[Enhancement]Expose subscribed audio track to CallParticipant#1136
Merged
ipavlidakis merged 2 commits intodevelopfrom Apr 24, 2026
Merged
Conversation
📝 WalkthroughWalkthroughAdds remote audio track exposure: Changes
Sequence DiagramsequenceDiagram
participant Peer as StreamRTC PeerConnection
participant Adapter as RemoteAudioMediaAdapter
participant Subject as Track Subject
participant State as WebRTC StateAdapter
participant Model as CallParticipant
Peer->>Adapter: AddedReceiverEvent(audio)
activate Adapter
Adapter->>Adapter: extract RTCAudioTrack, receiverId, trackId
Adapter->>Subject: send .added(id, trackId, .audio, track)
deactivate Adapter
Subject->>State: receive .added TrackEvent
activate State
State->>Model: assign audioTrack to participant
deactivate State
Peer->>Adapter: RemovedReceiverEvent(receiverId)
activate Adapter
Adapter->>Subject: send .removed(id, .audio)
deactivate Adapter
Subject->>State: receive .removed TrackEvent
activate State
State->>Model: clear audioTrack for participant
deactivate State
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
🧹 Nitpick comments (3)
Sources/StreamVideo/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter.swift (1)
17-35: Nit:AudioTrack.trackIdnaming is slightly misleading.The doc comment describes
trackIdas "the participant lookup id it belongs to", but the value assigned isstream.trackId(the stream/track lookup prefix). Consider renaming tolookupId(ortrackLookupId) to avoid confusion withRTCAudioTrack.trackId/event.receiver.receiverId, both of which are also in scope here.♻️ Optional rename for clarity
- private struct AudioTrack { - var receiverId: String - var trackId: String - var track: RTCAudioTrack + private struct AudioTrack { + var receiverId: String + /// Lookup id derived from the audio stream, used as the participant + /// track lookup key (not to be confused with `RTCAudioTrack.trackId`). + var lookupId: String + var track: RTCAudioTrack @@ - self.receiverId = event.receiver.receiverId - self.trackId = stream.trackId - self.track = audioTrack + self.receiverId = event.receiver.receiverId + self.lookupId = stream.trackId + self.track = audioTrackThen update
processAddedTrack/processRemovedTrackto emitid: trackEntry.lookupId.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@Sources/StreamVideo/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter.swift` around lines 17 - 35, Rename the AudioTrack property trackId to a clearer name (e.g., lookupId or trackLookupId) throughout the RemoteAudioMediaAdapter: update the private struct AudioTrack to declare lookupId and assign stream.trackId in its initializer, and then update all usages (notably processAddedTrack and processRemovedTrack) to emit id: trackEntry.lookupId; ensure any references to AudioTrack.trackId are replaced so the receiverId and RTCAudioTrack identifiers remain unambiguous.StreamVideoTests/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter_Tests.swift (1)
126-128: Optional:TrackEventRecorderwrite isn't fully atomic.
@Atomic var eventsguards individual get/set of the array reference, butrecorder.events.append($0)is read-modify-write and not atomic. In practice this is fine here because only a single Combine subscription writes to it, but if this recorder is ever reused from multiple sinks/threads it could race. If you want to keep it defensive:♻️ Optional tightening
private final class TrackEventRecorder: `@unchecked` Sendable { - `@Atomic` var events: [TrackEvent] = [] + `@Atomic` var events: [TrackEvent] = [] + + func record(_ event: TrackEvent) { + _events.mutate { $0.append(event) } + } }(Only relevant if
@Atomicin this codebase exposes amutateclosure; otherwise ignore.)🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@StreamVideoTests/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter_Tests.swift` around lines 126 - 128, The TrackEventRecorder's `@Atomic` var events only makes the getter/setter atomic but recorder.events.append(...) is a non-atomic read-modify-write; update TrackEventRecorder to perform mutations via a thread-safe API (e.g., use the `@Atomic-provided` mutate { } closure if available on events, or guard accesses with a dedicated serial DispatchQueue or NSLock) so all appends and reads happen inside the atomic/mutating block; update any call sites that do recorder.events.append(...) to use the chosen mutate/locked access pattern to avoid races.Sources/StreamVideo/WebRTC/v2/WebRTCStateAdapter.swift (1)
798-834: Combined audio/video presence logging looks fine.The four-way switch covers all combinations and produces a readable debug line. Minor nit (optional): in the
(false, false)case, the two names lists are appended independently, so a participant with both tracks appears in both clauses — acceptable for debug telemetry.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@Sources/StreamVideo/WebRTC/v2/WebRTCStateAdapter.swift` around lines 798 - 834, The logging in set(participants:) can list the same participant twice when they have both audio and video; update the generation of participantsWithAudioTracks and participantsWithVideoTracks (and/or the combined message in the (false, false) branch) to deduplicate names (e.g., use a Set or union of names) so each participant appears only once in the debug output; reference the participantsWithAudioTracks and participantsWithVideoTracks symbols and ensure the final logMessage still reports counts correctly.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Nitpick comments:
In
`@Sources/StreamVideo/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter.swift`:
- Around line 17-35: Rename the AudioTrack property trackId to a clearer name
(e.g., lookupId or trackLookupId) throughout the RemoteAudioMediaAdapter: update
the private struct AudioTrack to declare lookupId and assign stream.trackId in
its initializer, and then update all usages (notably processAddedTrack and
processRemovedTrack) to emit id: trackEntry.lookupId; ensure any references to
AudioTrack.trackId are replaced so the receiverId and RTCAudioTrack identifiers
remain unambiguous.
