Clients are usually disconnected in the streaming thread if their inactivity
is bigger than the timeout. If no new buffers are to be rendered in the sink,
these clients will never be disconnected and for that reason it should be
handled in the select() loop too.

Subtitle streams being parse can cause the pipeline to wait indefinitely
to PREROLL. This makes subtitle streams got to PAUSED even if no data is
available. This should not be a cause for concern as we don't expect to
get much data for subtitle streams other than language tags from the
container.

https://bugzilla.gnome.org/show_bug.cgi?id=632291

... otherwise we may end up without running clock in PLAYING.

Fixes #636886.

g_object_set is a varargs function. Save 7 g_obvject_calls (and the overhead of
them) by using it accordingly.

Mark 64-bit interger constant as such to avoid warnings such as:
gsttypefindfunctions.c:2152: error: integer constant is too large for ‘long’ type

Fixes #642504.

From 1de7f6a to 6aec6b9

Parsers are the only element class that are not changing the data and
could lead to an infinite loop. Other element classes like demuxers,
e.g. id3demux, can be used multiple times in a row and sometimes are.

This is going to lead to an infinite loop of this element and can easily
happen with parsers that accept their own src caps on the sinkpad.

Previously we only checked against the raw caps but we should also
check against the return value of autoplug-continue. Additionally fix
a thread-safety issue with accessing the raw caps.

Maps to GST_BUFFER_FLAG_MEDIA4. The purpose is to explicitly indicate
whether a telecined buffer is progressive or not without having to make
assumptions based on previous buffers.

Add "source-setup" signal for convenience and discoverability. No need
to figure out "notify::source", look up the notify callback signature,
then do an g_object_get() to get the source element..

https://bugzilla.gnome.org/show_bug.cgi?id=626152

Request pads have to be released by the caller and must be
unreffed after releasing them.

The GObject signal code assumes that the signal handlers return a
new reference or copy. Fixes bug #641927.

... rather than leave it as GST_BUFFER_OFFSET_NONE

Fix bug #642942.

Also fixes #642466.

... to avoid subsequent cleanup disposing an element not in NULL state.

It would otherwise be printed as (null) and mess up indentation (no \n).

This makes sure we maintain a ref on the discoverer object while the
async timeout callback is alive to prevent a potential crash if the
object is freed while the callback is pending.

https://bugzilla.gnome.org/show_bug.cgi?id=641706

We want to make sure the discoverer object passed to the various
callbacks doesn't become invalid if a callback is pending and the object
is free'd in the mean time.

https://bugzilla.gnome.org/show_bug.cgi?id=641706

https://bugzilla.gnome.org/show_bug.cgi?id=623846

https://bugzilla.gnome.org/show_bug.cgi?id=642949

They probably fail because ffmpegcolorspace can't handle those formats.

API: GstAudioRate::skip-to-first

... such as avoiding arithmetic mixing counts and ts, although latter
would typically be 0 so far.

This reverts commit 6d150873.

Depends on a core commit that was reverted.

Core now calls release_pad in finalize, which is usually after
the collectpads has been unreffed.