Skip to content

Committed .gitattributes has drifted from its emitting authority: witness_committed_matches_emit_holds is false — find what changed it without regenerating, and regenerate - #9384

Merged
briansrls merged 2 commits into
mainfrom
session/crisp-stag-142
Aug 27, 2026

Conversation

@briansrls

Copy link
Copy Markdown
Contributor

Auto-opened by session-dashboard for session crisp-stag-142.
Pushing to session/crisp-stag-142 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 2 commits August 26, 2026 08:40
…er it when a new hand-Rust stage0 source was registered (#9244)

witness_committed_matches_emit_holds was false. The cause is not an edit to
.gitattributes or to its emitter — neither has been touched since #9284, which
is HEAD. It is an input change one authority away.

The unset ("!merge") rows are a COMPLEMENT, derived from
v2.compiler.self_host.stage0_crate_layout hand_maintained_stage0_filenames()
minus the generated-artifact registry overlap
(gunbc.stage0_rust_source_lifecycle_scaffold
derived_hand_maintained_disposition_repo_paths). #9244 added a
SeedRetainedIntrinsicRegistration row for partition_crate_boundary_host, a new
hand-maintained stage0 source. That row is an emission input, so the committed
projection went stale without either file changing. #9244 and #9284 touch
disjoint paths, so nothing conflicted, and the generated-artifact drift gate is
on the floor cut's unguarded list, so nothing refused either.

Effect while stale: partition_crate_boundary_host.rs matched the
src/v1/stage0/src/*.rs glob with no unset row after it, so a hand-maintained
file silently carried merge=generated-artifact — exactly the transposed-row
hazard the emitted header warns about, arrived at from the other direction.

The bytes are the emitter's own candidate, not a hand-edit: built gunbc and ran
tools.generated_artifact_gate main_wet at 7de77a4 on a clean tree; it wrote one
insertion across the whole registry, reproduced here verbatim.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
@briansrls
briansrls marked this pull request as ready for review August 27, 2026 01:02
@briansrls
briansrls merged commit b19faff into main Aug 27, 2026
6 of 9 checks passed
@briansrls
briansrls deleted the session/crisp-stag-142 branch August 27, 2026 04:17
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.

1 participant