[Ffmpeg-devel-irc] ffmpeg-devel.log.20120325

[burek]

[00:00] <durandal_1707> clang fails here
[00:00] <durandal_1707> libswresample/swresample_internal.h:78:2: error: redefinition of typedef 'SwrContext' is invalid in C [-Wtypedef-redefinition]
[00:01] <durandal_1707> C99, C89, C2999 ?
[00:02] <michaelni> --std=c99
[00:03] <durandal_1707> gcc ignores standards?
[00:04] <Compn> always
[00:04] <Compn> :P
[00:06] <ubitux> just like ffmpeg, it is able to swallow any crap :)
[00:48] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r15890d67c1 10ffmpeg/libswresample/swresample.h: 
[00:48] <CIA-17> ffmpeg: swr: drop SWR_CH_MAX with the next major ABI bump
[00:48] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[00:48] <CIA-17> ffmpeg: 03Paul B Mahol 07master * rbcae7f432a 10ffmpeg/libavcodec/libaacplus.c: 
[00:48] <CIA-17> ffmpeg: libaacplus: switch to encode2()
[00:48] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[00:48] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[00:48] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rac6798dbc1 10ffmpeg/libswresample/swresample_internal.h: 
[00:48] <CIA-17> ffmpeg: swr: fix compilation failure with old versions of gcc.
[00:48] <CIA-17> ffmpeg: gcc 4.4 did not like the way i used typedefs
[00:48] <CIA-17> ffmpeg: Idea-by: beastd
[00:48] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[00:48] <Daemon404> ubitux, no
[00:48] <Daemon404> gcc is more like MS
[00:48] <Daemon404> makes up its own shit
[00:48] <Daemon404> gcc-isms
[00:50] <ubitux> :)
[01:15] <durandal_1707> michaelni: yay looks i will compare only last colummn of compn because recreating full compn is going to be impossible
[01:17] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rf4c380a5e6 10ffmpeg/ffmpeg.c: 
[01:17] <CIA-17> ffmpeg: ffmpeg: drop struct from SwrContext.
[01:17] <CIA-17> ffmpeg: Its a typdef now and doesnt need the struct anymore.
[01:17] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[01:20] <Compn> what ;)
[01:20] <Daemon404> its shorthand lol
[01:37] <durandal_1707> michaelni: so should be enough to write function in avutil which will compare only partial comp or i need to add flags for av_find_pix_fmt()
[01:37] <durandal_1707> ?
[01:38] <michaelni> anything that can find every pixfmt uniquely will surely work
[01:39] <durandal_1707> michaelni: problem is that openjpeg interface or jpeg2000 (or any image format) is flawed
[01:40] <durandal_1707> there is no enough information in opj_image_t to give uniquely match
[01:42] <durandal_1707> so to allow av_find_pix_fmt to be usefull in this case
[01:42] <durandal_1707> one need to adds flags
[01:42] <durandal_1707> if flag & 16 compare depth_minus1 comp
[01:54] <michaelni> ok, whatever works is fine of course
[03:10] <CIA-17> ffmpeg: 03Kostya Shishkov 07master * r494bce6224 10ffmpeg/libavcodec/ralf.c: 
[03:10] <CIA-17> ffmpeg: ralf: read Huffman code lengths without GetBitContext
[03:10] <CIA-17> ffmpeg: Those descriptions are stored in nibbles, so they are easy to extract.
[03:10] <CIA-17> ffmpeg: And this way we don't need to pad tables for possible bit reader overreads.
[03:10] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * r2ee01fbded 10ffmpeg/libavcodec/pthread.c: 
[03:10] <CIA-17> ffmpeg: pthread: free progress if buffer allocation failed.
[03:10] <CIA-17> ffmpeg: Else we run out of progress variables after a few failed buffer
[03:10] <CIA-17> ffmpeg: allocations.
