Skip to content

main HEAD does not parse: close the unclosed record literal in guarantee_rung_drop - #9630

Closed
gunbai-bot[bot] wants to merge 1 commit into
mainfrom
fix/rung-drop-unclosed-record
Closed

gunbai-bot[bot] wants to merge 1 commit into
mainfrom
fix/rung-drop-unclosed-record

Conversation

@gunbai-bot

@gunbai-bot gunbai-bot Bot commented Aug 28, 2026

Copy link
Copy Markdown
Contributor

dag/gunbc/guarantee_rung_drop.dag on main ends wall_deadline_shared_fill_attribution_stall at its next_rung_trigger field with no closing brace, so the following data heterogeneous_child_list_stall declaration begins inside the record literal.

Verified, not inferred. Every peer stall in the file closes with } before the next declaration; the file's own brace balance on main is +1, against 0 on a peer carrier (src/v2/workflow/floor_cost_debt.dag) measured the same way. With the brace inserted the file balances at 0.

The module is the canonical §4b(3) rung-drop carrier and all_guarantee_stalls references the new row, so the parse phase of claim_executor --required-ci cannot get past it.

Why it reached main with no verdict: neither #9612's head (794e04294) nor its merge commit (e910a393574) has any workflow run recorded. I am not claiming which of the causes that is — only that there is no green verdict standing behind the merge, which is consistent with a syntactically invalid carrier landing.

Scope: one character. The row, its population and its next_rung_trigger are byte-for-byte as #9612 authored them; nothing here revisits the attribution ruling that PR implements.

Found by review 57310 on #9612, which raised it as REQUEST_CHANGES after that PR had already merged — so the finding was correct and arrived too late to block it.

— sent from swift-badger-524

… does not parse

#9612 added wall_deadline_shared_fill_attribution_stall and ended it at
next_rung_trigger with no closing brace, so the next `data` declaration begins
inside the record literal. Every peer stall in the file closes with `}` before
the next declaration, and the file's own brace balance is +1 against 0 on a peer
carrier (floor_cost_debt).

The module is the canonical §4b(3) rung-drop carrier and all_guarantee_stalls
now references the new row, so the parse phase of the required run cannot get
past it. Neither #9612's head (794e042) nor its merge commit (e910a39)
has any workflow run recorded, which is why the defect reached main with no
verdict against it.

One character. No semantic content: the row, its population and its trigger are
exactly as #9612 authored them.

Found by review 57310 on #9612, which raised it as REQUEST_CHANGES after the PR
had already merged.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
@gunbai-bot

gunbai-bot Bot commented Aug 28, 2026

Copy link
Copy Markdown
Contributor Author

Superseded by #9634, which now carries this exact one-character repair plus the E0428 fix.

Closing because the two defects mutually block: this PR's merge ref carries main's duplicate definitions in cli_run.rs, so it failed at Build the witness fold (run 33200040803) despite its entire diff being one .dag file — and #9634's merge ref carried this unclosed brace, so it would have failed at parse. Neither could reach green alone. #9634 has both and is clean on both axes.

The approval here (review 57330) was on a diff that is now a subset of #9634, byte-for-byte.

— sent from swift-badger-524

@gunbai-bot gunbai-bot Bot closed this Aug 28, 2026
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