platform/upstream/libav.git
13 years agoswscale: fix JPEG-range YUV scaling artifacts. v0.7rc1
Michael Niedermayer [Mon, 13 Jun 2011 01:53:22 +0000 (21:53 -0400)]
swscale: fix JPEG-range YUV scaling artifacts.

YUV planes were marked as uint16_t, but they contained signed data.
Fixes issue 1108 and 675.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
13 years agobuild: move ALLFFLIBS to a more logical place
Mans Rullgard [Sun, 5 Jun 2011 21:18:06 +0000 (22:18 +0100)]
build: move ALLFFLIBS to a more logical place

Signed-off-by: Mans Rullgard <mans@mansr.com>
13 years agoARM: factor some repetitive code into macros
Mans Rullgard [Sun, 12 Jun 2011 23:33:19 +0000 (00:33 +0100)]
ARM: factor some repetitive code into macros

Signed-off-by: Mans Rullgard <mans@mansr.com>
13 years agoFix SVQ3 after adding 4:4:4 H.264 support
Jason Garrett-Glaser [Mon, 13 Jun 2011 17:21:46 +0000 (10:21 -0700)]
Fix SVQ3 after adding 4:4:4 H.264 support

13 years agoH.264: fix CODEC_FLAG_GRAY
Jason Garrett-Glaser [Thu, 9 Jun 2011 23:17:41 +0000 (16:17 -0700)]
H.264: fix CODEC_FLAG_GRAY

It was broken in 4:4:4, and still did chroma deblocking for no reason in 4:2:0.

13 years ago4:4:4 H.264 decoding support
Jason Garrett-Glaser [Fri, 3 Jun 2011 08:12:28 +0000 (01:12 -0700)]
4:4:4 H.264 decoding support

Note: this is 4:4:4 from the 2007 spec revision, not the previous (now deprecated) 4:4:4 mode in H.264.

13 years agoac3enc: fix allocation of floating point samples.
Justin Ruggles [Mon, 13 Jun 2011 21:44:50 +0000 (17:44 -0400)]
ac3enc: fix allocation of floating point samples.

sizeof(SampleType) is different for fixed and float encoders.

13 years agoutils: Drop pointless '#if 1' preprocessor directive.
Diego Biurrun [Fri, 10 Jun 2011 18:27:50 +0000 (20:27 +0200)]
utils: Drop pointless '#if 1' preprocessor directive.

13 years agoac3enc: remove empty ac3_float function that is never called
Justin Ruggles [Fri, 10 Jun 2011 19:17:55 +0000 (15:17 -0400)]
ac3enc: remove empty ac3_float function that is never called

13 years agoac3enc: split templated float vs. fixed functions into a separate file.
Justin Ruggles [Fri, 10 Jun 2011 18:57:19 +0000 (14:57 -0400)]
ac3enc: split templated float vs. fixed functions into a separate file.

Function pointers are used for templated functions instead of needlessly
duplicating many functions.

13 years agoac3enc: dynamically allocate AC3EncodeContext fields windowed_samples and mdct
Justin Ruggles [Fri, 10 Jun 2011 16:42:36 +0000 (12:42 -0400)]
ac3enc: dynamically allocate AC3EncodeContext fields windowed_samples and mdct

This will allow the same struct to be used for both the fixed and float ac3
encoders.

13 years agoac3enc: use function pointer to choose between AC-3 and E-AC-3 header output
Justin Ruggles [Fri, 10 Jun 2011 15:45:03 +0000 (11:45 -0400)]
ac3enc: use function pointer to choose between AC-3 and E-AC-3 header output
functions.

13 years agoRoll back 4:4:4 H.264 for now
Jason Garrett-Glaser [Mon, 13 Jun 2011 20:38:46 +0000 (13:38 -0700)]
Roll back 4:4:4 H.264 for now
Needs some ARM/PPC asm modifications.

13 years agoFix SVQ3 after adding 4:4:4 H.264 support
Jason Garrett-Glaser [Mon, 13 Jun 2011 17:21:46 +0000 (10:21 -0700)]
Fix SVQ3 after adding 4:4:4 H.264 support

13 years agoH.264: fix CODEC_FLAG_GRAY
Jason Garrett-Glaser [Thu, 9 Jun 2011 23:17:41 +0000 (16:17 -0700)]
H.264: fix CODEC_FLAG_GRAY

It was broken in 4:4:4, and still did chroma deblocking for no reason in 4:2:0.

