ffmpeg.git
8 years agoReplace all strcasecmp/strncasecmp usages.
Reimar Döffinger [Wed, 2 Nov 2011 19:17:25 +0000 (20:17 +0100)]
Replace all strcasecmp/strncasecmp usages.

All current usages of it are incompatible with localization.
For example strcasecmp("i", "I") != 0 is possible, but would
break many of the places where it is used.
Instead use our own implementations that always treat the data
as ASCII.

Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
8 years agolavc: Init AVFrame->opaque to AVCodecContext.opaque in avcodec_default_get_buffer()
Michael Niedermayer [Thu, 3 Nov 2011 13:54:15 +0000 (14:54 +0100)]
lavc: Init AVFrame->opaque to AVCodecContext.opaque in avcodec_default_get_buffer()
this should allow some simplifications in some applications.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agolavf: Do not set cur_dts to undetermined before first_dts has been calculated as...
Michael Niedermayer [Thu, 3 Nov 2011 13:31:01 +0000 (14:31 +0100)]
lavf: Do not set cur_dts to undetermined before first_dts has been calculated as this would upset the start_time and first_dts calculation.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoCosmetics: Reindent after last commit.
Carl Eugen Hoyos [Thu, 3 Nov 2011 13:30:00 +0000 (14:30 +0100)]
Cosmetics: Reindent after last commit.

8 years agoSupport new ALAC magic cookie in caf files.
Carl Eugen Hoyos [Thu, 3 Nov 2011 13:27:54 +0000 (14:27 +0100)]
Support new ALAC magic cookie in caf files.

8 years agoMerge remote-tracking branch 'qatar/master'
Michael Niedermayer [Thu, 3 Nov 2011 01:01:37 +0000 (02:01 +0100)]
Merge remote-tracking branch 'qatar/master'

* qatar/master: (44 commits)
  replacement Indeo 3 decoder
  gsm demuxer: do not allocate packet twice.
  flvenc: use first packet delay as global delay.
  ac3enc: doxygen update.
  imc: return error codes instead of 0 for error conditions.
  imc: return meaningful error codes instead of -1
  imc: do not set channel layout for stereo
  imc: validate channel count
  imc: check for ff_fft_init() failure
  imc: check output buffer size before decoding
  imc: use DSPContext.bswap16_buf() to byte-swap packet data
  rtsp: add allowed_media_types option
  libgsm: add flush function to reset the decoder state when seeking
  libgsm: simplify decoding by using a loop
  gsm: log error message when packet is too small
  libgsmdec: do not needlessly set *data_size to 0
  gsmdec: do not needlessly set *data_size to 0
  gsmdec: add flush function to reset the decoder state when seeking
  libgsmdec: check output buffer size before decoding
  gsmdec: log error message when output buffer is too small.
  ...

Conflicts:
Changelog
ffplay.c
libavcodec/indeo3.c
libavcodec/mjpeg_parser.c
libavcodec/vp3.c
libavformat/cutils.c
libavformat/id3v2.c
libavutil/parseutils.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoreplacement Indeo 3 decoder
Maxim Poliakovski [Wed, 2 Nov 2011 20:34:57 +0000 (21:34 +0100)]
replacement Indeo 3 decoder

The new decoder is much smaller and has better code quality.
Cleanup and fixes courtesy of Kostya Shishkov.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
8 years agogsm demuxer: do not allocate packet twice.
Justin Ruggles [Wed, 2 Nov 2011 22:30:49 +0000 (18:30 -0400)]
gsm demuxer: do not allocate packet twice.

fixes memleak with raw gsm demuxing.

8 years agoflvenc: use first packet delay as global delay.
Justin Ruggles [Wed, 26 Oct 2011 16:43:18 +0000 (12:43 -0400)]
flvenc: use first packet delay as global delay.

This keeps the streams sychronized. The packets must be interleaved per-DTS.

