Matthew Waters [Thu, 13 Jun 2019 07:05:44 +0000 (17:05 +1000)]
vulkan: add download element
Currently only downloads images into a host-visible buffer and
synchronises immediately.
Matthew Waters [Mon, 10 Jun 2019 16:05:32 +0000 (02:05 +1000)]
vulkan/identity: Split out most rendering code to a base class
A simple base class that renders a 2d fullscreen quad parallel to the
screen surface inside the view frustum.
Matthew Waters [Wed, 12 Jun 2019 08:25:00 +0000 (18:25 +1000)]
vkupload: fix a structure sType
Matthew Waters [Tue, 11 Jun 2019 08:37:59 +0000 (18:37 +1000)]
vulkan: ensure initialization of a couple of debug categories
Needed when some of the context querying functions can be called before
an instance has been created.
Matthew Waters [Tue, 21 May 2019 07:19:00 +0000 (17:19 +1000)]
vksink: Retrieve vulkan queue earlier
Allows using the swapper's queue over upstream's queue. The swapper
will check for the necessary presentation support that upstream may not
consider.
Sebastian Dröge [Wed, 19 Jun 2019 11:13:02 +0000 (14:13 +0300)]
mpegvideoparse: Pass through interlace-mode field from upstream if available
We generally always prefer the information from upstream for other
metadata (pixel-aspect-ration, etc.) and should also do so here.
Other parsers (h264parse) already do the same.
Sebastian Dröge [Wed, 19 Jun 2019 10:31:39 +0000 (13:31 +0300)]
zbar: Include running-time, stream-time and duration in the messages
The timestamp/PTS alone is meaningless without the segment and usually
applications care about the running-time or stream-time.
This also keeps the messages in sync with the spectrum and level
elements.
Seungha Yang [Sat, 4 Nov 2017 11:40:18 +0000 (20:40 +0900)]
tests: hls: Add a test case for EXT-X-MAP tag
https://bugzilla.gnome.org/show_bug.cgi?id=776928
Seungha Yang [Sun, 12 May 2019 10:21:23 +0000 (19:21 +0900)]
tests: Enable hls m3u8 unit test with meson build
Seungha Yang [Sat, 4 Nov 2017 11:39:39 +0000 (20:39 +0900)]
hlsdemux: Set fragment header uri if exists
To allow downloading fragment hearder, set its uri if there is
available "Media Initialization" parsed from EXT-X-MAP tag
https://bugzilla.gnome.org/show_bug.cgi?id=776928
Seungha Yang [Sat, 4 Nov 2017 11:15:33 +0000 (20:15 +0900)]
hls: m3u8: Parsing EXT-X-MAP tag to store initialization data
EXT-X-MAP tag informs media initialization data,
such as moov box in ISOBMFF case and PAT/PMT for MPEG TS stream.
https://bugzilla.gnome.org/show_bug.cgi?id=776928
Seungha Yang [Mon, 30 Jan 2017 05:13:06 +0000 (14:13 +0900)]
hlsdemux: Do not clear/advance fragment by finished header downloading
Header data must be forwarded to downstream, but if demux does not finish
to finding type (e.g., ts, mp4 and etc), this header data can be cleared
by _stream_clear_pending_data(). Moreover, although demux finish downloading
header data, still it has fragment date to be downloaded, fragment sequence
shouldn't be advanced yet at that moment.
https://bugzilla.gnome.org/show_bug.cgi?id=776928
Seungha Yang [Tue, 19 Feb 2019 12:46:43 +0000 (21:46 +0900)]
tests: insertbin: Don't use duplicated variable name
../subprojects/gst-plugins-bad/tests/check/libs/insertbin.c(46): warning C4273:
'mutex': ...
The "mutex" was declared in gstcheck.h already
Matthew Waters [Mon, 17 Jun 2019 04:58:50 +0000 (14:58 +1000)]
applemedia: add the '-fobjc-arc' for darwin/ios
We use ARC so we need to signal that to the compiler.