13 years ago4:4:4 H.264 decoding support
Jason Garrett-Glaser [Fri, 3 Jun 2011 08:12:28 +0000 (01:12 -0700)]
4:4:4 H.264 decoding support

Note: this is 4:4:4 from the 2007 spec revision, not the previous (now deprecated) 4:4:4 mode in H.264.

13 years agoh264_parser: Fix whitespace after previous change.
Philip Langdale [Tue, 29 Mar 2011 04:43:23 +0000 (21:43 -0700)]
h264_parser: Fix whitespace after previous change.

Signed-off-by: Philip Langdale <philipl@overt.org>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agoh264_parser: Fix behaviour when PARSER_FLAG_COMPLETE_FRAMES is set.
Philip Langdale [Tue, 29 Mar 2011 04:42:02 +0000 (21:42 -0700)]
h264_parser: Fix behaviour when PARSER_FLAG_COMPLETE_FRAMES is set.

Currently, the parser is buggy and only processes the stream extradata
when the flag is set. This fixes it to actually inspect the frames.

Whitespce will be fixed in a separate change.

Signed-off-by: Philip Langdale <philipl@overt.org>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agowav: remove an invalid free().
Carl Eugen Hoyos [Sun, 3 Apr 2011 20:45:16 +0000 (22:45 +0200)]
wav: remove an invalid free().

Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agolavf: initialise reference_dts in av_estimate_timings_from_pts.
Michael Niedermayer [Tue, 8 Mar 2011 21:39:14 +0000 (22:39 +0100)]
lavf: initialise reference_dts in av_estimate_timings_from_pts.

Fixes issue2437.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agoh264: don't be so picky on decoding pps in extradata.
Michael Niedermayer [Tue, 1 Mar 2011 14:59:00 +0000 (15:59 +0100)]
h264: don't be so picky on decoding pps in extradata.

Fixes issue2517

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agoavcodec.h: add or elaborate on some documentation comments.
Wim Lewis [Sat, 2 Apr 2011 22:30:24 +0000 (15:30 -0700)]
avcodec.h: add or elaborate on some documentation comments.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agoh264: change a few comments into error messages
Michael Niedermayer [Sat, 2 Apr 2011 15:57:53 +0000 (17:57 +0200)]
h264: change a few comments into error messages

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agoac3dec: fix doxy-style for comment ("///>" should be "///<" instead).
Reimar Döffinger [Fri, 1 Apr 2011 17:42:31 +0000 (19:42 +0200)]
ac3dec: fix doxy-style for comment ("///>" should be "///<" instead).

Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agoimg2: add .dpx to the list of supported file extensions.
Peter Ross [Fri, 1 Apr 2011 12:11:24 +0000 (23:11 +1100)]
img2: add .dpx to the list of supported file extensions.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agoffv1: fix undefined behavior with insane widths.
Michael Niedermayer [Fri, 1 Apr 2011 10:46:36 +0000 (12:46 +0200)]
ffv1: fix undefined behavior with insane widths.

The new tables is large enough to prevent this together with our image size checks.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agoARM: jrevdct_arm: simplify stack usage
Mans Rullgard [Mon, 13 Jun 2011 08:33:20 +0000 (09:33 +0100)]
ARM: jrevdct_arm: simplify stack usage

Signed-off-by: Mans Rullgard <mans@mansr.com>
13 years agoARM: jrevdct_arm: use push/pop mnemonics
Mans Rullgard [Mon, 13 Jun 2011 08:30:42 +0000 (09:30 +0100)]
ARM: jrevdct_arm: use push/pop mnemonics

Use push/pop instead of stmdb/ldmia for stack operations.  This
is the preferred syntax.

