gsmiller commented on PR #12089: URL: https://github.com/apache/lucene/pull/12089#issuecomment-1416895938
Thanks @rmuir! I'll have a look at the PRs tomorrow. I've got a bad case of jet lag kicking in at the moment and my brain is turning to mush. In the meantime, I took a pass at simplifying the proposal here in this PR. There was a lot of complexity (as you pointed out) for edge-cases and such. I tried to keep it focused on just the core idea of seeking some terms to get a better cost estimate, in a way that adds minimal complexity. Benchmark results are about the same as before, so quite a bit simpler and no real performance difference: [TiSBenchResults_Simplified.md.txt](https://github.com/apache/lucene/files/10610038/TiSBenchResults_Simplified.md.txt) Of course, this still relies on optimizing internally instead of leveraging `IndexOrDocValuesQuery`, so I recognize it's not getting at the core of your concern. I figured it would be good to at least simplify a bit though so we have something a little cleaner/simpler to discuss. Thanks again for the feedback and related PRs! -- 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