In `@Sources/StreamVideo/WebRTC/v2/WebRTCStateAdapter.swift`:
- Around line 798-834: The logging in set(participants:) can list the same
participant twice when they have both audio and video; update the generation of
participantsWithAudioTracks and participantsWithVideoTracks (and/or the combined
message in the (false, false) branch) to deduplicate names (e.g., use a Set or
union of names) so each participant appears only once in the debug output;
reference the participantsWithAudioTracks and participantsWithVideoTracks
symbols and ensure the final logMessage still reports counts correctly.
In
`@StreamVideoTests/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter_Tests.swift`:
- Around line 126-128: The TrackEventRecorder's `@Atomic` var events only makes
the getter/setter atomic but recorder.events.append(...) is a non-atomic
read-modify-write; update TrackEventRecorder to perform mutations via a
thread-safe API (e.g., use the `@Atomic-provided` mutate { } closure if available
on events, or guard accesses with a dedicated serial DispatchQueue or NSLock) so
all appends and reads happen inside the atomic/mutating block; update any call
sites that do recorder.events.append(...) to use the chosen mutate/locked access
pattern to avoid races.
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: ef2f14a7-868a-485a-91ea-475dbb57f811
📒 Files selected for processing (9)
CHANGELOG.mdSources/StreamVideo/Models/CallParticipant.swiftSources/StreamVideo/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter.swiftSources/StreamVideo/WebRTC/v2/PeerConnection/MediaAdapters/MediaAdapter.swiftSources/StreamVideo/WebRTC/v2/PeerConnection/RTCPeerConnectionCoordinator.swiftSources/StreamVideo/WebRTC/v2/WebRTCStateAdapter.swiftStreamVideoTests/Mock/CallParticipant_Mock.swiftStreamVideoTests/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter_Tests.swiftStreamVideoTests/WebRTC/v2/WebRTCStateAdapter_Tests.swift
added 2 commits
April 24, 2026 12:00
# Conflicts: # CHANGELOG.md
ipavlidakis
force-pushed
the
iliaspavlidakis/ios-1639-enhancementaudiotrack-on-callparticipant
branch
from
1427247 to
e76e67b
Compare
Public Interface public struct CallParticipant: Identifiable, Sendable, Hashable
- public var track: RTCVideoTrack?
+ public var audioTrack: RTCAudioTrack?
- public var trackSize: CGSize
+ public var track: RTCVideoTrack?
- public var screenshareTrack: RTCVideoTrack?
+ public var trackSize: CGSize
- public var showTrack: Bool
+ public var screenshareTrack: RTCVideoTrack?
- public var isSpeaking: Bool
+ public var showTrack: Bool
- public var isDominantSpeaker: Bool
+ public var isSpeaking: Bool
- public var sessionId: String
+ public var isDominantSpeaker: Bool
- public var connectionQuality: ConnectionQuality
+ public var sessionId: String
- public var joinedAt: Date
+ public var connectionQuality: ConnectionQuality
- public var audioLevel: Float
+ public var joinedAt: Date
- public var audioLevels: [Float]
+ public var audioLevel: Float
- public var pin: PinInfo?
+ public var audioLevels: [Float]
- public var pausedTracks: Set<TrackType>
+ public var pin: PinInfo?
- public var source: ParticipantSource
+ public var pausedTracks: Set<TrackType>
- public var userId: String
+ public var source: ParticipantSource
- public var name: String
+ public var userId: String
- public var profileImageURL: URL?
