[Ffmpeg-devel-irc] ffmpeg-devel.log.20120207

[burek]

[00:55] <CIA-40> ffmpeg: 03Paul B Mahol 07master * r6ba2505a0e 10ffmpeg/libavcodec/utvideo.c: 
[00:55] <CIA-40> ffmpeg: utvideo: remove ff_thread_finish_setup, it isn't needed at all
[00:55] <CIA-40> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[00:55] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[02:03] <CIA-40> ffmpeg: 03Andrey Utkin 07master * re7f0bc8c0f 10ffmpeg/libavfilter/vf_drawtext.c: 
[02:03] <CIA-40> ffmpeg: drawtext: add missing braces around an if() block.
[02:03] <CIA-40> ffmpeg: Prevents uninitialized read.
[02:03] <CIA-40> ffmpeg: Signed-off-by: Anton Khirnov <anton at khirnov.net>
[02:03] <CIA-40> ffmpeg: 03Andrey Utkin 07master * r2b43dfce36 10ffmpeg/libavfilter/vf_drawtext.c: 
[02:03] <CIA-40> ffmpeg: drawtext: fix text_{w, h} expression vars
[02:03] <CIA-40> ffmpeg: Before, {text_,}{w,h} vars hadn't got initialized
[02:03] <CIA-40> ffmpeg: Signed-off-by: Anton Khirnov <anton at khirnov.net>
[02:03] <CIA-40> ffmpeg: 03Andrey Utkin 07master * re496c45d9b 10ffmpeg/ (doc/filters.texi libavfilter/vf_drawtext.c): (log message trimmed)
[02:03] <CIA-40> ffmpeg: drawtext: add 'fix_bounds' option on coords fixing
[02:03] <CIA-40> ffmpeg: Before, drawtext filter deliberately altered given text coordinates if
[02:03] <CIA-40> ffmpeg: text didn't fully fit on the picture. This breaks the use case of
[02:03] <CIA-40> ffmpeg: scrolling large text, e.g. movie closing credits.
[02:03] <CIA-40> ffmpeg: Add 'fix_bounds', to make it usable in such cases (by setting its value to 0).
[02:03] <CIA-40> ffmpeg: Default behavior is not changed, and non-fitting text coords are fixed.
[02:03] <CIA-40> ffmpeg: 03Paul B Mahol 07master * rd73466f841 10ffmpeg/libavcodec/v210enc.c: 
[02:03] <CIA-40> ffmpeg: v210enc: check for coded_frame allocation failure
[02:03] <CIA-40> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[02:03] <CIA-40> ffmpeg: Signed-off-by: Anton Khirnov <anton at khirnov.net>
[02:03] <CIA-40> ffmpeg: 03Paul B Mahol 07master * r66fa2a1fb3 10ffmpeg/libavcodec/v210enc.c: 
[02:03] <CIA-40> ffmpeg: v210enc: do not set coded_frame->key_frame
[02:03] <CIA-40> ffmpeg: It is already set in avcodec_alloc_frame().
[02:03] <CIA-40> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[02:03] <CIA-40> ffmpeg: Signed-off-by: Anton Khirnov <anton at khirnov.net>
[02:03] <CIA-40> ffmpeg: 03Justin Ruggles 07master * r1bc035bc03 10ffmpeg/libavformat/ape.c: ape: return error if seeking to the current packet fails in ape_read_packet()
[02:03] <CIA-40> ffmpeg: 03Paul B Mahol 07master * rb5f50da593 10ffmpeg/libavcodec/v210enc.c: 
[02:03] <CIA-40> ffmpeg: v210enc: return proper AVERROR codes instead of -1
[02:04] <CIA-40> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[02:04] <CIA-40> ffmpeg: Signed-off-by: Anton Khirnov <anton at khirnov.net>
[02:04] <CIA-40> ffmpeg: 03Paul B Mahol 07master * r986c1c483b 10ffmpeg/libavcodec/v210enc.c: 
[02:04] <CIA-40> ffmpeg: v210enc: use stride as it is already calculated
[02:04] <CIA-40> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[02:04] <CIA-40> ffmpeg: Signed-off-by: Anton Khirnov <anton at khirnov.net>
[02:04] <CIA-40> ffmpeg: 03Justin Ruggles 07master * r66f7be3603 10ffmpeg/libavformat/ape.c: ape: return AVERROR_EOF instead of AVERROR(EIO) when demuxing is finished
[02:04] <CIA-40> ffmpeg: 03Kieran Kunhya 07master * r9814974d2e 10ffmpeg/libavcodec/avcodec.h: 
[02:04] <CIA-40> ffmpeg: avcodec: Clarify AVFrame member documentation.
