[Ffmpeg-devel-irc] ffmpeg-devel.log.20120621

[burek]

[00:06] <CIA-119> ffmpeg: 03Michael Niedermayer 07master * r104c914640 10ffmpeg/libavutil/ (time.c time.h): (log message trimmed)
[00:06] <CIA-119> ffmpeg: libavutil: fix author attribution
[00:06] <CIA-119> ffmpeg: Libav.org has removed this while moving the code around
[00:06] <CIA-119> ffmpeg: For reference, the code was originally added in:
[00:06] <CIA-119> ffmpeg: commit 9aeeeb63f7e1ab7b0b7bb839a5f258667a2d2d78
[00:06] <CIA-119> ffmpeg: Author: Fabrice Bellard <fabrice at bellard.org>
[00:06] <CIA-119> ffmpeg: Date: Wed Dec 20 00:02:47 2000 +0000
[00:06] <CIA-119> ffmpeg: 03Michael Niedermayer 07master * r5a90e55ece 10ffmpeg/libavfilter/transform.c: 
[00:06] <CIA-119> ffmpeg: libavfilter/transform: fix mirroring.
[00:06] <CIA-119> ffmpeg: mirror() borrowed from dwt.c, this could be moved to libavutil and made
[00:06] <CIA-119> ffmpeg: public API, Ill submit a patch for this
[00:06] <CIA-119> ffmpeg: Fixes Ticket1278
[00:06] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[04:55] <CIA-119> ffmpeg: 03Paul B Mahol 07master * rb18357326c 10ffmpeg/ (6 files in 2 dirs): 
[04:55] <CIA-119> ffmpeg: LucasArts SMUSH VIMA audio decoder
[04:55] <CIA-119> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[04:55] <CIA-119> ffmpeg: 03Paul B Mahol 07master * r69254f4628 10ffmpeg/ (7 files in 2 dirs): 
[04:55] <CIA-119> ffmpeg: LucasArts SMUSH video decoder
[04:55] <CIA-119> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[04:55] <CIA-119> ffmpeg: 03Paul B Mahol 07master * rbef8fd7099 10ffmpeg/ (6 files in 3 dirs): 
[04:55] <CIA-119> ffmpeg: LucasArts SMUSH demuxer
[04:55] <CIA-119> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[08:35] <Tjoppen> in mpeg12enc.c, what's up with "time_code = s->current_picture_ptr->f.coded_picture_number + s->avctx->timecode_frame_start;" ? timecode_frame_start is in that 25-bit format, so you can't just add the current frame counter to it
[11:09] <nevcairiel> Daemon404: happy now? :d
[11:13] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * r5bc87f1cd6 10ffmpeg/libavfilter/vf_blackdetect.c: lavfi/blackdetect: fix pts time debug printing
[11:13] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * reda4500866 10ffmpeg/libavfilter/vf_blackdetect.c: (log message trimmed)
[11:13] <CIA-119> ffmpeg: lavfi/blackdetect: add logic for handling stream termination
[11:13] <CIA-119> ffmpeg: Add request_frame() which checks the black duration when the end of the
[11:13] <CIA-119> ffmpeg: stream is signalled. Allow to detect black video segments at the end of
[11:13] <CIA-119> ffmpeg: the stream.
