Skip to content

feat(hang)!: video warmup leaves hang-04, and the players stop reading it - #8960

Merged
kixelated merged 3 commits into
mainfrom
quest/m2/hang04/video-warmup
Nov 16, 2026
Merged

kixelated merged 3 commits into
mainfrom
quest/m2/hang04/video-warmup

Conversation

@kixelated

@kixelated kixelated commented Nov 16, 2026 •

Copy link
Copy Markdown
Collaborator

Claims quest/m2/hang04/video-warmup.md and finishes it (the PR deletes the quest).

For the maintainer to confirm before merging. The 2026-11-16 pick ("delete the path, park #8857") was made on a wrong fact: moq-dev/moq#8857 ("a warmup rendition plays through one decoder") had already merged on 2026-11-15, about ten hours before the question. There was no PR to park, and the dual decoder was already gone. So what this PR deletes is the one-decoder path #8857 landed, which is the option the question offered as "land #8857 anyway" and you turned down. Merging this PR confirms the pick as recorded; closing it reverses it and keeps the one-decoder path on main.

Problem

moq-hang-04 promised video warmup (intra-refresh groups that start without a keyframe) with no producer that writes it, Firefox and Safari never run, and an H.265 refresh rendition that Chromium refuses. Worse, a 0.22 player that claims to read it would accept a later producer's rendition and mis-decode it instead of skipping it.

Approach

  • Catalog. warmup leaves hang::catalog::VideoConfig and @moq/hang's video schema, so a video warmup is an unknown field, parsed past and never written back. Both version computations lose the warmup trigger: a renamed track is hang-04's only reason for version 4. The MSF mappings only ever derived Opus warmup, so they are unchanged.
  • Browser (@moq/watch). The warmup hold in Recovery goes (its leading-picture refusal stays, back to its pre-feat(watch): run a min-delay and a max-delay decoder for warmup renditions #7555 shape), and so do the widened subscription, the consumer's warmup reach for video, the warmup decoder-config field, and insertRecoveryPoint, the H.264 recovery-point SEI rewrite Chromium needed. The doc/lib/js/watch.md paragraph goes.
  • Rust. moq_video::decode::Consumer no longer widens its subscription or hides pictures after a join; Start::Latest still waits for the live edge, as documented. moq play drops the buffer's warmup hold, Buffer::fresh, and the live-edge wait it only did for warmup. moq-gst passes no warmup for video.
  • Bindings. MoqVideo.warmup leaves moq-ffi (regenerated Dart bindings; Go's bridge.Video validator had nothing left to check, so it and its test go). moq-c's moq_video_config loses warmup_us and has_warmup; its media consumer applies a warmup for audio only.
  • Draft. The warmup section moves from Common Rendition Fields to Audio (and into AudioDecoderConfigExtensions), describing Opus pre-roll only, without the dual-decoder sentences. The group rule returns to "each group MUST start with a keyframe". The Version section and the hang-04 changelog drop video warmup.
  • What stays. moq_mux::container::Playout { delay, warmup }, the JS consumer's warmup reach, and every audio use. Done by hand, not with git revert, so the dual decoder feat(watch,moq-video,moq-cli): a warmup rendition plays through one decoder #8857 deleted (rs/moq-video/src/decode/select.rs) stays deleted.
  • Quests. The hang-04 epic's overlap note, shape.md, and entry-version.md drop their links to this quest; the intra-refresh epic and catalog warmup link this PR as the inverse to restore from.

Decisions

Taken while the maintainer was away; each is reversible in review.

  • moq-gst's warmup-update test becomes a unit test. a_warmup_update_keeps_the_reader_and_reaches_the_next_one drove the "a changed warmup never restarts the reader" rule through a video rendition, and moq-gst's sibling-instance fixtures (Publisher, Recorder) only publish and record video. Porting it to audio meant new fixtures for one rule, so a_warmup_is_not_part_of_the_shape checks the rule where it lives, Shape's equality, and a_rendition_subscribes_with_its_catalog_warmup keeps the audio subscription check.
  • The parity fixture keeps one video warmup case. The five video warmup bound cases (0, 2^53-1, 2^53, -1, 1.5) become one, video warmup is an unknown field, where -1 no longer refuses the catalog in either language. The audio bound cases are untouched.
  • Recovery returns to its pre-warmup shape in both languages. Recovery.restart() (JS) and the recover = false on an encoder end (Rust) came with feat(watch,moq-video,moq-cli): a warmup rendition plays through one decoder #8857 to keep a fresh encoder from warming up. Without warmup they only skip setting a leading-picture floor at a fresh encoder's first keyframe, where no leading picture exists, so both go.
  • The warmup draft section moves under Audio. It sat under Common Rendition Fields, which text renditions share too, and no schema listed it; now AudioDecoderConfigExtensions does.

