Skip to content

MachineShape construction wall may be dormant: test.claim.machine_shape_construction_wall matches no required_gate_prefixes entry, so its RED/GREEN controls may never execute on the required path - #9823

Merged
gunbai-bot[bot] merged 2 commits into
mainfrom
session/quick-eagle-249
Aug 31, 2026

Conversation

@gunbai-bot

@gunbai-bot gunbai-bot Bot commented Aug 31, 2026

Copy link
Copy Markdown
Contributor

Auto-opened by session-dashboard for session quick-eagle-249.
Pushing to session/quick-eagle-249 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.

@gunbai-bot
gunbai-bot Bot marked this pull request as ready for review August 31, 2026 14:28
@gunbai-bot
gunbai-bot Bot merged commit 5637e82 into main Aug 31, 2026
5 checks passed
@gunbai-bot
gunbai-bot Bot deleted the session/quick-eagle-249 branch August 31, 2026 16:24
@briansrls
briansrls restored the session/quick-eagle-249 branch August 31, 2026 16:24
briansrls pushed a commit that referenced this pull request Aug 31, 2026
…the lines git happened to print (#9829)

The changed-witness sublane selected ONE of the two sibling test fns in the
MachineShape construction wall that #9823 renamed into the required floor, and
the one it missed was the wall's discriminating RED.

The wall itself was fine: run 33413900349's own disposition receipt carries
`gate_red_synthetic_machine_shape_call` as `planned` / `passed`. It executed
via the ordinary Planned arm and only the projection line was absent.

The selection was not fine. `parse_unified_diff_added_paths` already rules that
a `rename to` destination is new-at-path -- "its declaration set is established
fresh at NEW" -- but that ruling only ever lifted the line-1 fail-closed
refusal. Nothing acted on the declaration set, so a moved file's identities were
attributed by which lines the move happened to edit. For a `/dev/null` add the
two agreed by accident (every line is a `+` line); for a rename they do not.

Measured over the 90 commits reachable from HEAD: 8 of 8 rename-destination
`.dag` files carrying test decls were under-enrolled, 87 of 103 identities
missed. A pure move with no content edit enrolls NOTHING.

It mattered beyond the projection line because changed-witness membership is
what OVERRIDES the cost-debt withhold and the outside-gate suppression, so the
same miss on a rostered identity is a silent decline of a witness whose author
is present -- the state v2.workflow.floor_changed_witness exists for.

The universe of the fix is the file's own parsed decl list, never the corpus:
the precise answer to "what does this path declare", not an absorbing widen
(DESIGN section 5). No refusal arm changed.

Two controls, run remote in release, in both directions:
  rename_destination_enrolls_every_test_decl_not_only_the_diff_touched_one
    feeds #9823's verbatim `git diff -U0` output and asserts enrolled ==
    declared. Fix disabled: FAILS. Proven, not asserted.
  in_place_modify_enrolls_only_the_touched_test_decl
    passes under both states, so the first test's red is discriminating rather
    than a build break, and the fix cannot be satisfied by widening.


Claude-Session: https://claude.ai/code/session_018aiTywconE9bm1hmUPjyft

Co-authored-by: Brian Searls <briansearls1@gmail.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
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