Skip to content

#454 docs: Dependabot rebase step checks the new head's base - #455

Merged
gregoryfoster merged 3 commits into
mainfrom
docs/454-rebase-base-check
Nov 20, 2026
Merged

gregoryfoster merged 3 commits into
mainfrom
docs/454-rebase-base-check

Conversation

@gregoryfoster

Copy link
Copy Markdown
Member

Closes #454

  • docs/DEPENDENCIES.md → How one lands: replaces "wait for its force-push" with a base check, gh pr view <N> --json mergeStateStatus (BLOCKED/CLEAN mean the head is on main, BEHIND/DIRTY mean not yet, UNKNOWN means read it again). The exact fallback is git merge-base --is-ancestor, with the PR ref fetched last: the issue's one-fetch form tests main against itself and always exits 0, which was measured against chore: bump fastapi from 0.163.0 to 0.163.1 #451's stale head (see the corrections).
  • One sentence of reason: after a landing, Dependabot may rebase the batch's other PRs onto the commit just landed, so a head that moved can already be behind.
  • --watch starts only once the base check passes, which is cosmetic because gh pr merge refuses regardless. The re-post timer counts from the request, and only a push onto current main answers it. The 6 minutes and the 6-of-20 history are unchanged.
  • tests/ci/test_dependabot.py: the docstring no longer says a Dependabot PR fast-forwards (Require pull requests into main: a protection ruleset plus shipping-work's pr merge mode #445). No assertion changed.

Docs only: no sync, no restart. DEPENDENCIES.md goes from 8,394 to 8,761 of 10,000 tokens (estimate).

🤖 Generated with Claude Code

gregoryfoster and others added 3 commits November 20, 2026 18:52
The recipe waited for a force-push after `@dependabot rebase`, but after a
landing Dependabot may auto-rebase the batch's other PRs onto the commit just
landed, so a moved head can already be behind (#451). The rebase step now
reads mergeStateStatus, with the ancestry form (PR ref fetched last) as the
exact fallback; --watch starts only once it passes; the re-post timer counts
from the request and only a push onto current main answers it. The
test_dependabot docstring no longer says Dependabot PRs fast-forward (#445).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…h, reply rule

- Ancestry check fetches the PR ref first in one call (measured on #451: 1,
  #455: 0, bad ref: 128) and names its exit codes.
- mergeStateStatus: only CLEAN reads as on main; BLOCKED (#455, up to date
  with checks pending) defers to the ancestry form.
- The cause sentence names the uv.lock conflict that triggers the auto-rebase.
- --watch also waits for the new head's runs to be listed.
- Re-post rule: a Dependabot reply is an answer, not silence.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@gregoryfoster
gregoryfoster merged commit 4102ae8 into main Nov 20, 2026
4 checks passed
@gregoryfoster
gregoryfoster deleted the docs/454-rebase-base-check branch November 20, 2026 19:34
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

docs: Dependabot rebase step checks the new head's base, not just a force-push

1 participant