Impact

Breaking, Rust, JS, bindings, and C:

  • hang: VideoConfig::warmup is removed. Drop any assignment; a catalog's video warmup is ignored on read.
  • moq-video: decode::Consumer no longer reads a warmup (there is none to read). Behavior only.
  • moq-ffi and every binding: MoqVideo.warmup is removed (Go media.Video.Warmup, Python/Kotlin/Swift/Dart Video.warmup). Drop it from video rendition literals; MoqAudio.warmup is unchanged.
  • moq-c: moq_video_config loses warmup_us and has_warmup. Remove them from initializers and rebuild against the new header; moq_audio_config's stay.
  • @moq/hang: VideoConfigSchema (and VideoConfig) lose warmup; a parsed catalog drops a video warmup as an unknown field.
  • @moq/watch: no public API change.
  • Release cut: the Unreleased "Catalog renditions carry their warmup (#7864)" notes in rs/moq-ffi, go/wrapper, kt/moq, py/moq-rs, swift, and dart/moq MIGRATE.md say the video record gains warmup too. Only the cut edits MIGRATE.md (check migrate), so the cut should rewrite them to name the audio record alone.

Wire: the hang-04 catalog JSON and container group rule, both unpublished. A video warmup is no longer defined and never raises version.

Validation

  • cargo check --workspace --all-targets.
  • just check (nextest over every affected crate, JS, drafts, bindings, Go, Kotlin, Dart): 5994 Rust tests pass. The only failure in the last run was py/moq-rs/tests/test_socket.py::test_abort_and_endpoint_close_are_errors, the known flake py-socket-abort that moq-dev/moq#8900 fixes. An earlier run timed out four moq-rtmp payload tests at load average 24; they pass alone in 30 ms and passed in the next run.
  • just test interop --all: all checks pass.
  • bun test: js/hang 563 pass, js/watch 626 pass, including the video decoder's join, gap, and leading-picture tests with the hold gone.
  • cargo test on hang, moq-video, moq-cli, moq-ffi, moq-mux, moq-gst, moq-c: all pass under nextest. Under plain cargo test, moq-c's raw_track_subscription_options_and_update and route_broadcast_survives_id_reuse fail intermittently, the same on main at 13e593c (3 of 4 runs), since its tests share one process's global state; CI runs nextest, where they pass.
  • Differential, Rust against JS: a catalog with video warmup: 500 and audio warmup: 80, decoded and re-encoded by @moq/hang (scratch script), writes the video entry without warmup, keeps the audio 80, and writes no version; hang's new a_video_warmup_is_ignored writes the same video entry, and the shared parity case video warmup is an unknown field is read identically by both languages.
  • No new behavior to exercise beyond that: the change removes a path no producer could reach.

Follow-ups

None.

Caused by

None: video warmup was deliberate work (moq-dev/moq#5212, moq-dev/moq#7555, moq-dev/moq#8857) that the pre-cut audit chose to move past hang-04, not a defect.

Models

Size: M
Implementer: Claude, claude-opus-5-5, high effort (arm B)
Implementer: Claude Code, claude-sonnet-5-5, unknown effort (iterate, S, arm B)

🤖 Generated with Claude Code

(Written by Claude Opus 5.5 with high effort)

Review

Rewritten by quest review after each round.

  • Head: 6dccb76

  • Base: origin/main at 38a1082

  • Risk: medium, the base has no RISK.md to name what is always high or customer-facing, so every change is medium

  • Classified: medium by default

  • Reviewer: Codex, gpt-6.1-sol, high effort

  • Session: 01a1e573-acaa-7392-ad82-76bc123e4216, round 2

  • Score: 9/10, The only new change removes a stray draft marker and introduces no actionable regression; builds and tests were not run as instructed.

  • Rounds: 2; first round 0; findings 0: fixed 0, refuted 0, open 0

  • Findings: none

kixelated and others added 2 commits November 16, 2026 08:32
…g it

moq-hang-04 defines `warmup` for audio renditions only. A video rendition's
`warmup` is an unknown field, every video group starts with a keyframe again,
and the renamed `track` is the only version-4 trigger. The one-decoder video
warmup path in @moq/watch, moq_video::decode, moq play, moq-gst, moq-ffi, and
moq-c is removed; audio warmup and the container consumer's reach stay.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@kixelated kixelated changed the title Video warmup leaves hang-04, and the players stop reading it feat(hang)!: video warmup leaves hang-04, and the players stop reading it Nov 16, 2026
@kixelated
kixelated marked this pull request as ready for review November 16, 2026 17:44

@kixelated kixelated left a comment

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Automated review by review (OpenAI)

Reviewed 59e48e1cbf515115ed43ea8324e66aac1e190a6a. No blocker found in the removal. Ordinary video keeps keyframe and leading-picture recovery, native Start::Latest still waits for the live edge, and audio pre-roll stays wired. I’d defer video warmup for 0.22 until a producer and the browser matrix are ready; a later return should mark the entry as needing a newer reader so older players can skip it. When resolving the overlap with #8950, keep renamed track as the only hang-04 entry-version trigger. Static review only; I did not rerun playback or interop.