[02:04] <CIA-40> ffmpeg: Signed-off-by: Diego Biurrun <diego at biurrun.de>
[02:04] <CIA-40> (36 lines omitted)
[06:09] <CIA-40> ffmpeg: 03Paul B Mahol 07master * r0b717e24f9 10ffmpeg/ffserver.c: 
[06:09] <CIA-40> ffmpeg: ffserver: use avcodec_get_context_defaults3()
[06:09] <CIA-40> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[06:09] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[09:40] <ubitux> who is Jeremy? (fate i386-darwin-llvm-gcc-4.2.1)
[13:56] <Granjow> Hello! How can I ensure that video is encoded with best quality? I have difficulties setting the bit rate ... when I set the AVCodecContext::bit_rate to 4M, the video quality is fine first but very blocky after a few seconds. With 400000000 it complains: [mpeg4 @ 0xfc73a0] bitrate tolerance too small for bitrate
[15:53] <CIA-40> ffmpeg: 03Stefano Sabatini 07master * r0349d61eef 10ffmpeg/libavfilter/vf_aspect.c: 
[15:53] <CIA-40> ffmpeg: lavfi/aspect: rename field aspect->aspect to aspect->ratio
[15:53] <CIA-40> ffmpeg: The new name is more sensible.
[15:53] <CIA-40> ffmpeg: 03Stefano Sabatini 07master * rd21b82fd3a 10ffmpeg/libavcodec/pcm.c: 
[15:53] <CIA-40> ffmpeg: lavc: provide aliases to the PCM A-law/mu-law codecs
[15:53] <CIA-40> ffmpeg: They are also known as G.711 codecs.
[17:33] <Compn> Daemon404 : diego asks why you arent maintaining the postproc switch changes that need to be done in mplayer now :P
[17:40] <Compn> Daemon404 : oh wait, hes looking at lavfi , nevermind
[17:40] <Compn> i'm mistaking things for different things
[17:40] Action: Compn blind
[17:40] <j-b> Daemon404: ping
[18:28] <CIA-40> ffmpeg: 03Carl Eugen Hoyos 07master * rbf5c0308f2 10ffmpeg/libavcodec/libutvideo.cpp: 
[18:28] <CIA-40> ffmpeg: Fix libutvideo compilation.
[18:28] <CIA-40> ffmpeg: Fixes ticket #985.
[19:05] <Daemon404> j-b:  mcrap i totally forgot. been buating my ass getting ready to go to UK
[19:06] <Daemon404> busting*
[19:12] <durandal_1707> Daemon404: have you wrote libutvideoenc?
[19:16] <Daemon404> i have in a rough form
[19:17] <Daemon404> it only supports yv12 atm
[19:17] <Daemon404> i wrote it in oct and just sorta... forgot
[19:18] Action: Daemon404 seems to have severe lack of time...
[19:18] <Daemon404> of only i could give up sleep like michealni ;)
[19:18] <Daemon404> if*
[19:20] Action: Daemon404 is tired all the time...
[19:22] <durandal_1707> looking at decoder it shouldn't be hard
[19:23] <Daemon404> it isnt
[19:23] <Daemon404> i ca  send you my thing from oct if u wanf
[19:23] <Daemon404> ... damn cellphone kb
[19:24] <durandal_1707> feel free to send whenever you can
[21:49] <j-b> https://trac.videolan.org/vlc/ticket/6000 security issue in libavcodec?
[22:14] <michaelni> j-b, cant reproduce a issue with valgrind ffmpeg seed/repro -f null -
[22:18] <ubitux> it plays pretty fine ;)
[22:20] <ubitux> michaelni: is there something bothering you about the 2nd patch? (lavfi polling simplification)
[22:20] <ubitux> i have a few more patches on top of this after that, if it is fine
[22:21] <michaelni> well , i know the code made some sense when it was written
[22:21] <michaelni> but i cant pinpoint a issue with the patch
[22:21] <michaelni> so maybe its ok maybe not
[22:21] <michaelni> fate passed with it though
[22:22] <ubitux> input_video_filter and output_video_filter are always "configured", what's bothering me though is the last line i drop
[22:22] <michaelni> dont count this as a objection to the patch ...
[22:22] <ubitux> (do we need to poll the frame before unref the buffer)
[22:22] <ubitux> i can wait more, i was just wondering why you avoided the patch :)
[22:22] <ubitux> if you have any idea what could go wrong
[22:23] <michaelni> i dont think we will find out before its applie d...
[22:23] <ubitux> haha
[22:23] <ubitux> well, i'm not fixing any bug, and i'm not adding any feature, so nothing urgent :)
[22:24] <ubitux> and overthinking might not be a bad idea at times :p
[22:25] <michaelni> well its costs time ...
[22:25] <michaelni> and if thats more time than (probability of bug) * (time to fix bug) then its a bad idea
[22:28] <ubitux> ok ok :)
[22:31] <CIA-40> ffmpeg: 03Joseph Artsimovich 07master * r950930b461 10ffmpeg/libavcodec/ (dnxhdenc.c options.c): 
[22:31] <CIA-40> ffmpeg: Set default qmax for VC-3/DNxHD to 1024 (maximum allowed by VC-3 spec).
[22:31] <CIA-40> ffmpeg: Also increase the global limit on qmax.
[22:31] <CIA-40> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[00:00] --- Wed Feb  8 2012