This one is tricky. Consider a graph with two sink links, A and B.
request_oldest() requests on A, and A returns EOF. This EOF causes
a filter to flush pending frames to B; they are unrequested.
Then request_oldest() moves on to B, and B returns EOF.

The doxy says, and the heap implementations assumes,
it is in AV_TIME_BASE units.

The image data is in BGR and not in RGB.

Signed-off-by: Paul B Mahol <onemda@gmail.com>

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

Reviewed-by: compn <tempn@twmi.rr.com>
Signed-off-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: Michael Niedermayer <michaelni@gmx.at>

channel_map is applied between in and postin.
Without this test, if the input format is an acceptable internal format,
the in->postin conversion is skipped. In the best case,
it ignores channel_map, in the worst case it causes an assert failure
if the number of channels is changed by the map.

Fixes ticket #1540

Reviewed-by: Derek Buitenhuis
Reviewed-by: Paul B Mahol

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

Signed-off-by: Paul B Mahol <onemda@gmail.com>

Signed-off-by: Paul B Mahol <onemda@gmail.com>

Prevents out of array reads.

Signed-off-by: Paul B Mahol <onemda@gmail.com>

This silence bunch of useless warnings like:

libavformat/mpeg.c:393:37: warning: variable 'dvdaudio_substream_type' is uninitialized when used within its own initialization [-Wuninitialized]

Signed-off-by: Paul B Mahol <onemda@gmail.com>

* qatar/master:
  rtmp: rtmp_parse_result() add case for video and audio packets to avoid undesired debug output.
  configure: Move the getaddrinfo function check into the network block
  configure: Remove an unused 'have' item
  mpeg: remove disabled code
  libfdk-aac: Check if cutoff value is valid
  network: Always use our version of gai_strerror on windows
  network: Undefine existing gai_strerror definitions
  network: Extend the fallback gai_strerror implementation to handle more error codes

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

Fixes Ticket1432

Thanks-to: Mike Scheutzow <mike.scheutzow@alcatel-lucent.com> for some of the bug analysis
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

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

Reviewed-by: Benjamin Larsson <benjamin@southpole.se>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Reviewed-by: Benjamin Larsson <benjamin@southpole.se>
Signed-off-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: Michael Niedermayer <michaelni@gmx.at>

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

Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>

This makes the two checks (general vs winsock) fit in better.

Signed-off-by: Martin Storsjö <martin@martin.st>

The actual size of uncompressed buffer returned by uncompress() may be
smaller than expected, so abort decoding in such cases.

Signed-off-by: Paul B Mahol <onemda@gmail.com>

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

* qatar/master:
  configure: Check for CommandLineToArgvW
  vc1dec: Do not use random pred_flag if motion vector data is skipped
  vp8: Enclose pthread function calls in ifdefs
  snow: refactor code to work around a compiler bug in MSVC.
  vp8: Include the thread headers before using the pthread types
  configure: Check for getaddrinfo in ws2tcpip.h, too
  vp8: implement sliced threading
  vp8: move data from VP8Context->VP8Macroblock
  vp8: refactor decoding a single mb_row
  doc: update api changes with the right commit hashes
  mem: introduce av_malloc_array and av_mallocz_array

Conflicts:
	configure
	doc/APIchanges
	libavcodec/vp8.c
	libavutil/mem.h
	libavutil/version.h

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

This is only used for checking for a certain library, but the code
doesn't need to know whether the function was found.

Signed-off-by: Martin Storsjö <martin@martin.st>

Passing a cutoff value < sample_rate/256 will cause a crash.
Also, values >20000 will have no effect and 20000 will be used anyway.

Signed-off-by: Mohammad Alsaleh <msal@tormail.org>
Signed-off-by: Martin Storsjö <martin@martin.st>

Even if linking directly to getaddrinfo, use our version of
gai_strerror instead of the system's version. Microsoft explicitly
documents that their version of gai_strerror is thread-unsafe.

Signed-off-by: Martin Storsjö <martin@martin.st>