* commit '61c6eef5':
  h264: prevent decoding of slice NALs in extradata
  doxy: Clarify what avpriv_set_pts_info does

Conflicts:
	libavcodec/h264.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'c35f0e84':
  au: Reorder code so that both muxer and demuxer are under #ifdefs
  fate: Move RALF test into lossless audio group
  cosmetics: Use consistent names for multiple inclusion guards.

Conflicts:
	libavformat/au.c
	tests/fate/lossless-audio.mak
	tests/fate/real.mak
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Fixes ticket #2025.
Signed-off-by: Peter Ross <pross@xvid.org>

This fixes files that store a somewhat incorrect value.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

It is not posible to call get_buffer during frame-mt codec
initialization. Libavformat might pass huge amounts of data as
extradata after parsing broken files. The 'extradata' for the fuzzed
sample sample_varPAR_s5374_r001-02.avi is 2.8M large and contains
multiple slices.

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

Fixes Ticket2040
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Ignore the other arguments, which have not to be processed by the POD
renderer.

This allows to clean the doc directory from the parent project dir.
The command:
make -C doc clean

does not work as long as the paths specified in doc/Makefile are relative
to the parent dir.

Also it is consitent with "testclean" and "distclean" targets.

With these commands it is possible to add HTML and non-HTML snippets,
which will be useful for the "See Also" manual chapters.

Fix a few warnings generated when creating the manuals.

Our current doc generation toolchain does not allow to reference external
documents.

Fixes out of array accesses

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Pass "title" metadata field to av_sdp_create (as in RTP muxer) in SAP
muxer for correct
session name in SAP SDP announcements.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Reviewed-by: Nicolas George <nicolas.george@normalesup.org>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

The "pts for a given stream" was nonsensical.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

The "pts for a given stream" was nonsensical.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>

This reduces code size when either piece has been disabled.

* ramiro/dshow:
  dshow: call CoUninitialize() on dshow_read_close()
  dshow: handle events in graph
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master:
  avresample: use valid log context in mixing functions
  lavr: remove automatic context close/open for resampling compensation
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit 'f322b207':
  lavr: only save/restore the mixing matrix if mixing is being done
  rtpdec_vp8: Cosmetics: Fix bad alignment/indentation
  rtpenc: Allow including a SDES/CNAME block in RTCP SR packets
Merged-by: Michael Niedermayer <michaelni@gmx.at>

* commit '65e05327':
  rtpenc_vp8: Include the picture number in VP8 packets
Merged-by: Michael Niedermayer <michaelni@gmx.at>

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

Signed-off-by: Peter Ross <pross@xvid.org>

Signed-off-by: Peter Ross <pross@xvid.org>

Signed-off-by: Peter Ross <pross@xvid.org>

Signed-off-by: Peter Ross <pross@xvid.org>

Every call to CoInitialize should have a matching CoUnititalize.

Based on patch by Don Moir <donmoir@comcast.net>.

Handling DirectShow events prevents infinite loops when there is an error in
the graph, such as a device being disconnected. This makes it possible for
dshow to return an error to the caller and run the cleanup code.

Based on patch by Don Moir <donmoir@comcast.net>.

Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>

Fixes Ticket1865
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

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

It adds unnecessary complication for insignificant usability improvement.
The user really should know if they'll need resampling compensation before
opening the context.

Note that only the documentation has changed. The current functionality will
still work until the next major bump.

This prevents a null pointer dereference

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>