Vivia Nikolaidou [Tue, 28 Aug 2018 12:04:53 +0000 (15:04 +0300)]
decklink: Only use drop-frame flag for 29.97 and 59.94 FPS
Otherwise it can cause errors in gstvideotimecode.c because drop-frame
is not defined for any other frame rate (e.g. 24000/101)
https://blog.frame.io/2017/07/17/timecode-and-frame-rates/
https://bugzilla.gnome.org/show_bug.cgi?id=797037
Johan Bjäreholt [Tue, 28 Aug 2018 07:11:32 +0000 (09:11 +0200)]
audiobuffersplit: Fix format string warning
We have a GST_DEBUG_OBJECT, which prints a guint64 with %lu which gave a
compiler warning. Used G_GUINT64_FORMAT instead.
https://bugzilla.gnome.org/show_bug.cgi?id=797036
Philipp Zabel [Fri, 17 Aug 2018 09:12:26 +0000 (11:12 +0200)]
kmssink: configure mode setting from video info
drmModeGetFB returns -EINVAL for multi-planar framebuffers. Instead of
depending on the framebuffer dimensions to select the mode, use width
and height from GstVideoInfo, which was used to create the framebuffer
in the first place. This enables kmssink to display multi-planar
formats such as I420 or NV12 with modesetting enabled.
https://bugzilla.gnome.org/show_bug.cgi?id=796985
Jerome Laheurte [Thu, 16 Aug 2018 10:35:50 +0000 (12:35 +0200)]
ksvideosrc: fix device enumeration when hotplugging a camera
Since both audio and video capture devices declare the KSCATEGORY_CAPTURE interface,
plugging a camera that supports both could result in an audio device being mistaken
for a video one.
https://bugzilla.gnome.org/show_bug.cgi?id=796958
Tim-Philipp Müller [Sun, 19 Aug 2018 14:50:25 +0000 (15:50 +0100)]
meson: build musepack plugin
Tim-Philipp Müller [Sun, 19 Aug 2018 14:44:16 +0000 (15:44 +0100)]
musepack: remove support for the 'old' API
Tim-Philipp Müller [Sun, 19 Aug 2018 13:39:53 +0000 (14:39 +0100)]
freeverb: update for g_type_class_add_private() deprecation
Not that the private struct is really needed here.
Tim-Philipp Müller [Sun, 19 Aug 2018 00:11:38 +0000 (01:11 +0100)]
meson: build neonhttpsrc
Tim-Philipp Müller [Sat, 18 Aug 2018 20:32:11 +0000 (21:32 +0100)]
meson: add options to disable gobject cast checks and glib asserts
... and define G_DISABLE_DEPRECATED for development versions,
like we do in autotools.
Sebastian Dröge [Fri, 17 Aug 2018 13:37:45 +0000 (16:37 +0300)]
audiobuffersplit: Add a gapless mode which inserts silence/drops samples on disconts
The output is always a continguous stream without any gaps.
Sebastian Dröge [Fri, 17 Aug 2018 13:37:31 +0000 (16:37 +0300)]
audiobuffersplit: Always set DISCONT flag after resyncs
Sebastian Dröge [Fri, 17 Aug 2018 13:33:52 +0000 (16:33 +0300)]
audiobuffersplit: Keep track of resync time separately
If we drain after a discont, the discont time given by the stream
synchronizer is already the time after the discontinuity. But we need to
drain all pending data based on the previous discont time instead.
Sebastian Dröge [Fri, 17 Aug 2018 11:56:51 +0000 (14:56 +0300)]
audiobuffersplit: Update output buffer size after each buffer to compensate for accumulated errors
https://bugzilla.gnome.org/show_bug.cgi?id=796981
Vivia Nikolaidou [Thu, 16 Aug 2018 14:47:55 +0000 (17:47 +0300)]
avwait: Start video and audio together if audio starts late
Also add test to meson
https://bugzilla.gnome.org/show_bug.cgi?id=796977
Nirbheek Chauhan [Thu, 16 Aug 2018 21:29:26 +0000 (02:59 +0530)]
meson: host_system is 'ios' when building for iOS
The cross file sets this value, and we use 'ios' in Cerbero.
