Instead of querying the Qt include path from the dependency or from
qmake, rely on the qt5qml_dep to set the include path to QtGui
correctly, and look for the header inside the private includedir.

Then we can use that path to include the header directly.

Reported in gstreamer/gst-plugins-base#780 (comment 548092)

shout2send caps depend on what the libshout2
version in question supports, but the
documentation caps should always be the same.

Part-of: <gstreamer/gst-plugins-good!635>

Jan Schmidt authored 5 years ago and GStreamer Marge Bot committed 4 years ago

Move some code out of the enormous qtdemux.c into a separate
qtdemux_tags helper, and make some structs available via qtdemux.h
to accommodate that.

Part-of: <gstreamer/gst-plugins-good!634>

Jan Schmidt authored 5 years ago and GStreamer Marge Bot committed 4 years ago

Reduce a tiny bit of the bulk of qtdemux.c by moving some
agnostic helper functions out.

Part-of: <gstreamer/gst-plugins-good!634>

Jan Schmidt authored 5 years ago and GStreamer Marge Bot committed 4 years ago

Move the SVMI stereoscopic atom parsing out to a helper
function to shrink qtdemux_parse_trak a bit.

Add a bounds check that the received atom is large enough
before parsing it.

Add a note to the atom parser that svmi comes from the
MPEG-A spec 23000-11.

Part-of: <gstreamer/gst-plugins-good!634>

Guillaume Desmottes authored 4 years ago and GStreamer Marge Bot committed 4 years ago

Fix regression introduced in 7bc5e28d
preventing the device provider to send the device-added message for new
devices.
By early returning the patch was discarding add/remove events.

Fix #735

Part-of: <!626>

Would get "Tried to create target "qt5-qmlsink_qrc", but a
target of that name already exists." with older meson versions.

Work around that by renaming the qrc file.

Part-of: <gstreamer/gst-plugins-good!633>

Fixes gstreamer/gst-plugins-good#751

Part-of: <gstreamer/gst-plugins-good!631>

Expected return value for unhandled query is FALSE

Part-of: <!629>

Measured to be about 3.4x faster than C

Part-of: <gst-plugins-good!621>

Part-of: <gstreamer/gst-plugins-good!621>

Matthew Waters authored 4 years ago and GStreamer Marge Bot committed 4 years ago

Reset to the original OpenGL state as required by the GStreamer OpenGL
API contract.  Fixes output with a glimagesink element downstream.

Part-of: <gstreamer/gst-plugins-good!623>

Matthew Waters authored 4 years ago and GStreamer Marge Bot committed 4 years ago

Qt may replace the drawable with its own which breaks output if Qt is
not displaying the resulting video as used with e.g. glimagesink.

Part-of: <gstreamer/gst-plugins-good!623>

It was accidentally changed in gstreamer/gst-plugins-good!436

Part-of: <gstreamer/gst-plugins-good!624>

gst_element_class_set_metadata is meant to only be used with
static or inlined strings, which isn't the case for the 2 elements
here resulting in use-after-free later on.

https://gstreamer.freedesktop.org/documentation/gstreamer/gstelement.html?gi-language=c#gst_element_class_set_static_metadata

Part-of: <gstreamer/gst-plugins-good!622>

Sebastian Dröge authored 4 years ago and GStreamer Marge Bot committed 4 years ago

This reverts commit 54810bf4

Part-of: <gstreamer/gst-plugins-good!620>

Define g_queue_clear_full if glib < 2.60.

Fixes #747

Part-of: <gstreamer/gst-plugins-good!619>

Set up our plugin include list for tests in such a way that
we don't pull in *all* plugins from -bad but only the one
used in the splitmuxsink unit test, i.e. the timecode plugin,
so we don't accidentally use other encoders/decoders such as
nvenc/dec for example.

Part-of: <gstreamer/gst-plugins-good!617>

While the caller should make sure this does not happen, make sure timer
collision are not silently ignored and leaked.

Fixes #726

Part-of: <gstreamer/gst-plugins-good!616>

Until now, do_expected_timeout() was shortly dropping the JBUF_LOCK in order
to push RTX event event without causing deadlock. As a side effect, some
CPU hung would happen as the timerqueue would get filled while looping over
the due timers. To mitigate this, we were processing the lost timer first and
placing into a queue the remainign to be processed later.

In the gap caused by an unlock, we could endup receiving one of the seqnum
present in the pending timers. In that case, the timer would not be found and
a new one was created. When we then update the expected timer, the seqnum
would already exist and the updated timer would be lost.

In this patch we remove the unlock from do_expected_timeout() and place all
pending RTX event into a queue (instead of pending timer). Then, as soon as
we have selected a timer to wait (or if there is no timer to wait for) we send
all the upstream RTX events. As we no longer unlock, we no longer need to pop
more then one timer from the queue, and we do so with the lock held, which
blocks any new colliding timers from being created.