Thibault Saunier [Mon, 3 Jun 2019 16:18:13 +0000 (12:18 -0400)]
uvch264: Implement device provider
Thibault Saunier [Mon, 3 Jun 2019 16:17:22 +0000 (12:17 -0400)]
uvch264: Factor out checking if v4l2device is uvc compatible
Nicola Murino [Mon, 8 Apr 2019 17:24:00 +0000 (19:24 +0200)]
h265parse: update parser state and header flag when using fallback sps
When sps parsing fails we use a fallback sps from the caps, since we
have got an sps we need to update parser state and header as in the case the
sps was successfully parsed
Nicola Murino [Mon, 26 Nov 2018 16:23:21 +0000 (17:23 +0100)]
h264parse: update parser state and header flag when using fallback sps
When sps parsing fails we use a fallback sps from the caps, since we
have got an sps we need to update parser state and header as in the case the
sps was successfully parsed
Closes #503
Thibault Saunier [Fri, 31 May 2019 14:12:54 +0000 (10:12 -0400)]
h264parse: Post a WARNING when data is broken
Seungha Yang [Wed, 12 Jun 2019 06:47:52 +0000 (15:47 +0900)]
h265parse: Add more string representations of extension profiles
Seungha Yang [Wed, 12 Jun 2019 06:32:17 +0000 (15:32 +0900)]
h265parser: Add more profiles to known type
"High Throughput", "Multiview", "Scalable", "3D", "Screen Content Coding",
and "Scalable format range extensions" profiles can be supported
via h265parser APIs now.
Dong Il Park [Thu, 18 Apr 2019 09:12:34 +0000 (18:12 +0900)]
h265parse: Update framerate when we found vps_timing_info
The timing_info was described at vps or vui parameter.
So we can update the framerate field of GstCaps when we could
parse vps_timing_info parameters.
Dong Il Park [Thu, 18 Apr 2019 07:54:51 +0000 (16:54 +0900)]
h265parser: Use vps_timing_info when not present in vui
The same timing_info will be present at vps or vui.
When the timeing_info is present in the VPS, vui_timing_info
, when present, shall be equal to vps_timing_info, and when
not present, is inferred to be equal to vps_timing_info.
Seungha Yang [Sat, 8 Jun 2019 13:57:21 +0000 (22:57 +0900)]
vulkan: Add support WIN32 for Windows
It's almost a fork of glwindow_win32 implementation.
To build on Windows, Vulkan SDK (at https://vulkan.lunarg.com/sdk/home)
and VK_SDK_PATH environment are required. Note that VK_SDK_PATH environment
setting is a part of the SDK installation.
Arun Raghavan [Wed, 12 Jun 2019 08:12:14 +0000 (10:12 +0200)]
avdtpsrc: Honour initial transport volume setting before connection
We make the binding go from the avdtpsrc to the transport, so that any
initial setting before a connection made is used.
Thibault Saunier [Tue, 11 Jun 2019 16:10:13 +0000 (12:10 -0400)]
uvch264src: Make sure we set our segment
We were not setting self->segment and we are using it
when notifying downstream that we handled a REQUEST_KEY_UNIT
event, leading to all sort of criticals.
Seungha Yang [Tue, 11 Jun 2019 05:28:22 +0000 (14:28 +0900)]
h265parse: Don't miss constraint indicator flags in codec data
Set more unhandled flags to general_constraint_indicator_flags field.
The field is required for building "Codecs" parameter as defined
ISO/IEC 14496-15 Annex E. The resulting "Codecs" string might be used
in various places (e.g., HLS/DASH manifest, browser, player, etc)
Marc Leeman [Fri, 7 Jun 2019 14:12:25 +0000 (14:12 +0000)]
rtpmanagerbad: fix the plugin registration
After compilation, the compiled library needs to be added to the list
of plugin libraries.
.
Also, fix for static builds
Tim-Philipp Müller [Thu, 6 Jun 2019 17:22:43 +0000 (18:22 +0100)]
Remove VDPAU plugin
It's been replaced by NVENC/NVDEC and even NVIDIA doesn't
support VDPAU any longer and hasn't for quite some time.
The plugin has been unmaintained and unsupported for a very
long time, and given the track record over the last 10 years
it seems highly unlikely anyone is going to make it work well,
not to mention adding plumbing for proper zero-copy or
gst-gl integration.
