Right now this mostly demonstatest what not works. That is seeking with
start-type = NONE to only update the rate and playing backwards. Also
it shows that non-flushing seeks tend to lockup adder. Separate unit tests
for the issues follow.
XMP metadata can be embedded in many media container formats. Implement own
parser and formatter that can be used to convert between an xpacket and a
GstTagList. Add unit tests.
Previously we hardcoded the target rectangle passes to Xv(Shm)PutImage. Extend
the implementation to use a full rectangle and don't assume 0,0 for top,left.
Add set_render_rectangle() vmethod to the interface to better support windowless
toolkits (e.g. qt graphicsview or video on canvas in general). Right now we
always fill the widget to 100%. With the patch we can use a rectangular target
region. Fixes#610249.
API: GstXOverlay::set_render_rectangle()
We can update the geometry in ConfigureNotify (unless we disable event-
handling). If event handling is disabled, one should use _expose() to trigger a
redraw and update the geometry.
This just replaces every "$ERROR_CFLAGS" usage with a usage of
"$WARNING_CFLAGS $ERROR_CFLAGS" to get the same functionality as
previously.
Actually using that separation will happen later.
Change playbin2 to not error out if there are subtitles and audio
but no video. If visualizations are enabled the subtitles are rendered on top
of the visualization stream, otherwise the subtitles are not linked at all and
only the audio is played (and a warning message is posted).
If there are only subtitles but neither audio nor video an error message is
still posted.
Fixes bug #610866.
If we first check for >= 4.0 the second check for >= 4.6 will just
short-cut since we are using the same prefix for the variables for
both checks, and they've already been set previously. So the examples
requiring >= 4.6 were built even in the >= 4.0 case.
We update the passed begintime argument to narrow our search region in the
binary search. This means that it does not always contain the chain begin time
after a couple of bisects. Use the real chain->begin_time to bring the
granuletime to the time in the chain instead.
Fixes#610005