Nirbheek Chauhan [Thu, 16 Aug 2018 20:19:07 +0000 (01:49 +0530)]
meson: Add an option for tests
This is needed because we don't always have gstreamer-check available,
for instance inside Cerbero on iOS.
Nicolas Dufresne [Thu, 16 Aug 2018 19:21:00 +0000 (15:21 -0400)]
dashdemux: Always create the adapter
Worst case it will be empty. This fixes a crash when the base class
calls data_received() when the stream is neither is_isobmff or
has_isoff_ondemand_profile.
https://bugzilla.gnome.org/show_bug.cgi?id=796745
Sebastian Dröge [Thu, 16 Aug 2018 14:07:06 +0000 (17:07 +0300)]
compositor: Define crossfade-ratio to have range [0.0,1.0]
Previously negative values had the same effect as 0.0, which was
confusing.
https://bugzilla.gnome.org/show_bug.cgi?id=796845
Wim Taymans [Thu, 16 Aug 2018 09:42:25 +0000 (11:42 +0200)]
curlhhtpsrc: avoid invalid memory references
gst_curl_http_src_remove_queue_item() can free qelement and then
we get an invalid memory reference when we do qelement->next a
couple of lines below. Take the next pointer earlier so that we can
safely free.
Wim Taymans [Thu, 16 Aug 2018 09:20:54 +0000 (11:20 +0200)]
rfbdecoder: don't free decoder data
The decoder data is freed when we read more data.
Sebastian Dröge [Thu, 16 Aug 2018 07:02:42 +0000 (10:02 +0300)]
videoaggregator: Make sure to hold object lock while iterating sink pads
They might otherwise just change while we iterate.
Nirbheek Chauhan [Wed, 15 Aug 2018 14:14:17 +0000 (19:44 +0530)]
meson: Define HAVE_OSX on macOS and iOS
Needed by the shm, ladspa, and lv2 plugins. Same as Autotools.
Nirbheek Chauhan [Wed, 15 Aug 2018 14:04:31 +0000 (19:34 +0530)]
meson: Fix shm librt dep check on macOS, iOS, and *BSD
Nicolas Dufresne [Tue, 14 Aug 2018 19:59:52 +0000 (15:59 -0400)]
makefile: opencv: Fix header names in noinst_HEADERS
This should fix missing gstcameracalibrate.h and gstcameraundistort.h
missing into generated tarballs.
Nirbheek Chauhan [Tue, 14 Aug 2018 14:18:28 +0000 (19:48 +0530)]
meson: Add an option for the applemedia plugin
Michael Drake [Wed, 18 Jul 2018 10:11:17 +0000 (11:11 +0100)]
assrender: fix multiple subtitles on screen simultaneously
This fixes an issue with SSA/ASS subtitles, where subtitles
would fail to appear if there was already a subtitle on screen.
This was because `struct _GstAssRender` had a single
`GstBuffer *subtitle_pending` member. This meant that
the assrender context could only be aware of one subtitle
at a time.
This patch changes the subtitle_pending member to a
linked list of pending subtitles.
The `gst_ass_render_chain_text` function no longer needs
to care about whether there are already subtitles pending,
it simply appends new subtitles to the list.
The `gst_ass_render_chain_video` function has been modified
to handle the list of pending subtitles.