[03:10] <CIA-17> ffmpeg: Signed-off-by: Anton Khirnov <anton at khirnov.net>
[03:10] <CIA-17> ffmpeg: 03Janne Salonen 07master * r14ba7472dc 10ffmpeg/libavcodec/vp8.c: 
[03:10] <CIA-17> ffmpeg: vp8: fix update_lf_deltas in libavcodec/vp8.c
[03:10] <CIA-17> ffmpeg: lf_delta.ref[i] and lf_delta.mode[i] were incorrectly reset to 0 if
[03:10] <CIA-17> ffmpeg: specific delta value was not updated. Fixed to keep the previous value
[03:10] <CIA-17> ffmpeg: if flag indicates that element in question is not updated.
[03:10] <CIA-17> ffmpeg: Signed-off-by: Janne Salonen <jsalonen at google.com>
[03:10] <CIA-17> ffmpeg: Signed-off-by: Ronald S. Bultje <rsbultje at gmail.com>
[03:10] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * rbc1ef85520 10ffmpeg/ (avconv.c libavcodec/pthread.c libavcodec/utils.c): 
[03:10] <CIA-17> ffmpeg: lavc/avconv: support changing frame sizes in codecs with frame mt.
[03:10] <CIA-17> ffmpeg: conversion function.
[03:11] <CIA-17> ffmpeg: Signed-off-by: Martin Storsjö <martin at martin.st>
[04:59] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r4f16153d35 10ffmpeg/libswresample/ (swresample.c swresample_internal.h): 
[04:59] <CIA-17> ffmpeg: swr: allow flushing in multiple steps if the output is too small.
[04:59] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[04:59] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r86e3289ffd 10ffmpeg/libavformat/mov.c: 
[04:59] <CIA-17> ffmpeg: mov: fix heap buffer overflow
[04:59] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[04:59] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[06:34] <durandal_1707> michaelni: what puts back s->flushed to zero?
[06:54] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r19342bc725 10ffmpeg/libswresample/swresample_test.c: 
[06:54] <CIA-17> ffmpeg: swr: test partial flushing
[06:54] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[06:54] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r8b11ae8a9a 10ffmpeg/libswresample/swresample.c: 
[06:54] <CIA-17> ffmpeg: swr: Fix partial flushing.
[06:54] <CIA-17> ffmpeg: Found-by: durandal_1707
[06:54] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[10:52] <ubitux> michaelni: why aren'y you dropping {in,out}_sample_rate init line 129-132 in 19342bc?
[10:52] <ubitux> unless i'm missing sth you're initializing them for nothing the first time
[10:56] <floater> GMplayer Playing ./Cowboyok.jpg.
[10:56] <floater> Seek failed
[10:56] <floater> libavformat file format detected.
[10:56] <floater> lold
[10:59] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r7e592379d9 10ffmpeg/libswresample/resample.c: 
[10:59] <CIA-17> ffmpeg: resample: Fix boundary handling with fragments.
[10:59] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[11:05] <michaelni> ubitux, hmmm
[11:06] <michaelni> looks like a stray hunk that somehow was created by some misapplied patch maybe
[11:06] <michaelni> iam not sure
[11:10] <michaelni> ill kill that hunk im my next push, i remember where it come from ...
[12:24] <CIA-17> ffmpeg: 03Reimar Döffinger 07master * r535eed9301 10ffmpeg/libavcodec/vqavideo.c: 
[12:24] <CIA-17> ffmpeg: vqavideo: fix decoding of samples with no partial updates.
[12:24] <CIA-17> ffmpeg: Fixes playback of
[12:24] <CIA-17> ffmpeg: http://samples.mplayerhq.hu/game-formats/vqa/landsoflore/B0EB8711.VQA
[12:24] <CIA-17> ffmpeg: (trac issue #710).
[12:24] <CIA-17> ffmpeg: Signed-off-by: Reimar Döffinger <Reimar.Doeffinger at gmx.de>
[12:24] <CIA-17> ffmpeg: 03Reimar Döffinger 07master * rfeda266db5 10ffmpeg/libavcodec/dvdata.c: 
[12:24] <CIA-17> ffmpeg: DV: check coded_width/height instead of width/height.
