Repository navigation
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
PR #6334 fixes a local-overlap inventory that normalizes CR/CRLF Git names to LF, but its author explicitly left POSIX execution and symlink coverage open. This independent audit closes that bounded environment gap at
e579ddc0e0b4614184b869f8d20cffdcc26380a6and supplies a self-contained differential replay.The new package constructs real CR, LF and CRLF filenames, executable notes and symlinks. Across seven filesystem states, it runs both the pinned canonical-main helper and #6334's helper through their offline JSON CLI. In five incomplete states, main falsely reports complete coverage while #6334 lists the exact two missing paths. Real-note restoration, untracked notes, leaf/parent/dangling aliases, the result limit and unchanged cache bytes are checked. The implementation repair remains credited to josecarlosgonzalezroque-design; no shared helper or selected result is changed here.
Validation
macOS 26.5 arm64, CPython 3.12.16, Apple Git 2.50.1:
-Oin optimized mode. The unchanged predecessor local-suite driver's children remain nonoptimized, as documented.git diff --checkpass.The unrelated current-main
test_research_current_status.pyhas one reproduced failure out of six tests: CI documentation still names frontier6030 while selected metadata names frontier7000. Existing #7276 addresses that repair. This submission does not claim globally green tests or perform a scientific cold replay.Hosted CI result
Actions run 43600296628 completed with failure at the generic onboarding stage: 760 tests, 1 failure, 24 platform skips (735 passed). The sole failure is exactly the pre-existing
CurrentCiNavigationTests.test_current_ci_section_matches_selected_recordmismatch reproduced above. The package-specific hosted replay was therefore not reached. This does not replace the successful local four-command registered replay and Git-history-free source export.Scope, evidence and coordination
README and exact evidence, source origins, original test logs, machine receipts, and explicit replay hooks are included. Base:
e54523517d0f03e58352e647c403a5af3e525a57, priority revision 8. The portable-replay card is retained with an explicit scope change to #6334's handoff. This is evidence for the exact pending head, not the eventual merge result; integration with later main changes remains a separate check.Advance claim. Handoff-first and targeted PR/Discussion searches found no matching POSIX audit. The overlap helper refreshed 901 open PR bodies and 7,512 Discussion records. The final PR refresh succeeded; the final Discussion refresh failed with EOF and retained the earlier 13:20 UTC complete snapshot. Sparse local and closed-PR coverage remain limited; no exhaustive novelty claim.
Credit josecarlosgonzalezroque-design (#6334), lee-phys (#4419), Ali Romman (#5725), Nick Petalas (#6161), Douglas Colkitt and retained tool contributors. Prepared by Hao Yan (@FanPUYen) with substantial OpenAI Codex assistance and separately scoped agents for overlap review, test execution, package scaffolding and review. Apache-2.0. No theorem, supplier admission, merge, or selected-frontier change.