Finally, the `gst_ass_render_pop_text` function has been
modified to pop the entire list of pending subtitles.
https://bugzilla.gnome.org/show_bug.cgi?id=735944
Víctor Manuel Jáquez Leal [Tue, 14 Aug 2018 11:53:00 +0000 (13:53 +0200)]
check: update gitignore
Víctor Manuel Jáquez Leal [Tue, 14 Aug 2018 10:42:02 +0000 (12:42 +0200)]
closedcaption: comment out unused function
When compiling with clang-6 this error raises:
raw_decoder.c:411:1: error: unused function 'cpr1204_crc'
[-Werror,-Wunused-function]
This patch only comments it out.
https://bugzilla.gnome.org/show_bug.cgi?id=796957
Víctor Manuel Jáquez Leal [Tue, 14 Aug 2018 10:36:53 +0000 (12:36 +0200)]
closedcaption: avoid implicit convertion of enums
When compiling with clang-6 this error pops out:
raw_decoder.c:1011:62: error: implicit conversion from enumeration
type 'const vbi_modulation' to different enumeration type
'vbi3_modulation' [-Werror,-Wenum-conversion]
This is because function vbi3_bit_slicer_set_params() sets
vbi3_modulation as enum type parameter, nonetheless vbi_modulation
enum is passed. Both enums looks semantically equal, thus the fix is a
simple cast.
https://bugzilla.gnome.org/show_bug.cgi?id=796957
Nirbheek Chauhan [Fri, 10 Aug 2018 22:42:46 +0000 (04:12 +0530)]
meson: Add gnustl dep for webrtcdsp on Android
Needed for C++ headers. Without this, webrtcdsp can't find <algorithm>
Bastian Köcher [Fri, 10 Aug 2018 11:57:47 +0000 (12:57 +0100)]
meson: fix install dir for generated header files
Nixos installs into a non-standard includedir, so need
to take account of the 'includedir' option instead of
just hard-coding 'include' here.
https://bugzilla.gnome.org/show_bug.cgi?id=794856
Nirbheek Chauhan [Tue, 7 Aug 2018 19:01:51 +0000 (00:31 +0530)]
wasapi: Fix build with Windows 8.1 SDK
With the Windows 8.1 SDK, the v1 of the AUDCLNT_STREAMOPTIONS enum is
defined which only has NONE and RAW, so it's not only defined when
AudioClient3 is available.
Add a meson check for the symbol. This is not needed for Autotools
because there we build against the MinGW audioclient.h which is still
at v1 of the AudioClient interface.
George Kiagiadakis [Fri, 3 Aug 2018 12:23:24 +0000 (15:23 +0300)]
tests/check/Makefile.am: add GST_AUDIO_CFLAGS also on the planar audio adapter unit test
Nicolas Dufresne [Fri, 3 Aug 2018 12:16:19 +0000 (08:16 -0400)]
badaudio: Fix typo, ADUIO vs AUDIO
Nicolas Dufresne [Fri, 3 Aug 2018 11:57:50 +0000 (07:57 -0400)]
GstPlanarAudioAdapter: Add audio library in Makefile.am
This fixes a build regression.
Nicolas Dufresne [Fri, 3 Aug 2018 11:57:02 +0000 (07:57 -0400)]
webrtcdsp: Avoid redefinition of GST_USE_UNSTABLE_API
George Kiagiadakis [Mon, 19 Feb 2018 16:30:13 +0000 (18:30 +0200)]
webrtcdsp: add support for using F32/non-interleaved buffers
This is the native format that is in use by the webrtc audio processing
library internally, so this avoids internal {de,}interleaving and
format conversion (S16->F32 and back)
https://bugzilla.gnome.org/show_bug.cgi?id=793605
George Kiagiadakis [Mon, 19 Feb 2018 09:48:01 +0000 (11:48 +0200)]
GstPlanarAudioAdapter: copy pts, dts and offset tracking from GstAdapter
https://bugzilla.gnome.org/show_bug.cgi?id=793605
George Kiagiadakis [Fri, 16 Feb 2018 13:55:45 +0000 (15:55 +0200)]
libs: audio: add new GstPlanarAudioAdapter class
This is a GstAdapter, but for planar audio buffers.
