virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2026-06-19 19:03:00 +02:00
Code Issues Releases Activity
19,380 Commits 39 Branches 425 Tags
a00f440f378715e527d34aa844cc509252a10f34
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Ramiro Polla 44be8d4278 mingw32: FFmpeg now passes -fno-common automatically. 
Originally committed as revision 18510 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-04-14 15:25:11 +00:00
Ramiro Polla 1894e741f5 win32, dos: Work around a bug in the GNU toolchain. 
The bug[1] is exposed when gcc decides to put some data in a common
section (i.e. data that will be used by more than one object). It will
suggest the data to be aligned, but binutils >= 2.19 will fail to
properly align it. Thus if the data requires any alignment greater
than 4, the application will crash.
The workaround prevents gcc from putting data in a common section,
instead putting it in a properly aligned section.

[1] http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37216

Originally committed as revision 18501 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-04-13 22:54:21 +00:00
Ramiro Polla e90f5b5ab1 configure: Add --enable-runtime-cpudetect 
Originally committed as revision 18380 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-04-08 20:26:18 +00:00
Ramiro Polla 10a7d21672 Rename RUNTIME_CPUDETECT to CONFIG_RUNTIME_CPUDETECT and always define it. 
Originally committed as revision 29154 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-04-08 20:21:21 +00:00
Ramiro Polla 75428fa447 mlpdec: Filters and matrices may change only once per substream per access unit. 
Originally committed as revision 18338 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-04-06 00:58:03 +00:00
Ramiro Polla 89e39be3b6 mlpdec: There must be no extraword for MLP. 
Originally committed as revision 18337 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-04-05 20:54:19 +00:00
Ramiro Polla cc9c512638 mlpdec: Validate non-restart bit from the substream header. 
Originally committed as revision 18336 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-04-05 20:46:53 +00:00
Ramiro Polla 0091d8a164 mlpdec: matrix_out_ch must not be greater than max_matrix_channel, and not max_channel. 
Originally committed as revision 18335 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-04-05 20:11:40 +00:00
Ramiro Polla dd4090256c swscale: Remove X86 commented out code. 
Originally committed as revision 29149 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-04-05 19:17:53 +00:00
Ramiro Polla 40fa514033 swscale: Use function pointers for swScale functions. 
Originally committed as revision 29145 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-04-04 22:45:02 +00:00
Ramiro Polla 14d5c18c98 swscale: {}-related cosmetics. 
Originally committed as revision 29144 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-04-04 22:43:24 +00:00
Ramiro Polla 7ac4032784 swscale: Add const to some swScale functions' parameters. 
Originally committed as revision 29143 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-04-04 22:41:46 +00:00
Ramiro Polla 95b5770bc7 swscale: Remove mmx2 params from h[yc]scale(). 
Originally committed as revision 29129 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-04-02 18:34:11 +00:00
Ramiro Polla 392b656764 swscale: Split h[yc]scale_fast() into their own functions. 
Originally committed as revision 29128 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-04-02 18:32:49 +00:00
Ramiro Polla 5b7c7dd329 swscale: Execute sfence and emms depending on runtime flags. 
Originally committed as revision 29127 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-04-02 18:31:22 +00:00
Ramiro Polla ab79fa44f1 mlpdec: Simplify check for substream_parity_present. 
Originally committed as revision 18240 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-30 03:46:52 +00:00
Ramiro Polla 01aaf09274 mlpdec: Simplify no restart header seen error. 
Originally committed as revision 18239 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-30 03:44:20 +00:00
Ramiro Polla d7952be311 mlpdec: Simplify substream length mismatch error. 
Originally committed as revision 18238 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-30 03:42:40 +00:00
Ramiro Polla d544dcdf38 mlpdec: Prettify substream parity check. 
Originally committed as revision 18237 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-30 03:37:19 +00:00
Ramiro Polla 7b18e13a7c mlpdec: Cleaner and better termination word check. 
Originally committed as revision 18236 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-30 03:34:15 +00:00
Ramiro Polla 125cf771f6 mlpdec: More validation for read_channel_params() 
Originally committed as revision 18235 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-30 03:20:01 +00:00
Ramiro Polla 06e34be88f mlpdec: whitespace cosmetics. 
Originally committed as revision 18234 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-30 03:12:39 +00:00
Ramiro Polla a44b9f9574 mlpdec: Simplify &foo[0] to foo and use index (which has just been initialized 
to MAX_BLOCKSIZE) instead of MAX_BLOCKSIZE, so both copies to/from state data
look alike.

