Repository navigation
docs(release): archive cutover is 0.0.44; runbook and client copy follow - #318
Conversation
…copy follow 0.0.43 shipped through the old npm pipeline, so the last npm version is 0.0.43 and the first release archive is 0.0.44. The J5 Release resolve gate, release notes, FORK.md cases 15/40/41, user migration/install/update/service docs, install.sh's error and code comments now say so. The macOS packaging runbook's GitHub Actions section described npm trusted publishing; it now describes the J5 Release archive flow. Two new client-runtime strings named T3 Code; they are product-neutral now and listed in BRANDING.md's pin-advance audit sites. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
Navigate logical layers of code changes, visualize relationships, and explore their blast radius. No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Repository: Jacksondr5/j5code/.coderabbit.yaml Review profile: CHILL Plan: Essentials Run ID: 📒 Files selected for processing (6)
Included review availability: This review used your included allowance. 3 included reviews remain after this review. Your included PR review attempts over the past 7 days set your current allowance at 5 reviews per hour. 📝 WalkthroughWalkthroughThe release workflow and documentation set 0.0.44 as the first release-archive version and identify versions through 0.0.43 as npm-era. Legacy-service references use the updated range. Shared-client error messages use product-neutral wording. ChangesRelease archive transition
Product-neutral client messages
Priority: ➖ Normal Estimated code review effort: 2 (Simple) | ~10 minutes Change: Other Suggested reviewers: Merge Risk: ⚪ Minimal · up to No current release-boundary or client-behavior issue warrants blocking merge. The 0.0.43 archive example is clearly a local test record, not a claim of a public release. 🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
Full details: Docstring CoverageExplanation Docstring coverage is 50.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 2 functions across 11 files. (6 skipped: 6 unsupported.)
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…erges the bump back Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
0.0.43 went out through the old npm pipeline before #262 merged, so 0.0.44 is J5's first release-archive release. The docs, the J5 Release gate and some messages still named 0.0.43 as the cutover, the release runbook still described npm publishing, and two client-runtime messages still said "T3 Code".
J5 Releaseresolve gate and release notes, FORK.md cases 15, 40 and 41, the install, updating, background-service and migration guides, theinstall.sherror, and code comments. The only version comparison in code is the release gate (patch < 44). Other version strings are unrelated provider ranges or test fixtures, and are unchanged.docs/j5/runbooks/macos-packaging.mdnow describes the archive flow: bump, pushj5/release-<version>,J5 Signed macOS Build, dispatchJ5 Releasewith the run ID, what gets published, and the resolve rules. It no longer describes npm trusted publishing.j5/main: the 0.0.44 bump (e359d62346) only lived onj5/release-0.0.44, soj5/mainstill reported 0.0.43. This PR carries that bump, and the runbook gains a step to bring each release's bump back toj5/main.README.mdalso names 0.0.43 as the last npm release.compatibility.tsandpullRequestDiffHttp.tsuse product-neutral wording ("Update the server on …", "quit and reopen the app"), and both are recorded in BRANDING.md's audit sites.Verified: tests for resolver, pullRequestDiffHttp, update, bootService and install (75 passed),
tscclean for client-runtime, server and scripts, format clean,bash -n scripts/install.sh, and the workflow YAML parses and its script passesnode --check.🤖 Generated with Claude Code
Summary by CodeRabbit