Closes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/issues/828
Tim-Philipp Müller [Thu, 6 Jun 2019 17:36:00 +0000 (18:36 +0100)]
autotools: fix distcheck
Tim-Philipp Müller [Wed, 5 Jun 2019 16:10:19 +0000 (17:10 +0100)]
autotools: remove gst/rtp/ from cruft dir list
Tim-Philipp Müller [Wed, 5 Jun 2019 15:58:32 +0000 (16:58 +0100)]
rtp: fix autotools build some more
Philippe Normand [Wed, 5 Jun 2019 11:47:16 +0000 (12:47 +0100)]
wpe: Fix build with -Werror enabled
Including gl.h from WPEThreadedView.h leads to GST_LEVEL_DEFAULT detected as
redefined. The proposed fix is to include config.h from the CPP implementation
file and disable gl.h inclusion in the header, by using forward declarations.
Sebastian Dröge [Wed, 5 Jun 2019 08:46:49 +0000 (11:46 +0300)]
avwait: Allow start and end timecode to be set back to NULL
And check everywhere if they're NULL before accessing them.
Niels De Graef [Wed, 5 Jun 2019 06:12:10 +0000 (08:12 +0200)]
Use G_DEFINE_AUTOPTR_CLEANUP_FUNC unconditionally
Since we started depending on GLib 2.44, we can be sure this macro is
defined (it will be a no-op on compilers that don't support it).
Matthew Waters [Tue, 21 May 2019 04:04:07 +0000 (14:04 +1000)]
tests/vulkan: allow instance creation failures
Usually means that there is no vulkan implementation available.
We skip tests in that case
Matthew Waters [Mon, 20 May 2019 03:54:56 +0000 (13:54 +1000)]
test/vulkan: add simple memory test
Matthew Waters [Mon, 20 May 2019 03:48:27 +0000 (13:48 +1000)]
vkupload: remove debugging error log
Matthew Waters [Mon, 20 May 2019 03:46:56 +0000 (13:46 +1000)]
vulkan: implement the correct memory barriers necessary
Vulkan validation layers are finally silent!
Matthew Waters [Thu, 16 May 2019 12:39:51 +0000 (22:39 +1000)]
vulkan: remove uneeded buffer inheritence information
buffer inheritence informationi is only required for secondary command
buffers which we are not creating
Matthew Waters [Thu, 16 May 2019 09:48:24 +0000 (19:48 +1000)]
vulkan: add a new image copying element
Copies using the graphics pipeline, an input image and renders to an
output image.
Matthew Waters [Thu, 16 May 2019 09:39:20 +0000 (19:39 +1000)]
vkmemory: fix allocation of sizes less than the alignment
Avoids allocating a 0-sized device memory which is invalid vulkan usage
Matthew Waters [Thu, 16 May 2019 09:38:35 +0000 (19:38 +1000)]
vk*pool: expand usage hints to cover use in a renderpass
Matthew Waters [Thu, 16 May 2019 09:36:18 +0000 (19:36 +1000)]
vkformat: fix format conversion for matching vulkan formats
Other formats still need a swizzle or conversion.
Matthew Waters [Thu, 16 May 2019 09:35:11 +0000 (19:35 +1000)]
vkbuffermemory: bufferview's are only for texture-like memory
Matthew Waters [Thu, 16 May 2019 09:34:41 +0000 (19:34 +1000)]
vkupload: add raw->vulkanimage uploader
Matthew Waters [Thu, 16 May 2019 09:33:06 +0000 (19:33 +1000)]
vktrash: add more destruction of vulkan types
Matthew Waters [Thu, 9 May 2019 11:50:47 +0000 (21:50 +1000)]
vulkan: use c99 designated initializers where possible
Makes the code much easier to read and allows removing our terrible
macros for structure initialization.
