Repository navigation
Bind review source fingerprints to full-precision priority inputs - #7
Merged
Merged
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Swapping two scores or amounts one binary64 ULP apart changes the selected voucher, but the existing
%.12gsource serialization produces the same fingerprint. Reproduced both failures before the fix.Record fingerprint format
selection-v2, retain 17 significant digits for binary64 inputs, quote every source field and name the ledger/rule sections. The source fingerprint now detects these selection-relevant changes; the separate queue checksum continues to bind the displayed CSV bytes. Refresh only the committed selection record and explain how to retain old CSV/record pairs.Validation: all 503 tests pass, including two new one-ULP ranking regressions; Ruff and diff checks pass. The full 30,128-voucher pipeline ran locally. Relative to its same-environment baseline, only
review_plan_summary.jsonchanges; the 300-row queue, selection counts, random sampling, model/rule reports and synthetic benchmark remain byte-identical. Repeated review-plan generation leaves every report byte unchanged. Every uploaded file's Git blob hash was checked against its local bytes.