8 years agoac3enc: doxygen update.
Justin Ruggles [Wed, 2 Nov 2011 14:38:10 +0000 (10:38 -0400)]
ac3enc: doxygen update.

Add some parameters to existing function documentation.
Remove some unneeded documentation.
Convert some static function documentation to non-doxygen style.

8 years agoimc: return error codes instead of 0 for error conditions.
Justin Ruggles [Sat, 29 Oct 2011 15:16:01 +0000 (11:16 -0400)]
imc: return error codes instead of 0 for error conditions.

This fixes a bug where the whole buffer was returned as decoded audio due to
*data_size not being set to zero and the return value being >= 0.

8 years agoimc: return meaningful error codes instead of -1
Justin Ruggles [Fri, 28 Oct 2011 22:36:28 +0000 (18:36 -0400)]
imc: return meaningful error codes instead of -1

8 years agoimc: do not set channel layout for stereo
Justin Ruggles [Fri, 28 Oct 2011 22:32:40 +0000 (18:32 -0400)]
imc: do not set channel layout for stereo

we only support decoding of mono imc

8 years agoimc: validate channel count
Justin Ruggles [Fri, 28 Oct 2011 22:31:21 +0000 (18:31 -0400)]
imc: validate channel count

ask for a sample if not mono

8 years agoimc: check for ff_fft_init() failure
Justin Ruggles [Fri, 28 Oct 2011 22:25:49 +0000 (18:25 -0400)]
imc: check for ff_fft_init() failure

8 years agoimc: check output buffer size before decoding
Justin Ruggles [Fri, 28 Oct 2011 22:24:03 +0000 (18:24 -0400)]
imc: check output buffer size before decoding

8 years agoimc: use DSPContext.bswap16_buf() to byte-swap packet data
Justin Ruggles [Fri, 28 Oct 2011 22:16:31 +0000 (18:16 -0400)]
imc: use DSPContext.bswap16_buf() to byte-swap packet data

8 years agovf_overlay: adopts FAST_DIV255 macro for calculating destination alpha
Stefano Sabatini [Mon, 31 Oct 2011 10:32:11 +0000 (11:32 +0100)]
vf_overlay: adopts FAST_DIV255 macro for calculating destination alpha

Ease readability, improve exactness.

8 years agovf_fade: fade to correct CCIR601/709 black level
Mark Himsley [Wed, 2 Nov 2011 20:39:55 +0000 (21:39 +0100)]
vf_fade: fade to correct CCIR601/709 black level

Current implementation fades to 0. This implementation fades to 16 for
YUV formats that contain CCIR601/709 video levels. RGB and YUVJ
formats are not altered.

Signed-off-by: Stefano Sabatini <stefasab@gmail.com>
8 years agompegtsenc: EAC3 support
Mean [Wed, 2 Nov 2011 19:41:24 +0000 (20:41 +0100)]
mpegtsenc: EAC3 support

Reviewed-by: Baptiste Coudurier <baptiste.coudurier@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agortsp: add allowed_media_types option
John Brooks [Sat, 29 Oct 2011 21:41:24 +0000 (15:41 -0600)]
rtsp: add allowed_media_types option

Streams from RTSP or SDP that do not match an allowed type will
be skipped entirely, which allows video-only or audio-only
streaming from servers that provide both.

Signed-off-by: Martin Storsjö <martin@martin.st>
8 years agovc1dec: Fix CODEC_FLAG_LOW_DELAY
Mean [Wed, 2 Nov 2011 18:35:22 +0000 (19:35 +0100)]
vc1dec: Fix CODEC_FLAG_LOW_DELAY

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agompeg4videodec: export quarter_sample & divx_packed
Michael Niedermayer [Wed, 2 Nov 2011 18:48:26 +0000 (19:48 +0100)]
mpeg4videodec: export quarter_sample & divx_packed