Originally committed as revision 18233 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-30 03:08:55 +00:00
Ramiro Polla e71365f427 mlpdec: Split filter_state_buffer into [fi]irbuf and fp to [fi]ir. 
Originally committed as revision 18232 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-30 03:05:38 +00:00
Ramiro Polla 0ae59bb01a mlpdec: Unroll copying filter state data and filtering for the two filters. 
Originally committed as revision 18231 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-30 02:59:45 +00:00
Ramiro Polla 0c5670a0e5 mlpdec: Max filter orders for FIR and IIR are 8 and 4 respectively. 
Originally committed as revision 18230 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-30 02:54:19 +00:00
Ramiro Polla 4b51fb1ce0 mingw32: Restore creation of import libraries. They are still not being 
installed, but they're good for creating -dev packages.

Originally committed as revision 18214 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-29 03:42:32 +00:00
Ramiro Polla fa937faba4 mlpdec: Remove few random dprintf()s. 
Originally committed as revision 18212 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-28 00:53:52 +00:00
Ramiro Polla b864098c16 mlpdec: Check for blocksize in proper range. 
Originally committed as revision 18211 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-27 23:55:38 +00:00
Ramiro Polla 8e9ee714fc mlpdec: quant_step_size can be any value from 0 to 0xF. 
Originally committed as revision 18210 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-27 23:48:08 +00:00
Ramiro Polla 6fae7cbd7f mlpdec: output_shift can be any value from -8 to 7. 
Originally committed as revision 18209 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-27 23:47:18 +00:00
Ramiro Polla af04802640 mlpdec: output_shift is signed 
Originally committed as revision 18208 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-27 23:45:52 +00:00
Ramiro Polla 43ee5fe035 mlpdec: Split read_channel_params() into its own function. 
Originally committed as revision 18207 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-27 23:42:22 +00:00
Ramiro Polla f8e6293bde mlpdec: Split read_matrix_params() into its own function. 
Originally committed as revision 18206 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-27 23:32:32 +00:00
Ramiro Polla 17a86e87ab Add myself as mlp* maintainer. 
Originally committed as revision 18205 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-27 23:27:16 +00:00
Ramiro Polla d5e9bc0636 Move bfin specific code to its subdir. 
Originally committed as revision 29079 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-03-27 22:56:56 +00:00
Ramiro Polla d4efacff64 Use intptr_t when casting pointers to int. 
Originally committed as revision 18192 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-26 01:34:02 +00:00
Ramiro Polla befa8e665c Move yuv2rgb code to subdirs. 
Originally committed as revision 29063 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-03-26 01:30:10 +00:00
Ramiro Polla b2cfd7e360 Create swscale subdirs for out-of-tree builds. 
Originally committed as revision 18191 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-26 01:29:20 +00:00
Ramiro Polla 1bbeb23356 Clear some more variables that are used on subdir Makefiles. 
Originally committed as revision 18180 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-25 00:50:56 +00:00
Ramiro Polla 9731e7f13e mlp, truehd: support non 1:1 channel mapping. 
Originally committed as revision 18074 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-20 13:07:09 +00:00
Ramiro Polla cbf3cf19f3 Support "next parameter flags present" flag. 
Originally committed as revision 18073 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-20 13:04:14 +00:00
Ramiro Polla 932cee5d67 truehd: support up to 3 substreams. 
Originally committed as revision 18072 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-20 13:02:15 +00:00
Ramiro Polla dd35beb29b Allocate executable memory with VirtualAlloc() in Windows. 
Originally committed as revision 29006 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-03-19 21:52:44 +00:00
Ramiro Polla 7c5bfe7add Check for VirtualAlloc() so it may be used to allocate executable memory in 
Windows.

Originally committed as revision 18050 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-19 21:52:01 +00:00
Ramiro Polla 16a3717f8d Drop unnecessary cast and cosmetically align. 
Originally committed as revision 29005 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-03-19 21:50:10 +00:00
Ramiro Polla 23d9cc4563 Support raw TrueHD files 
Originally committed as revision 18049 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-19 21:46:56 +00:00
Ramiro Polla 9ba4821d93 Split TrueHD decoder from MLP 
Originally committed as revision 18045 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-19 21:23:39 +00:00
Ramiro Polla 9a034dde8b swscale-example: use LFG instead of random() 
Originally committed as revision 28993 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-03-18 17:12:44 +00:00
Ramiro Polla d0ce212a7a Do not assume long is same width as x86 register. 
Originally committed as revision 28992 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
 2009-03-18 17:07:30 +00:00
Ramiro Polla e05c8d0682 use intptr_t to cast pointers to int in codecs maintained by benjamin larsson 
Originally committed as revision 18027 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-17 20:47:31 +00:00
Ramiro Polla 06abe7f5b4 mingw32: call lib.exe instead of lib to facilitate running it through wine. 
Originally committed as revision 17995 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-03-15 23:15:57 +00:00
Ramiro Polla da2dc39e74 Revert part of r16593. 
network.h should only be included when CONFIG_NETWORK is set.