@pullfrog pullfrog Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ℹ️ No critical issues — one stray marker in the draft, inline.

Reviewed changes

Reviewed the full removal of video warmup from moq-hang-04 across the catalog, players, bindings, and draft, with audio warmup left intact.

  • Catalog. VideoConfig::warmup leaves Rust hang and @moq/hang, and both version computations now raise version to 4 only for a renamed track. A video warmup parses as an unknown field, and the parity fixture covers it in both languages.
  • Browser player. Recovery keeps only its leading-picture refusal. Rendition drops the widened subscription, the warmup hold, and the H.264 recovery-point rewrite. Dropping Recovery.restart() is safe: a fresh encoder's first keyframe already clears #needed and the floor.
  • Rust players. moq_video::decode::Consumer still waits for the live edge on Start::Latest. moq play's buffer drops present, joined, fresh(), and warmed(). moq-gst passes no warmup for video.
  • Bindings and C. MoqVideo.warmup is gone from moq-ffi, along with the Dart bindings and Go's bridge.Video. moq-c's moq_video_config loses warmup_us and has_warmup, and Media::warmup now defaults to zero, which only audio overrides.
  • Draft. The warmup section moves under Audio and into AudioDecoderConfigExtensions. The group rule goes back to keyframe-only, and the Version section and changelog are updated to match.

Pullfrog  | Fix all ➔ | Fix 👍s ➔ | View workflow run | Using claude-opus-5-5 | 𝕏

Comment thread drafts/draft-lcurley-moq-hang.md Resolved
@coderabbitai

coderabbitai Bot commented Nov 16, 2026 •

Copy link
Copy Markdown
Contributor

Explain Change →

No actionable comments were generated in the recent review.

Walkthrough

This PR removes video warmup from catalog definitions, bindings, and playback code. Catalog version logic no longer treats video warmup as a version 4 trigger. Video schema and config types in JS, Rust, C, Dart, and FFI bindings drop the field. JS watch and Rust video playback paths stop widening subscriptions, suppressing decoded frames, or rewriting payloads for video warmup. Audio warmup remains documented and tested. Related tests, fixtures, generated API docs, and planning documents were updated or removed.

Merge Risk: Minimal · up to 6dccb

This removes the unused video warmup option and keeps audio warmup. Existing catalogs that still carry a video warmup are parsed and the field is ignored. I found no concrete merge-blocking risk.

Pre-merge checks: Passed 4

Check name Status Explanation
Title check Passed The title "feat(hang)!: video warmup leaves hang-04, and the players stop reading it" matches the main change. The changeset removes video warmup from the hang catalog schema and from the browser and …
Description check Passed The description states the purpose (claiming and finishing the video-warmup quest), the approach (catalog, browser, Rust, bindings, C, draft, and quest changes), the breaking-change impact, and valida…
Linked Issues check Passed Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check Passed Check skipped because no linked issues were found for this pull request.
Full details: Title check

Explanation

The title "feat(hang)!: video warmup leaves hang-04, and the players stop reading it" matches the main change. The changeset removes video warmup from the hang catalog schema and from the browser and Rust consumers, while audio warmup stays. The title uses a conventional-commit prefix and a short, specific summary.

Full details: Description check

Explanation

The description states the purpose (claiming and finishing the video-warmup quest), the approach (catalog, browser, Rust, bindings, C, draft, and quest changes), the breaking-change impact, and validation steps. These match the file changes in the summaries, so the description is clearly related to the changeset.

Finishing Touches

Simplify code

  • Commit to this branch
  • Create a new PR
  • Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

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

Recent review info

Run configuration

  • Source mode: Auto
  • Repository: Unavailable
  • Branch: Unavailable
  • File: No configuration file from this repository applied
  • Commit: Unavailable
  • Layers: Organization UI
  • Review profile: CHILL
  • Plan: Advanced
  • Run ID: 4efa358a-273a-4fc5-a09f-3c2b465d530d
Commits

Reviewing files that changed from the base of the PR and between 59e48e1 and 6dccb76.

Files selected for processing: 1

  • drafts/draft-lcurley-moq-hang.md
Files with no reviewable changes: 1

  • drafts/draft-lcurley-moq-hang.md

Included review availability: This review used your included allowance. Your plan provides up to 4 included reviews per hour; 0 remain after this review.

…as cut

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

@pullfrog pullfrog Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

✅ No new issues found.

Reviewed changes

Reviewed the delta since the prior Pullfrog review at 59e48e1.

  • Draft cleanup. Removed the stray >>> left where the old warmup section was cut, so jitter flows straight into delay. This addresses the prior inline finding.

Pullfrog  | View workflow run | Using claude-sonnet-5-5 | 𝕏

@kixelated
kixelated added this pull request to the merge queue Nov 16, 2026
Merged via the queue into main with commit 5296b34 Nov 16, 2026
7 checks passed
@kixelated
kixelated deleted the quest/m2/hang04/video-warmup branch November 16, 2026 21:18
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant