Skip to content

refactor(engine): late-start measurement reads from names, not comments - #6744

Closed
miguel-heygen wants to merge 1 commit into
mainfrom
refactor/late-start-names
Closed

miguel-heygen wants to merge 1 commit into
mainfrom
refactor/late-start-names

Conversation

@miguel-heygen

Copy link
Copy Markdown
Collaborator

What

The late-start measurement in extractMediaMetadata carries its meaning in names (startsLate, lastPacket) instead of four comment lines. No behaviour change.

Why

The change that measures late-starting videos to their last packet raised packages/engine/src/utils/ffprobe.ts's comment share from 13.5% to 13.7%, which the comment ratchet refuses. It merged with that check red; this brings the file back under its share.

Related work

Refs #6736.

How

Drops the doc line on resolvePlayableVideoDuration and the three-line comment above the late-start read. The reasoning (ffprobe's durations do not all run from the first frame in Matroska and fragmented MP4; a failed read keeps ffprobe's duration) is in #6736's description.

Test plan

  • comment-ratchet.mjs against the merge base: clean.
  • streamStartOffset.test.ts: 24 of 24, 3 runs in a row; engine frame extractor and ffprobe suites: 410 passed, 1 skipped.

@github-actions

Copy link
Copy Markdown
Contributor

Edit accuracy: accurate 271 (base branch 271), smooth 110 of those

The gate passes.
Smoothness is reported in the artifact, not gated. A case fails only if it fails 2 of 3 runs.

Quarantined, measured but not gated (0)

Unstable (3)

  • seqplainkeys-none-center-r0-root-z100: tracking 0.06, pressJump 0, drop 0, reload 0, render 0.03, renderKey -, undo false, teleport true / tracking 0.06, pressJump 0, drop 0, reload 0, render 0.03, renderKey -, undo true, teleport true / tracking 0.06, pressJump 0, drop 0, reload 0, render 0.03, renderKey -, undo true, teleport true
  • seqplainkeys-none-px-r0-root-z100: tracking 0.06, pressJump 0, drop 0, reload 0, render 0.03, renderKey -, undo false, teleport true / tracking 0.06, pressJump 0, drop 0, reload 0, render 0.03, renderKey -, undo true, teleport true / tracking 0.06, pressJump 0, drop 0, reload 0, render 0.03, renderKey -, undo true, teleport true
  • seqplainkeys-none-pct-r0-root-z100: tracking 0.06, pressJump 0, drop 0, reload 0, render 0.03, renderKey -, undo false, teleport true / tracking 0.06, pressJump 0, drop 0, reload 0, render 0.03, renderKey -, undo true, teleport true / tracking 0.06, pressJump 0, drop 0, reload 0, render 0.03, renderKey -, undo true, teleport true

@heygengenesis heygengenesis 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.

Verdict: APPROVE · 0 blocker · 0 important · 0 nits · 6b7fad08

The behaviour is the same: startsLate and lastPacket take the place of the inline condition and last, and only comments are removed. Comments (the ratchet) and CI pass.

Required Tests on windows-latest is red, but this PR didn't cause it. streamStartOffset.test.ts:389 (fragment-cut.mp4 3.54 vs 3.375, late.webm 4.04 vs 3.52) fails the same way on main at the merge base 22b53c0 (#6736, run 40873362926). The other failure, autoUpdate.test.ts in cli, passes on main and is unrelated, so it looks like a flake. This PR can't merge until main's Windows late-start measurement is fixed or an admin merges it.

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