https://bugzilla.gnome.org/show_bug.cgi?id=793605
Alessandro Decina [Fri, 21 Jul 2017 07:53:11 +0000 (17:53 +1000)]
meson: enable applemedia
Thibault Saunier [Fri, 3 Aug 2018 01:05:24 +0000 (21:05 -0400)]
rsvg: Also accept </svg:svg> as ending tag
Some SVG files created by inkscape use that, such as:
https://gitlab.gnome.org/GNOME/pitivi/uploads/
8dd8d9d988b5eb6cc38f871196caac6f/Titel-Tafel3.2_anim.svg
This is a first patch for https://bugzilla.gnome.org/show_bug.cgi?id=796909
but it is not enough to support animated gifs.
Nicolas Dufresne [Thu, 2 Aug 2018 15:19:07 +0000 (11:19 -0400)]
opencv: Fixup include to follow new standard
This uses the new path for OpenCV headers. OpenCV now have
master headers files per modules, which reduce the amount of
required includes. Note that HIGHGUI was included to get the
imgcodecs includes, which I fixed, though the master header is
missing the C headers, so I included that directly. All the
image stuff should be ported to C++ eventually. Finally, this
patch also update the header checks to reflect the modules that
are really being used.
Nicolas Dufresne [Thu, 2 Aug 2018 15:03:47 +0000 (11:03 -0400)]
opencv: Updated to use new header path
Nicolas Dufresne [Thu, 2 Aug 2018 14:39:14 +0000 (10:39 -0400)]
opencv: Bump requirement to 3.0.0+
And removes all the ifdef code to support the 2.X APIs.
Sebastian Dröge [Thu, 2 Aug 2018 13:18:12 +0000 (16:18 +0300)]
ceaccoverlay: Initialize debug category
Nirbheek Chauhan [Thu, 2 Aug 2018 12:54:02 +0000 (18:24 +0530)]
wasapisrc: Correctly disable provide-clock
`#ifdef` will, of course, evaluate to 1 in this case. We want `#if`.
Nicolas Dufresne [Thu, 2 Aug 2018 12:13:08 +0000 (08:13 -0400)]
cameracalibrtate: Fix opencv2 core.hpp path
This is to support older release of OpenCV2
https://bugzilla.gnome.org/show_bug.cgi?id=789075
Christoph Reiter [Sun, 27 May 2018 12:31:55 +0000 (14:31 +0200)]
wasapisink: fix regression in shared mode segment size
In commit
fd806628a8 (
839cc3926 in the stable branch) I changed the
segment size to match exactly the buffer size. I missed that this is
only valid in exclusive mode and in shared mode the buffer size is
a multiple of the device period.
Revert the logic for the shared mode.
https://bugzilla.gnome.org/show_bug.cgi?id=796354
https://bugzilla.gnome.org/show_bug.cgi?id=796858
Philippe Renon [Mon, 16 Oct 2017 20:29:01 +0000 (22:29 +0200)]
opencv: new cameracalibrate and cameraundistort elements
https://bugzilla.gnome.org/show_bug.cgi?id=789075
Vivia Nikolaidou [Wed, 1 Aug 2018 12:11:09 +0000 (15:11 +0300)]
avwait: Don't wait if audio_running_time_to_wait_for is NONE
The case is properly handled a few lines below by dropping the buffer.
We shouldn't perpetually block the audio chain function until the
target-timecode is reached.
https://bugzilla.gnome.org/show_bug.cgi?id=796906
Nirbheek Chauhan [Tue, 31 Jul 2018 22:13:31 +0000 (03:43 +0530)]
meson: Add option for winks plugin
This was accidentally missed.