Matthew Waters [Wed, 8 May 2019 11:26:17 +0000 (21:26 +1000)]
vkswapper: support displaying memory:VulkanImage
Matthew Waters [Wed, 8 May 2019 04:50:29 +0000 (14:50 +1000)]
vkupload: implement buffer to image uploader
Matthew Waters [Wed, 8 May 2019 04:48:43 +0000 (14:48 +1000)]
vkqueue: unly unref the context query if it succeeds
Matthew Waters [Wed, 8 May 2019 04:47:43 +0000 (14:47 +1000)]
vulkansink: Respond to queue queries
Matthew Waters [Thu, 11 Apr 2019 06:52:54 +0000 (16:52 +1000)]
vulkan: large docs update
Matthew Waters [Wed, 10 Apr 2019 04:27:26 +0000 (14:27 +1000)]
vulkan: add a command pool object for tracking
Matthew Waters [Wed, 10 Apr 2019 01:56:59 +0000 (11:56 +1000)]
vkswapper: don't output duplicate formats
Matthew Waters [Mon, 8 Apr 2019 11:35:22 +0000 (21:35 +1000)]
vulkan: make a gstvulkan library out of the existing API
Matthew Waters [Mon, 8 Apr 2019 08:43:50 +0000 (18:43 +1000)]
vkbufferpool: Fix multiplaner allocations
Use the plane width/height and the sizes required by vulkan
Fixes allocation of:
videotestsrc ! video/x-raw,format=I420 ! vulkanupload ! fakesink
Matthew Waters [Mon, 8 Apr 2019 08:42:12 +0000 (18:42 +1000)]
vk*memory: explicitly error out for driver NPOT alignment
Matthew Waters [Thu, 4 Apr 2019 08:30:28 +0000 (19:30 +1100)]
vulkan/image: initialize the requirements struct before using it
Matthew Waters [Thu, 4 Apr 2019 08:27:55 +0000 (19:27 +1100)]
vulkan: don't require every element to have a display
Only sink elements really care about a valid display
Matthew Waters [Thu, 4 Apr 2019 07:38:13 +0000 (18:38 +1100)]
vulkan: remove unused X11 window system references
We use XCB instead
Matthew Waters [Thu, 4 Apr 2019 06:33:34 +0000 (17:33 +1100)]
vulkan: remove unused layer enablement
This is possible now via the vulkan loader
Ali Yousuf [Mon, 20 May 2019 11:34:55 +0000 (16:34 +0500)]
webrtc: Fix log when adding stun server
Matthew Waters [Tue, 4 Jun 2019 05:35:24 +0000 (15:35 +1000)]
webrtc: fix the location of signalling-state change notification
1. The spec indicates that the notification should occur near the end of
'setting the description' processing
2. The current location with the drop of the lock could cause the 'check
if negotiation is needed' logic to execute and become confused about
the state of the webrtcbin's current local descriptions.
In the bad case, the following assertions could be hit:
g_assert (trans->mline < gst_sdp_message_medias_len (webrtc->current_local_description->sdp));
g_assert (trans->mline < gst_sdp_message_medias_len (webrtc->current_remote_description->sdp));
Moving the signalling state change later in the set description task
means that checking for a renegotiation will early abort as the
signalling state is not STABLE before the session description and
transceivers have been updated.
Nicolas Dufresne [Tue, 4 Jun 2019 00:29:18 +0000 (20:29 -0400)]
make: rtp: Remove spurious header file
This header file no longer exist.
Marc Leeman [Mon, 3 Jun 2019 20:08:23 +0000 (20:08 +0000)]
rtpmanagerbad: add RTP streaming elements
This is a re-implementation of the RTP elements that are submitted in
2013 to handle RTP streams. The elements handle a correct connection
for the bi-directional use of the RTCP sockets.
https://bugzilla.gnome.org/show_bug.cgi?id=703111
The rtpsink and rtpsrc elements add an URI interface so that streams
can be decoded with decodebin using the rtp:// interface.