[12:24] <CIA-17> ffmpeg: This makes the workaround also work with lowres.
[12:24] <CIA-17> ffmpeg: Fixes issue #633.
[12:24] <CIA-17> ffmpeg: Signed-off-by: Reimar Döffinger <Reimar.Doeffinger at gmx.de>
[12:24] <CIA-17> ffmpeg: 03Reimar Döffinger 07master * r31ff2cb266 10ffmpeg/libavcodec/dvdata.c: 
[12:24] <CIA-17> ffmpeg: DV: detect another case of broken DV files.
[12:24] <CIA-17> ffmpeg: See trac issue #217.
[12:25] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r691b92474a 10ffmpeg/libswresample/swresample_test.c: 
[12:25] <CIA-17> ffmpeg: swr: revert stray hunk
[12:25] <CIA-17> ffmpeg: Found-by:ubitux
[12:25] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[12:26] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * re0c7128302 10ffmpeg/libswresample/swresample.c: 
[12:26] <CIA-17> ffmpeg: swr: fix segfault when 0 out size is used.
[12:26] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[12:26] <CIA-17> ffmpeg: 03Philip Langdale 07master * r1318b14379 10ffmpeg/libavcodec/crystalhd.c: 
[12:26] <CIA-17> ffmpeg: CrystalHD: Remove EXPERIMENTAL flag for known good formats.
[12:26] <CIA-17> ffmpeg: With the flag in place, it's hard to actually use the decoder, and
[12:26] <CIA-17> ffmpeg: I'm happy with how it works, with the exception of DivX3 where I've
[12:26] <CIA-17> ffmpeg: never found a sample that worked that I was confident actually
[12:26] <CIA-17> ffmpeg: matched what the hardware claimed to support.
[12:26] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[12:26] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r680981832f 10ffmpeg/ (6 files in 3 dirs): 
[12:26] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[12:26] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[12:26] <CIA-17> ffmpeg: 03Marton Balint 07master * r5420523ae3 10ffmpeg/libavcodec/pthread.c: 
[12:26] <CIA-17> ffmpeg: pthread: return proper error code on pthread_create failure
[12:26] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[13:29] <Compn> floater : mplayer mf://file.jpg for images
[15:58] <transplant> hi. I've just installed latest HEAD from git after some months. I have trouble with some mp3s. while the previous version detected only one stream of type "mp3", now two streams are detected. #0.0 mp3 and #0.1 video mjpeg, which is the cover art
[16:00] <JEEB> yes, to add support for more metadata. You can make it ignore the cover art if you wish, there's a flag for that at least in the API as far as I know.
[16:01] <ubitux> http://ffmpeg.org/ffmpeg.html#mp3
[16:14] <transplant> ok. thanks
[19:30] <durandal_1707> michaelni: i wrote av_find_pix_fmt and i get strange behaviour when using it with libopenjpeg decoder
[19:31] <durandal_1707> finding rgb and yuv pix_fmts is working, but gray one works only with ffprobe
[19:31] <durandal_1707> and ffmpeg -i input output fails for strange reason
[19:32] <durandal_1707> so perhaps something corrupts av_pix_fmt_descriptors[] ?
[19:40] <michaelni> durandal_1707, hmm, you could check with memcpy & memcmp if the table changed
[20:20] <durandal_1707> mp fileter vf_screenshot is pointless shouldn't it be removed?
[20:21] <Daemon404> the idea was to remove mp completely.
[20:34] <durandal_1707> idea is to port mp filters and remove it
[20:34] <durandal_1707> porting screenshot filter is pointless idea
[20:43] <durandal_1707> and there is mirror filter is it same as hflip ?
