jpountz opened a new pull request, #12838: URL: https://github.com/apache/lucene/pull/12838
This is a new iteration on #12810, which I had to revert because of test failures when docFreq is a multiple of 128. We currently have a hack when the doc freq is not a multiple of 128 in order to not write skip data for the last block. I suspect it made sense in the past when postings were only about advancing postings, but storing this last skip data point also helps simplify advancing because postings enums can then always assume that the last doc ID of a full block is equal to the next skip doc. This also helps dynamic pruning by also storing impacts for this last block. See the second commit of this PR for more details. -- 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
