Originally committed as revision 16747 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16746 to svn://svn.ffmpeg.org/ffmpeg/trunk

data exists in alloc_table_1 and alloc_table_3

Originally committed as revision 16745 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16744 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16743 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16742 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16741 to svn://svn.ffmpeg.org/ffmpeg/trunk

fixes hang on EOF

Originally committed as revision 16740 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16739 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16738 to svn://svn.ffmpeg.org/ffmpeg/trunk

subsequent a= lines from the m= block to be applied to the previous
m= line, thus breaking otherwise functional RTP streams. See discussion in
[PATCH] RTSP-MS 7/15: parse and allow unknown m= line codes" thread on
mailinglist.

Originally committed as revision 16737 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16736 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16735 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16734 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16733 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16732 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16731 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16730 to svn://svn.ffmpeg.org/ffmpeg/trunk

http://av.watch.impress.co.jp/docs/20040929/xf.asf

Originally committed as revision 16729 to svn://svn.ffmpeg.org/ffmpeg/trunk

Instead move relevant fields into MPADecodeHeader and use it
where appropriate.

Originally committed as revision 16728 to svn://svn.ffmpeg.org/ffmpeg/trunk

Should fix regression tests & fate.

Originally committed as revision 16727 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16726 to svn://svn.ffmpeg.org/ffmpeg/trunk

Spotted by Stephen Warren.

Originally committed as revision 16725 to svn://svn.ffmpeg.org/ffmpeg/trunk

to AVStream, this is required if we want to allow the user to pull frames
from specific streams at some point in the future.

Originally committed as revision 16724 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16723 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16722 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16721 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16720 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16719 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16718 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16717 to svn://svn.ffmpeg.org/ffmpeg/trunk

libavcodec/audioconvert.c:190: warning: initialization discards qualifiers from pointer target type

Originally committed as revision 16716 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16715 to svn://svn.ffmpeg.org/ffmpeg/trunk

this prevents using negative index and possibly crashing with broken files

Originally committed as revision 16714 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16713 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16712 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16711 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16710 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 16709 to svn://svn.ffmpeg.org/ffmpeg/trunk

size. Thus, make demuxer and decoder handle it as a whole.

Originally committed as revision 16708 to svn://svn.ffmpeg.org/ffmpeg/trunk