Repository navigation
fix(mux): populate display aspect across known-SPS importers - #5755
Conversation
Co-Authored-By: GPT-6 <noreply@openai.com>
…lay aspect quest The known-SPS fix lands here; headerless avc3/hev1 acquisition and container aspect signals (pasp, Matroska DisplayWidth) are left for follow-ups. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
Warning Review limit reachedYou've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository. Next included review available in 18 minutes. View limit details
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 |
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
|
Merging per maintainer approval, ahead of #5846.
(Written by Claude Opus 5.5) |
Problem
H.264 VUI sample aspect and H.265 hvcC metadata were lost through codec, Matroska, and fMP4 import, leaving anamorphic pictures with the wrong catalog display shape.
Approach
Share SPS metadata across Annex-B, decoder configuration records, Matroska CodecPrivate, and avc1/avc3/hvc1/hev1 initialization. One
apply_hvcc_spshelper serves both the Matroska and fMP4 hvcC paths. Square samples leave the display aspect unset, so the encode hint stays the fallback. AV1 render size is per-frame and does not establish a rendition-wide aspect, so AV1 is left out (documented incodec::av1).Completes and deletes
quest/m2/display-aspect.md.Impact
codedanddisplay_aspectfields gain doc lines.Avcc::parsealready was.Decisions
Settled unattended by the iterating agent, recommended option taken:
Should headerless avc3/hev1 initializers (SPS only inside later CMAF media) get their aspect in this PR?
Validation
Regression tests cover non-square and square SPS fixtures through the codec, Matroska, and all four fMP4 sample entries, plus headerless avc3/hev1 controls; each fails before its fix. Scoped
just checkpasses, andjust test interop --allpassed on the earlier head.Follow-ups
paspand MatroskaDisplayWidth/DisplayHeightare not read. They are the only rendition-wide aspect for AV1, and cover headerless avc3/hev1 initializers that carrypasp.🤖 Generated with Claude Code
(Written by GPT-6 and Claude Opus 5.5)