GitLab

If the context isn't fixed yet or if it doesn't exist then always
restrict the caps.

Also restrict the maximum channels for G726 and G722 to 1 channel

Whether a frame is interlaced or not is unknown at the time of
buffer allocation, so caps on the buffer in opaque will have
a previous frame's interlaced flag set. So if interlacedness
changes, we update the buffer (if any) caps with the correct
interlaced flag once we know.

https://bugzilla.gnome.org/show_bug.cgi?id=656155

There seems to be a little bit of non obvious logic where
the QoS logic can decide to not drop decoding a frame and
pass it to ffmpeg, telling it to drop.
In this case, the QoS logic does not drop the frame itself,
but since the frame will end up being dropped, we still
mark it as dropped and post a QoS message.

https://bugzilla.gnome.org/show_bug.cgi?id=657950

Not needed to make anything work, just seems to make sense.

https://bugzilla.gnome.org/show_bug.cgi?id=589361

Since then it just errors out because it's too small, as it
wants the flac headers as codec data if there is codec data.

https://bugzilla.gnome.org/show_bug.cgi?id=589361

Just like deinterlace.

https://bugzilla.gnome.org/show_bug.cgi?id=656328

This include decoder for ASS, SRT, PGS, DVD and DVB. Those are generated
with bad capabilities and are not usable in GStreamer. Other elements exist
to handle those subtitle formats.

https://bugzilla.gnome.org/show_bug.cgi?id=658019

If extradata is created for LATM, libav will use it for initialising AAC
decoder and decoding will fail.

Fixes bug #652812.

Too many suspicious errors/warnings pop up when more than 1 thread
is used.

Fixes #653649

This still causes issues with bad (missing?) downmixing.

They can be handled by existing a52dec/dtsdec

See Bug #608892

Fixes #653717.

... by _init'ing them as locals rather than _new'ing them.

Fixes #653648.

Add a max-threads property, which defaults to '0 = auto'
Add a utility function taken from libschroedinger which sets
the ffmpeg worker thread count to match the computer processor
count by default.

Fixes bug #651714.

https://bugzilla.gnome.org/show_bug.cgi?id=651625

Fixes bug #651447.

Avoids bogus timestamps for AVCHD-lite streams

Fixes bug #650596.

Also add the stream-format fields to the CODEC_ID_AAC caps.

Fixes bug #650695.

When input buffer timestamps are invalid, next timestamp are used for
audio. Then, the next out timestamp is updated with the used timestamp
and the calculated duration. However, if the used timestamp is invalid,
it should not be used. Otherwise, the next buffer will use a wrong
timestamp that is not in the clipped segment, making the buffer to be
dropped.

This fixes playback with SBTVD MPEG TS streams, using AAC LATM.

A keyframe may be quite a while in the future, and the decoder
has no way of knowing this. A poor decision could mean quite some
time with no video output.
This decision should be left to the upstream element: a demuxer
might know about incoming keyframes, or some other element might
be able to request a keyframe.

Fixes bug #649372.

We might want to expose the various properties in the future

Right now it doesn't use any of the extra fields AVPacket provides.
It might be wise to investigate the pts/dts ones to see if we can finally
get rid of the timing-related cruft we have.

We just use qmin/qmax instead

It is now gone and we need to switch to the new API

See #566605

It was always a bogus one, since it used the external libxvid for encoding.

Idea provided by David Schleef
Support provided by Michael Smith and Tim Mueller
Food and Drinks provided by Toronado