Seungha Yang [Sun, 29 Jul 2018 11:55:26 +0000 (20:55 +0900)]
srt: Allow the host name "localhost"
Add support "srt://localhost:port" style uri, and change the
default host to "localhost"
https://bugzilla.gnome.org/show_bug.cgi?id=796842
Seungha Yang [Sat, 28 Jul 2018 06:59:04 +0000 (15:59 +0900)]
srtclientsink: Fix SRT socket option setting
SRTClientSink is sender
https://bugzilla.gnome.org/show_bug.cgi?id=796842
Seungha Yang [Fri, 27 Jul 2018 06:04:29 +0000 (15:04 +0900)]
srt: Do not ignore SRT socket error event
... and set SRT_EPOLL_IN flag if the client is not sender
(i.e., source element) since waiting readable event for
srt_recvmsg() makes more sense.
https://bugzilla.gnome.org/show_bug.cgi?id=796842
Seungha Yang [Fri, 27 Jul 2018 04:23:43 +0000 (13:23 +0900)]
srtbasesrc: Use GstBaseSrc's timestamp impl.
... instead of doing it ourselves. Otherwise, we should add more
logic here (such as checking GstClock and etc) which was already provided by
GstBaseSrc.
https://bugzilla.gnome.org/show_bug.cgi?id=796842
Seungha Yang [Wed, 25 Jul 2018 01:44:20 +0000 (10:44 +0900)]
srt: Add "const" keyword to a function argument
Given "passphrase" shouldn't be modified
https://bugzilla.gnome.org/show_bug.cgi?id=796842
Seungha Yang [Fri, 20 Jul 2018 07:02:26 +0000 (16:02 +0900)]
srt: Remove duplicated code for setting server socket
https://bugzilla.gnome.org/show_bug.cgi?id=796842
Seungha Yang [Sat, 28 Jul 2018 09:06:08 +0000 (18:06 +0900)]
srt: Fix signal args mismatch
client-added and client-{removed,closed} signals were defined
to be emitted with two arguments (socket fd and address)
https://bugzilla.gnome.org/show_bug.cgi?id=796842
Seungha Yang [Wed, 25 Jul 2018 08:36:03 +0000 (17:36 +0900)]
srtserversrc: Emit client-closed signal in _stop()
... instead of client-added
https://bugzilla.gnome.org/show_bug.cgi?id=796842
Seungha Yang [Wed, 25 Jul 2018 06:16:32 +0000 (15:16 +0900)]
srtclientsink: Do not leak string
https://bugzilla.gnome.org/show_bug.cgi?id=796842
Seungha Yang [Mon, 23 Jul 2018 05:28:05 +0000 (14:28 +0900)]
srtclient: Fix bind-port and rendez-vous property getter
https://bugzilla.gnome.org/show_bug.cgi?id=796842
Seungha Yang [Fri, 20 Jul 2018 05:05:32 +0000 (14:05 +0900)]
srt: Reduce useless call depth
Providing simplified _client_connect() is not required for now.
https://bugzilla.gnome.org/show_bug.cgi?id=796842
Seungha Yang [Fri, 20 Jul 2018 03:10:21 +0000 (12:10 +0900)]
srt: Remove white space
https://bugzilla.gnome.org/show_bug.cgi?id=796842
Nicolas Dufresne [Sat, 28 Jul 2018 03:00:53 +0000 (23:00 -0400)]
jpegparse: Generate timestamp if framerate is known
This change allow setting timestamp on streams that would otherwise have
no timestamp. This is useful to make a video from bunch of JPEG files. An
example of such pipeline would be:
gst-launch-1.0 multifilesrc location=%05d.jpeg caps=image/jpeg,framerate=30/1 \
! jpegparse ! fakesink silent=0 -v
Nicolas Dufresne [Sat, 28 Jul 2018 02:36:40 +0000 (22:36 -0400)]
meson: mjpegtools_api should be 20K for mplex too
This error was duplicated.
Nicolas Dufresne [Sat, 28 Jul 2018 02:33:47 +0000 (22:33 -0400)]
meson: mjpegtools_api should have been 20K not 2K
This was miss-ported when the options got added. The requires version was
properly set to 2000 before.
Nicolas Dufresne [Sat, 28 Jul 2018 02:17:56 +0000 (22:17 -0400)]
meson: Generate bluez.h instead of bluez-interface.h
This is a recent regression, the meson code was ported to use the meson
gnome module helpers, but the generated file was changed to
bluez-interface.h.