Part-of: <gstreamer/gst-plugins-good!616>

Test was not enforcing a video format on videotestsrc. I420 was picked
as it was the first format in GST_VIDEO_FORMATS_ALL which will no longer
be true (gst-plugins-base!689).

Part-of: <gstreamer/gst-plugins-good!615>

Edward Hervey authored 4 years ago and Edward Hervey committed 4 years ago

When a GST_EVENT_FLUSH_START reaches the jitterbuffer, there is a chance that
our task is currently blocking waiting for a timer.

There was two problems:
* That wait wasn't checking for flushing situations
* The flushing handling wasn't waking up that conditional (to check whether it
should abort)

Part-of: <gstreamer/gst-plugins-good!608>

`last_out` would be used uninitialized if the element has no `set-active`
signal. Initialize it to -1 as that's what the "default" value is
further below.

CID 1455443

Fixes gstreamer/gst-plugins-good#727

Part-of: <gstreamer/gst-plugins-good!613>

Mikhail Fludkov authored 9 years ago and GStreamer Marge Bot committed 4 years ago

As part of this also change the default bitrate value to 0. The default
value was 256000 previously. In reality, if the property was not set the
bitrate value would be scaled according to the resolution which is not
very intuitive behavior. It is better to use 0 for this purpose. Now
together with newly introduced property "bits-per-pixel" 0 means to
assign the bitrate according to resolution/framerate.

The default bitrates are now
 - 1.2Mbps for VP8 720p@30fps
 - 0.8Mbps for VP9 720p@30fps
and scaled accordingly for different resolutions/framerates.

Previously the default bitrate was also not scaled according to the
framerate but only took the resolution into account.

This also fixes the side effect of setting bitrate to 0. Previously
encoder would not produce any data at all.

Addition from Sebastian Dröge <sebastian@centricular.com> to assume
30fps if no framerate is given in the caps instead of not calculating
any bitrate at all.

Part-of: <gstreamer/gst-plugins-good!611>

Stian Selnes authored 7 years ago and Sebastian Dröge committed 4 years ago

For VP8 it's possible to signal width or height to be 0, but it does
not make sense to do so. For VP9 it's impossible. Hence, we most
likely have a corrupt stream. Trying to negotiate caps downstream with
either width or height as 0 will fail with something like

gst_video_decoder_negotiate_default: assertion 'GST_VIDEO_INFO_WIDTH (&state->info) != 0' failed

Part-of: <gstreamer/gst-plugins-good!610>

Use speex_header_free() to free memory which was allocated by
library. Cross-CRT issue should not happen on 1.17 Cerbero build
but might happen custom build or so.

Part-of: <gstreamer/gst-plugins-good!606>

It's an integer property and rtpbin also expects an integer. Passing it
as a GstClockTime (guint64) to g_object_set() will cause problems, and
on big endian MIPS apparently causes crashes.

Fixes #737

Part-of: <!605>

If core is built as a subproject (e.g. as in gst-build), make sure to use
the gst-plugin-scanner from the built subproject. Without this, gstreamer
might accidentally use the gst-plugin-scanner from the install prefix if
that exists, which in turn might drag in gst library versions we didn't
mean to drag in. Those gst library versions might then be older than
what our current build needs, and might cause our newly-built plugins
to get blacklisted in the test registry because they rely on a symbol
that the wrongly-pulled in gst lib doesn't have.

This should fix running of unit tests in gst-build when invoking
meson test or ninja test from outside the devenv for the case where
there is an older or different-version gst-plugin-scanner installed
in the install prefix.

In case no gst-plugin-scanner is installed in the install prefix, this
will fix "GStreamer-WARNING: External plugin loader failed. This most
likely means that the plugin loader helper binary was not found or
could not be run. You might need to set the GST_PLUGIN_SCANNER
environment variable if your setup is unusual." warnings when running
the unit tests.

In the case where we find GStreamer core via pkg-config we use
a newly-added pkg-config var "pluginscannerdir" to get the right
directory. This has the benefit of working transparently for both
installed and uninstalled pkg-config files/setups.

Part-of: <gstreamer/gst-plugins-good!603>

And let it rety twice.

Fixes #717

Part-of: <!601>

Nicolas Dufresne authored 4 years ago and GStreamer Marge Bot committed 4 years ago

The profiles and levels were applied to the common caps instead of the copy.
That had the side effect of setting profiles/level from one CODEC onto
another. Leaving to encoder not being registered or not-negotiated errors.

Part-of: <gstreamer/gst-plugins-good!599>

Nicolas Dufresne authored 4 years ago and GStreamer Marge Bot committed 4 years ago

The levels and profiles probe function returned a dynamically allocated GValue
that was leaked. Simplify this by using a stack allocated GValue and a boolean
return value.

Part-of: <gstreamer/gst-plugins-good!599>

Nicolas Dufresne authored 4 years ago and GStreamer Marge Bot committed 4 years ago

There is only one user of that function and the split only increase
complexicity.

Part-of: <gstreamer/gst-plugins-good!599>