add participant to pcm

[tschellenbach]

@tbarbugli

Summary by CodeRabbit

• Bug Fixes
  • Preserve participant metadata across audio frame transformations (resample, copy, conversion) so participant info remains attached throughout the audio pipeline.
  • Improve audio queue overflow diagnostics by including contextual details (queue capacity and buffer duration) in warning messages for easier troubleshooting.

[coderabbitai]

Walkthrough

Added a participant attribute to PcmData and propagated it through PcmData creation/transformation paths; updated the audio track write() method's queue-overflow warning to include max_queue_size and pcm.duration_ms.

Changes

Cohort / File(s)	Change Summary
Audio queue overflow logging getstream/video/rtc/audio_track.py	Enhanced write() warning to include max_queue_size and pcm.duration_ms in the queue-overflow log message.
PCM data participant tracking getstream/video/rtc/track_util.py	Added public participant: Any attribute to PcmData (__init__ signature updated) and propagated participant through to_float32, to_int16, copy, head, tail, resample, and other PcmData-producing paths to preserve participant context.

Sequence Diagram(s)

sequenceDiagram
    participant Writer as AudioStreamTrack.write()
    participant Queue as InternalQueue
    participant Logger as Logger
    participant PCM as PcmData

    Writer->>PCM: receive pcm (has .participant, .duration_ms)
    alt queue full
        Writer->>Logger: warn("queue overflow", max_queue_size, pcm.duration_ms)
        Writer-->>Queue: drop/skip pcm
    else enqueue
        Writer-->>Queue: push pcm
    end

Loading

sequenceDiagram
    participant Source as PcmData (source)
    participant Transform as to_float32 / resample / copy / head / tail
    participant Result as PcmData (result)

    Source->>Transform: transform(samples, ...)
    Transform->>Result: construct PcmData(..., participant=Source.participant)
    note right of Result: participant propagated

Loading

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

• Pay attention to all places that construct new PcmData to ensure participant is passed consistently.
• Verify no tests rely on PcmData constructor signature unchanged (default None behavior).
• Check formatted log message does not raise issues when pcm is None or missing duration_ms.

Possibly related PRs

• More audio utils #173 — Changes the same audio subsystem and PcmData/AudioStreamTrack paths; likely overlaps in participant propagation and AudioStreamTrack.write logging.

Poem

🐰 I hop through bytes and samples bright,
I tuck each name in PCM by night,
When queues complain, I add the clue —
size and ms to see it through. 🎧

Pre-merge checks and finishing touches

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'add participant to pcm' directly and accurately describes the main change: adding a participant attribute to the PcmData class.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch participant

---

📜 Recent review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

Disabled knowledge base sources:

• Linear integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

📥 Commits

Reviewing files that changed from the base of the PR and between 85bd8ef and 65a67e4.

📒 Files selected for processing (1)

• getstream/video/rtc/track_util.py (11 hunks)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (4)

• GitHub Check: Tests (3.11)
• GitHub Check: Tests (3.12)
• GitHub Check: Tests (3.13)
• GitHub Check: Tests (3.10)

🔇 Additional comments (7)

getstream/video/rtc/track_util.py (7)

100-100: Participant attribute properly added to PcmData.

The new participant attribute is well-integrated with complete documentation, appropriate type hints, and correct instance assignment. Using Any type provides flexibility for various participant context types.

Also applies to: 112-113, 125-125, 173-173

---

659-659: Participant correctly propagated in to_float32().

---

723-723: Participant correctly propagated in to_int16().

---

884-884: Participant correctly propagated in copy().

---

1299-1299: Participant correctly propagated in tail().

---

1408-1408: Participant correctly propagated in head().

---

1494-1494: Participant correctly propagated in Resampler.resample().

The comment accurately describes the preservation of participant context, and the implementation correctly passes it through.

Also applies to: 1503-1503

---

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.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (1)

getstream/video/rtc/track_util.py (1)

101-101: Consider initializing participant in __init__ for clarity.

The class-level attribute definition works but may be confusing since it's not explicitly set in __init__. While Python allows this pattern (instance assignments shadow the class attribute), initializing it in __init__ would be more explicit and maintainable.

Apply this diff to initialize in __init__:

-    participant: Any = None
-
     def __init__(
         self,
         sample_rate: int,
@@ -111,6 +110,7 @@
         time_base: Optional[float] = None,
         channels: int = 1,
     ):
+        self.participant: Any = None
         """
         Initialize PcmData.

📜 Review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

Disabled knowledge base sources:

• Linear integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

📥 Commits

Reviewing files that changed from the base of the PR and between b9966c3 and 85bd8ef.

📒 Files selected for processing (2)

• getstream/video/rtc/audio_track.py (1 hunks)
• getstream/video/rtc/track_util.py (2 hunks)

🧰 Additional context used

🧬 Code graph analysis (1)

getstream/video/rtc/audio_track.py (1)

getstream/video/rtc/track_util.py (1)

• duration_ms (236-238)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (4)

• GitHub Check: Tests (3.13)
• GitHub Check: Tests (3.10)
• GitHub Check: Tests (3.11)
• GitHub Check: Tests (3.12)

🔇 Additional comments (1)

getstream/video/rtc/audio_track.py (1)

94-94: LGTM! Enhanced logging provides valuable context.

The addition of max_queue_size and pcm.duration_ms to the overflow warning improves debugging by showing both the queue limit and the duration of the PCM data being processed.