alessandrobenedetti commented on PR #12314: URL: https://github.com/apache/lucene/pull/12314#issuecomment-1557354463
I pushed a commit with the query time simplification (only MAX strategy is supported). The diff is simpler but I am not convinced it's better. I also remembered a bit more why I chose to explicitly differentiate Index/query time behavior for multi-valued search: The main reason is that at the moment the multi-valued approach requires some extra work both at indexing and query time, from data structures and processing perspective. With a clear distinction, I tried to impact the single-valued approach as less as possible (literally using different code when multi-valued occurs). For example, at query time this is visible in the org.apache.lucene.util.hnsw.NeighborQueue. With the latest commit, I am pretty sure single valued search gets a performance regression (that shouldn't be the case with the 'NONE' strategy as it was before). I anyway think it was a good exercise to do the simplification and we can iterate/revert from there. -- 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
