[Ffmpeg-devel-irc] ffmpeg-devel.log.20120518

[burek]

[00:06] <gnafu> Compn: But while you may not be able to type English today, perhaps you can type it tomorrow when tomorrow becomes today.
[00:13] <Compn> never!
[01:01] <CIA-105> ffmpeg: 03Stefano Sabatini 07master * r47ef261288 10ffmpeg/libavutil/parseutils.h: lavu/parseutils: fix specification for av_parse_time()
[01:01] <CIA-105> ffmpeg: 03Stefano Sabatini 07master * rcfde7395b1 10ffmpeg/libavfilter/asrc_aevalsrc.c: lavfi/aevalsrc: remove limitation on the number of channels
[01:01] <CIA-105> ffmpeg: 03Stefano Sabatini 07master * re1a8df70a2 10ffmpeg/ (3 files in 2 dirs): 
[01:01] <CIA-105> ffmpeg: lavfi/aevalsrc: add option channel_layout
[01:01] <CIA-105> ffmpeg: Allow to explicitly specify the channel layout to select.
[01:01] <CIA-105> ffmpeg: 03Stefano Sabatini 07master * r17d4233c9c 10ffmpeg/libavfilter/asrc_aevalsrc.c: lavfi/aevalsrc: reindent after the previous commit
[01:05] <ubitux> saste: hey :)
[01:05] <ubitux> saste: i was working on testing lavfi right now
[01:05] <ubitux> do you have a nice aevalsrc in mind? :)
[01:06] <ubitux> i was trying do to some random piano keys
[01:07] <ubitux> with sth like sin(3000*t)*exp(-4*t)
[01:07] <ubitux> i don't get very satisfying results :')
[01:07] <ubitux> :(
[02:04] <ubitux> it's fun to see how better ffplay -f lavfi 'aevalsrc=sin(440*PI*2*t)*exp(-4*t),aformat=sample_fmts=s16' is instead of just the aevalsrc (which get u8 as default output)
[02:04] <ubitux> can't we have a list of preferences?
[04:52] <CIA-105> ffmpeg: 03Marton Balint 07master * r89080a0a5e 10ffmpeg/ffplay.c: 
[04:52] <CIA-105> ffmpeg: ffplay: fix video_thread when no frame is returned in get_video_frame
[04:52] <CIA-105> ffmpeg: Affects only ffplay build with avfilter disabled.
[04:52] <CIA-105> ffmpeg: Signed-off-by: Marton Balint <cus at passwd.hu>
[04:53] <CIA-105> ffmpeg: 03Marton Balint 07master * r84087b243e 10ffmpeg/ffplay.c: 
[04:53] <CIA-105> ffmpeg: ffplay: use stream sample_aspect_ratio if available in source frames
[04:53] <CIA-105> ffmpeg: When we are using filter chains we have to set the aspect ratio of the source
[04:53] <CIA-105> ffmpeg: to the best known value, we use the av_guess_sample_aspect_ratio function to
[04:53] <CIA-105> ffmpeg: determine that.
[04:53] <CIA-105> ffmpeg: Fixes ticket 1228.
[04:53] <CIA-105> ffmpeg: Signed-off-by: Marton Balint <cus at passwd.hu>
[04:53] <CIA-105> ffmpeg: 03Marton Balint 07master * r843509e264 10ffmpeg/ffplay.c: 
[04:53] <CIA-105> ffmpeg: ffplay: use AVFrame::width and height instead of using codec or filter settings
[04:53] <CIA-105> ffmpeg: Codec values may not reflect the actual frame size, and it also enables us to
[04:53] <CIA-105> ffmpeg: simplify code in the avfilter enabled and the avfilter disabled case.
[04:53] <CIA-105> ffmpeg: Signed-off-by: Marton Balint <cus at passwd.hu>
[04:53] <CIA-105> ffmpeg: 03Michael Niedermayer 07master * rdb6ef054f5 10ffmpeg/libavfilter/af_aresample.c: 
[04:53] <CIA-105> ffmpeg: af_aresample: 10l fix, wrong timebase for next_pts.
