rmuir commented on PR #12055: URL: https://github.com/apache/lucene/pull/12055#issuecomment-1370211309
> Thanks @rmuir, you're right. The way rewrite methods are decoupled from the core protocol term intersection logic in MultiTermQuery did slip my mind. The curse of having a poor memory for things I'm not actively working on. But yes, agreed and I take back that part of my comment. I didn't really mean you specifically. I meant in the file itself. The `FILTER` rewrite is in danger of becoming a "wonder-do-it-all-rewrite-method" trying to solve all possible situations. But it doesn't need to be the default for all queries. I think the skipping logic proposed here makes total sense for stuff like WildcardQuery, PrefixQuery, RegexQuery, etc. But for TermInSetQuery? I think the rewritemethod there need not optimize for "dense" terms at all, but could focus on "PK lookups" (e.g. lots of terms with one posting) since we all know, people use it as a `JOIN` :( It could just be a default for the query, someone could always setRewriteMethod to a different one. -- 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 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