Originally committed as revision 17009 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-02-05 18:51:29 +00:00
Ramiro Polla 2db5da97e2 Do not misuse unsigned long to store pointers. 
Originally committed as revision 16986 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-02-04 05:56:39 +00:00
Ramiro Polla f7ad5aa2e8 indentation 
Originally committed as revision 16828 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-01-27 20:06:13 +00:00
Ramiro Polla b1549d6113 mingw: don't check for minimum mingw runtime and w32api versions for mingw-w64 
there is no vfw in mingw-64, and it has different (not yet stable) version
defines

Originally committed as revision 16827 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-01-27 20:03:22 +00:00
Ramiro Polla d52481b9ce Pass subarch instead of arch to YASMFLAGS 
Originally committed as revision 16560 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-01-11 23:15:13 +00:00
Ramiro Polla b507ebd179 Use setmode() if it exists in <io.h>, and not based on O_BINARY. 
Originally committed as revision 16553 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-01-11 22:05:43 +00:00
Ramiro Polla 1bb04d5a44 configure: allow to disable sse code. 
Based on patch by Michael Kostylev <mik at it-1 dot ru>

Originally committed as revision 16490 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-01-07 23:38:54 +00:00
Ramiro Polla 1f91cdce0b Use posix_memalign() if available. 
Originally committed as revision 16488 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2009-01-07 23:36:34 +00:00
Ramiro Polla 2014b51cae documentation: update documentation for minimal w32api version. 
Originally committed as revision 16030 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-07 12:14:44 +00:00
Ramiro Polla bc4b722143 mingw: update w32api to 3.13 
Originally committed as revision 16029 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-07 05:17:57 +00:00
Ramiro Polla c687643c68 mlp: cosmetics: remove {}s over one-lined if. 
Originally committed as revision 16027 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-07 00:58:41 +00:00
Ramiro Polla ee1a8f6290 mlp: support bit-depths greater than 16 by default. 
Originally committed as revision 16026 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-12-07 00:57:26 +00:00
Ramiro Polla a4fa192c81 configure: vertical alignment cosmetics. 
Originally committed as revision 15958 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-11-30 00:20:19 +00:00
Ramiro Polla 632948bd5a configure: Use check_lib2 for bzlib. 
Functions are exported as WINAPI in win32, so it is not picked up by check_lib.

Originally committed as revision 15957 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-11-30 00:19:16 +00:00
Ramiro Polla 3ddb448d0a Support absolute DOS paths for preset parameters 
Originally committed as revision 15624 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-10-15 14:12:59 +00:00
Ramiro Polla f2e5786771 nellymoserenc: flush PutBitContext after use. 
Originally committed as revision 15596 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-10-11 14:52:00 +00:00
Ramiro Polla bc574408dc Only special-case absolute DOS paths on systems that support them. 
Originally committed as revision 15594 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-10-10 16:59:37 +00:00
Ramiro Polla 3defa81fc4 MinGW runtime 3.15 has a wrapper for usleep(). 
Originally committed as revision 15522 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-10-02 17:43:27 +00:00
Ramiro Polla 126cadc723 Specify minimum version for some MinGW packages. 
Originally committed as revision 15521 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-10-02 17:39:30 +00:00
Ramiro Polla 4a0266a0ec Spelling... 
Originally committed as revision 15327 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-14 22:15:11 +00:00
Ramiro Polla ab5ce4aeae Update link to FFmpeg Windows Help website. 
Originally committed as revision 15312 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-09-13 18:50:53 +00:00
Ramiro Polla cc940caf4d alacenc: Use user-specified min and max prediction order. 
Originally committed as revision 14844 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-19 16:12:05 +00:00
Ramiro Polla 0d962ecb70 Import more OKed parts of ALAC encoder from GSoC repo. 
Originally committed as revision 14843 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-19 16:07:58 +00:00
Ramiro Polla ca04826627 Import more ok'd parts of ALAC encoder from GSoC repo. 
Originally committed as revision 14820 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-17 22:47:40 +00:00
Ramiro Polla 83e92ab6b8 alacenc: NULL_IF_CONFIG_SMALL long_name. 
Originally committed as revision 14801 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-17 12:25:01 +00:00
Ramiro Polla 6b93147684 Import ok'd parts of ALAC encoder from GSoC repo. 
Originally committed as revision 14798 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-17 04:36:06 +00:00
Ramiro Polla e9e3c9801e lpc: cosmetics: vertically align declarations and definitions. 
Originally committed as revision 14792 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-16 17:28:29 +00:00
Ramiro Polla c50da3ad6c flacenc, lpc: Move LPC code from flacenc.c to new lpc.[ch] files. 
Originally committed as revision 14790 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-16 17:18:20 +00:00
Ramiro Polla 51c796d012 flacenc: Allow more flexible shift calculation in LPC. 
Originally committed as revision 14789 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-16 17:03:57 +00:00
Ramiro Polla 1e8bd10c57 mlp: Define End-of-Stream code in common header file and use it in decoder. 
Originally committed as revision 14763 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-14 19:38:39 +00:00
Ramiro Polla a7cc783d71 mlp: split simple inline function that xors 4 bytes into one. 
Originally committed as revision 14747 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-14 03:58:05 +00:00
Ramiro Polla 06cddea060 mlp: Cosmetics: Close comments in an extra line and 
remove empty lines between doxy comments and function declarations.