[11:13] <CIA-119> ffmpeg: Note that the reported last black video segment duration is not very
[11:13] <CIA-119> ffmpeg: accurate, since we still miss the duration information in the video
[11:13] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * r98701be356 10ffmpeg/libavfilter/vf_ass.c: lavfi/ass: update use of deprecated functions, fix warnings
[11:15] <ayad02> hello everybody
[11:16] <ayad02> can you help me to get on the sony camera the video mpeg4 and to convert to .mp4 
[11:16] <ayad02> here's my comman but i don't have a good resultat 
[11:16] <ayad02> ffmpeg -i http://@IP/image -b 1024 -s 640x480 - t 30 -vcodec mpeg4 image.mp4
[11:16] <ayad02> but the encoding take many time and no finish 
[12:37] <CIA-119> ffmpeg: 03Hendrik Leppkes 07master * r07d49437d5 10ffmpeg/libavcodec/wavpack.c: 
[12:37] <CIA-119> ffmpeg: wavpack: set bits_per_raw_sample for S32 samples to properly identify 24-bit
[12:37] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[12:38] <CIA-119> ffmpeg: 03Hendrik Leppkes 07master * rf6943f2379 10ffmpeg/libavcodec/lagarith.c: 
[12:38] <CIA-119> ffmpeg: lagarith: frame multithreading
[12:38] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[12:38] <CIA-119> ffmpeg: 03Hendrik Leppkes 07master * r2d61f0040d 10ffmpeg/libavcodec/utvideo.c: 
[12:38] <CIA-119> ffmpeg: utvideo: mark interlaced frames as such
[12:38] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[12:38] <CIA-119> ffmpeg: 03Hendrik Leppkes 07master * r6d887ecac2 10ffmpeg/libavformat/ (avformat.h utils.c): 
[12:38] <CIA-119> ffmpeg: avformat: add 48 fps to the NTSC framerates
[12:38] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[12:49] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * r065851ffac 10ffmpeg/libswresample/rematrix.c: 
[12:49] <CIA-119> ffmpeg: lswr/rematrix: specify explicitly the name of the unsupported channel layouts
[12:49] <CIA-119> ffmpeg: Help debugging.
[12:49] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * rc0b3c0ffb7 10ffmpeg/libavfilter/audio.c: 
[12:49] <CIA-119> ffmpeg: lavfi/audio: remove unused variable in ff_filter_samples()
[12:49] <CIA-119> ffmpeg: Fix warning:
[12:49] <CIA-119> ffmpeg: libavfilter/audio.c: In function ff_filter_samples:
[12:49] <CIA-119> ffmpeg: libavfilter/audio.c:193:13: warning: unused variable size [-Wunused-variable]
[12:49] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * rcb08b6efc9 10ffmpeg/libavfilter/vf_scale.c: 
[12:49] <CIA-119> ffmpeg: lavfi/scale: remove unused variable out_link
[12:49] <CIA-119> ffmpeg: Fix warning:
[12:49] <CIA-119> ffmpeg: libavfilter/vf_scale.c: In function start_frame:
[12:49] <CIA-119> ffmpeg: libavfilter/vf_scale.c:284:23: warning: unused variable out_link [-Wunused-variable]
[13:35] <CIA-119> ffmpeg: 03Hendrik Leppkes 07master * r24e7a22e1f 10ffmpeg/libavformat/riff.c: 
[13:35] <CIA-119> ffmpeg: riff: don't overwrite bps from WAVEFORMATEX if EXTENSIBLE doesn't contain that data.
[13:35] <CIA-119> ffmpeg: According to the specification on the MSDN [1], 0 is valid for that particular field, and it should be ignored in that case.
[13:35] <CIA-119> ffmpeg: [1]: http://msdn.microsoft.com/en-us/library/windows/desktop/dd757714(v=vs.85).aspx
[13:35] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[13:35] <CIA-119> ffmpeg: 03Hendrik Leppkes 07master * rf77fd34bc3 10ffmpeg/libavcodec/aacdec.c: 
[13:35] <CIA-119> ffmpeg: aac: move sample_fmt selection before decoder configuration.
[13:35] <CIA-119> ffmpeg: The decoder configuration initializes the sbr mdct, which needs to know about the sample format to properly take the scale factor into account.
[13:35] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[14:21] <saste> ubitux: are you by chance working on -af support in ffplay?
[14:21] <saste> just in case so we don't duplicate effort
[15:13] <ubitux> saste: no i'm not
[15:15] <saste> too bad
[15:16] <ubitux> :)
[15:34] <brocatz> the ffmpeg documention changes need to be modified to set the selection color to something strong
[15:34] <brocatz> so when you're using a find-string that honors the css it's actually visible
[15:35] <brocatz> instead of the current palid blue bullshit
[15:35] <brocatz> (red?)
