mos-dt-0andClaude Opus 5
0d5aae2c84
Registered before the reviewer read the diff. Review 66 and the merge-gate GO are void at the new head per D-50; Jason's approval is standing and executes once this re-gates. Verified independently first so the review does not re-spend on it: range-diff shows commits 2-5 patch-equivalent with only commit 1 differing, and that difference is the test:framework-shell chain resolved as a union preserving main's terminal-green contract harness AND all four RM-03 suites plus the pre-existing branch-absent test. Nothing dropped either way. CI 2199 is 9/9 including clone, the machine check is bound and clean, and mergeable is now true. R1 is the deliverable: the guard must actually FAIL on asserted non-readiness. RM-03 exists because the queue guard is zero-information — it returns pass for every possible input — so proving its tests pass is not the same as proving the guard can fail. R2 is the risk: confirm no queue-guard semantic moved in the rebase, because one resolved quietly re-breaks exactly what this fixes. R4 checks the tri-state can represent an indeterminate result without presenting it as a pass, which is the defect verbatim (state=unknown exit 0). Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
43 lines
4.7 KiB
Markdown
43 lines
4.7 KiB
Markdown
# RM-03 / PR #1032 — PRE-REGISTERED RE-REVIEW (post-freshening)
|
||
|
||
**Subject head (exact):** `868b9b871a5118762aa5b8aafecd2f0592f7ab5c`
|
||
**Rebased onto:** `main` @ `f4fd5967` **Registered by:** `mos-remediation` before the reviewer read the
|
||
diff. **Reviewer:** `rev-974`.
|
||
|
||
**Why this exists:** review 66 APPROVED and the merge-gate GO were bound to `78ec47cd`; `main` drifted
|
||
under a long hold and the PR went `mergeable=false` (**D-50 — a GO has a shelf life**). Both verdicts
|
||
are VOID at the new head. Jason's merge approval is **standing** and executes once this re-gates.
|
||
|
||
## What the orchestrator already verified (do not re-spend here)
|
||
|
||
- `git range-diff 78ec47cd...868b9b87`: commits **2–5 are `=` (patch-equivalent)**; only commit 1 differs.
|
||
- That single difference is the `test:framework-shell` chain, resolved as a **UNION** — main's
|
||
`test-terminal-green-contract.sh` **and** RM-03's `tristate`, `github-checks`, `merge-queue-branch`,
|
||
`merge-head-pin`, plus the pre-existing `branch-absent`. **Nothing dropped from either side.**
|
||
- CI #2199 at the exact head: 9 children, 9 success incl. `clone`; machine check exit 0, bound, no anomalies.
|
||
- `mergeable` is now **true**.
|
||
|
||
## Checks
|
||
|
||
| id | check |
|
||
| ------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
||
| **R1** | ★ **THE GUARD CAN FAIL.** RM-03 exists because the queue guard is ZERO-INFORMATION (D-23) — it returns pass for every possible input. Prove the rebased head's guard **actually fails** on asserted non-readiness, not merely that its tests pass. This is the whole deliverable. |
|
||
| **R2** | ★ **NO QUEUE-GUARD SEMANTIC MOVED IN THE REBASE.** Confirm the range-diff equivalence independently. A queue-guard semantic resolved quietly re-breaks the thing this PR fixes. |
|
||
| **R3** | **Union is complete in BOTH directions:** no RM-03 suite lost to main's version, no main suite (esp. `test-terminal-green-contract.sh`) lost to RM-03's. Enumeration count reconciles. |
|
||
| **R4** | **Tri-state is real** (`verified` / `written-unverified` / `failed`) — P-WRAPPER-001. Verify an indeterminate result cannot present as a pass; that is the exact defect (`state=unknown exit 0`). |
|
||
| **R5** | The guard's **exit-asserting non-null-case** tests are genuine — each fails for its own stated reason against pre-fix code, not merely present (D-8). |
|
||
| **R6** | `test:framework-shell` is **runnable to completion in CI** (canonical env). Known: it exits 97 on some hosts via a Bash 5.2.15 `BASH_LINENO` assertion — if you hit that, report it as an environment limitation, do not treat it as a finding and do not let it mask one. |
|
||
| **R7** | Exact-head CI terminal-green, full `-f json`, counts stated incl. `clone`; machine-checked with `--expect-commit 868b9b871a5118762aa5b8aafecd2f0592f7ab5c`. |
|
||
| **R8** | `Closes #1019` present in the body (D-38c delivery-issue rule). |
|
||
| **R9** | Nothing weakened by the rebase: no test, case, or assertion removed or relaxed to make it land. |
|
||
|
||
## Reviewer instruction
|
||
|
||
Verdict bound to `868b9b871a5118762aa5b8aafecd2f0592f7ab5c`, evidence enumerated, posted durably under
|
||
your own identity. If a check is unrunnable, **say so**.
|
||
|
||
**R1 is the deliverable and R2 is the risk.** Attack outside this set. **Do not cite the queue guard's
|
||
own green as evidence of anything** — it remains inert until this very PR lands (D-23).
|
||
|
||
**No one dispatching this review may state a conclusion on an open check (D-39).**
|