilter_copy_frame_props() was already defined in
[11:46] <CIA-101> ffmpeg: libavfilter/avcodec.h, and keeping the lavc/lavfi API glue localized in a
[11:46] <CIA-101> ffmpeg: specific file should ease maintainance and help the ones which use
[11:46] <CIA-101> ffmpeg: libavfilter without depending on libavcodec.
[11:46] <CIA-101> ffmpeg: 03Stefano Sabatini 07master * r8fb03b4d70 10ffmpeg/ (7 files in 3 dirs): (log message trimmed)
[11:46] <CIA-101> ffmpeg: lavfi: port tinterlace filter from MPlayer
[11:46] <CIA-101> ffmpeg: Port MPlayer tinterlace filter from MPlayer, with some ideas taken
[11:46] <CIA-101> ffmpeg: from the FFmbc/libavfilter port, with the following main differences:
[11:46] <CIA-101> ffmpeg: * added support for full-scale YUVJ formats
[11:46] <CIA-101> ffmpeg: * added support for YUVA420P
[11:46] <CIA-101> ffmpeg: * request_frame() on the filter is forced to return a frame
[11:53] <CIA-101> ffmpeg: 03Jean First 07master * r81980bf740 10ffmpeg/libavdevice/lavfi.c: 
[11:53] <CIA-101> ffmpeg: lavd/lavfi: fix compiler warning for uninitialized variables
[11:53] <CIA-101> ffmpeg: Signed-off-by: Jean First <jeanfirst at gmail.com>
[11:53] <CIA-101> ffmpeg: Signed-off-by: Stefano Sabatini <stefasab at gmail.com>
[15:58] <CIA-101> ffmpeg: 03Michael Niedermayer 07master * re6362f3a3c 10ffmpeg/libavformat/utils.c: 
[15:58] <CIA-101> ffmpeg: lavf: ffio_limit(), treat avio_size()==0 as error.
[15:58] <CIA-101> ffmpeg: This works around issues arising from inputs that claim to have a
[15:58] <CIA-101> ffmpeg: filesize of 0.
[15:58] <CIA-101> ffmpeg: Reported-by: buzz_
[15:58] <CIA-101> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[16:26] <CIA-101> ffmpeg: 03Nicolas George 07master * re90a69e955 10ffmpeg/ (5 files in 3 dirs): lavfi: add astreamsync audio filter.
[16:27] <CIA-101> ffmpeg: 03Nicolas George 07master * r4962edf889 10ffmpeg/ (6 files in 3 dirs): lavfi: add amerge audio filter.
[16:31] <CIA-101> ffmpeg: 03Reimar Döffinger 07master * r42a1f1d7a8 10ffmpeg/ (libavcodec/lcldec.c libavutil/lzo.h): 
[16:31] <CIA-101> ffmpeg: Avoid uninitialized data in lcldec when ofs is 0 in MSZH.
[16:31] <CIA-101> ffmpeg: Signed-off-by: Reimar Döffinger <Reimar.Doeffinger at gmx.de>
[16:34] <buzz_> michaelni, i assume the problem in this case, is the size of the input is not known ? 
[16:35] <michaelni> avIo_size should return an error and not 0 in that case
[16:36] <buzz_> hmm ok. ill take a look at our code, and see if elupus can help (he wrote it)
[16:40] <michaelni>  maybe an issue with AVSEEK_SIZE or SEEK_END handling
[16:41] <buzz_> i emailed Joakim Plate (elupus) in the hope he can shed some light/explain the issue
[17:11] <CIA-101> ffmpeg: 03Nicolas George 07master * r0098e79f8a 10ffmpeg/libavcodec/libvorbis.c: libvorbis: proper error logging and return.
[17:15] <CIA-101> ffmpeg: 03Carl Eugen Hoyos 07master * rf69a766d11 10ffmpeg/libavcodec/pamenc.c: Fix mono* pam encoding.
[17:17] <CIA-101> ffmpeg: 03Michael Niedermayer 07master * r810a14dbcc 10ffmpeg/libavcodec/aacdec.c: 
[17:17] <CIA-101> ffmpeg: aacdec: Evaluate further PCE elements.
[17:17] <CIA-101> ffmpeg: Fixes Music\ Station\ Super\ Live\ -\ 2011.12.23.mkv
[17:17] <CIA-101> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[17:17] <CIA-101> ffmpeg: 03Michael Niedermayer 07master * rbd553d5ea9 10ffmpeg/libavformat/mxg.c: 
[17:17] <CIA-101> ffmpeg: mxg: fix compiler warning for uninitialized variables
[17:17] <CIA-101> ffmpeg: Based on work by: Jean First <jeanfirst at gmail.com>
[17:17] <CIA-101> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[17:17] <CIA-101> ffmpeg: 03Jean First 07master * racc88f07d4 10ffmpeg/libavformat/id3v2.c: 
[17:17] <CIA-101> ffmpeg: id3v2: fix compiler warning for uninitialized variables
[17:17] <CIA-101> ffmpeg: Signed-off-by: Jean First <jeanfirst at gmail.com>
[17:17] <CIA-101> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[18:35] <durandal_1707> does anyone can actually get mplayer to display bt20 videos?
[19:24] <durandal_1707> is it fine to do flipping at decoder level?
[20:40] <CIA-101> ffmpeg: 03Michael Niedermayer 07master * r9d9316ae34 10ffmpeg/libavcodec/h264_refs.c: 
[20:40] <CIA-101> ffmpeg: h264: Fix seeking in 00017.MTS
[20:40] <CIA-101> ffmpeg: Fixes Ticket852
[20:40] <CIA-101> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[20:47] <CIA-101> ffmpeg: 03Carl Eugen Hoyos 07master * r714d2f9796 10ffmpeg/libavcodec/tiff.c: Only expand raw and packed 1bpp tiff horizontally if it is palettised.
[20:53] <buzz_> happy new year michaelni and all other ffmpeggers. have a good one!
[22:37] <CIA-101> ffmpeg: 03Carl Eugen Hoyos 07master * r2cba62bd84 10ffmpeg/libavformat/asf.c: 
[22:37] <CIA-101> ffmpeg: Do not map asf metadata key WM/Track to track.
[22:37] <CIA-101> ffmpeg: WM/Track is 0-based (and deprecated in favour of WM/TrackNumber),
[22:37] <CIA-101> ffmpeg: track is 1-based.
[22:37] <CIA-101> ffmpeg: Fixes ticket #825.
[23:40] <Granjow> Why do I get a «Could not open file» with av_write_frame() here on line 113? http://codepad.org/GL3N1Lhr
[23:41] <funman> Granjow: you should ask on #ffmpeg (see topic)
[00:00] --- Sun Jan  1 2012