[15:35] <brocatz> (spelling? pallid derp)
[16:44] <Daemon404> nevcairiel, yeaaaaaaap
[17:02] <Daemon404> nevcairiel, btw, lagarith technicallt isnt intar-only
[17:02] <Daemon404> intra*
[17:03] <nevcairiel> technically? :d
[17:03] <Daemon404> it has null frames
[17:03] <CIA-119> ffmpeg: 03Stefano Sabatini 07master * r1bb30d0295 10ffmpeg/libavfilter/vf_deshake.c: 
[17:03] <CIA-119> ffmpeg: lavfi/deshake: fix crash in uninit()
[17:03] <CIA-119> ffmpeg: Add NULL check on avctx before calling avcodec_close() on it.
[17:03] <CIA-119> ffmpeg: Fix trac ticket #1472.
[17:03] <nevcairiel> anything that causes problems with frame threading?
[17:07] <Daemon404> im not sure how it interacts
[17:07] <Daemon404> but iric lags' null frames were just "use th last decoded frame"
[17:08] <Daemon404> howevery many times
[17:09] <nevcairiel> reusing the last frame is complicated with frame threading, so probably implemented just by skipping :p
[17:11] <Daemon404> nevcairiel, aye
[17:11] <Daemon404> nevcairiel, btw i see you nabbed my TC
[17:11] <nevcairiel> yeah, works just fine
[17:12] <Daemon404> xkhm rejected it because "different layout"
[17:13] <nevcairiel> his reasons are weird
[17:13] <Daemon404> pretty sure he just does drugs
[17:22] <durandal_1707> nice, qatar is going to bump lavfi major
[17:40] <durandal_1707> michaelni: how to handle strange issue in smush demuxer? First audio packet is bigger that any other and this confuses ffmpeg.
[17:41] <durandal_1707> it is mostly noticable with mplayer - it shows ??? for A-V sync 
[17:41] <durandal_1707> and audio is not sync with video when samples are played with mplayer
[17:41] <michaelni> are timestamps & duration set correctly ?
[17:41] <durandal_1707> strangely encoded files do not have such issue
[17:42] <durandal_1707> audio and video sets correct timestamps wit avpriv_set_pts....
[17:42] <durandal_1707> for duration do you mean duration per packet?
[17:43] <michaelni> yes AVPacket.dts/pts/duration
[17:44] <durandal_1707> hmm in that case part of VIMA decoder code should be copyied to demuxer 
[17:44] <durandal_1707> the one which sets nb_frames
[17:59] <durandal_1707> how much big files should be for fate?
[18:06] <michaelni> files in the fate test suite should be as small as possible
[18:07] <michaelni> the fate suite is 600mb currently
[18:33] <durandal_1707> michaelni: i set pkt->duration and mplayer A-V sync issues are resolved, but -acodec, vcodec copy is still broken - it lists only audio packets
[18:39] <CIA-119> ffmpeg: 03Paul B Mahol 07master * r3821632e1b 10ffmpeg/libavformat/smush.c: 
[18:39] <CIA-119> ffmpeg: smush: set audio packet duration
[18:39] <CIA-119> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[18:48] <Compn> copy is broken in a lot of stuff durandal_1707 
[18:48] <durandal_1707> but this does not allow demuxer test for smush
[18:49] <durandal_1707> because only audio frames shows up and video frames dissapears in /dev/null
[18:56] <CIA-119> ffmpeg: 03Paul B Mahol 07master * rf4da707f58 10ffmpeg/doc/general.texi: 
[18:56] <CIA-119> ffmpeg: Replace Libav with FFmpeg
[18:56] <CIA-119> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[19:15] <durandal_1707> and btw when using -f framemd5 - output lines may be corrupted with s at the end, do you also get this?
[19:17] <Compn> sounds like something that got fixed recently
[19:17] <durandal_1707> not here
[19:18] <durandal_1707> strangely it does not happen if i add -a/vcodec copy
[19:20] <durandal_1707> it is bug in some kind of terminal handling...