[22:17] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r1b0fcf33b8 10ffmpeg/libswresample/ (swresample.c swresample.h): 
[22:17] <CIA-17> ffmpeg: swr: More flexible and convenient buffering
[22:17] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[22:17] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r59b038b4a4 10ffmpeg/libswresample/swresample.c: 
[22:17] <CIA-17> ffmpeg: swr: realloc_audio: support packed audio buffers
[22:17] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[22:17] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r8d7f2db60a 10ffmpeg/libswresample/swresample_test.c: 
[22:17] <CIA-17> ffmpeg: swr-test: test the new flexible buffering
[22:17] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[22:17] <CIA-17> ffmpeg: 03Julien Ramseier 07master * re479013ae4 10ffmpeg/ffmpeg.c: 
[22:17] <CIA-17> ffmpeg: Fix SVCD scan_offset option
[22:17] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[23:42] <CIA-17> ffmpeg: 03Ronald S. Bultje 07master * rec0ed97b04 10ffmpeg/libavcodec/utvideo.c: 
[23:42] <CIA-17> ffmpeg: utvideo: port header reading to bytestream2.
[23:42] <CIA-17> ffmpeg: Fixes crash during slice size reading if slice_end goes negative.
[23:42] <CIA-17> ffmpeg: Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
[23:42] <CIA-17> ffmpeg: CC: libav-stable at libav.org
[23:42] <CIA-17> ffmpeg: 03Alexander Strange 07master * r6ef4063957 10ffmpeg/libavcodec/h264_ps.c: 
[23:42] <CIA-17> ffmpeg: h264: Add check for invalid chroma_format_idc
[23:42] <CIA-17> ffmpeg: Fixes a crash when FF_DEBUG_PICT_INFO is used.
[23:42] <CIA-17> ffmpeg: Signed-off-by: Ronald S. Bultje <rsbultje at gmail.com>
[23:42] <CIA-17> ffmpeg: 03Alexander Strange 07master * r147ee4cf06 10ffmpeg/libavcodec/pthread.c: 
[23:42] <CIA-17> ffmpeg: pthread: Immediately release all frames in ff_thread_flush()
[23:42] <CIA-17> ffmpeg: Before this, they were only added to the delayed release queue and not
[23:42] <CIA-17> ffmpeg: freed until later. This could lead to unnecessary memory use or buffer
[23:42] <CIA-17> ffmpeg: exhaustion.
[23:42] <CIA-17> ffmpeg: Signed-off-by: Ronald S. Bultje <rsbultje at gmail.com>
[23:43] <CIA-17> ffmpeg: 03Martin Storsjö 07master * r68893afe1d 10ffmpeg/libavformat/movenc.c: 
[23:43] <CIA-17> ffmpeg: 03Diego Biurrun 07master * r3b54912113 10ffmpeg/libavcodec/x86/dsputil_mmx.c: x86: K&R prettyprinting cosmetics for dsputil_mmx.c
[23:43] <CIA-17> ffmpeg: 03Diego Biurrun 07master * r62ce9defb8 10ffmpeg/libavcodec/x86/dsputil_mmx.c: x86: dsputil: prettyprint gcc inline asm
[23:43] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r9621646eb3 10ffmpeg/: (log message trimmed)
[23:43] <CIA-17> ffmpeg: Merge remote-tracking branch 'qatar/master'
[23:43] <CIA-17> ffmpeg: * qatar/master:
[23:43] <CIA-17> ffmpeg:  x86: dsputil: prettyprint gcc inline asm
[23:43] <CIA-17> ffmpeg:  x86: K&R prettyprinting cosmetics for dsputil_mmx.c
[23:43] <CIA-17> ffmpeg:  x86: conditionally compile H.264 QPEL optimizations
[23:43] <CIA-17> ffmpeg:  dsputil_mmx: Surround QPEL macros by "do { } while (0);" blocks.
[23:43] <CIA-17> ffmpeg: 03Alexander Strange 07master * rcb34867780 10ffmpeg/libavcodec/h264_ps.c: 
[23:44] <CIA-17> ffmpeg: h264: fix memleak in error path.
[23:44] <CIA-17> ffmpeg: Signed-off-by: Ronald S. Bultje <rsbultje at gmail.com>
[23:44] <CIA-17> (1 lines omitted)
[00:00] --- Mon Mar 26 2012