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agolibgsm: add flush function to reset the decoder state when seeking
Justin Ruggles [Fri, 28 Oct 2011 14:35:15 +0000 (10:35 -0400)]
libgsm: add flush function to reset the decoder state when seeking

8 years agolibgsm: simplify decoding by using a loop
Justin Ruggles [Fri, 28 Oct 2011 14:28:41 +0000 (10:28 -0400)]
libgsm: simplify decoding by using a loop

8 years agogsm: log error message when packet is too small
Justin Ruggles [Fri, 28 Oct 2011 14:18:07 +0000 (10:18 -0400)]
gsm: log error message when packet is too small

8 years agolibgsmdec: do not needlessly set *data_size to 0
Justin Ruggles [Fri, 28 Oct 2011 14:10:12 +0000 (10:10 -0400)]
libgsmdec: do not needlessly set *data_size to 0

8 years agogsmdec: do not needlessly set *data_size to 0
Justin Ruggles [Fri, 28 Oct 2011 14:09:45 +0000 (10:09 -0400)]
gsmdec: do not needlessly set *data_size to 0

8 years agogsmdec: add flush function to reset the decoder state when seeking
Justin Ruggles [Fri, 28 Oct 2011 14:07:41 +0000 (10:07 -0400)]
gsmdec: add flush function to reset the decoder state when seeking

8 years agolibgsmdec: check output buffer size before decoding
Justin Ruggles [Fri, 28 Oct 2011 04:52:36 +0000 (00:52 -0400)]
libgsmdec: check output buffer size before decoding

8 years agogsmdec: log error message when output buffer is too small.
Justin Ruggles [Fri, 28 Oct 2011 04:34:52 +0000 (00:34 -0400)]
gsmdec: log error message when output buffer is too small.

also return AVERROR(EINVAL) instead of -1

8 years agogsm: use av_get_bytes_per_sample() in frame_bytes calculation
Justin Ruggles [Fri, 28 Oct 2011 04:31:11 +0000 (00:31 -0400)]
gsm: use av_get_bytes_per_sample() in frame_bytes calculation

8 years agoCreate separate functions for the raw GSM demuxer.
Justin Ruggles [Fri, 28 Oct 2011 04:07:32 +0000 (00:07 -0400)]
Create separate functions for the raw GSM demuxer.

Put the new raw GSM demuxer in its own file.
Fixes raw GSM demuxing.

8 years agoid3v2enc: include strings.h for strcasecmp()
Michael Niedermayer [Wed, 2 Nov 2011 18:10:49 +0000 (19:10 +0100)]
id3v2enc: include strings.h for strcasecmp()

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoopt: remove attribute_deprecated from several functions that do not have adequante...
Michael Niedermayer [Wed, 2 Nov 2011 17:41:21 +0000 (18:41 +0100)]
opt: remove attribute_deprecated from several functions that do not have adequante replacements.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoh264: add nal_length_size and is_avc to the publically readable fields
Michael Niedermayer [Wed, 2 Nov 2011 17:39:42 +0000 (18:39 +0100)]
h264: add nal_length_size and is_avc to the publically readable fields

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agolavc: fix avcodec_get_type()
Michael Niedermayer [Wed, 2 Nov 2011 16:04:46 +0000 (17:04 +0100)]
lavc: fix avcodec_get_type()

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoHWAccel: adds Video Decoder Acceleration (VDA) module for Mac OS X.
Sebastien Zwickert [Tue, 1 Nov 2011 23:55:09 +0000 (00:55 +0100)]
HWAccel: adds Video Decoder Acceleration (VDA) module for Mac OS X.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoMAINTAINERS: update release manager
Michael Niedermayer [Tue, 1 Nov 2011 11:58:03 +0000 (12:58 +0100)]
MAINTAINERS: update release manager

Found-by: Compn
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoid3v2enc: add support for year and day/month tags when writing id3v2 version 3 metadata
Tobias Rapp [Wed, 2 Nov 2011 10:50:14 +0000 (11:50 +0100)]
id3v2enc: add support for year and day/month tags when writing id3v2 version 3 metadata

Adds support for year (TYER) and day/month (TDAT) tags when writing
id3v2 version 3 metadata by splitting the "date" tag. The date tag
should have a format of "YYYY-MM-DD" or "YYYY".