[19:36] <ubitux> oh nice the json captions
[19:49] <CIA-119> ffmpeg: 03Nicolas George 07master * r7681d0eaca 10ffmpeg/libavdevice/lavfi.c: 
[19:49] <CIA-119> ffmpeg: lavd/lavfi: do not set the channel layout list.
[19:49] <CIA-119> ffmpeg: The list was set to avfilter_all_channel_layouts,
[19:49] <CIA-119> ffmpeg: which is really all common channel layouts.
[19:49] <CIA-119> ffmpeg: Not setting it leaves an empty list,
[19:49] <CIA-119> ffmpeg: which now means really all.
[19:49] <CIA-119> ffmpeg: 03Michael Niedermayer 07master * rb829da3639 10ffmpeg/libavcodec/smacker.c: 
[19:49] <CIA-119> ffmpeg: smacker: add sanity check for length in smacker_decode_tree()
[19:49] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[19:49] <CIA-119> ffmpeg: 03Hendrik Leppkes 07master * r14ac47fcc7 10ffmpeg/libavcodec/vc1dec.c: 
[19:49] <CIA-119> ffmpeg: vc1: signal interlaced and tff flag to the consumer
[19:49] <CIA-119> ffmpeg: Reviewed-by: Derek Buitenhuis <derek.buitenhuis at gmail.com>
[19:49] <CIA-119> ffmpeg: Reviewed-by: Mashiat Sarker Shakkhar <mashiat.sarker at gmail.com>
[19:49] <CIA-119> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[20:34] <durandal_1707> i always forgot how to upload fate samples
[20:44] <llogan> i should be more consistent with my new lines in my patches.
[20:47] <durandal_1707> so where to upload them?
[21:01] <durandal_1707> http://ffmpeg.org/developer.html have some characters at right end lost
[21:01] <durandal_1707> section 1.5 Development Policy
[21:03] <CIA-119> ffmpeg: 03Nicolas George 07master * r9e63c30daa 10ffmpeg/libavformat/ (Makefile rawenc.c srtenc.c): 
[21:03] <CIA-119> ffmpeg: lavf: add a real SRT muxer.
[21:03] <CIA-119> ffmpeg: This muxer supports CODEC_ID_SRT with the timestamps in the packet data
[21:03] <CIA-119> ffmpeg: and CODEC_ID_TEXT with the timestamps in the packet fields.
[21:03] <CIA-119> ffmpeg: Makes -scodec copy work from Matroska.
[21:09] <michaelni> durandal_1707, if you want to add something to fatesamples, i or bcoudurier can do that
[21:09] <michaelni> should be documented in fate.texi but maybe its not
[21:09] <durandal_1707> where to upload one sample so i can sent patch to ml?
[21:10] <michaelni> if you give me a url i can copy it to the server 
[21:10] <michaelni> also i need to know the directory!
[21:16] <durandal_1707> michaelni: i uploaded it to ftp://upload.ffmpeg.org/MPlayer/incoming/smush/
[21:19] <CIA-119> ffmpeg: 03Paul B Mahol 07master * rfd0b8c6ad5 10ffmpeg/libavcodec/sanm.c: 
[21:19] <CIA-119> ffmpeg: sanm: silence warnings
[21:19] <CIA-119> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[21:21] <michaelni> durandal_1707, uploaded
[21:35] <llogan> is ffmpeg-devel ml the preferred place for people to submit logo?
[21:36] <durandal_1707> isn't there policy to submit logos on trac?
[21:39] <llogan> where is Carl, Master of Trac, when you need him?
[21:41] <Compn> llogan : more people look at ml
[21:41] <Compn> but if you make a trac, its easier to make a collection of them :)
[21:41] <Compn> so do both
[21:43] <beastd> the logo should definitely be uploaded on trac with type art. you can advertise it on the ML though to catch more attention
[21:46] <llogan> i'm writing a news entry about Victoria's logo with instructions for others who may be interested in doing the same
[21:46] Action: llogan forgot about "[FFmpeg-devel] Logo/Banner rules (try #1)"
[21:46] <beastd> llogan: great
[21:46] <beastd> llogan: also look at an email i sent about a few month ago
[21:47] <beastd> in response to jagannath
[21:50] <llogan> i think this requires a wiki page with instructions.
