epotyom commented on PR #13559: URL: https://github.com/apache/lucene/pull/13559#issuecomment-2228524276
@msokolov thanks for looking into it! > I wonder if DocIdSetBuilder would help? I guess it doesn't support accessing as a BitSet while under construction though. Yes, I think this is exactly the reason we don't use it. > Still maybe a similar class inspired from that, or even a fork of SparseFixedBitset could be helpful, although I do wonder how often SPFBS has a large number of trailing zeros. Yeah I was curious if making in SparseFixedBitset makes sense, but I haven't found similar use cases in Lucene code. Also looking at benchmark results, it is probably a case for creating a fork. > Another idea -- would it help your use case? -- would be to support nextSetBit(start, end) . We could do this without adding any additional tracking in existing SparseBitSet methods. Basically instead of asking SBS to keep track, the client might be able to do so? Ah, I like this idea! Working on it. Not sure if it worth adding an abstract method to the base BitSet class, so I'll start with SparseFixedBitSet itself and see how it goes. -- 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
