Write the agreed release-candidate rules into the roadmap - #90
Merged
Merged
Conversation
On 2026-09-29 nomologR proposed two details for release-candidate day
(nomologR#53), and a check of the plan found gaps it did not cover. The
v1.0.0 section now records what the two packages agreed and what the
maintainer decided:
- Versions: DESCRIPTION 0.99.N for tag v1.0.0-rc.(N+1), so no two
candidates share a version or a fixture name (maintainer, 2026-09-29).
Candidate tags are annotated, never moved or reused, and never get a
GitHub release of any kind, since R-universe publishes releases.
- Step 1: the 0.99.0 commit is merged, then checked as it sits on
origin/master (a squash merge changes the SHA): green CI including
--as-cran and pkgdown, and the release gate. Merges pause until the tag.
Fixtures come from that commit, recorded as "untagged".
- Step 2: nomologR's fixture PR merges only after its suite passes; a
failure that needs a contentvalidR change restarts step 1.
- Step 3: tag, then nomologR checks the tag with ls-remote ^{} before
recording it and tagging.
- Step 4: after a tag, only README, ROADMAP, and NEWS prose change; no
.9000 bump while a candidate is open; the README keeps its stable-release
line and adds one release-candidate line (maintainer, 2026-09-29).
- 1.0.0 day repeats the cycle, with GitHub releases last.
- The compatibility table uses the "1.x" wording from the release candidate
on (maintainer, 2026-09-29), each README stating its own column; the
shared example is named and prints nothing that varies with the version
or date.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Records the release-candidate rules agreed with nomologR on 2026-09-29 (nomologR#53) in the ROADMAP's v1.0.0 section, with the maintainer's three decisions. ROADMAP only: no code, and no user-facing documentation change.
Maintainer decisions (2026-09-29)
DESCRIPTION0.99.N for tagv1.0.0-rc.(N+1). So 0.99.0 is rc.1 and 0.99.1 is rc.2, and no two candidates share a version or a fixture name.Rules agreed with nomologR
A check of the plan before replying found gaps; nomologR accepted these fixes:
origin/master. A squash merge changes the SHA, so they never come from a PR head.--as-cranand pkgdown) and pass the release gate.git ls-remote … 'refs/tags/v1.0.0-rc.1^{}'. A plain rev-parse returns the tag object and looks like a mismatch..9000bump. Code, data, and tests wait for the next candidate. nomologR moved its stored walkthrough-handoff update into its pre-tag PR because of this.nomo_screen(nomo_demo_walkthrough, items = h). Our README shows it without running it. It prints nothing that varies with the version or the date.Verification
tools/release-gate.Ron this PR's head (fa6bcde), run in a detached worktree: build, smoke, and check all pass; R CMD check 0 errors, 0 warnings, 0 notes. Two earlier runs failed on the environment, not the change: the intermittent pak "Subprocess is busy" startup error, then github.com URLs timing out (GitHub was answering in 3.5-3.8 s against the 5 s limit).ROADMAP.mdis build-ignored.🤖 Generated with Claude Code