In lots of our event handlers we return the event status.
This is so we can know if the callee has finished. In lots
of cases we don't care so get rid of them to simplify the code.
A change was recently made to the SetLocal and SetRemoteDescription
functions for PeerConnection that starts ICE gathering for the trickle
method only when generating an answer. This means clients generating an
offer to initiate signaling will stall indefinitely.
Before a:setup=actpass was showing up twice in each media description.
It is already being added in the addTransceiverSDP and
addDataMediaSection methods. This commit removes the extra call.
Take into consideration if remote is running DTLS as a
client/server. Before we ignored this value and we could
enter cases where DTLS would never connect.
Resolves#494
This fixes the echo program so it works properly on Safari
and Firefox, where the preferred offered dynamic media type
is not 96/VP8. It loads MediaEngine with codecs found in the
offer and then uses the payload type of the offer's preferred
video codec in the answer.
This behavior is fine. RTCP is lossy communication, so the caller
should should be ready to send packets multiple times if reception
is mandatory
Relates to #106
These are generated at runtime (depending if we are doing Plan-B or
Unified Plan). Currently this field is actually unused, so just getting
rid of it.
Relates to #106
We will implement these eventually, but for now just delete them.
We need to explore how we can automatically figure out what parts
of WebRTC we haven't implemented yet.
Relates to #106
Fix RemoteDescription parsing so that it parses all sections when
looking for a fingerprint, before it would fail if the first section
did not contain one
This only does parsing, does not generate them.
In the future this option will be exposed via the SettingEngine,
including the ability to opt out of mDNS completely for security/scaling
concerns
Relates to pion/webrtc#699
Users are unable to use the callbacks inside `internal/ice`.
Even though we alias things like OnSelectedCandidatePairChange
are unusable since in the package they use `ice.Candidate` instead of
`ICECandidate`
Fix inconsistency with error handling when we have no candidate pairs.
Before we had custom code in RTP handling that would discard errors
if it was because we had no candidate pairs. Move this logic into the
mux so we have consistent behavior with Datachannels
This can be expected and is a soft failure. Every subsystem is expected
to handle lossy communication.
Resolves#706
Moved ice related files to internal/ice. Added ice.go and
ice_go.go files containing aliases to internal/ice. Added
internal/tools/gen/genaliasdocs.go helper script for generating the
documentation for these aliases. Dropped ice
prefix from file names in internal/ice package.
Data channel creation checks if the sctp transport is already
established. Because this check was incorrect, in some cases it returned
an error instead of delaying opening the data channel until the sctp
connection is established. Fixed the check. Added locking to
CreateDataChannel to avoid race condition. Added test.
Before our RTPTransceiver logic would only allow Sendrecv to
match with another Sendrecv. Instead allow a SendRecv to be used
with a local Recvonly
Co-authored-by: Sean DuBois <sean@siobud.com>
Previously the PeerConnection would create invalid SDPs
when answering an offer that contained codecs it didn't
know about. Now it rejects them by setting the media format
to 0.
Related to #449
By default, we should be using unified-plan as the
SDP format of choice. This patch adds a PeerConnection
configuration option to allow the user to specify that
they want plan-b only, unified-plan only, or a special
compatibility mode where a plan-b answer will be generated
IFF a plan-b offer is received.
Instead of printing the error to stdout return the error to the user.
This may not be a hard error (as later certificates would have passed)
but it never is good to be in a state where you have certificates in a
broken state.
Resolves#586