[21:50] <llogan> and now I'm digging myself a bigger hole. this was supposed to be a simple news entry.
[21:51] <beastd> llogan: yes. make a news page and add a link to a wiki page with detailed instructions
[22:41] <CIA-119> ffmpeg: 03Anton Khirnov 07master * ra50b756280 10ffmpeg/libavcodec/options_table.h: (log message trimmed)
[22:41] <CIA-119> ffmpeg: lavc: remove stats_in from AVCodecContext options table.
[22:41] <CIA-119> ffmpeg: Its documentation states that it is allocated/freed by the caller, but
[22:41] <CIA-119> ffmpeg: it is declared as an AV_OPT_TYPE_STRING AVOption. Since
[22:41] <CIA-119> ffmpeg: 367732832faaf1bac4ece37cf7fef8c911e16312 the AVOptions system frees
[22:41] <CIA-119> ffmpeg: strings automatically. This can be considered an API break, since it
[22:41] <CIA-119> ffmpeg: won't work when the caller doesn't use av_malloc() to allocate the
[22:41] <CIA-119> ffmpeg: 03Anton Khirnov 07master * r394a2a0452 10ffmpeg/libavfilter/ (Makefile vf_fifo.c): 
[22:41] <CIA-119> ffmpeg: lavfi: rename vf_fifo.c -> fifo.c
[22:41] <CIA-119> ffmpeg: It will be used for audio too.
[22:41] <CIA-119> ffmpeg: 03Anton Khirnov 07master * r27570e211a 10ffmpeg/libavfilter/fifo.c: fifo: fix parenthesis placement.
[22:41] <CIA-119> ffmpeg: 03Anton Khirnov 07master * r4a97ff2fa7 10ffmpeg/libavfilter/ (Makefile allfilters.c fifo.c): lavfi/fifo: add audio version of the fifo filter.
[22:41] <CIA-119> ffmpeg: 03Anton Khirnov 07master * re026c9b390 10ffmpeg/libavfilter/audio.c: 
[22:41] <CIA-119> ffmpeg: * qatar/master:
[22:41] <CIA-119> ffmpeg:  fix hardcoded tables compililation caused by missing math constants
[22:42] <CIA-119> ffmpeg:  lavf: Make codec_tag arrays constant
[22:42] <CIA-119> ffmpeg:  twinvq: give massive struct a name.
[22:42] <CIA-119> ffmpeg:  lavf, lavu: version bumps and APIchanges for av_gettime() move
[22:42] <CIA-119> ffmpeg: 03Janne Grunau 07master * r1d01fee980 10ffmpeg/ (4 files in 2 dirs): 
[22:42] <CIA-119> ffmpeg: fix hardcoded tables compililation caused by missing math constants
[22:42] <CIA-119> ffmpeg: Add -D_XOPEN_SOURCE=600 to host cflags to make the constants in math.h
[22:42] <CIA-119> ffmpeg: available. Include math.h where necessary and remove redundant M_PI
[22:42] <CIA-119> ffmpeg: defines.
[22:53] <llogan> ubitux: +1 on the SVN removal. i was thinking of doing the same (I still see people attempting to use it).
[22:57] <ubitux> :)
[22:57] Action: ubitux hates doing that stuff, but it seems no one else want to :(
[23:28] <cbsrobot> ubitux: thanks for siwtching back to a serif font !
[23:28] <ubitux> np :p
[23:31] <durandal_1707> ubitux: gonna fix 1.5 section of developer.html?
[23:32] <ubitux> it looks like you are willing to do it
[23:33] <ubitux> (and i'm not)
[23:33] <ubitux> durandal_1707: what do you want to change?
[23:34] <durandal_1707> it is rendered incorrectly in opera
[23:34] <durandal_1707> characters at right end are hidden
[23:38] <ubitux> oO
[00:00] --- Fri Jun 22 2012