+ public var name: String
- public var isPinned: Bool
+ public var profileImageURL: URL?
- public var isPinnedRemotely: Bool
+ public var isPinned: Bool
- public var shouldDisplayTrack: Bool
+ public var isPinnedRemotely: Bool
-
+ public var shouldDisplayTrack: Bool
-
+
- public init(id: String,userId: String,roles: [String],name: String,profileImageURL: URL?,trackLookupPrefix: String?,hasVideo: Bool,hasAudio: Bool,isScreenSharing: Bool,showTrack: Bool,track: RTCVideoTrack? = nil,trackSize: CGSize = CGSize(width: 1024, height: 720),screenshareTrack: RTCVideoTrack? = nil,isSpeaking: Bool = false,isDominantSpeaker: Bool,sessionId: String,connectionQuality: ConnectionQuality,joinedAt: Date,audioLevel: Float,audioLevels: [Float],pin: PinInfo?,pausedTracks: Set<TrackType>,source: ParticipantSource = .webRTCUnspecified)
+
-
+ public init(id: String,userId: String,roles: [String],name: String,profileImageURL: URL?,trackLookupPrefix: String?,hasVideo: Bool,hasAudio: Bool,isScreenSharing: Bool,showTrack: Bool,audioTrack: RTCAudioTrack? = nil,track: RTCVideoTrack? = nil,trackSize: CGSize = CGSize(width: 1024, height: 720),screenshareTrack: RTCVideoTrack? = nil,isSpeaking: Bool = false,isDominantSpeaker: Bool,sessionId: String,connectionQuality: ConnectionQuality,joinedAt: Date,audioLevel: Float,audioLevels: [Float],pin: PinInfo?,pausedTracks: Set<TrackType>,source: ParticipantSource = .webRTCUnspecified)
-
+
- public static func ==(lhs: CallParticipant,rhs: CallParticipant)-> Bool
+
- public func withUpdated(trackSize: CGSize)-> CallParticipant
+ public static func ==(lhs: CallParticipant,rhs: CallParticipant)-> Bool
- public func withUpdated(track: RTCVideoTrack?)-> CallParticipant
+ public func withUpdated(trackSize: CGSize)-> CallParticipant
- public func withUpdated(screensharingTrack: RTCVideoTrack?)-> CallParticipant
+ public func withUpdated(audioTrack: RTCAudioTrack?)-> CallParticipant
- public func withUpdated(audio: Bool)-> CallParticipant
+ public func withUpdated(track: RTCVideoTrack?)-> CallParticipant
- public func withUpdated(video: Bool)-> CallParticipant
+ public func withUpdated(screensharingTrack: RTCVideoTrack?)-> CallParticipant
- public func withUpdated(screensharing: Bool)-> CallParticipant
+ public func withUpdated(audio: Bool)-> CallParticipant
- public func withUpdated(showTrack: Bool)-> CallParticipant
+ public func withUpdated(video: Bool)-> CallParticipant
- public func withUpdated(trackLookupPrefix: String)-> CallParticipant
+ public func withUpdated(screensharing: Bool)-> CallParticipant
- public func withUpdated(isSpeaking: Bool,audioLevel: Float)-> CallParticipant
+ public func withUpdated(showTrack: Bool)-> CallParticipant
- public func withUpdated(dominantSpeaker: Bool)-> CallParticipant
+ public func withUpdated(trackLookupPrefix: String)-> CallParticipant
- public func withUpdated(connectionQuality: ConnectionQuality)-> CallParticipant
+ public func withUpdated(isSpeaking: Bool,audioLevel: Float)-> CallParticipant
- public func withUpdated(pin: PinInfo?)-> CallParticipant
+ public func withUpdated(dominantSpeaker: Bool)-> CallParticipant
- public func withPausedTrack(_ trackType: TrackType)-> CallParticipant
+ public func withUpdated(connectionQuality: ConnectionQuality)-> CallParticipant
- public func withUnpausedTrack(_ trackType: TrackType)-> CallParticipant
+ public func withUpdated(pin: PinInfo?)-> CallParticipant
+ public func withPausedTrack(_ trackType: TrackType)-> CallParticipant
+ public func withUnpausedTrack(_ trackType: TrackType)-> CallParticipant |
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (2)
Sources/StreamVideo/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter.swift (1)
59-71: Consider symmetric scheduling of the two subscriptions.The added-receiver chain runs
compactMap(AudioTrack.init)beforereceive(on: processingQueue), while the removed-receiver chain runscompactMapafter. The added branch's mapping therefore executes on whichever thread the peer connection publisher emits from (dereferencingevent.receiver.track). Movingreceive(on:)earlier would make both flows serialize on the same queue and keep WebRTC track inspection on a predictable executor:♻️ Proposed reorder