The code can be used as follows
```
gst-launch-1.0 videotestsrc ! x264enc ! rtph264pay config-interval=3 ! rtpsink uri=rtp://239.1.1.1:1234
gst-launch-1.0 videotestsrc ! x264enc ! rtph264pay config-interval=1 ! rtpsink uri=rtp://239.1.2.3:5000
gst-launch-1.0 rtpsrc uri=rtp://239.1.2.3:5000?encoding-name=H264 ! rtph264depay ! avdec_h264 ! videoconvert ! xvimagesink
gst-launch-1.0 videotestsrc ! avenc_mpeg4 ! rtpmp4vpay config-interval=1 ! rtpsink uri=rtp://239.1.2.3:5000
gst-launch-1.0 rtpsrc uri=rtp://239.1.2.3:5000?encoding-name=MP4V-ES ! rtpmp4vdepay ! avdec_mpeg4 ! videoconvert ! xvimagesink
```
rtpmanagerbad: add pkg-config
rtpmanagerbad: Rtp should be uppercase
rtpmanagerbad: add G_OS_WIN32 for shielding unix headers
rtpmanagerbad: remove Since from documentation
rtpmanagerbad: rename lib name from nrtp to rtpmanagerbad
rtpmanagerbad: sync meson.build with other modules
rtpmanagerbad: add Makefile.am
rtpmanagerbad: use GstElement to count pads
rtpmanagerbad: use gst_bin_set_suppressed_flags
rtpmanagerbad: check element creation
rtpmanagerbad: post message when trying to access missing rtpbin
rtpmanagerbad: return FALSE with g_return tests
rtpmanagerbad: use gsocket multicast check
rtpmanagerbad: use gst_caps_new_empty_simple iso gst_caps_from_string
rtpmanagerbad: sync with gstrtppayloads.h
rtpmanagerbad: correct media type X-GST
rtpmanagerbad: test if a compatible pad was found
rtpmanagerbad: remove evil copy of GstRTPPayloadInfo
rtpmanagerbad: add gio_dep to meson
rtpmanagerbad: revert to old glib boilerplate
GStreamer 1.16 does not yet support the newer GLib templates, so revert.
rtpmanagerbad: return GST_STATE_CHANGE_NO_PREROLL for live sources
for live sources, NO_PREROLL should be returned for PLAYING->PAUSED and
READY->PAUSED transitions.
rtpmanagerbad: use GstElement pad counting
rtpmanagerbad: just use template name to request pad
rtpmanagerbad: remove commented code
rtpmanagerbad: use funnel to send multiple streams on one socket
rtpmanagerbad: avoid beaches
beaches should only be used during the summer, so rewrite the code to
return explicitly and avoid beaches during the winter.
rtpmanagerbad: add copyright to test code
rtpmanagerbad: g_free is NULL safe
rtpmanagerbad: do not trace rtpbin
rtpmanagerbad: return NULL explitly
rtpmanagerbad: warn when data port is not even
According to RFC 3550, RTP data should be sent on even ports, while RTCP
is sent on the following odd port.
rtpmanagerbad: document port allocation in rtpsink/src
rtpmanagerbad: improve uri description
rtpmanagerbad: add comment re-use socket
rtpmanagerbad: rename gst_object_set_properties_from_uri_query
rtpmanagerbad: loan prop/val setter from rist
rtpmanagerbad: rtpsrc: fix unitialised pointer
rtpmanagerbad: fix silly typo
rtpmanagerbad: test for empty key/value
rtpmanagerbad: rtpsrc: deprecate ssrc collision to INFO
rtpmanagerbad: sync debug with rist
rtpmanagerbad: small strings allocated on stack
rtpmanagerbad: correct rename
rtpmanagerbad: add locking on prop setters/getters
Locking is added because the URI allows to access the properties too.
rtpmanagerbad: allow for RTCP through NAT
rtpmanagerbad: move gio to header file
rtpmanagerbad: free small strings too
rtpmanagerbad: ttl_mc for ttl on dynudpsink
rtpmanagerbad: add comments on the URI registered
rtpmanagerbad: correct macro after file rename
rtpmanagerbad: code style
rtpmanagerbad: handle wrong URIs in setter
rtpmanagerbad: nit URI notation correction
In an URI, the first key/value pair should not have an ampersand, the
parser did not die though.
Niels De Graef [Fri, 31 May 2019 21:06:33 +0000 (23:06 +0200)]
meson: Bump minimal GLib version to 2.44
This means we can use some newer features and get rid of some
boilerplate code using the G_DECLARE_* macros.
As discussed on IRC, 2.44 is old enough by now to start depending on it.
Alex Ashley [Tue, 11 Dec 2018 14:43:21 +0000 (14:43 +0000)]
tests/dash_mpd: take account of Period start in expected timestamps
The start of each segment is relative to the Period start, minus
the presentation time offset.
