gsmiller commented on PR #841: URL: https://github.com/apache/lucene/pull/841#issuecomment-1154142817
> Anyway, let's benchmark it, but with the analysis above, I also agree we should actually start with the long[] API, and replace it with a byte[] one only if actually performs better. +1 to starting with `long[]` and then benchmarking a `byte[]` version when time permits. > If I understand your change correctly, then it creates a new long[] in each call to matches() right? I see two main problems here Yeah, good callouts. I put this together pretty quickly as a sketched out idea, and didn't think super deeply about it. I was going for an approach that would let users extend the long-based API as the common approach, but allow extending the byte-based API if they really care about performance (but maybe it's not even more performant... TBD!). At this point, I'm convinced we should go with the long-based API for the initial version. Let's get this functionality shipped and then we can benchmark, optimize, etc. -- 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