peerConnection .publisher(eventType: StreamRTCPeerConnection.AddedReceiverEvent.self) - .compactMap(AudioTrack.init) .receive(on: processingQueue) + .compactMap(AudioTrack.init) .sink { [weak self] in self?.processAddedTrack($0) } .store(in: disposableBag)🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@Sources/StreamVideo/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter.swift` around lines 59 - 71, The two peerConnection publisher subscriptions should schedule onto processingQueue before mapping to keep execution symmetric: in the added-receiver chain that uses peerConnection.publisher(eventType: StreamRTCPeerConnection.AddedReceiverEvent.self) and compactMap(AudioTrack.init) then .receive(on: processingQueue) (which currently runs the compactMap on the publisher thread), move .receive(on: processingQueue) earlier so the chain becomes publisher(...).receive(on: processingQueue).compactMap(AudioTrack.init).sink { [weak self] in self?.processAddedTrack($0) } .store(in: disposableBag) so both added and removed flows serialize on processingQueue and track inspection happens on the predictable executor; keep processAddedTrack, processRemovedTrack, audioReceivers and disposableBag references intact.StreamVideoTests/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter_Tests.swift (1)
33-85: Consider adding negative-path coverage.Both existing tests exercise the happy path. The adapter's
AudioTrack.init?silently drops receivers whose first matching stream isn't audio (or whose track isn't anRTCAudioTrack), and the removed pipeline'scompactMapdrops unknownreceiverIds. A small test each would lock down those filters against future regressions (e.g., if someone changes the stream-selection predicate inAudioTrack.init?).Suggested additions:
test_addedReceiver_nonAudioStream_doesNotPublishTrackEvent()— send anAddedReceiverEventwhose streamtrackTypeis.video, assert recorder stays empty.test_removedReceiver_unknownReceiver_doesNotPublishTrackEvent()— sendRemovedReceiverEventwith a receiver that was never added, assert recorder stays empty.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@StreamVideoTests/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter_Tests.swift` around lines 33 - 85, Add two negative-path tests in RemoteAudioMediaAdapter_Tests.swift: implement test_addedReceiver_nonAudioStream_doesNotPublishTrackEvent() that creates a TrackEventRecorder, a receiver via makeAudioReceiver(), and a stream whose trackType is .video (use makeAudioStream but adjust to video or create a stream with video track type), call observeEvents(with: recorder), send StreamRTCPeerConnection.AddedReceiverEvent(receiver: receiver, streams: [stream]) via mockPeerConnection.subject.send and then assert the recorder has no events; and implement test_removedReceiver_unknownReceiver_doesNotPublishTrackEvent() that creates a TrackEventRecorder, a receiver that was never added, calls observeEvents(with: recorder), sends StreamRTCPeerConnection.RemovedReceiverEvent(receiver: receiver) and then asserts the recorder remains empty. Use the same helpers used in other tests (TrackEventRecorder, observeEvents(with:), mockPeerConnection.subject.send, waitForEvents/wait helpers) and name the tests exactly as suggested.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@Sources/StreamVideo/WebRTC/v2/WebRTCStateAdapter.swift`:
- Around line 801-833: The current logic computes participantsWithAudioTracks
and participantsWithVideoTracks as joined String values and then uses isEmpty on
those strings, which falsely treats a single participant with an empty name as
"no track"; change to first compute filtered arrays (e.g., audioParticipants =
participants.filter { $0.value.audioTrack != nil }.map(\.value.name) and
videoParticipants = participants.filter { $0.value.track != nil
}.map(\.value.name)), branch on audioParticipants.isEmpty and
videoParticipants.isEmpty in the switch, and only call joined(separator: ",")
when building logMessage (e.g., join audioParticipants/videoParticipants for the
message), updating references to
participantsWithAudioTracks/participantsWithVideoTracks accordingly and
preserving the logMessage construction and log.debug call.