Signed-off-by: Mans Rullgard <mans@mansr.com>
13 years agoARM: jrevdct_arm: misc cleanup
Mans Rullgard [Mon, 13 Jun 2011 08:24:27 +0000 (09:24 +0100)]
ARM: jrevdct_arm: misc cleanup

- use 'const' macro to define coeff table
- add missing endfunc
- remove superflous directives

Signed-off-by: Mans Rullgard <mans@mansr.com>
13 years agoARM: optimised mpadsp_apply_window_fixed
Mans Rullgard [Sat, 11 Jun 2011 21:53:32 +0000 (22:53 +0100)]
ARM: optimised mpadsp_apply_window_fixed

Signed-off-by: Mans Rullgard <mans@mansr.com>
13 years agoAdd some (important) changelog entries
Reinhard Tartler [Mon, 13 Jun 2011 06:56:54 +0000 (08:56 +0200)]
Add some (important) changelog entries

13 years agocmdutils: add missing NULL check in parse_options()
Stefano Sabatini [Sun, 12 Jun 2011 11:54:22 +0000 (13:54 +0200)]
cmdutils: add missing NULL check in parse_options()

Fix ffplay -i FILE, which was recently broken.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agox11grab: remove a memory allocation and the associated memcpy.
Sven C. Dack [Wed, 30 Mar 2011 15:02:29 +0000 (17:02 +0200)]
x11grab: remove a memory allocation and the associated memcpy.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agobuild: fix "make install" with documentation disabled
Mans Rullgard [Sat, 11 Jun 2011 18:54:23 +0000 (19:54 +0100)]
build: fix "make install" with documentation disabled

Signed-off-by: Mans Rullgard <mans@mansr.com>
13 years agobuild: simplify some conditional targets
Mans Rullgard [Tue, 7 Jun 2011 09:11:21 +0000 (10:11 +0100)]
build: simplify some conditional targets

Use intermediate targets instead of variables for conditional
parts of "all" and "install" targets.

Signed-off-by: Mans Rullgard <mans@mansr.com>
13 years agobitstream: Properly promote av_reverse values before shifting.
Alex Converse [Tue, 26 Apr 2011 16:08:26 +0000 (09:08 -0700)]
bitstream: Properly promote av_reverse values before shifting.

13 years agolibavutil/swscale: YUV444P10/YUV444P9 support.
Ronald S. Bultje [Thu, 9 Jun 2011 02:16:39 +0000 (19:16 -0700)]
libavutil/swscale: YUV444P10/YUV444P9 support.

Also add missing glue code for recently added YUV422P10 formats
to swscale.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
13 years agoH.264: Fix high bit depth explicit biweight
Jason Garrett-Glaser [Fri, 10 Jun 2011 01:20:19 +0000 (18:20 -0700)]
H.264: Fix high bit depth explicit biweight

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
13 years agoh264: Fix 10-bit H.264 x86 chroma v loopfilter asm.
Oskar Arvidsson [Fri, 10 Jun 2011 00:40:10 +0000 (02:40 +0200)]
h264: Fix 10-bit H.264 x86 chroma v loopfilter asm.

The tc variable was not splatted correctly.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
13 years agoReplace DEBUG_SEEK/DEBUG_SI + av_log combinations by av_dlog.
Diego Biurrun [Tue, 7 Jun 2011 11:18:12 +0000 (13:18 +0200)]
Replace DEBUG_SEEK/DEBUG_SI + av_log combinations by av_dlog.

13 years agoUpdate copyright year for ac3enc_opts_template.c.
Justin Ruggles [Fri, 10 Jun 2011 16:58:08 +0000 (12:58 -0400)]
Update copyright year for ac3enc_opts_template.c.

The code was originally committed to Libav on March 25, 2011.

13 years agoadts: Adjust frame size mask to follow the specification.
Kieran Kunhya [Wed, 8 Jun 2011 06:26:20 +0000 (01:26 -0500)]
adts: Adjust frame size mask to follow the specification.

