gsmiller opened a new issue, #12174: URL: https://github.com/apache/lucene/issues/12174
### Description This rewrite method (implemented in `MultiTermQueryConstantScoreBlendedWrapper`) relies on `DefaultBulkScorer` when there are more than 16 terms (with 16 or fewer, a `BooleanQuery` gets created which has custom bulk scoring logic already). This is inefficient since the bulk scoring will require managing 17 postings on a heap. A simple improvement would be to always fully rewrite to a bitset filter when bulk scoring and wrap that with a `DefaultBulkScorer` (like `MultiTermQuery#CONSTANT_SCORE_REWRITE`). This can probably be done pretty simply in the abstract base class (`AbstractMultiTermQueryConstantScoreWrapper`). There may be other ways to do this—possibly relying on the logic already in `BooleanScorer` to make better use of memory—but we can probably start simple here as well. I'm not sure how frequently users would actually have a top-level term-in-set query they want to score, but it would be nice to make this better. -- 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: issues-unsubscr...@lucene.apache.org.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