---
Nitpick comments:
In
`@Sources/StreamVideo/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter.swift`:
- Around line 59-71: The two peerConnection publisher subscriptions should
schedule onto processingQueue before mapping to keep execution symmetric: in the
added-receiver chain that uses peerConnection.publisher(eventType:
StreamRTCPeerConnection.AddedReceiverEvent.self) and compactMap(AudioTrack.init)
then .receive(on: processingQueue) (which currently runs the compactMap on the
publisher thread), move .receive(on: processingQueue) earlier so the chain
becomes publisher(...).receive(on:
processingQueue).compactMap(AudioTrack.init).sink { [weak self] in
self?.processAddedTrack($0) } .store(in: disposableBag) so both added and
removed flows serialize on processingQueue and track inspection happens on the
predictable executor; keep processAddedTrack, processRemovedTrack,
audioReceivers and disposableBag references intact.
In
`@StreamVideoTests/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter_Tests.swift`:
- Around line 33-85: Add two negative-path tests in
RemoteAudioMediaAdapter_Tests.swift: implement
test_addedReceiver_nonAudioStream_doesNotPublishTrackEvent() that creates a
TrackEventRecorder, a receiver via makeAudioReceiver(), and a stream whose
trackType is .video (use makeAudioStream but adjust to video or create a stream
with video track type), call observeEvents(with: recorder), send
StreamRTCPeerConnection.AddedReceiverEvent(receiver: receiver, streams:
[stream]) via mockPeerConnection.subject.send and then assert the recorder has
no events; and implement
test_removedReceiver_unknownReceiver_doesNotPublishTrackEvent() that creates a
TrackEventRecorder, a receiver that was never added, calls observeEvents(with:
recorder), sends StreamRTCPeerConnection.RemovedReceiverEvent(receiver:
receiver) and then asserts the recorder remains empty. Use the same helpers used
in other tests (TrackEventRecorder, observeEvents(with:),
mockPeerConnection.subject.send, waitForEvents/wait helpers) and name the tests
exactly as suggested.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: ba534fc3-4726-46e0-859b-71103c0d22d5
📒 Files selected for processing (9)
CHANGELOG.mdSources/StreamVideo/Models/CallParticipant.swiftSources/StreamVideo/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter.swiftSources/StreamVideo/WebRTC/v2/PeerConnection/MediaAdapters/MediaAdapter.swiftSources/StreamVideo/WebRTC/v2/PeerConnection/RTCPeerConnectionCoordinator.swiftSources/StreamVideo/WebRTC/v2/WebRTCStateAdapter.swiftStreamVideoTests/Mock/CallParticipant_Mock.swiftStreamVideoTests/WebRTC/v2/PeerConnection/MediaAdapters/LocalMediaAdapters/RemoteAudioMediaAdapter_Tests.swiftStreamVideoTests/WebRTC/v2/WebRTCStateAdapter_Tests.swift
✅ Files skipped from review due to trivial changes (4)
- CHANGELOG.md
- StreamVideoTests/Mock/CallParticipant_Mock.swift
- Sources/StreamVideo/WebRTC/v2/PeerConnection/RTCPeerConnectionCoordinator.swift
- Sources/StreamVideo/Models/CallParticipant.swift
🚧 Files skipped from review as they are similar to previous changes (2)
- Sources/StreamVideo/WebRTC/v2/PeerConnection/MediaAdapters/MediaAdapter.swift
- StreamVideoTests/WebRTC/v2/WebRTCStateAdapter_Tests.swift
Collaborator
SDK Size
|
Collaborator
StreamVideo XCSize
|
Collaborator
StreamVideoSwiftUI XCSize
Show 11 more objects
|
|
martinmitrevski
approved these changes
Apr 24, 2026
Contributor
martinmitrevski
left a comment
martinmitrevski
left a comment
There was a problem hiding this comment.
nice stuff, let's test this next week
ipavlidakis
deleted the
iliaspavlidakis/ios-1639-enhancementaudiotrack-on-callparticipant
branch
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
🔗 Issue Links
Resolves https://linear.app/stream/issue/IOS-1639/enhancementaudiotrack-on-callparticipant
🎯 Goal
Expose each participant's subscribed remote
RTCAudioTrackso consumers can access remote audio tracks directly fromCallParticipant.📝 Summary
CallParticipant.audioTrack.🛠 Implementation
Remote audio is surfaced by WebRTC through
RTCRtpReceivercallbacks rather than media stream add/remove callbacks. This PR adds a remote audio adapter that observes audio receiver additions and removals, maps each receiver to the participant track lookup id, and emits the existing track add/remove flow.WebRTCStateAdapternow assigns stored audio tracks to matching participants alongside video and screenshare tracks.☑️ Contributor Checklist
Summary by CodeRabbit
New Features
Bug Fixes
Tests