Nirbheek Chauhan [Fri, 27 Jul 2018 13:29:23 +0000 (18:59 +0530)]
Add feature options for almost all plugins
The only plugins remaining are those that haven't been ported to Meson
yet, and msdk. Also, the tests are still automagic.
https://bugzilla.gnome.org/show_bug.cgi?id=795107
Nirbheek Chauhan [Wed, 25 Jul 2018 11:42:00 +0000 (17:12 +0530)]
meson: Add feature options for many plugins
The rest will be converted later, these are necessary for gst-build to
set options correctly.
https://bugzilla.gnome.org/show_bug.cgi?id=795107
Olivier Crête [Thu, 26 Jul 2018 15:19:41 +0000 (11:19 -0400)]
av1enc: Implement latency setting
Olivier Crête [Thu, 26 Jul 2018 14:48:34 +0000 (10:48 -0400)]
av1enc: Propagate flags from encoder to GstBuffer
Olivier Crête [Thu, 26 Jul 2018 13:55:48 +0000 (09:55 -0400)]
av1enc: Implement finish for EOS
Call the encode function repeatedly until nothing comes out.
https://bugzilla.gnome.org/show_bug.cgi?id=796727
Olivier Crête [Thu, 26 Jul 2018 13:49:02 +0000 (09:49 -0400)]
av1enc: Return downstream error from push
Sebastian Dröge [Wed, 25 Jul 2018 21:20:02 +0000 (00:20 +0300)]
compositor: Don't leak all buffers while crossfading and not all pads are crossfading
Gary Bisson [Wed, 25 Jul 2018 11:44:48 +0000 (13:44 +0200)]
kmssink: Add support for mxsfb-drm driver
The mxsfb-drm driver has been added to the kernel long ago and will now
be the default display driver for NXP i.MX28, i.MX6SX and i.MX7D
processors so now is a good time to add it to kmssink.
Also, this is used in the upcoming i.MX8MQ and i.MX8MM processors.
https://bugzilla.gnome.org/show_bug.cgi?id=796873
Matteo Valdina [Tue, 10 Jul 2018 01:20:59 +0000 (20:20 -0500)]
waylandsink: relaxed wl_shell check and added zwp_fullscreen_shell.
Relaxed the wl_shell interface constrains, so application that
pass via GstContext the wl_surface can use waylandsink in a
compositor without wl_surface and zwp_fullscreen_shell.
Added support for zwp_fullscreen_shell.
https://bugzilla.gnome.org/show_bug.cgi?id=796772
Vivia Nikolaidou [Thu, 19 Jul 2018 15:34:40 +0000 (18:34 +0300)]
avwait: Add recording property
It works like a valve in front of the actual avwait. When recording ==
TRUE, other rules are then examined. When recording == FALSE, nothing is
passing through.
https://bugzilla.gnome.org/show_bug.cgi?id=796836
Sebastian Dröge [Fri, 20 Jul 2018 13:25:02 +0000 (16:25 +0300)]
compositor: Use 255 as maximum alpha instead of 256
255 will easily become 0 in the blending function as they expect
the maximum value to be 255.
Can be reproduce with
gst-launch-1.0 videotestsrc pattern=ball ! c.sink_0 \
videotestsrc pattern=snow ! c.sink_1 \
compositor name=c \
sink_0::zorder=0 sink_1::zorder=1 sink_0::crossfade-ratio=0.5 \
background=black ! \
videoconvert ! xvimagesink
crossfade-ratio +/- 0.001 makes it work correctly and the same happens
at e.g. 0.25, 0.75, N*0.0625
https://bugzilla.gnome.org/show_bug.cgi?id=796846
Jan Schmidt [Sun, 22 Jul 2018 00:34:06 +0000 (10:34 +1000)]
ladspa: Only scan LADSPA subdir APPDATA and COMMONPROGRAMFILES
Don't scan the entire APPDATA and COMMONPROGRAMFILES
path tree, just the LADSPA sub-directory inside them, if any.
https://bugzilla.gnome.org/show_bug.cgi?id=796450
Jan Schmidt [Sat, 21 Jul 2018 23:52:23 +0000 (09:52 +1000)]
ladspa: Don't try and load every file as a plugin
When scanning paths for LADSPA plugins, don't try and load
every random file as a module, as g_module_open ends up throwing
errors on Windows.
