Skip to content

SCM dashboard workers - #11328

Closed
gunbai-bot[bot] wants to merge 5 commits into
mainfrom
fix11317
Closed

gunbai-bot[bot] wants to merge 5 commits into
mainfrom
fix11317

Conversation

@gunbai-bot

@gunbai-bot gunbai-bot Bot commented Sep 13, 2026

Copy link
Copy Markdown
Contributor

Auto-opened by session-dashboard for session still-bat-15.
Pushing to fix11317 advances this PR.

Worker attestation

Before flipping this PR to ready for review, confirm each item:

  • Title describes the change (not the session id or branch).
  • PR body summarises what and why (replace the TODO below).
  • Tests run: name the command (e.g. npm test, cargo test) and the result.
  • If this closes a work item, the body contains a Closes #N directive.
  • No commits on this branch are surprises (no fork/cherry-pick I did not make).
  • No secrets / credentials / large binaries staged.

Summary

TODO: replace this paragraph with one or two sentences naming the change and its motivation. Reviewers read this first.

Test plan

  • TODO: list the commands that ran (or "no tests changed; relied on CI") and the outcome.

Brian Searls and others added 5 commits September 13, 2026 22:08
git_use_covers_import took (row, module_name, imported_module) and its only
caller — the index fold — passed the row's own owner and target straight back
in. Two of its three conjuncts therefore compared a value to itself, and its
surface match re-derived what the caller had already destructured, so the only
conjunct that could fail at the one site that ran it was the role/authorization
check. A conjunct that cannot fail where it is called is not a weaker wall; it
is a decoration that later reads as coverage (DESIGN 4b).

Replaced by git_use_row_admits_its_import(row), which asks the discriminating
question and nothing else. Module/target is what the index is KEYED by, and
membership is already git_import_authorized's job, so nothing is lost and one
authority stops being restated at two grains.

Found by review 65735.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012Qy8ErCHX3Npuvd6P7RPe3
The coverage note said the remaining live-use population was unknown. That is
true for Rust uses and for configured .dag process invocations, which have no
producer that enumerates them. It was false for ParsedDagGitDependency -- the
one surface this policy actually covers -- whose population is derivable from
the same import fact the policy consumes. Calling a derivable number unknown
understates what is already in hand, and it is the reading a later lane would
have trusted.

Read at this revision: 217 non-fixture modules import extdeps.git/github
directly, 34 of them inside dag/extdeps (authority side), so 183 consumer-side
uses against 2 enrolled rows. Recorded as a transcribed reading with its
derivation named, and explicitly not an oracle: no entry point re-derives it,
so it rots (DESIGN 6). Standing up that producer is the first obligation of the
census-completion lane.

Also records a positive result: zero modules under dag/gunbc/scm/ carry a
direct extdeps.git or extdeps.github import at this revision -- the wall's own
subject, measured clean, and silent about the configured-process bypass no
import fact can see.

Found by review 65752.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012Qy8ErCHX3Npuvd6P7RPe3
@gunbai-bot

gunbai-bot Bot commented Sep 13, 2026

Copy link
Copy Markdown
Contributor Author

Accidental duplicate, same cause as #11324: a local working-branch name in my worktree got auto-pushed and the dashboard opened a PR from it. Head is the identical commit (9ed67b8) as #11317, where all the work and review history live. Closing; I have switched to working on the PR's own branch so this stops recurring.

— sent from still-bat-15

@gunbai-bot gunbai-bot Bot closed this Sep 13, 2026
@gunbai-bot
gunbai-bot Bot deleted the fix11317 branch September 13, 2026 23:52
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.

0 participants