As specified in section 5.3.9.6 of the MPEG DASH specification:
The value of the @t attribute minus the value of the
@presentationTimeOffset specifies the MPD start time of
the first Segment in the series.
Several tests use a Period@start value of 10 seconds, which either
needs to be taken into account when calculating expected timestamps
or have that attribute removed.
This commit uses a mix of updating the timestamps and removing the
start attribute, so that both the case of its presence and absence
is tested.
Alex Ashley [Tue, 16 Oct 2018 15:57:30 +0000 (16:57 +0100)]
dashdemux: include both Period start and presentationTimeOffset in segment start
The start of each segment is relative to the Period start, minus
the presentation time offset.
As specified in section 5.3.9.6 of the MPEG DASH specification:
The value of the @t attribute minus the value of the
@presentationTimeOffset specifies the MPD start time of
the first Segment in the series.
dashdemux was not taking account of presentationTimeOffset and in
some methods was not taking into account the Period start time.
This commit modifies the segment->start value to always be
relative to the MPD start time (zero for VOD,
availabilityStartTime for live streams). This makes all uses of
the segment list consistent.
Fixes #841
Mathieu Duponchelle [Fri, 31 May 2019 23:42:51 +0000 (01:42 +0200)]
docs: document gstreamer-bad-audio
And unprefix subproject paths, making a special case for
webrtc, to not conflict with the webrtc plugin
Mathieu Duponchelle [Fri, 31 May 2019 23:42:06 +0000 (01:42 +0200)]
libs: build a gir file for gstreamer-bad-audio
Vivia Nikolaidou [Tue, 2 Oct 2018 08:25:14 +0000 (11:25 +0300)]
avwait: Don't print warnings for every buffer passed
Haihao Xiang [Fri, 31 May 2019 06:09:40 +0000 (14:09 +0800)]
msdk: return a right pointer for VUYA format
The first channel in memory is V for VUYA format, note
GST_VIDEO_FORMAT_VUYA is mapped to MFX_FOURCC_AYUV in this plugin
Tim-Philipp Müller [Wed, 29 May 2019 16:05:34 +0000 (17:05 +0100)]
docs: update plugin doc cache and add more plugins
Tim-Philipp Müller [Wed, 29 May 2019 10:28:00 +0000 (11:28 +0100)]
meson: add more plugins to plugins list
Makes sure their path gets added to the uninstalled environment
and makes sure they get included in the docs.
Mathieu Duponchelle [Wed, 22 May 2019 19:44:44 +0000 (21:44 +0200)]
basetsmux: don't reset pad on flush_stop
This was mistakenly added when porting to aggregator, this
restores the old behaviour, by only resetting them when the
muxer itself is reset
Mathieu Duponchelle [Wed, 22 May 2019 19:44:22 +0000 (21:44 +0200)]
mpegtssection: events don't necessarily have a structure
Mathieu Duponchelle [Wed, 22 May 2019 19:44:02 +0000 (21:44 +0200)]
basetsmux: fix send_event by chaining up
Mathieu Duponchelle [Wed, 29 May 2019 16:27:09 +0000 (18:27 +0200)]
mpegtsmux: add SECTION comment
We include an example for injecting sections in the transport
stream in the documentation
Mathieu Duponchelle [Wed, 29 May 2019 15:03:59 +0000 (17:03 +0200)]
tests: add example for injecting MPEG-TS sections
Mathieu Duponchelle [Wed, 29 May 2019 15:01:40 +0000 (17:01 +0200)]
basetsmux: preserve user-specified sections across resets
As sections can be provided by the user through send_event
when the element state is NULL, their lifetime is expected
to match that of the muxer, and they must be preserved when
the state changes
Mathieu Duponchelle [Wed, 22 May 2019 23:44:19 +0000 (01:44 +0200)]
atscmux: send empty RRT / MGT / STT tables
These are mandated by A/65, their absence gets flagged by
stream analyzers. Users can of course provide filled up
versions through the send_event API.
Mathieu Duponchelle [Wed, 22 May 2019 22:03:18 +0000 (00:03 +0200)]
tsmux: maintain packet counters in a global array
We can have multiple TsMuxPacketInfo objects for the same PID
with user-provided sections, for example ATSC requires multiple
tables with the same PID.