This fixes ADTS detection for at least one sample.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
13 years agomovenc: Add RTP muxer/hinter options
Martin Storsjö [Sat, 21 May 2011 11:58:43 +0000 (14:58 +0300)]
movenc: Add RTP muxer/hinter options

Signed-off-by: Martin Storsjö <martin@martin.st>
13 years agomovenc: Pass the RTP AVFormatContext to the SDP generation
Martin Storsjö [Wed, 18 May 2011 12:41:38 +0000 (15:41 +0300)]
movenc: Pass the RTP AVFormatContext to the SDP generation

Signed-off-by: Martin Storsjö <martin@martin.st>
13 years agortspenc: Add RTP muxer options
Martin Storsjö [Sat, 21 May 2011 12:03:48 +0000 (15:03 +0300)]
rtspenc: Add RTP muxer options

Signed-off-by: Martin Storsjö <martin@martin.st>
13 years agortspenc: Add an AVClass for setting muxer specific options
Martin Storsjö [Sat, 21 May 2011 12:03:35 +0000 (15:03 +0300)]
rtspenc: Add an AVClass for setting muxer specific options

Signed-off-by: Martin Storsjö <martin@martin.st>
13 years agortpenc_chain: Pass the rtpflags options through to the chained muxer
Martin Storsjö [Wed, 18 May 2011 12:42:53 +0000 (15:42 +0300)]
rtpenc_chain: Pass the rtpflags options through to the chained muxer

Signed-off-by: Martin Storsjö <martin@martin.st>
13 years agortpenc: Declare the rtp flags private AVOptions in rtpenc.h
Martin Storsjö [Fri, 27 May 2011 19:38:36 +0000 (22:38 +0300)]
rtpenc: Declare the rtp flags private AVOptions in rtpenc.h

This allows other muxers that chain a RTP muxer to declare
the same options easily.

Signed-off-by: Martin Storsjö <martin@martin.st>
13 years agosdp: Reindent after the previous commit
Martin Storsjö [Wed, 18 May 2011 12:48:20 +0000 (15:48 +0300)]
sdp: Reindent after the previous commit

Signed-off-by: Martin Storsjö <martin@martin.st>
13 years agortpenc: MP4A-LATM payload support
Juan Carlos Rodriguez [Wed, 18 May 2011 12:00:03 +0000 (15:00 +0300)]
rtpenc: MP4A-LATM payload support

This is enabled with an AVOption on the RTP muxer. The SDP
generator looks for a latm flag in the rtpflags field.

Signed-off-by: Martin Storsjö <martin@martin.st>
13 years agoavoptions: Add an av_opt_flag_is_set function for inspecting flag fields
Martin Storsjö [Fri, 27 May 2011 19:21:40 +0000 (22:21 +0300)]
avoptions: Add an av_opt_flag_is_set function for inspecting flag fields

Signed-off-by: Martin Storsjö <martin@martin.st>
13 years agosdp: Allow passing an AVFormatContext to the SDP generation
Martin Storsjö [Wed, 18 May 2011 12:41:06 +0000 (15:41 +0300)]
sdp: Allow passing an AVFormatContext to the SDP generation

Options from the AVFormatContext can be read for modifying
the generated SDP.

Signed-off-by: Martin Storsjö <martin@martin.st>
13 years agomov: Fix wrong timestamp generation for fragmented movies that have time offset cause...
Yusuke Nakamura [Fri, 13 May 2011 14:46:48 +0000 (23:46 +0900)]
mov: Fix wrong timestamp generation for fragmented movies that have time offset caused by the first edit list entry.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agompeg12: more advanced ffmpeg mpeg2 aspect guessing code.
Michael Niedermayer [Fri, 25 Mar 2011 00:13:08 +0000 (01:13 +0100)]
mpeg12: more advanced ffmpeg mpeg2 aspect guessing code.

Fixes issue1613, 621, 562 simultaneously

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agoswscale: split YUYV output out of yuv2packed[12X]_c().
Ronald S. Bultje [Thu, 9 Jun 2011 22:13:53 +0000 (18:13 -0400)]
swscale: split YUYV output out of yuv2packed[12X]_c().

This is part of the Great Evil Plan to simplify swscale.

13 years agolavf: make compute_pkt_fields2() return meaningful error values
Stefano Sabatini [Tue, 15 Mar 2011 11:34:55 +0000 (12:34 +0100)]
lavf: make compute_pkt_fields2() return meaningful error values

Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agomatroskadec: set timestamps for RealAudio packets.
Reimar Döffinger [Sat, 26 Feb 2011 11:52:01 +0000 (12:52 +0100)]
matroskadec: set timestamps for RealAudio packets.

