aneesh-db opened a new pull request, #16700: URL: https://github.com/apache/lucene/pull/16700
### Description QueryRescorer sorts first-pass hits by document ID and advances a scorer to each target. For two-phase scorers, advancing the filtered scorer iterator confirms approximation candidates until the next full match, even though only the known first-pass documents need to be checked. This change advances the two-phase approximation directly and calls matches only when the approximation lands on a first-pass document. It preserves the existing path for single-phase scorers and caches confirmations for duplicate ScoreDoc inputs. A deterministic regression test verifies that only the requested documents are confirmed and that the resulting scores are unchanged. ### Tests - Direct JUnit run of TestQueryRescorer: 9 tests passed - Google Java Format 1.30.0 dry run - git diff --check -- 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]