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoReplace vendor string in Ogg and FLAC muxers.
Diego Biurrun [Sun, 30 Oct 2011 20:30:32 +0000 (21:30 +0100)]
Replace vendor string in Ogg and FLAC muxers.

8 years agoReplace some forgotten FFmpeg references by Libav.
Diego Biurrun [Sun, 30 Oct 2011 17:25:48 +0000 (18:25 +0100)]
Replace some forgotten FFmpeg references by Libav.

8 years agovsrc_testsrc: fix mailing list reference URL
Diego Biurrun [Sun, 30 Oct 2011 17:04:22 +0000 (18:04 +0100)]
vsrc_testsrc: fix mailing list reference URL

8 years agoReplace ffmpeg references with more accurate libav* references.
Diego Biurrun [Sun, 30 Oct 2011 17:02:42 +0000 (18:02 +0100)]
Replace ffmpeg references with more accurate libav* references.

8 years agoReplace outdated references to ffmpeg tool with avconv.
Diego Biurrun [Sun, 30 Oct 2011 16:59:01 +0000 (17:59 +0100)]
Replace outdated references to ffmpeg tool with avconv.

8 years agoRemove some stray unnecessary ffmpeg references.
Diego Biurrun [Sun, 30 Oct 2011 16:56:57 +0000 (17:56 +0100)]
Remove some stray unnecessary ffmpeg references.

8 years agovp3: remove some pointless comments
Diego Biurrun [Sun, 30 Oct 2011 16:52:11 +0000 (17:52 +0100)]
vp3: remove some pointless comments

8 years agoid3v2: fix type of ID3v2EMFunc.free()
Anton Khirnov [Wed, 2 Nov 2011 08:47:59 +0000 (09:47 +0100)]
id3v2: fix type of ID3v2EMFunc.free()

8 years agolavc: use designated initialisers for parsers.
Anton Khirnov [Wed, 2 Nov 2011 08:34:41 +0000 (09:34 +0100)]
lavc: use designated initialisers for parsers.

8 years agodsputil: fix typo (referece -> reference).
Clément Bœsch [Wed, 2 Nov 2011 06:54:44 +0000 (07:54 +0100)]
dsputil: fix typo (referece -> reference).

8 years agojpeg: rgb support
Michael Niedermayer [Wed, 2 Nov 2011 01:26:28 +0000 (02:26 +0100)]
jpeg: rgb support
Fixes Ticket297

Based on a patch by Carl
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agog726dec: add flush() function to reset state when seeking
Justin Ruggles [Fri, 28 Oct 2011 01:33:18 +0000 (21:33 -0400)]
g726dec: add flush() function to reset state when seeking

8 years agog726: don't pass index to g726_reset()
Justin Ruggles [Fri, 28 Oct 2011 01:31:24 +0000 (21:31 -0400)]
g726: don't pass index to g726_reset()

calculate it from c->code_size instead.

8 years agog726enc: add private option for setting code size directly.
Justin Ruggles [Fri, 28 Oct 2011 01:19:41 +0000 (21:19 -0400)]
g726enc: add private option for setting code size directly.

This is an easy alternative to setting bit_rate. This patch also selects the
closest bit_rate to the requested one rather than requiring an exact value.