Improves seeking in ffplay with
http://samples.mplayerhq.hu/Matroska/RA_missing_timestamps.mkv

Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agointelh263dec: aspect ratio processing fix.
Anton Khirnov [Thu, 9 Jun 2011 20:36:31 +0000 (22:36 +0200)]
intelh263dec: aspect ratio processing fix.

patch submitted by xvid_fan freenet de

13 years agointelh263dec: fix "Strict H.263 compliance" file playback
Anton Khirnov [Thu, 9 Jun 2011 20:29:40 +0000 (22:29 +0200)]
intelh263dec: fix "Strict H.263 compliance"  file playback

fixes issue2449

patch submitted by xvid_fan freenet de

13 years agooss,sndio: simplify by using FFMIN.
Reimar Döffinger [Tue, 29 Mar 2011 18:18:45 +0000 (20:18 +0200)]
oss,sndio: simplify by using FFMIN.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agoswscale: extract monowhite/black output from yuv2packed[12X]_c().
Ronald S. Bultje [Thu, 9 Jun 2011 20:25:21 +0000 (16:25 -0400)]
swscale: extract monowhite/black output from yuv2packed[12X]_c().

This is part of the Great Evil Plan to simplify swscale.

13 years agoswscale: de-macro'ify RGB15/16/32 input functions.
Ronald S. Bultje [Thu, 9 Jun 2011 17:59:21 +0000 (13:59 -0400)]
swscale: de-macro'ify RGB15/16/32 input functions.

Inline functions are easier to read, maintain, modify and test,
which justifies the slightly increased source size. This patch
also adds support for non-native endianness RGB15/16 and fixes
isSupportedOutput() to no longer claim that we support writing
non-native RGB565/555/444.

13 years agoswscale: rearrange code.
Ronald S. Bultje [Thu, 9 Jun 2011 14:00:09 +0000 (10:00 -0400)]
swscale: rearrange code.

This way the code in the file is less cluttered all-over-the-
place.

13 years agomovdec: Add support for the 'wfex' atom.
Alex Converse [Mon, 6 Jun 2011 17:59:46 +0000 (10:59 -0700)]
movdec: Add support for the 'wfex' atom.

The 'wfex' is just a Microsoft WaveFormatEx struct.

13 years agoffmpeg.c: Add a necessary const qualifier
Alex Converse [Thu, 9 Jun 2011 16:52:47 +0000 (09:52 -0700)]
ffmpeg.c: Add a necessary const qualifier

13 years agoriff: Fix potential memleak.
Alex Converse [Thu, 9 Jun 2011 16:50:38 +0000 (09:50 -0700)]
riff: Fix potential memleak.

Make ff_get_wav_header() free existing extradata before allocing a new
buffer.

13 years agoswscale: change 48bit RGB input macros to inline functions.
Ronald S. Bultje [Wed, 8 Jun 2011 19:12:32 +0000 (15:12 -0400)]
swscale: change 48bit RGB input macros to inline functions.

Inline functions are slightly larger in source code, but
are easier to handle in source code editors. The binary code
generated is the same.

13 years agoswscale: change 9/10bit YUV input macros to inline functions.
Ronald S. Bultje [Wed, 8 Jun 2011 18:50:49 +0000 (14:50 -0400)]
swscale: change 9/10bit YUV input macros to inline functions.

Inline functions are slightly larger in source code, but
are easier to handle in source code editors. The binary code
generated is the same.

13 years agoswscale: extract gray16 output functions from yuv2packed[12X]().
Ronald S. Bultje [Wed, 8 Jun 2011 17:29:09 +0000 (13:29 -0400)]
swscale: extract gray16 output functions from yuv2packed[12X]().

This is part of the Great Evil Plan to simplify swscale.

13 years agoswscale: use standard clipping functions.
Ronald S. Bultje [Tue, 7 Jun 2011 15:22:54 +0000 (11:22 -0400)]
swscale: use standard clipping functions.

This generates better code on some non-x86 architectures.

13 years agoswscale: merge macros that are used only once.
Ronald S. Bultje [Tue, 7 Jun 2011 15:06:44 +0000 (11:06 -0400)]
swscale: merge macros that are used only once.

This reduces source code size without affecting the binary.