[04:53] <CIA-105> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[04:53] <CIA-105> ffmpeg: 03Michael Niedermayer 07master * rae1ab8265b 10ffmpeg/: (log message trimmed)
[04:53] <CIA-105> ffmpeg: Merge remote-tracking branch 'cus/stable'
[04:53] <CIA-105> ffmpeg: * cus/stable:
[04:53] <CIA-105> ffmpeg:  ffplay: put aspect ratio info to the VideoPicture struct
[04:53] <CIA-105> ffmpeg:  ffplay: use AVFrame::width and height instead of using codec or filter settings
[04:53] <CIA-105> ffmpeg:  ffplay: use stream sample_aspect_ratio if available in source frames
[04:53] <CIA-105> ffmpeg:  ffplay: fix video_thread when no frame is returned in get_video_frame
[04:54] <CIA-105> ffmpeg: 03Marton Balint 07master * r825ec16da9 10ffmpeg/ffplay.c: 
[04:54] <CIA-105> ffmpeg: ffplay: put aspect ratio info to the VideoPicture struct
[04:54] <CIA-105> ffmpeg: Also use av_guess_sample_aspect_ratio for determining aspect ratio of the video
[04:54] <CIA-105> ffmpeg: frame if not using avfilter.
[04:54] <CIA-105> ffmpeg: Signed-off-by: Marton Balint <cus at passwd.hu>
[12:11] <ubitux> eval question: how can i initialize a register only once?
[12:12] <ubitux> sth like : if(isset(0), st(0, 10))
[12:16] <ubitux> mmh it should be set to zero by default
[12:18] <ubitux> oh, it works.
[12:26] <ubitux> http://pastie.org/private/8dbsbfcfiy2p7v77ddi4q
[12:27] <ubitux> here it is, my first random melody
[15:35] <CIA-105> ffmpeg: 03Nicolas George 07master * rc0c2424f73 10ffmpeg/tests/ (Makefile fate/probe.mak): fate: ffprobe tests require SAMPLES.
[18:29] <Daemon404> any reason why ffplay would show nothing but a black screen for an avc file
[18:31] <Compn> does it work in mplayer?
[18:32] <Daemon404> lol i think i see the problem
[18:34] <Daemon404> MP4Box video stdout: AVC-H264 import - frame size 320 x 240 at 30000.000 FPS
[18:34] <Daemon404> yeah.
[18:34] <Daemon404> mp4box does not liek rationals.
[18:35] <Compn> lol
[18:35] <Compn> ffplay -r 30 ?
[18:36] <Compn> ffplay doesnt warn on fps over 9000 ?
[18:36] <Compn> :P
[18:36] <Compn> not to mention mp4box bug
[18:37] <Daemon404> mp4box is just shitty
[18:37] <Compn> why u no use mkvtoolsnix ?
[18:37] <Compn> mp4 itself is ... not good
[18:37] <Daemon404> >streaming
[18:37] <Compn> ?
[18:37] <Compn> you mean broken-hack-hack-mp4 streaming?
[18:38] <Daemon404> you mean the thing that every streamign site in existence uses?
[18:38] <Daemon404> and is basically required at this point?
[18:38] <Compn> oh flash and stuff
[18:38] <Compn> :P
[18:38] <Daemon404> yes
[18:40] <gnafu> 30000 fps?  Pfft!  29999 fps should be enough for anybody.
[21:45] Action: Daemon404 jabs tg2
[23:16] <CIA-105> ffmpeg: 03Michael Niedermayer 07master * r96ac8663de 10ffmpeg/libavfilter/af_aresample.c: 
[23:16] <CIA-105> ffmpeg: af_aresample: Consider the swresample delay during calculating timestamps.
[23:16] <CIA-105> ffmpeg: The difference from this should be pretty small.
[23:16] <CIA-105> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[23:16] <CIA-105> ffmpeg: 03Michael Niedermayer 07master * r4def5d2b64 10ffmpeg/libswresample/ (resample.c swresample.h): 
[23:16] <CIA-105> ffmpeg: swr: add swr_get_delay() to find the exact delay the swresampler introduces.