Mathieu Duponchelle [Tue, 21 May 2019 00:53:20 +0000 (02:53 +0200)]
mpegts: extend support for ATSC tables
Adds constructors for the following sections:
STT: System Time Table
MGT: Master Guide Table
RRT: Rating Region Table
Also adds parsing code for RRT
Matthew Waters [Fri, 24 May 2019 14:02:05 +0000 (00:02 +1000)]
webrtc: only set sctp ports if they are different
SCTPassociation will complain if we do that while running and resetting
is not something we support at the moment
Matthew Waters [Thu, 7 Mar 2019 13:39:59 +0000 (00:39 +1100)]
tests/webrtc: wait until the SDP has been set before continuing
If we renegotiate, then it is currently possible for an added stream to
be added to webrtcbin before the SDP is complete. This causes an
internal inconsistency as there is a 'pending sink transceiver' without
a corresponding media section in the sdp. It also does not have an
associated transport stream and will fail in _connect_input_stream().
Matthew Waters [Thu, 7 Mar 2019 13:37:39 +0000 (00:37 +1100)]
tests/webrtc: fix racy test with a prenegotiated data channel
If both data channels become ready simultaneously, then the two integer
read-add-update cycles can execute concurrently and only ever increment
once instead of the required twice. Use an atomic add instead.
Matthew Waters [Thu, 7 Mar 2019 13:32:53 +0000 (00:32 +1100)]
webrtc/dc: move some code from webrtcbin into the datachannel
Matthew Waters [Mon, 4 Mar 2019 10:44:33 +0000 (21:44 +1100)]
webrtc: hold onto any unknown ICE candidates until the next SDP set
It is very possible for badly behaving signalling or peers to send
us ICE candidates before we receive an SDP. While we had consideration
for that on the first set SDP, subsequent SDP's could result in
misconfigured ICE transports. Expand the previous code to also take
into account reconfigurations.
Matthew Waters [Wed, 28 Nov 2018 06:23:31 +0000 (17:23 +1100)]
webrtc: Initial support for stream addition/removal
Limitations:
- No transport changes at all (ICE, DTLS)
- Codec changes are untested and probably don't work
- Stream removal doesn't remove transports (i.e. non-bundled transports
will stay around until webrtcbin is shutdown)
- Unified Plan SDP only. No Plan-B support.
Matthew Waters [Wed, 5 Dec 2018 07:36:27 +0000 (18:36 +1100)]
tests/webrtc: a couple of debug/error string fixes
Matthew Waters [Tue, 4 Dec 2018 11:49:05 +0000 (22:49 +1100)]
tests/webrtc: rewrite bundle checks for separate validate_sdp passes
Improves reusability
Matthew Waters [Tue, 4 Dec 2018 09:38:19 +0000 (20:38 +1100)]
tests/webrtc: add helper for getting the offer/answer element
Matthew Waters [Tue, 4 Dec 2018 09:32:56 +0000 (20:32 +1100)]
tests/webrtc: only check audio/video for direction attributes
Matthew Waters [Thu, 6 Dec 2018 12:28:18 +0000 (23:28 +1100)]
webrtcbin: expose the transceiver as a pad property
Matthew Waters [Thu, 6 Dec 2018 12:25:54 +0000 (23:25 +1100)]
webrtc/transceiver: add a set_direction function
Matches the setDirection() from the W3C spec and allows changing the
transceiver direction at the next negotiation cycle.
Matthew Waters [Wed, 5 Dec 2018 08:17:45 +0000 (19:17 +1100)]
webrtc: track and log more rtpbin state
like bye's timeouts, validation, activation, etc
Matthew Waters [Thu, 6 Dec 2018 05:59:52 +0000 (16:59 +1100)]
webrtc: add support for intersecting inactive transceiver directions
Matthew Waters [Tue, 4 Dec 2018 09:30:49 +0000 (20:30 +1100)]
webrtc: mark remote/local-description as readonly
Matthew Waters [Tue, 27 Nov 2018 08:27:27 +0000 (19:27 +1100)]
webrtc: don't reuse stopped transceivers at all