Fixes Ticket1817
Based on patch by: Przemyslaw Wesolek <przemyslaw.wesolek@go.art.pl>
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

Fixes: CID733746
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

Fixes CID717497
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

This should make no difference but might help static analyzers
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

Fixes CID700574
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

Fixes CID717752
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

The field is intended to overflow and have only its low 16bits stored.
This commit avoids the implicit truncation and clarifies that its
intended and not a bug

S326m section 7.6 ("Continuity count"):
> The continuity count word consists of 2 bytes allow-
> ing a number to be created by a modulo 65536
> counter (bits C15 to C0 in figure 7). The continuity
> count shall increment by 1 for each newly transmit-
> ted content package with the same SDTI source and
> destination addresses. The continuity count may
> be used to detect whether the content package
> sequence has been broken by an operation such as
> a routing switch.

Approved-by: Tjoppen
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

Partly fixes Ticket1664
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

Fixes CID733709.
Signed-off-by: NPaul B Mahol <onemda@gmail.com>

Fixes CID733173.
Signed-off-by: NPaul B Mahol <onemda@gmail.com>

Fixes null pointer dereference. Fixes CID732242.
Signed-off-by: NPaul B Mahol <onemda@gmail.com>

Fixes integer overflow. Fixes CID732223.
Signed-off-by: NPaul B Mahol <onemda@gmail.com>

Fixes CID703626.
Signed-off-by: NPaul B Mahol <onemda@gmail.com>

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

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

This fixes playback in some circumstances (like webm in firefox).
Regression after 2c34367b.

It is also matching the Matroska specifications:
http://matroska.org/technical/specs/notes.html, "The quick eye will
notice that if a Cluster's Timecode is set to zero, it is possible to
have Blocks with a negative Raw Timecode. Blocks with a negative Raw
Timecode are not valid."

Based on libcompface code by James Ashton <James.Ashton@anu.edu.au>, and
relicensed to LGPL with the author's consent.

Fixes Ticket1065
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

If total number of samples is -1 it means stream is of unknown length.

Fixes ticket #1777.
Signed-off-by: NPaul B Mahol <onemda@gmail.com>

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

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

missed these in my previous search and replace
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

Optional sdp speex payload parameter is outputed only when
data is encoded. It's not printed in case of stream copy.
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

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

Use native tags instead of avi ones, simplifies a lot raw video codecs
handling.

The native tags will be used when available.

Fixes ticket #1807.

Failure of this assert could lead to hard to debug behavior
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

Fixes CID718002
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

This prevents a double free

Fixes CID718285
Signed-off-by: NMichael Niedermayer <michaelni@gmx.at>

parse_packed_headers() returns either zero or a negative error code
so its return type must be signed.
Signed-off-by: NMans Rullgard <mans@mansr.com>

The 'ret' variable is used for negative error codes so it should
be a signed type.
Signed-off-by: NMans Rullgard <mans@mansr.com>

Based on swf specs v10, page 271-273.

We now only return in the middle of the refill in case of read error, so
inflate can be re-called if zlib needs an empty (z.avail_in=0) inflate
call for flushing.