Repository navigation
chore: release v0.8.185 - #6047
Conversation
|
Preview deployment for your docs. Learn more about Mintlify Previews.
💡 Tip: Enable Automations to automatically generate PRs for you. |
jrusso1020
left a comment
There was a problem hiding this comment.
Reviewed exact release head abf7ccc4da200e18a981ece57fbaafaf9f709e3c.
Release scope and notes: the commit is a direct child of release base 25126bca6; all 21 changed files are the 14 package manifests, five plugin manifests and two changelog files. Each manifest changes only version, consistently 0.8.184 → 0.8.185. The notes match exactly these seven merged PRs/commits since v0.8.184: #6037, #6031, #6043, #6018, #5967, #5968 and #6045. No skill-rewrite change from #5651 is included. Reviewed the complete release files and the included changes' relevant contracts; no release-code addition or main reversion found.
Approval-history exceptions, not an all-green historical approval claim: five included PRs have valid final-head approvals before merge. #5967's approval at final head 254e20a5 was dismissed on base change before merge; #5968's approval at 3a35ca04 was dismissed before its final rebase d2489eb8, with no subsequent exact-head review. I checked those underlying revisions rather than treating stale approvals as current: #5967's merged motion code preserves its reviewed implementation; #5968's final motion files are byte-identical to its reviewed version, with only main's already-reviewed #6035 root-duration fix added in init.ts. All ten final #5968 files match its merge commit.
A fresh independent source audit and 108 authored lifecycle/compositor/discovery/pose/WAAPI tests against the assembled motion-blur sources pass. Opt-in remains exact data-motion-blur="on"; default rendering returns before allocations, and cleanup, sharp fallback and device-scale propagation remain intact. These current-source checks support this release approval; they do not retroactively create missing final-push approval records. Earlier exact-head reviews/controls cover the audio correction, GSAP exemption, browser-close handling and 4K alpha paths; no new render or performance benchmark claimed here.
Publication: v0.8.184 is a published, non-draft/non-prerelease release. Verified its versions/integrities on npm for all 14 actual publish names: the CLI is published as hyperframes (the workflow deliberately renames monorepo @hyperframes/cli), plus the 13 scoped packages. There is no prior-release publication gap.
Current CI snapshot: Build and its verify:packed-manifests step executed successfully. Latest snapshot has 33 successes, six skips, one neutral and 36 jobs still running, with no failed/cancelled/timed-out result. Thus CI is not yet fully green. I found no separate PR publish dry-run job; the publish workflow runs after release landing. I did not publish, merge, enqueue, deploy or run the full render matrix.
Verdict: APPROVE
Reasoning: The release contains only consistent version/notes changes, the exact seven included changes match the notes, and substantive review/parity checks cover the two historical approval-record exceptions. Pending CI remains independently enforced; it is not represented as passed.
— hf-review
Edit accuracy: accurate 271 (base branch 271), smooth 132 of thoseThe gate passes. Quarantined, measured but not gated (0) |
hyperframes cloud render --resolution 4know works for transparent MOV and WebM. A render no longer fails when its audio peak stalls under correction,checkaccepts projects that inline the GSAP library, and cloud render chunks start their capture browsers while video frames extract. Capture motion blur arrives as an opt-in on the composition root.Changes
release:prepare 0.8.185 --from v0.8.184 --no-tag; the release workflow tags on merge).releases/v0.8.185.md: 7 PRs, every one merged since v0.8.184. Checked againstgit log v0.8.184..main: none missing.docs/changelog.mdx: the matching v0.8.185 entry.No visible change
Version bump and changelog only; each change was captured in its own PR.