Otherwise when changing the target caps (e.g. for reducing quality)
there is a race condition between buffers between the converter elements
and renegotiation.
For example, videoconvertscale might've output a 1920x1080 buffer, then
the capsfilter is configured to 1280x720, the buffer arrives in
videorate, videorate notices that renegotiation is pending, tries to
renegotiate and ends up with EMPTY caps because it can only change the
framerate but not the resolution.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1949>
It is possible that in unprepare(), waiting for a task to complete while
holding the state lock, that task may be waiting to acquire the state lock and
result in a deadlock.
This is quick to reproduce when starting and stopping webrtcsink in very quick
succession.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1931>
The GstWebRTC API web interface defaults to receiving an SDP offer and
generating an answer, but this can be overridden by entering "offer
options" before clicking to open the remote stream. The Python
webrtcsink-custom-signaller.py example failed in this mode as it was
coded to only generate an offer and receive an answer. Fix this by
implementing support for receiving an offer and sending an answer.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1883>
The InPlace/Taken logic was introduced to avoid using an extra lock
around the session, but it places expectations that are not always
obvious to meet around when a session is expected to be taken or not.
Any code that expects to have access to the sessions at all times thus
needs either extra logic in the session wrapper, or to maintain the
state of the session outside of the session (eg mids).
This commit removes the logic, and wraps sessions in Arc<Mutex>>.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1852>
Since 6a23ae168f, the chain function
of webrtcsink adds a custom meta on input buffers.
That custom meta was registered only by the class_init of the subclasses
of BaseWebRTCSink, but the custom signaller example uses
BaseWebRTCSink::with_signaller() directly.
Fix by registering the meta in BaseWebRTCSink::class_init()
Fixes: #610
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1854>
When congestion control is used for a session with multiple encoders,
the default implementation simply divides the overall bitrate equally
between encoders.
This is not always desirable, and this patch exposes a new signal
that users can register to, with two arguments:
* The overall bitrate to allocate
* A structure with an encoder.stream_name -> bitrate mapping
Handlers should return a similar structure with a custom mapping.
An example is also provided.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1792>
RFC7273 related attributes are set in the SDP offer by passing them via the
transceiver `codec-preferences` signal. These attributes are intended to be set
at the media level so they must be prefixed by `a-` in the `Caps` argument to
the signal. Otherwise they end up under `a=fmtp`.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1810>
It may be desirable for the frontend to receive ancillary information
over the control channel.
Such information includes but is not limited to time code metas, support
for other metas (eg custom meta) might be implemented in the future, as
well as downstream events.
This patch implements a new info message, probes buffers that arrive at
nicesink to look up timecode metas and potentially forwards them to the
consumer when the `forward-metas` property is set appropriately.
Internally, a "dye" meta is used to trace the media identifier the
packet we are about to send over relates to, as rtpfunnel bundles all
packets together.
The example frontend code also gets a minor update and now logs info
messages to the console.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-rs/-/merge_requests/1749>
