chore(release): v0.2.1 #15
No reviewers
Labels
No labels
bump
major
bump
minor
bump
patch
kind/bug
kind/chore
kind/docs
kind/feature
priority/critical
priority/high
priority/low
priority/medium
size/L
size/M
size/S
size/XL
No milestone
No project
No assignees
2 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
frankenbit/release-toolkit!15
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "release-prep/v0.2.1"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Added
Fixed
Sudo-aware install-deps step in
_release-prep.yml+_release-draft.yml+_manifest-check.yml. v0.2.0 hard-codedsudo apt-get installfor the yq/jq/curl fallback, which failed opaquely withsudo: command not foundon minimal root-only CI images (a common shape: Debian-derived images that run as root with nosudopackage installed). v0.2.1 replaces the baresudocalls with a 3-shape handler: (1) if EUID=0,apt-getdirectly; (2) else ifsudois on PATH,sudo apt-get; (3) else fail loud with::error::annotations naming the missing tools + actionable remedies ("pre-bake into your runner image, run as root, or install sudo")._release-publish.ymlis unchanged — it has no install-deps step (pure tag-format validation). Closes #10.list_fragmentswarns on unrecognized basenames instead of silently skipping. v0.2.0 dropped fragments named<id>.<kind>.<suffix>.md(e.g.,3.fixed-2.md) without any signal — the parser extracts the LAST dot-segment as kind (fixed-2), which isn't a recognized kind, and the file was treated as "not a fragment" and quietly excluded from the cut. This was caught manually during the v0.2.0 release when a file-count probe surfaced a missing fragment. v0.2.1 emitswarning: <path> does not match \..md` (unknown kind in basename) — skippingto stderr so the operator sees the dropped file before the cut consumes the workspace. A smarter parser that accepts...md` shapes is deferred to v0.3 (release-toolkit#9 disposition B). Closes #9.workflows: sudo-aware install-deps step in 3 reusables (closes #10)
scripts/lib: list_fragments warns on unrecognized basenames
The auto-generated v0.2.1 cut produced a CHANGELOG section with: 1. An `### Added` bullet ("toolkit's own consumer workflows for v0.2.1+ dogfood") from PR #6's `feat:` commit — internal substrate, not external-consumer-facing surface. Same issue release-toolkit#14 covers. 2. Duplicate commit-subject bullets at the end of `### Fixed` that restate what the long-form fragment prose above already says. ## Cleanup - Move the "toolkit's own consumer workflows" entry from `### Added` to a new `### Internal` section + reframe as "internal substrate-of- record" (acknowledges the work shipped without overstating the user-facing impact — matches docs/conventions.md's `internal` kind semantics: emitted but no version impact). - Drop the two duplicate commit-subject bullets ("workflows: sudo-aware install-deps step in 3 reusables" + "scripts/lib: list_fragments warns on unrecognized basenames"). The long-form fragment prose is the canonical CHANGELOG entry; per docs/conventions.md "No deduplication in v0.1; review prep PR + dedupe manually if it matters." Substrate-honest disclosure: the Internal-section text I'm adding is HAND-WRITTEN, not derived from a fragment. The toolkit's auto-generated section was cosmetically wrong for the actual shape of the change. v0.2.1 still ships the same code as the auto-cut; only the CHANGELOG narrative changes. Tracker: frankenbit/release-toolkit (v0.2.1 cut cleanup; related to #14)release-draft.ymlpasses emptyinputs.versionto reusable on pull_request.closed (new bug exposed by #16's v0.3.1 fix) #41quartermaster referenced this pull request2026-06-27 14:43:40 +02:00