Skip to content

Require history-backed navigation checks - #779

Merged
d6g8k5htny-coder merged 2 commits into
mainfrom
codex/cbz-navigation-history-20261109
Nov 9, 2026
Merged

d6g8k5htny-coder merged 2 commits into
mainfrom
codex/cbz-navigation-history-20261109

Conversation

@d6g8k5htny-coder

@d6g8k5htny-coder d6g8k5htny-coder commented Nov 9, 2026 •

Copy link
Copy Markdown
Owner

The navigation workflow used a shallow checkout, so historical-source reader tests reported skipped checks while the job succeeded. Fetch full history and run the two history-backed reader classes as an additional fail-closed step: the child must succeed, each selected control must report ok, and the final unittest summary must have a positive count and no skips. The existing discovery run and test bodies remain intact.

Local verification on the committed candidate: 25 focused workflow/discovery/guard controls passed in each Python mode; a real full-history checkout passed 20 selected tests, while a real depth-one checkout recorded 12 skips and was rejected. Nine designated workflow mutations were killed in both modes, with traceback code-origin controls rejecting subject, helper, and nested-callback assertions as oracle failures. The initial guard's false acceptances of forged earlier summary lines were reproduced and corrected. Evidence: Library file libfile_fde3_f-f232855b2dc081919315ab3c31a32f56_59a0f9f234e48191bc19fddcc0f80648, SHA-256 6fa1b43e564baa27270e311f44b56cd2e580b2564c8253805789ab17f4c39fcf.

This is a scoped workflow repair, not a full regression or CBZ-0 qualification. Other observed skip categories remain unresolved, and hosted checks and artifacts must be inspected at the current head. Canonical status records the evidence and the existing checkpoint/archive version split.

Hosted qualification at tested merge ca1154a06b4ad3afc31227846c87479b1361ff25: navigation artifact 13695969227 records 2,469 discovery tests with 25 skips and a separate 20-test history-reader run without skips. All five PR-triggered workflows returned success; the protected merge was rejected because required public-intake has not yet reported. Its normal trusted-base check must run and succeed before integration. The five sandbox, two browser and two Lean cases skipped by navigation also ran successfully in their separate hosted jobs; these observations do not turn the skipped discovery into full regression 100%.

A separate adverse finding concerns the existing node-status freshness mutation grader: a subject-origin AssertionError was wrongly counted as a kill in both child modes. Evidence is retained in libfile_fde3_f-606b8b159414819198b7a213ca93d32c_59a0f9f234e48191bc19fddcc0f80648, SHA-256 af7e53894196fd12ad6b0e42e0c022656358f4aa4e5f556c21b9b3b0c92f5c74; a disjoint repair is underway. This PR does not change that grader or claim its mutation evidence is provenance-qualified. The prior trusted public-intake runs failed to start during GitHub's scheduled pull_request_target brownout ending 18:00 UTC; all failed runs and the rejected merge remain recorded.

Copy link
Copy Markdown
Owner Author

Additional scoped execution at local commit e28d070b6a6c55d7c5e9c330f87becc97e83831a, tree-identical to hosted head 2d4403838f6d697e671834f12a9a87b379a93ad6: the G11 census passed in normal and optimized modes with 1,356 files, 69 tools, zero failing tools and zero production failures; both reports SHA-256 dcf86e70f06696fc9177f2eabfc51f80e7f143d49f288d4d54646f01da936f3d. Separately, Python 3.12.14 in an isolated environment ran the ten existing pytest bytecode-policy controls in each mode: 10/10, no skips. After the pinned wheel download failed, pytest 9.1.1, pluggy 1.7.0 and iniconfig 2.3.1 were built offline from recorded upstream Git commits; this is not a PyPI-wheel or hosted-run claim. Actual child receipts, logs and source/build provenance were read back from evidence file libfile_fde3_f-989fd5685bf481919f1e635d1c4a35ac_59a0f9f234e48191bc19fddcc0f80648, SHA-256 4b28a4aeea674545e070c7d511c833a2ad7597f4974af7bdd100e491063e0b42. The separate runs do not erase the historical full-discovery skips or award full regression or final G11.

Copy link
Copy Markdown
Owner Author

The exact-head navigation run 42666720029 / job 129332230105 completed successfully. Its checkout was tested merge ca1154a06b4ad3afc31227846c87479b1361ff25, parents 5ebe5771… and 2d440383…, tree df5509f1f25dc406bad95ab4b88c288300901a98, matching the local candidate tree. Raw artifact 13695969227, SHA-256 89a30aaa6980a2301cecc0ed41b79edc9ee7da846093317fa40b0de49f79809b, reports Python 3.11.18, 2,469 discovery tests with 25 skips, and the separate history-reader suite 20/20 with no skips. The twelve history-object skip events from #778's artifact are absent; the 25 other skip events are retained and full regression is not awarded. The raw artifact, tested-merge Git bundle and exact diff have been independently read back from libfile_fde3_f-a002871fd92481918f8b82129c824c43_59a0f9f234e48191bc19fddcc0f80648, SHA-256 3769f0f76deae7b5b07b7c34c0596790a67fda24ac9b6fadca3a441c70d1da57. Other required jobs and final integration remain subject to actual fresh checks.

@d6g8k5htny-coder
d6g8k5htny-coder merged commit 6ab126b into main Nov 9, 2026
12 checks passed
@d6g8k5htny-coder
d6g8k5htny-coder deleted the codex/cbz-navigation-history-20261109 branch November 9, 2026 18:01
d6g8k5htny-coder added a commit that referenced this pull request Nov 9, 2026
Preserve Main #779 and every decision row. The current Main decision ledger remains an exact prefix; the two recovery decisions follow it unchanged. The inventory tool, its tests, recovery guide and plan are byte-identical to the reviewed candidate. No proof or scientific disposition changes.

Actual composition: OpenAI/Codex. Local equivalent merge 3579831c has the identical tree.
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