8 years agog726: wrap the decoder functions with a CONFIG_ADPCM_G726_DECODER check
Justin Ruggles [Fri, 28 Oct 2011 00:51:40 +0000 (20:51 -0400)]
g726: wrap the decoder functions with a CONFIG_ADPCM_G726_DECODER check

8 years agog726: group the g726_encoder AVCodec with the other encoding functions
Justin Ruggles [Fri, 28 Oct 2011 00:50:41 +0000 (20:50 -0400)]
g726: group the g726_encoder AVCodec with the other encoding functions

8 years agog726: return AVERROR(EINVAL) instead of -1 for invalid channel count
Justin Ruggles [Fri, 28 Oct 2011 00:49:04 +0000 (20:49 -0400)]
g726: return AVERROR(EINVAL) instead of -1 for invalid channel count

8 years agog726enc: use av_assert0() for sample_rate validation
Justin Ruggles [Fri, 28 Oct 2011 00:47:08 +0000 (20:47 -0400)]
g726enc: use av_assert0() for sample_rate validation

This should never happen, but the check avoids a divide-by-zero.

8 years agog726: treat sample rates other than 8kHz as unofficial.
Justin Ruggles [Fri, 28 Oct 2011 00:45:24 +0000 (20:45 -0400)]
g726: treat sample rates other than 8kHz as unofficial.

8 years agog726dec: remove the sample_rate validation
Justin Ruggles [Fri, 28 Oct 2011 00:22:14 +0000 (20:22 -0400)]
g726dec: remove the sample_rate validation

8 years agog726: use bits_per_coded_sample instead of bitrate to determine mode
Justin Ruggles [Fri, 28 Oct 2011 00:16:45 +0000 (20:16 -0400)]
g726: use bits_per_coded_sample instead of bitrate to determine mode

This requires some workarounds in the WAV muxer and demuxer. We need to write
the correct bits_per_coded_sample and block_align in the muxer. In the
demuxer, we cannot rely on the bits_per_coded_sample value, so we use the bit
rate and sample rate to determine the value.

This avoids having the decoder rely on AVCodecContext.bit_rate, which is not
required to be set by the user for decoding according to our API.

8 years agog726: split the init function for the encoder and decoder
Justin Ruggles [Thu, 27 Oct 2011 22:13:47 +0000 (18:13 -0400)]
g726: split the init function for the encoder and decoder

This also allows for not having a decoder close function.

8 years agog726: pre-calculate the number of output samples.
Justin Ruggles [Thu, 27 Oct 2011 18:36:41 +0000 (14:36 -0400)]
g726: pre-calculate the number of output samples.

Allows for checking output buffer size and simplification of decoding loop.

8 years agog726: use int16_t instead of short
Justin Ruggles [Thu, 27 Oct 2011 18:31:42 +0000 (14:31 -0400)]
g726: use int16_t instead of short

8 years agoMerge remote-tracking branch 'qatar/master'
Michael Niedermayer [Tue, 1 Nov 2011 23:07:44 +0000 (00:07 +0100)]
Merge remote-tracking branch 'qatar/master'

* qatar/master:
  Enable w32threads automatically unless explicitly disabled

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agovf_transpose: remove pix_fmts which can currently not be supported.
Michael Niedermayer [Tue, 1 Nov 2011 22:45:28 +0000 (23:45 +0100)]
vf_transpose: remove pix_fmts which can currently not be supported.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agodirac: drop unused END_HORIZONTAL
Michael Niedermayer [Tue, 1 Nov 2011 20:56:27 +0000 (21:56 +0100)]
dirac: drop unused END_HORIZONTAL

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agodirac: fix horizontal_compose_dd97i_ssse3
Michael Niedermayer [Tue, 1 Nov 2011 20:55:37 +0000 (21:55 +0100)]
dirac: fix horizontal_compose_dd97i_ssse3

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agodirac: Fix mmx/sse haar wavelet compose
Michael Niedermayer [Tue, 1 Nov 2011 20:41:01 +0000 (21:41 +0100)]
dirac: Fix mmx/sse haar wavelet compose