[23:16] <CIA-105> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[23:16] <CIA-105> ffmpeg: 03Michael Niedermayer 07master * r6ba692f8a7 10ffmpeg/ (4 files in 3 dirs): 
[23:16] <CIA-105> ffmpeg: af_aresample: fix rounding that led to sample accumulation in the buffers.
[23:16] <CIA-105> ffmpeg: This fixes a regression that apparently was missed when switching to the
[23:16] <CIA-105> ffmpeg: in af resampler
[23:16] <CIA-105> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * rd1562d32cf 10ffmpeg/libavdevice/dshow.h: lavd/dshow: protect header.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * rcb9aed10f7 10ffmpeg/libavdevice/Makefile: lavd/dshow: fix make checkheaders.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * rf8f3deb731 10ffmpeg/libavfilter/transform.h: lavfi/transform: fix make checkheaders.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r1308fa42ad 10ffmpeg/libavformat/swf.h: lavf/swf: fix make checkheaders.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r7b78cb3a70 10ffmpeg/libavcodec/g729postfilter.h: lavc/g729postfilter: fix make checkheaders.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r5b70e28c6e 10ffmpeg/libavcodec/libutvideo.h: lavc/libutvideo: protect header.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * rc8a7dd03f8 10ffmpeg/libavcodec/jacosub.h: lavc/jacosub: fix make checkheaders.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * rb56f885a24 10ffmpeg/libavcodec/Makefile: lavc/libutvideo: fix make checkheaders.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r9ae47fbd9e 10ffmpeg/libavcodec/dcaenc.h: lavc/dcaenc: fix make checkheaders.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r08ad9f9a78 10ffmpeg/libavcodec/Makefile: lavc/os2threads: fix make checkheaders.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r3dba3b265c 10ffmpeg/libavcodec/Makefile: lavc/vda: fix make checkheaders.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r9e2cee5770 10ffmpeg/ffmpeg.c: 
[23:45] <CIA-105> ffmpeg: ffmpeg: do not deprecate -vol and -map_channel.
[23:45] <CIA-105> ffmpeg: These helpers are handy, just notice the user it is done using lavfi.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r48f30595f5 10ffmpeg/libavfilter/af_pan.c: 
[23:45] <CIA-105> ffmpeg: pan: use ff_parse_channel_layout() instead of av_get_channel_layout().
[23:45] <CIA-105> ffmpeg: This allow specifying hexadecimal channel layout, and also handle the
[23:45] <CIA-105> ffmpeg: error printing internally.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * rfa29ca5f55 10ffmpeg/ (ffmpeg.c tests/Makefile): ffmpeg: resurrect -map_channel.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r7c2bb79f3b 10ffmpeg/libavutil/error.h: 
[23:45] <CIA-105> ffmpeg: lavu/error: fix make checkheaders.
[23:45] <CIA-105> ffmpeg: The errors need to be defined before including functions depending on
[23:45] <CIA-105> ffmpeg: them. See av_size_mult() for instance. stddef.h is included for the
[23:45] <CIA-105> ffmpeg: prototype of av_sterror (use of size_t).
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r7f8bd16103 10ffmpeg/ffmpeg.c: ffmpeg: remove unused variable i.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r1913989f56 10ffmpeg/libavcodec/Makefile: build: add codec_names.h to SKIPHEADERS to fix make checkheaders.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r64ff8a7670 10ffmpeg/libavfilter/ (all_channel_layouts.h formats.c): 
[23:45] <CIA-105> ffmpeg: lavfi: rename all_channel_layouts.h to all_channel_layouts.inc.
[23:45] <CIA-105> ffmpeg: all_channel_layouts.h is not a real header file and breaks make checkheaders.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r978c829431 10ffmpeg/libavcodec/g723_1_data.h: lavc/g723.1: fix make checkheaders.
[23:45] <CIA-105> ffmpeg: 03Clément BSsch 07master * r02908a4bd0 10ffmpeg/libavcodec/g723_1_data.h: lavc/g723: protect header.
[23:47] <ubitux> michaelni: there is a suspicious warning in ffmpeg.c
[23:48] <ubitux> ffmpeg.c: In function choose_pix_fmts:
[23:48] <ubitux> ffmpeg.c:710:9: warning: return from incompatible pointer type [enabled by default]
[00:00] --- Sat May 19 2012