Reordering some functions to avoid forward declarations.
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
These were left behind in the transition to 1.0.
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Some decoders and files fail to find any valid frames during test-play, so
don't wait for data we might never receive.
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Source pads can be created asynchronously, so they might not be created until
during the call to gst_element_set_state(). Therefore don't copy This->ppPins
into a local variable.
This condition can be triggered by failing to demux a stream.
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Especially since pads might be exposed before failure is reported.
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Otherwise the push_data() thread will run and try to use resources we
never allocated.
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Signed-off-by: Bruno Jesus <bjesus@codeweavers.com>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This reverts commit 976ccb92ea.
This is what I get for not reading the documentation...if you call
g_error_free(NULL), it prints a nasty error message, which I am now
getting frequently.
Signed-off-by: Alex Henrie <alexhenrie24@gmail.com>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This fixes video playback issue in a certain application.
The approach is very similar to 068593b238.
Signed-off-by: Akihiro Sagawa <sagawa.aki@gmail.com>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Creating an element and placing it in a bin gives overship of the
element refcount to the bin. Don't explicitly unref it after that.
Signed-off-by: Jan Schmidt <jan@centricular.com>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
The videoflip element doesn't handle all formats, so some formats like
Intel Indeo 3 will cause errors and fail.
Add videoconvert before videoflip to do format conversion when it's
needed.
Signed-off-by: Jan Schmidt <jan@centricular.com>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Once placed inside the container, decodebin will be managed and
destroyed by the container. There's no need to store it, or (especially)
unref it.
Signed-off-by: Jan Schmidt <jan@centricular.com>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Don't replace the bus on the decodebin inside our container, as it
causes problems with internal GStreamer element management.
Put the bus only on the top-level container.
Signed-off-by: Jan Schmidt <jan@centricular.com>
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Gstreamer handles video top-down, but Windows's dshow handles it
bottom-up. So let's insert a videoflip filter to fix the discrepancy
instead of relying on videorenderer to flip it like we did before.
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Zebediah Figura