It gets resolved during translation.

Fixes: https://github.com/jitsi/lib-jitsi-meet/issues/2055

* feat: Add videoType to sources signaled in jingle.

* fix(SDP): don't separate fmtp parameter with spaces

* adjusted tests accordingly to not expect spaces anymore

Reusing(removing it and adding it again) an SSRC in the
remote SDP is leading to the issue where remote video is received but
not rendered. That's why we shouldn't remove the receive only SSRCs
from the SDP. This way the sender won't send any source-remove and
source-add for the same SSRCs.

Jicofo by default now strips the simulcast SSRCs and sends only the primary SSRC and RTX to all the receiving endpoints in the call. Therefore remote SDP munging is not required on the clients anymore.

When source-name signaling is enabbled on plan-b clients, the sources to MSID map needs to be cleared after every transformation since browser in plan-b mode produces a new set of SSRCs and trackID when a track is removed and new track is added to the peerconnection. This fixes an issue where the track index in the source-name gets incremented after a track replace operation.

Do this even if the browser provides a non '-' string for streamId since we rely on this to generate the source name.

* fix(multi-stream-support) Support muting of desktop track.

* fix(multi-stream) Always initiate responder renegotiation even for p2p.
This is needed since the new m-line is always added to the remote description.

* fix(multi-stream) Support multi-stream only on Unfied plan endpoints.

* fix(multi-stream) Advertise source-name signaling to Jicofo when supported.

* fix(JitsiLocalTrack): Remove unnecessary reject.

* fix: add the correct source name attribute for the second video track.

* squash: Address review comments.

* feat(multi-stream-support) Add the support for multiple local video streams.
This feature is behind 'sendMultipleVideoStreams' config.js flag and is currently supported only on clients running in Unified plan mode.

* squash: Use the track index from source name to find the matching transceiver on unmute.

* feat(multi-stream-support) Handle SDP munging for multiple local/remote streams/ep.
A new SDPSimulcast class is added for handling SDP munging for local and remote descriptions to handle multiple streams for both local and remote endpoints. This new class will be used only for unified plan implentation. The sdp-simucast npm package will be used for plan-b and deprecated later when all the clients switch to unified plan.

* fix(build) fix building with TS code

Use the Babel preset for the webpack bundle since Babel 7 already
understands TS and ts-loader uses tsc instead, which we configure
differently.

* squash: Add unit test and address review comments.

Co-authored-by: Saúl Ibarra Corretgé <saghul@jitsi.org>

* fix(sdp): provide SCTP streams, because the XMPP parser expects them

* fixed linting error

This PR switches to generating sctp-port based SDP.
It also adds support for parsing sctp-port, but can handle
fallback to old-style sctpmap as well.
Fixes: #1778

the protocol was switched back in Chrome 44 in 2016:
  https://bugs.chromium.org/p/webrtc/issues/detail?id=2796
the port switch more than five years ago too.

...to advertise track's muted state and the video type.

For now, if the source name signaling flag is enabled, both legacy and
new <SourceInfo> element will be used at the same time. This is to be
able to interoperate with legacy clients and start testing the new
format at the same time. Whenever possible <SourceInfo> will be used
as main source of truth with the fallback to legacy <audiomuted/>,
<videomuted/> and <videotype/> elements.

The issue was caused by incorrect ordering of the mixedmslabel mline.
sdp-interop is expecting that this mline was the first one but in
reality we were placing it last. This was causing the direction of a
user related m line to be sendrecv instead of sendonly. In certain
situation this issue was causing 2 MediaStreamTracks associated with a
single MediaStream/participant.

This is the first step in adding support for multiple
streams per endpoint. Each source(stream?) needs to have
an identifier. For now always generate the 0 index name
until the machinery for sending more than 1 stream is put
in place.

* send only MSID attribute

* add feature flag for source name signaling

* log a msg if source name signaling is enabled

This fixes an issue where the bridge doesn't forward the HD stream from Firefox to other users in the call. The order of the ssrcs produced by the browser is from Highest resolution to lowest whereas the bridge assumes it to be from lowest to highest as is the case in Chrome and Safari.

... is stopped. If there's any sRD/sLD cycle happening
while the screen sharing stops, the local track
returns muted and it was injecting fake video SSRCs
which results in invalid SSRC description.

At the time when it happens there's error log in jicofo:
"Error adding SSRCs from X"
and the client will get bad-request error in response
to the source-add request.

Hurts mobile performance. They can still be enabled by setting the level at the
right value (debug / trace).

In unified plan, re-use of m-line (i.e., adding an SSRC, removing it and then adding it back) causes the browser to not render the media on Chrome and Safari. The WebRTC spec is not clear as to how browsers have to behave, this doesn't cause any issues on Firefox. As a workaround, only change the media direction and leave the ssrc in the remote desc. This automatically triggers a 'removetrack' event on the associated MediaStream and the track can be removed from the UI.

Avoid configuring the encodings on chromium immediately after replace track since the encoding params are read-only until the renegotation is done.

This is needed only for JVB connections. Add unit tests for LocalSdpMunger.

If the msid attribute is missing, then remove the ssrc from the transformed description so that a source-remove is signaled to Jicofo. This happens when the direction of the transceiver (or m-line) is set to 'inactive' or 'recvonly' on Firefox. Not signaling these source updates creates issues with remote track handling on the other endpoints in the call.

Get the correct media type when generating the source identifier.

When a msid attribute is missing in the 'a=ssrc' line, use the local endpoint id as an indentifier. Move this generation logic to LocalSdpMunger. Also suppress notifying Jicfo of a ssrc change on Firefox when the change is a result of the media being suspended on the jvb connection.

Get rid of noisy logs related to SDP transformations which are redundant. Fix formatting and add missing information.

In certain cases, when RTCPeerconnection#addTrack is used, the browser doesn't produce the streamid in the 'a=msid' line which is being used by Jicofo as a unique identifier for the associated ssrc. Whenever the msid is missing in the local description, generate an ID using the local endpoint id.

SDP utility classes are spread across RTC and XMPP directories now, moving these class files to a 'sdp' directory.