feat(reports): leftover-map rank on grouping comparison leftover-pair payload (v2.62.0) - #843
Conversation
… payload (v2.62.0)
Return persisted leftover_map_rank on GET /api/reports/compare/{period}
leftover pairs (ADR 0305). Rank 0 is shown when stored. A missing rank
stays null and omits that leftover-map comparison rank badge. Do not
invent leftover-map rank from leftover-map coordinates, leftover-map
distance, leftover-map reconstruction, leftover-map axis share, or the
count of unused axes. No new columns.
|
Important Review skippedAuto reviews are disabled on base/target branches other than the default branch. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
seonghobae
left a comment
There was a problem hiding this comment.
Parent-convergence finding: the PR metadata was still pinned to obsolete #842 12e40c93... while feat/leftover-map-compare-axis-share-v2610 is now 93f52af35e5643e77cfe5dff4501c5ee06ac4932. The base has been retargeted to the current branch head. GitHub now correctly reports this Draft child as conflicting; compare is ahead 1 / behind 31 from merge base 12e40c93.... Preserve the single ADR 0305/rank-payload delta, but do not treat its old test merge or predecessor checks as current-parent evidence. #842 itself has unresolved authorization and ADR-authority REDs, so reconstruct #843 non-force only from the repaired #842 product head; do not resolve conflicts by dropping parent security/governance changes or by force-rebasing.
seonghobae
left a comment
There was a problem hiding this comment.
Current-head repair finding: ADR 0305 is still Decision status: Accepted while this PR is Draft and has no qualifying current-head independent approval. The ADR also cites the superseded axis-share authority path 0304-leftover-map-compare-axis-share.md; repaired parent #842 now owns that decision as ADR 0368/Proposed. Preserve the rank-payload delta, but reconstruct it non-force on #842f013dfb, converge the parent reference to ADR 0368, and return ADR 0305 to Proposed before exact-head validation. Do not transfer predecessor GREEN/approval evidence.
seonghobae
left a comment
There was a problem hiding this comment.
Reconstruction run 34007566889 did not surface a product defect. The backend validation command selected zero tests (135 deselected / 0 selected, pytest exit 5) because the harness used -k period_comparison, while this child’s rank-payload assertion is added to test_seed_period_report_surfaces_on_get_reports in backend/tests/test_api.py. Keep the existing rank-payload/ADR findings, but classify this run failure as a harness selector defect. I repaired the isolated reconstruction lane non-force at 83ec341cec3e7708e70463adcf66dc59c0fd1cd8 to target that actual API regression; do not promote until the new run is GREEN and a clean tree is produced against exact parent f013dfb... and exact child 2a5ab4d....
Summary
Return persisted leftover-map rank on
GET /api/reports/compare/{period}leftover pairs (ADR 0305 / v2.62.0). Preserve the grouping-comparison axis-share, coordinates, reconstruction/share identities,Y/E/R/U, distance, and coverage contracts. Missing rank remains null; rank 0 remains valid. No schema/new columns, invented score, theta, inner product, cosine, or length.Live authority
#842@f013dfb53b255e8e0be1dfb783cb41dca7eeb83e2a5ab4d735a1240997150578b151c6b6492e2f43The one-commit rank-payload delta remains valid and must not be closed or silently dropped. The repaired parent owns mixed-visibility fail-close and axis-share ADR 0368/Proposed.
Current-head findings
Review
5123876360records two product repair findings on this exact child head: ADR 0305 is prematurely Accepted while Draft, and it still cites superseded axis-share ADR 0304 instead of current-parent ADR 0368. Preserve ADR 0305 and the rank-payload fixture/contract, return the decision to Proposed, and converge the parent reference to ADR 0368.Non-force reconstruction RCA
The first reconstruction run
34007566889on staging699040b23501eca36cb1f00cce1a4a2ceb3fd39esuccessfully verified the exact parent, reapplied the child rank-payload delta, resolved only reviewed overlap, converged ADR 0368 references, restored ADR 0305/Proposed, and verified v2.62.0 identity. It then failed in the backend validation harness, not in a backend assertion:pytest backend/tests/test_api.py -k 'period_comparison'selected zero tests (135 deselected) and exited with pytest code 5. Frontend validation and clean-tree commit were therefore correctly skipped.A concurrent non-force repair has been adopted rather than overwritten: staging is now
83ec341cec3e7708e70463adcf66dc59c0fd1cd8, changing the selector to the actual current regressiontest_seed_period_report_surfaces_on_get_reports. Follow-up run34010029163is queued. This is a harness correction only; no product GREEN or product-head movement is claimed.If the repaired run produces a clean tree while parent and child remain unchanged, recreate only that verified tree as a new two-parent commit with current child head first and #842 exact parent second, then ordinary fast-forward this product branch and refresh descendants immediately. Keep Draft until exact-product-head required/browser/a11y evidence and qualifying independent approval are fresh. No force-push, destructive rebase, self-approval, gate weakening, or predecessor GREEN transfer.