(cherry picked from commit 3197968bedf9f0d6d3cccf11cfbd19c23063493b)
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master: (29 commits)
  doc: update libavfilter documentation
  tls: Use the URLContext as logging context
  aes: Avoid illegal read and don't generate more key than we use.
  mpc7: Fix memset call in mpc7_decode_frame function
  atrac1: use correct context for av_log()
  apedec: consume the whole packet when copying to the decoder buffer.
  apedec: do not needlessly copy s->samples to nblocks.
  apedec: check output buffer size after calculating actual output size
  apedec: remove unneeded entropy decoder normalization.
  truespeech: use memmove() in truespeech_update_filters()
  vorbisdec: remove AVCODEC_MAX_AUDIO_FRAME_SIZE check
  vorbisdec: remove unneeded buf_size==0 check
  vorbisdec: return proper error codes instead of made-up ones
  http: Don't add a Range: bytes=0- header for POST
  sunrast: Check for invalid/corrupted bitstream
  http: Change the chunksize AVOption into chunked_post
  http: Add encoding/decoding flags to the AVOptions
  avconv: remove some codec-specific hacks
  crypto: add decoding flag to options.
  tls: use AVIO_FLAG_NONBLOCK instead of deprecated URL_FLAG_NONBLOCK
  ...

Conflicts:
	doc/libavfilter.texi
	libavcodec/atrac1.c
	libavcodec/sunrast.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Update the reference to the conversion tool to use avconv and
make sure the example line works as supposed.
Remove the paragraph pointing to the svn repo

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

Fixes Ticket600
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fixes Ticket613
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Fixes ticket #632.

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

This avoids artifically consuming a partial packet but ignoring remaining data
in subsequent calls.

also move nblocks to the local scope where it is used.

The decoder already skips data at the end of the packet without this.
Also remove 2 APEContext fields that were only used for the end-of-frame
normalization.

The user could provide a larger buffer, which is already checked separately
before writing output.

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

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

That header simply doesn't make sense in that context.
Signed-off-by: Martin Storsjö <martin@martin.st>

Signed-off-by: Janne Grunau <janne-libav@jannau.net>

The chunksize internal variable has two different uses - for
reading, it's the amount of data left of the current chunk
(or -1 if the server doesn't send data in chunked mode), where
it's only an internal state variable. For writing, it's used
to decide whether to enable chunked encoding (by default), by
using the value 0, or disable chunked encoding (value -1).

This, while consistent, doesn't make much sense to expose
as an AVOption. This splits the usage of the internal variable
into two variables, chunksize which is used for reading (as
before), and chunked_post which is the user-settable option,
with the values 0 and 1, where 1 is default.
Signed-off-by: Martin Storsjö <martin@martin.st>

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

The problem they are supposed to fix is handled in riff.c, so those
hacks are pointless.

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

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

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

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

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

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

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

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

Based on reimars suggestion
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>