Repository navigation
Hoist a body annotation to module scope: main's parse phase refuses it, leaving every PR's floor with no declaration index - #11822
Merged
Conversation
…t (§4c), which leaves every PR's floor with no declaration index dag/test/claim/roadmap/roadmap_review_function_witness_test.dag carried an 8-line // block inside the_aggregate_regroups_by_file_counts_the_overlap_and_names_the_incomplete_calls (landed by #11761). The required parse phase refuses body-grain annotations (only module-item grain is modeled), so the parse-phase declaration index is never lent and the floor refuses with ArmSetConsumerPlanningUnavailable on every PR whose merge ref includes it (seen on #11744). The block is rationale, not a claim, so it moves above the declaration it describes; the witness runs unchanged (5 PASS). Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
gunbai-bot Bot
pushed a commit
that referenced
this pull request
Sep 20, 2026
The floor job was GREEN while its log refused: parse FAIL at mac.dag:182 'source annotation sits inside a declaration body', then floor refused cause=ArmSetConsumerPlanningUnavailable -- no parse-phase declaration index was lent to the floor. That is the fail-open #11836 is fixing, and the parse failure was mine: an explanatory comment inside parse_mac_octet_at's body. DESIGN 4c admits annotations only on module-scope declarations, which is the class #11822 just repaired on main. The rationale now sits above parse_mac_octet_at, the declaration it describes. Every changed file was scanned for in-body annotations: none remain. Co-Authored-By: Claude Opus 5 (1M context) <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
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.
What
dag/test/claim/roadmap/roadmap_review_function_witness_test.dagcarried an 8-line//block insidethe_aggregate_regroups_by_file_counts_the_overlap_and_names_the_incomplete_calls(landed by #11761). DESIGN §4c admits only module-item-grain annotations; the required parse phase refuses body-grain ones — 8parse FAIL … source annotation sits inside a declaration bodyrows — so no parse-phase declaration index is lent and the floor refuses withArmSetConsumerPlanningUnavailableon every PR whose merge ref includes it (observed on #11744's fleet-lane run 35495424607).The block is rationale ("the count is derived, not transcribed"), not a claim, so it moves above the declaration it describes. No semantic change; the witness still runs: 5 PASS locally on the built seed.
Evidence
The fleet lane's
floorjob on this PR is the executing check: parse phase green, floor planned. Third main-side red found since the build-only gate (#11776, #11790 before it); this one was invisible to the PR path until #11791 restored the fleet lane.🤖 Generated with Claude Code