Use a G_MODULE_SUFFIX and GST_EXTRA_MODULE_SUFFIX suffix check as
we do for GStreamer plugins.
https://bugzilla.gnome.org/show_bug.cgi?id=796450
Sam Gigliotti [Wed, 18 Jul 2018 13:26:17 +0000 (14:26 +0100)]
webrtc: fix memory leak
When it parses SDP, it doesn't free the error object.
https://bugzilla.gnome.org/show_bug.cgi?id=796830
Seungha Yang [Wed, 18 Jul 2018 11:05:26 +0000 (20:05 +0900)]
compositor: Update conversion info in property setter
... not in getter. Otherwise, video-converter will not be updated
with new width/height
https://bugzilla.gnome.org/show_bug.cgi?id=796828
Jonathan Karlsson [Mon, 18 Jun 2018 11:01:50 +0000 (13:01 +0200)]
curlhttpsink: Support for multipart/form-data Content-Type
Add support to set multipart/form-data as Content-Type by sending a
caps event to the curlhttpsink with a field named "boundary".
https://bugzilla.gnome.org/show_bug.cgi?id=796618
Jan Schmidt [Sun, 15 Jul 2018 13:05:26 +0000 (23:05 +1000)]
webrtc: Add some ssrc to caps in the unit test.
The unit test uses incomplete caps to test webrtcbin,
causing some weirdness generating RTX stream mappings.
https://bugzilla.gnome.org/show_bug.cgi?id=796810
Jan Schmidt [Sat, 14 Jul 2018 13:15:02 +0000 (23:15 +1000)]
webrtc: Add a warning in sdp_media_from_transceiver()
When generating caps with no ssrc, at least throw a
warning instead of using an uninitialised stack variable
https://bugzilla.gnome.org/show_bug.cgi?id=796810
Jan Schmidt [Sat, 14 Jul 2018 13:05:53 +0000 (23:05 +1000)]
webrtc: Fix memory leak
Fix a leaked string when building RTX info.
Jan Schmidt [Sat, 14 Jul 2018 13:04:30 +0000 (23:04 +1000)]
webrtc: Clean up and fix transportsendbin
Refactor transportsendbin, and change the way
pads are blocked on dtlssrtpenc so that they
don't interfere with state changes.
As well as being easier to read, this fixes
spurious failures shutting down webrtcbin
if DTLS negotiation hasn't completed yet.
Jan Schmidt [Sat, 14 Jul 2018 13:02:05 +0000 (23:02 +1000)]
webrtc: Move dtlssrtpenc state management
Move the errant piece of dtlssrtpenc state change
management from dtlstransport in the Webrtc libs,
into the transportsendbin that does the rest of
the element management so it's all in one place.
Jan Schmidt [Mon, 2 Jul 2018 09:55:41 +0000 (19:55 +1000)]
webrtc/dtlstransport: Add more debug. Rename category
Rename the dtlstransport debug category to webrtcdtlstransport.
Thibault Saunier [Fri, 13 Jul 2018 18:42:28 +0000 (14:42 -0400)]
opencv: Fix build for opencv >= 3.4.2
The `CV_RGB` macro is now in `imgproc.hpp`.
