gsmiller commented on pull request #422: URL: https://github.com/apache/lucene/pull/422#issuecomment-957946153
I really like that you're experimenting with this idea! I commented over on the associated Jira issue in more detail, but just for completeness, I'll add a brief note here as well: I'm not sure we need the complexity of handling document adds that are out-of-order. Certainly for the purpose of pre-populating bitsets for the purpose of caching, we can assume docs will be visited in a strictly increasing order. So that allows some simplifications/optimizations (which I sketched out in a patch file attached to the Jira issue). Looking forward to seeing if this "moves the needle" in any way! While I'm not sure how often users will hit these cases (of strictly contiguous blocks of doc IDs), but it seems like a win-win to me in cases where it occurs (less memory and should be faster to intersect with other iterators). -- 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
