mikebridge commented on PR #43490: URL: https://github.com/apache/superset/pull/43490#issuecomment-5536525332
⚠️ **Please do not merge on the current green badge** — a fix landed after it. @sadpandajoe raised three over-deletion findings on the deletion predicate. Adjudicated each by building the exact scenario as an integration test against head, with reverted-fix controls (replies on each thread): - **Finding 1** (pending tied to the later block at MySQL second granularity): **COVERED** — the `<=` unresolved-attempt tie protects it; control proves it load-bearing. - **Finding 2** (reason transition amid repeats): **COVERED** — the reason discrimination preserves the transition; control proves it. - **Finding 3** (force-purge record erased as a duplicate): **REAL over-deletion bug** — now fixed in `ac9bb3bf34`. `_repeats_an_earlier_block` excludes force-trigger rows so a force attempt is never collapsed. Added tests; 34 unit + 117 integration green; changed-file pre-commit clean. Because `ac9bb3bf34` adds a real deletion-logic fix on top of the previously-approved head, the approval is stale — the new commit needs a fresh committer look. There is also **one open policy call for the committer** on thread 3: a force block in a *resolved* streak still ages on the operational window (taxonomy-consistent; scoped in the docs and pinned by a test). If it should instead be fully immortal, that's a one-line extension I'll make on your word. A full 4-lens review of the fix (python/sqlalchemy/clean-code/tidy-first) is filed in the spec repo. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
