Sligthly improve readability, reduce redundancy.

Use the value specified in the codec context for setting the
filterchain sample aspect ratio, when it is not specified in the
stream context.

Consistent with the ffmpeg behavior.

Fix trac issue #398.

* qatar/master:
  movenc: change AV_LOG_INFO to AV_LOG_WARNING for some warnings
  avconv: remove a write-only variable
  avconv: remove pointless parameter from new_*_stream().
  avconv: cosmetics, move code
  avconv: cosmetics -- move copy_chapters().
  avconv: cosmetics -- move parse_forced_key_frames().
  lavc: add audio flag to the 'b' option, deprecate 'ab'.
  avconv: rename sameq to same_quant
  doc/avconv: add forgotten end of chapter.
  Changelog: document avconv incompatibilities with ffmpeg.
  avconv: replace -vcodec/-acodec/-scodec with a better system.
  avconv: remove presets.
  svq3: propagate codec memory allocation failure in context init

Conflicts:
	Changelog
	avconv.c
	libavcodec/options.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

avcodec_set_dimensions should be used for size changes to ensure
compatibility with future changes.
avctx->width/avctx->height may not be set to display-only dimensions.
Even more so since vc1dec.c would later set coded_width/height based
on this.
coded_width/coded_height should be used instead of width/height for
decoder setup.
This fixes playback of e.g. zz-mcr-nsqr.vc1 sample (containing
display width/height settings) in MPlayer and should fix a crash
with MPC: http://forum.doom9.org/showthread.php?t=162221.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>

strsep() is not POSIX compliant, and thus not supported on some platform.
Fix compilation on Solaris.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

New streams are always added to the last output file.

Move find_codec_or_die(), choose_codec(), new_output_stream() and
read_avserver_streams() below transcode() where they belong.

It's only used in opt_output_file(), so move it right above
opt_output_file().

parse_forced_key_frames() is only used in new_video_stream(), so move it
right above new_video_stream().

Its only reason for its existence was ffmpeg's inability to properly assign
AVOptions to streams. Now this is not a problem anymore, so 'ab' should
go.

It's often mistakenly used as 'same quality', emphasize that it's not
true in the manual.

The new option doesn't depend on its placement wrt -new* options (which
don't exist anymore) and works in a similar way as per-stream AVOptions.

-[vas]codec remain as aliases to -codec:[vas]

Nothing except x264 uses them and native x264 presets should be used
with private options.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master:
  Fix NASM include directive
  dsputil_mmx: Honor HAVE_AMD3DNOW
  lavf,lavd: remove all usage of AVFormatParameters from demuxers.
  jack: add 'channels' private option.
  VC-1: fix reading of custom PAR.
  Remove redundant and dubious video codec detection by its extradata
  mpeg12: remove repeat-field code disabled since May 2002
  patch checklist: suggest fate instead of regression tests
  Turn on resampling on sudden size change instead of bailing out during recode.
  avtools: reinitialise filter chain when input video stream changes dimensions

Conflicts:
	Makefile
	avconv.c
	doc/developer.texi
	ffplay.c
	libavcodec/x86/dsputil_mmx.c
	libavdevice/libdc1394.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Anton Khirnov <anton@khirnov.net>

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>

AVFormatParameters are converted into corresponding private options in
av_open_input_file/stream() compat wrappers, so accessing them from
demuxers is redundant.

Get rid of AVFormatParameters usage.

Custom PAR num/denum are in 1-256 range.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
Signed-off-by: Diego Biurrun <diego@biurrun.de>

Signed-off-by: Diego Biurrun <diego@biurrun.de>

Signed-off-by: Diego Biurrun <diego@biurrun.de>

Signed-off-by: Anton Khirnov <anton@khirnov.net>

Signed-off-by: Anton Khirnov <anton@khirnov.net>

Signed-off-by: Anton Khirnov <anton@khirnov.net>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Since SDL has no audio buffer fullness info, one can get a much precise audio
clock based on the last time of the audio callback and the elapsed time since.

To achieve this I introduced the audio_current_pts and audio_current_pts_drift
variables (similar to video_current_pts and video_current_pts_drift) and
calculate them in the end of the audio callback, when VideoState->audio_clock
is already updated. The reference time I use is from the start of the audio
callback, because this way the amount of time used for audio decoding is not
interfereing with calculation.

I also replaced the audio_write_get_buf_size function with a calculated
variable because when the audio frame decoding is in progress audio_buf_size
and audio_buf_index are not stable, so using them from other threads are not a
good idea.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>