Originally committed as revision 14735 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-13 19:43:18 +00:00
Ramiro Polla ce15710f55 mlp: Split common code from parser and decoder to be used by encoder. 
Originally committed as revision 14733 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-13 18:47:03 +00:00
Ramiro Polla ee5b34d56e mlp_parser: Initialize crc data in its own function. 
Originally committed as revision 14722 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-13 02:44:38 +00:00
Ramiro Polla 5dae0dd22c mlpdec: Put some doxy comments to the right of vars instead of interleaved. 
Originally committed as revision 14721 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-13 01:41:06 +00:00
Ramiro Polla f53acb7b17 mlpdec: Split channel parameters from context into their own struct. 
Originally committed as revision 14720 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-13 01:36:01 +00:00
Ramiro Polla db01fa131b mlpdec: whitespace and line-breaking cosmetics. 
Originally committed as revision 14717 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-12 18:05:58 +00:00
Ramiro Polla d6aa052dc5 mlpdec: Split filter parameters from context into their own struct. 
Originally committed as revision 14716 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-08-12 17:53:59 +00:00
Ramiro Polla eb2cd99c73 mlpdec: Check for bits left before each read of End-of-Stream indicator and 
Substream parity check independently. This allows decoding of streams that
have EOS but don't have Substream parity check.

Originally committed as revision 14206 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-07-13 14:59:39 +00:00
Ramiro Polla 6647ab80e3 bitstream: move put_sbits() from flacenc.c to bitstream.h and use it 
throughout libavcodec.

Originally committed as revision 14204 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-07-13 14:27:48 +00:00
Ramiro Polla d08e3e91af Prevent users from setting -vol and -acodec copy together. 
They are incompatible since the frames are not decoded, so they can't be preprocessed.

Originally committed as revision 14178 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-07-12 11:43:21 +00:00
Ramiro Polla 28ce183917 Add a checklist about new codecs or formats to the documentation. 
Originally committed as revision 14159 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-07-11 00:59:27 +00:00
Ramiro Polla 822cc8f71c Update Changelog and documentation for MVI demuxer and decoder. 
Originally committed as revision 14158 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-07-10 18:01:21 +00:00
Ramiro Polla b517af05b9 MLP/TrueHD decoder. 
Originally committed as revision 14067 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-07-04 15:44:13 +00:00
Ramiro Polla 5faf3a4449 Clarify comment about parity nibble. 
Originally committed as revision 14047 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-07-02 12:24:50 +00:00
Ramiro Polla 49de8d1e9c Make ff_mlp_read_major_sync() take a GetBitContext instead of buffers. 
Originally committed as revision 14044 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-07-01 01:36:16 +00:00
Ramiro Polla ae10e1fd3d Seek the file forwards instead of backwards when sync is lost. 
Originally committed as revision 14014 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-06-27 21:36:01 +00:00
Ramiro Polla c624c2195f Make it a little easier to spot that the code is not dealing only with 
substream headers.

Originally committed as revision 13961 to svn://svn.ffmpeg.org/ffmpeg/trunk
 2008-06-25 12:08:39 +00:00