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoMAINTAINERS: add myself as mailinglists maintainer
Lou Logan [Tue, 1 Nov 2011 18:02:52 +0000 (10:02 -0800)]
MAINTAINERS: add myself as mailinglists maintainer

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agogxf: add timecode information to metadata
Matthieu Bouron [Mon, 31 Oct 2011 18:54:31 +0000 (19:54 +0100)]
gxf: add timecode information to metadata

Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
8 years agoconfigure: fix arch x86_32
Michael Niedermayer [Tue, 1 Nov 2011 13:46:18 +0000 (14:46 +0100)]
configure: fix arch x86_32

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoEnable w32threads automatically unless explicitly disabled
Martin Storsjö [Tue, 1 Nov 2011 14:22:56 +0000 (16:22 +0200)]
Enable w32threads automatically unless explicitly disabled

Signed-off-by: Martin Storsjö <martin@martin.st>
8 years agoFix compilation of dct-test on ia32.
Carl Eugen Hoyos [Tue, 1 Nov 2011 14:47:15 +0000 (15:47 +0100)]
Fix compilation of dct-test on ia32.

8 years agoTwo extra samples are required before top of the in buffer rather than out
Vladimir Voroshilov [Tue, 1 Nov 2011 08:08:01 +0000 (15:08 +0700)]
Two extra samples are required before top of the in buffer rather than out

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agommsh: add return value for mmsh_open()
Michael Niedermayer [Tue, 1 Nov 2011 11:52:19 +0000 (12:52 +0100)]
mmsh: add return value for mmsh_open()

Found-by: cptspiff
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoMAINTAINERS: add libstagefright
Michael Niedermayer [Tue, 1 Nov 2011 01:56:25 +0000 (02:56 +0100)]
MAINTAINERS: add libstagefright

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoMAINTAINERS: update project server
Michael Niedermayer [Tue, 1 Nov 2011 11:57:51 +0000 (12:57 +0100)]
MAINTAINERS: update project server

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoMerge remote-tracking branch 'qatar/master'
Michael Niedermayer [Tue, 1 Nov 2011 01:02:20 +0000 (02:02 +0100)]
Merge remote-tracking branch 'qatar/master'

* qatar/master:
  cmdutils: Rename read_file to cmdutils_read_file

Conflicts:
ffmpeg.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agodirac: disable ff_horizontal_compose_haar0i_sse2() the function is not bitexact
Michael Niedermayer [Mon, 31 Oct 2011 23:43:05 +0000 (00:43 +0100)]
dirac: disable ff_horizontal_compose_haar0i_sse2() the function is not bitexact

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agommsh: do not try to seek to negative tiimestamps, it does not work
Michael Niedermayer [Mon, 31 Oct 2011 22:40:04 +0000 (23:40 +0100)]
mmsh: do not try to seek to negative tiimestamps, it does not work

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agomp3enc: avoid truncating id3v1 tags by one byte
Tobias Rapp [Mon, 31 Oct 2011 14:11:37 +0000 (15:11 +0100)]
mp3enc: avoid truncating id3v1 tags by one byte

Avoid writing the trailing null-byte for id3v1 tags if length reaches max length.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agommsh_read_seek: skip header
Michael Niedermayer [Mon, 31 Oct 2011 21:59:56 +0000 (22:59 +0100)]
mmsh_read_seek: skip header

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agommsh:_implement a minimal mmsh_seek() so the stream position can be found out.
Michael Niedermayer [Mon, 31 Oct 2011 21:43:38 +0000 (22:43 +0100)]
mmsh:_implement a minimal mmsh_seek() so the stream position can be found out.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoasfdec: Check packet_replic_size earlier
Michael Niedermayer [Mon, 31 Oct 2011 21:42:45 +0000 (22:42 +0100)]
asfdec: Check packet_replic_size earlier

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agommsh: mmsh_read_seek()
Michael Niedermayer [Mon, 31 Oct 2011 19:50:04 +0000 (20:50 +0100)]
mmsh: mmsh_read_seek()

