Repository navigation
links.yml's decline drops a false uniqueness clause (issue #1180) - #1185
Merged
Merged
Conversation
`links.yml`'s permissions comment declined `issues: write` for two reasons in one sentence: the sound one, that a write would only save a maintainer from reading a failure notification already received, and a uniqueness claim that this would become the only workflow here that can write anything. `claude-review.yml` already grants `pull-requests: write` and `id-token: write` to the job that calls `reusable-claude-review.yml` by path, so the claim is false. The false clause is deleted; the sound reason stays, and the decline of `issues: write` itself is unchanged. Other trees carry the same false clause and are not touched here. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Xjn8bQapGHmhvLKMSy8b7d
There was a problem hiding this comment.
Sorry @fametrano, you've used your own review budget of 250,000 diff characters for the last 7 days.
You can request another review in 3 days and 7 hours by commenting @sourcery-ai review. Upgrade to get a review now.
Reviewer's guide (collapsed on small PRs)Reviewer's GuideThe PR corrects the File-Level Changes
Tips and commandsInteracting with Sourcery
Customizing Your ExperienceAccess your dashboard to:
Getting Help
|
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Part of ISS 1180 —
no closing keyword. Four of the issue's six sites are settled with this
one:
portanode(PR 553),bitcoin-core-rpc(PR 489),btclib-org.github.io(PR 71)and this tree. The two that remain are both in
btclib-node, whoselinks.ymlandvendored-vectors.ymleach carry the clause, and thatbranch lands last with the closing citation.
What the comment claimed
links.yml's comment besidepermissions:said that grantingissues: writewould make it the only workflow here that can writeanything. Measured in this tree, six workflows already carry an
uncommented
<scope>: write:claude-review.yml(
pull-requests: write,id-token: write),reusable-attest.yml,reusable-claude-review.yml,reusable-codeql.yml,reusable-github-release.ymlandreusable-scorecard.yml.The clause was born false at every one of the issue's six sites —
two of them in the very commit that wrote them — rather than true when
written and stale later. The issue body's own first reading was the
staleness story; the correction is a comment on it.
What the branch does
Deletes the false clause and keeps the sound reason beside it,
converging on the wording
btclib-benchmarksalready carries ratherthan inventing a fourth: "an automatic 'link rot' issue is the usual
next step, to save a maintainer from reading a failure notification they
already get."
Gates
At
b79aedd, in the worktree:uvx pre-commit run --all-filesexit 0(40 hooks, none failed), and
BTCLIB_INTEGRATION=1 uv run --no-default-groups --group test pytest661 passed, 110 skipped, exit 0. The tree was clean after the fixers.
The rebase
The branch was rebased onto
a99e31aafter the sibling landed, and theunion merge driver swallowed the blank line above the new
###heading, as it has on every changelog rebase of this campaign. It was
restored and the result proved from the object store rather than from
the repair's own account of itself: the added lines are byte-identical
to the pre-rebase hunk at
9c8f0f9, the seam alarm prints nothing overthe repaired file and printed line 8852 over the damaged one, the
blank-pair count is 0 in branch and base alike, and the entry is last in
the open section at exactly the hook's three body lines.
🤖 Generated with Claude Code
https://claude.ai/code/session_01Xjn8bQapGHmhvLKMSy8b7d
Summary by Sourcery
Remove the inaccurate uniqueness claim from the
links.ymlpermissions guidance while preserving the valid link-rot rationale.Bug Fixes:
links.ymlpermissions comment by removing its false claim that decliningissues: writemakes it the only workflow able to write.Documentation:
issues: writein the changelog.