13 years agoswscale: fix function declarations in swscale.c.
Ronald S. Bultje [Wed, 8 Jun 2011 16:31:11 +0000 (12:31 -0400)]
swscale: fix function declarations in swscale.c.

Remove inline keyword from functions that are never inlined.
Use av_always_inline for functions that should be force-inlined
for performance reasons. Use av_cold for init functions.

13 years agoswscale: fix function declaration keywords in x86/swscale_template.c.
Ronald S. Bultje [Wed, 8 Jun 2011 16:12:50 +0000 (12:12 -0400)]
swscale: fix function declaration keywords in x86/swscale_template.c.

Remove inline keyword for functions that are only called through
their function pointers (and thus cannot be inlined); add av_cold
keyword to init function, and use av_always_inline instead of
inline for functions that must be inlined for performance reasons.

13 years agocrypto: Use av_freep instead of av_free
Etienne Buira [Wed, 8 Jun 2011 00:20:53 +0000 (02:20 +0200)]
crypto: Use av_freep instead of av_free

This fixes a potential double free.

Signed-off-by: Martin Storsjö <martin@martin.st>
13 years agolavf: don't try to free private options if priv_data is NULL.
Anton Khirnov [Mon, 6 Jun 2011 11:14:05 +0000 (13:14 +0200)]
lavf: don't try to free private options if priv_data is NULL.

This might happen if there was an error before priv_data was allocated
and result in segfault.

13 years agoswscale: fix types of assembly arguments.
Ronald S. Bultje [Wed, 8 Jun 2011 15:39:26 +0000 (11:39 -0400)]
swscale: fix types of assembly arguments.

This prevents the following compiler warnings: "warning:
initialization from incompatible pointer type". Since the
variables are only ever used in inline assembly, their type
is actually irrelevant (so the part where it was wrong did
not invoke any buggy behaviour).

13 years agoswscale: move two macros that are only used once into caller.
Ronald S. Bultje [Wed, 8 Jun 2011 15:33:46 +0000 (11:33 -0400)]
swscale: move two macros that are only used once into caller.

This way, they look like regular code, which is easier to
understand.

13 years agoswscale: remove unused function.
Ronald S. Bultje [Wed, 8 Jun 2011 15:03:11 +0000 (11:03 -0400)]
swscale: remove unused function.

Use of this wrapper was removed in a previous patch, but I
forgot to actually remove the function itself.

13 years agooptions: Add missing braces around struct initializer.
Diego Biurrun [Wed, 8 Jun 2011 12:59:47 +0000 (14:59 +0200)]
options: Add missing braces around struct initializer.

This fixes the warning:
libavformat/options.c:62:1: warning: missing braces around initializer [-Wmissing-braces]

13 years agomov: Remove leftover crufty debug statement with references to a local file.
Diego Biurrun [Sun, 5 Jun 2011 14:10:17 +0000 (16:10 +0200)]
mov: Remove leftover crufty debug statement with references to a local file.

13 years agodvbsubdec: Fix compilation of debug code.
Diego Biurrun [Mon, 6 Jun 2011 12:10:22 +0000 (14:10 +0200)]
dvbsubdec: Fix compilation of debug code.

13 years agoRemove all uses of now deprecated metadata functions.
Anton Khirnov [Sun, 22 May 2011 10:46:29 +0000 (12:46 +0200)]
Remove all uses of now deprecated metadata functions.

13 years agoMove metadata API from lavf to lavu.
Anton Khirnov [Sun, 22 May 2011 10:45:00 +0000 (12:45 +0200)]
Move metadata API from lavf to lavu.

Rename it to AVDictionary, since it will be used as such.  Tags
documentation and metadata conversion API is lavf-specific, so remains
there.

13 years agoRemove some non-compiling debug messages.
Diego Biurrun [Sun, 5 Jun 2011 13:53:34 +0000 (15:53 +0200)]
Remove some non-compiling debug messages.

13 years agoffplay: Fix non-compiling debug printf and replace it by av_dlog.
Diego Biurrun [Tue, 7 Jun 2011 22:41:12 +0000 (00:41 +0200)]
ffplay: Fix non-compiling debug printf and replace it by av_dlog.