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agommsh: split out mmsh_open_internal()
Michael Niedermayer [Mon, 31 Oct 2011 19:49:23 +0000 (20:49 +0100)]
mmsh: split out mmsh_open_internal()

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agommsh: move location storage into context
Michael Niedermayer [Mon, 31 Oct 2011 19:45:12 +0000 (20:45 +0100)]
mmsh: move location storage into context

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoRevert "Remove detection of mmsh protocol in ffplay."
Michael Niedermayer [Mon, 31 Oct 2011 17:04:39 +0000 (18:04 +0100)]
Revert "Remove detection of mmsh protocol in ffplay."

This reverts commit 2f642393381659f4a198d9c5a96896eebb23de18.

8 years agodirac: alignment for dirac_hpel_filter_v has been fixed
Michael Niedermayer [Mon, 31 Oct 2011 16:03:03 +0000 (17:03 +0100)]
dirac: alignment for dirac_hpel_filter_v has been fixed

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agodirac: alignment for add_dirac_obmc has been fixed
Michael Niedermayer [Mon, 31 Oct 2011 16:00:59 +0000 (17:00 +0100)]
dirac: alignment for  add_dirac_obmc has been fixed

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agodirac: alignment for add_rect_clamped has been fixed
Michael Niedermayer [Mon, 31 Oct 2011 16:00:40 +0000 (17:00 +0100)]
dirac: alignment for add_rect_clamped has been fixed

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agodirac: alignment for put_signed_rect_clamped has been fixed
Michael Niedermayer [Mon, 31 Oct 2011 16:00:03 +0000 (17:00 +0100)]
dirac: alignment for put_signed_rect_clamped has been fixed

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agodirac: alignment for dirac_hpel_filter_v() has been fixed
Michael Niedermayer [Mon, 31 Oct 2011 15:59:23 +0000 (16:59 +0100)]
dirac: alignment for dirac_hpel_filter_v() has been fixed

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agodirac: fix stride alignment
Michael Niedermayer [Mon, 31 Oct 2011 15:46:50 +0000 (16:46 +0100)]
dirac: fix stride alignment

This code was already added by Yuvi in c82cbea68273c6f08c4d0e94fc9fd50bfdea4e2b
It was subsequently lost somehow

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agodirac: fix segfault in horizontal_compose_haar
Michael Niedermayer [Mon, 31 Oct 2011 15:34:26 +0000 (16:34 +0100)]
dirac: fix segfault in horizontal_compose_haar

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agocmdutils: Rename read_file to cmdutils_read_file
Martin Storsjö [Mon, 7 Feb 2011 09:49:37 +0000 (11:49 +0200)]
cmdutils: Rename read_file to cmdutils_read_file

This symbol name clashes with a symbol in gnutls, if linking
statically to that library.

Signed-off-by: Martin Storsjö <martin@martin.st>
8 years agolavfi: bump micro after recent changes in the overlay filter
Stefano Sabatini [Mon, 31 Oct 2011 10:19:53 +0000 (11:19 +0100)]
lavfi: bump micro after recent changes in the overlay filter

8 years agovf_overlay: add support to alpha pre-multiplication in the RGBA path
Stefano Sabatini [Fri, 28 Oct 2011 22:29:25 +0000 (00:29 +0200)]
vf_overlay: add support to alpha pre-multiplication in the RGBA path

Based on the work of Mark Himsley <mark@mdsh.com>.

See thread:
Subject: [FFmpeg-devel] libavfilter: extending overlay filter
Date: Sun, 13 Mar 2011 14:18:42 +0000

8 years agovf_overlay: enable RGB path
Stefano Sabatini [Fri, 28 Oct 2011 22:10:43 +0000 (00:10 +0200)]
vf_overlay: enable RGB path

Add option rgb which forces the RGB path.