Fixes:
../subprojects/gst-plugins-bad/ext/opencv/gsthanddetect.cpp:497:40: error: ‘CV_RGB’ was not declared in this scope
cvCircle (img, center, radius, CV_RGB (0, 0, 200), 1, 8, 0);
^~~~~~
Mathieu Duponchelle [Sat, 7 Jul 2018 01:40:49 +0000 (03:40 +0200)]
find_codec_preferences: use received caps
When negotiation is triggered by receiving caps on our sink pad
probes, we could encounter a race condition where need-negotiation
is emitted and the application requires the creation of an offer
before the current caps were actually updated.
This led to retrieving incomplete caps when creating the offer,
using find_codec_preferences -> pad_get_current_caps.
Instead, as we save the caps in the probe callback anyway, it is better
and thread safe to use these if they were set.
https://bugzilla.gnome.org/show_bug.cgi?id=796801
Matthew Waters [Thu, 12 Jul 2018 02:48:39 +0000 (12:48 +1000)]
glvideomixer: fix default placement when different sized output
i.e. when expanding from 320x240 to 800x600, the resulting frame should
appear in the top left corner, not the middle.
https://bugzilla.gnome.org/show_bug.cgi?id=794401
Daniel Klamt [Wed, 25 Apr 2018 14:36:21 +0000 (16:36 +0200)]
glvideomixer: Moves the objects to zero on z axis
Matches the output from a similar glimagesink pipeline when
rotating from an upstream gltransformation passed through
the affine transformation meta with xpos/ypos being set.
https://bugzilla.gnome.org/show_bug.cgi?id=794401
Nicolas Dufresne [Tue, 10 Jul 2018 13:45:31 +0000 (09:45 -0400)]
glvideomixer: Fix coding style
Nicolas Dufresne [Tue, 10 Jul 2018 13:42:47 +0000 (09:42 -0400)]
glvideomixer: Add missing string.h include
Matthew Waters [Sun, 8 Jul 2018 14:54:04 +0000 (09:54 -0500)]
decklink: keep the internal time we send to decklink always increasing
Otherwise decklink seems to hold onto the latest frame and may cause a
flash in the output if the played sequence has a framerate less than the
video output.
Scenario is play for N seconds, pause, flushing seek to some other position,
play again. Each of the play sequences would normally start at 0 with
the decklink time. As a result, the latest frame from the previous sequence
is kept alive waiting for it's timestamp to pass before either dropping
(if a subsequent frame in the new sequence overrides it) or displayed
causing the out of place frame to be displayed.
This is also supported by the debug logs from the decklink video sink
element where a ScheduledFrameCompleted() callback would not occur for
the frame until the above had happened.
It was timing related as to whether the frame was displayed based
on the decklink refresh cycle (which seems to be 16ms here),
when the frame was scheduled by the sink and the difference between
the 'time since vblank' of the two play requests (and thus start times
of scheduled playback).
Matthew Waters [Tue, 10 Jul 2018 10:12:15 +0000 (20:12 +1000)]
gl/utils: fix matrix conversion for column majorness
Companion to
https://cgit.freedesktop.org/gstreamer/gst-plugins-base/commit/?id=
e4bf9ed8f060021151cd57e2b00493ed696cb47b
https://cgit.freedesktop.org/gstreamer/gst-plugins-base/commit/?id=
16cba63d43f319ac33bc993eca2b1b51679727b3
Daniel Klamt [Wed, 25 Apr 2018 14:39:34 +0000 (16:39 +0200)]
Moved the pad offset and aspect ratio to a matrix so it will be added in view space and not in world space
https://bugzilla.gnome.org/show_bug.cgi?id=794401
Arun Raghavan [Mon, 9 Jul 2018 01:10:29 +0000 (06:40 +0530)]
closedcaption: Fix compilation on older compilers
For when -std=c99 (or later) is not the default.
Arun Raghavan [Mon, 9 Jul 2018 01:10:24 +0000 (06:40 +0530)]
bluez: Expose transport volume as a property on avdtpsrc
This can be handy if we want to track Bluetooth volume changes from the
device, or if we want to set it.
https://bugzilla.gnome.org/show_bug.cgi?id=787020