13 years agoH264: x86 predict init cosmetics.
Daniel Kang [Tue, 7 Jun 2011 14:31:51 +0000 (10:31 -0400)]
H264: x86 predict init cosmetics.

Change indentation and whitespace; also move HAVE_YASM blocks.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
13 years agoac3enc: Fix linking of AC-3 encoder without the E-AC-3 encoder.
Diego Biurrun [Tue, 7 Jun 2011 22:06:19 +0000 (00:06 +0200)]
ac3enc: Fix linking of AC-3 encoder without the E-AC-3 encoder.

The AC-3 encoder unconditionally references some symbols from the E-AC-3
encoder; make those references conditional to fix linking.

13 years agoMove E-AC-3 encoder functions to a separate eac3enc.c file.
Justin Ruggles [Tue, 7 Jun 2011 16:47:09 +0000 (12:47 -0400)]
Move E-AC-3 encoder functions to a separate eac3enc.c file.

13 years agoac3enc: remove convenience macro, #define DEBUG
Justin Ruggles [Tue, 7 Jun 2011 15:50:38 +0000 (11:50 -0400)]
ac3enc: remove convenience macro, #define DEBUG

13 years agoac3enc: remove unused #define
Justin Ruggles [Tue, 7 Jun 2011 14:53:44 +0000 (10:53 -0400)]
ac3enc: remove unused #define

13 years agovc1: re-initialize tables after width/height change.
John Stebbins [Tue, 7 Jun 2011 17:53:31 +0000 (19:53 +0200)]
vc1: re-initialize tables after width/height change.

read_sequence_header can change width/height; therefore, re-initialize
all tables if width/height changed

Signed-off-by: Anton Khirnov <anton@khirnov.net>
13 years agoAPIchanges: fill-in git commit hash for av_get_bytes_per_sample() addition
Justin Ruggles [Tue, 7 Jun 2011 17:04:06 +0000 (13:04 -0400)]
APIchanges: fill-in git commit hash for av_get_bytes_per_sample() addition

13 years agosamplefmt: add av_get_bytes_per_sample()
Stefano Sabatini [Sun, 5 Jun 2011 23:14:50 +0000 (01:14 +0200)]
samplefmt: add av_get_bytes_per_sample()

Deprecate av_get_bits_per_sample_fmt(), which was a misnamed function.

For the moment we don't have sample formats with a non-integer number
of bytes, in that case we may need to create a new
av_get_bits_per_sample() function. In the meanwhile we prefer to adopt
this variant, since avoids divisions by 8 all over the place.

13 years agoiirfilter: fix biquad filter coefficients.
Justin Ruggles [Mon, 6 Jun 2011 21:10:11 +0000 (17:10 -0400)]
iirfilter: fix biquad filter coefficients.

The current filter implementation should only have the cx coefficients
divided by gain in order to give the correct output scale.

13 years agoswscale: remove duplicate conversion routine in swScale().
Ronald S. Bultje [Mon, 6 Jun 2011 02:54:30 +0000 (22:54 -0400)]
swscale: remove duplicate conversion routine in swScale().

13 years agoswscale: add yuv2planar/packed function typedefs.
Ronald S. Bultje [Mon, 6 Jun 2011 02:31:11 +0000 (22:31 -0400)]
swscale: add yuv2planar/packed function typedefs.

13 years agoswscale: integrate yuv2nv12X_C into yuv2yuvX() function pointers.
Ronald S. Bultje [Mon, 6 Jun 2011 02:27:45 +0000 (22:27 -0400)]
swscale: integrate yuv2nv12X_C into yuv2yuvX() function pointers.

13 years agoswscale: reindent x86 init code.
Ronald S. Bultje [Mon, 6 Jun 2011 01:48:51 +0000 (21:48 -0400)]
swscale: reindent x86 init code.

13 years agoswscale: extract SWS_FULL_CHR_H_INT conditional into init code.
Ronald S. Bultje [Mon, 6 Jun 2011 01:46:16 +0000 (21:46 -0400)]
swscale: extract SWS_FULL_CHR_H_INT conditional into init code.

13 years agoswscale: cosmetics.
Ronald S. Bultje [Mon, 6 Jun 2011 01:40:04 +0000 (21:40 -0400)]
swscale: cosmetics.