iverase opened a new pull request, #979: URL: https://github.com/apache/lucene/pull/979
This PR proposes to add a new method to SortedDocValues that helps users to advance an iterator to the next document that contains a different term that the current document, which can be specially useful when the index is sorted by this field. The method contains a default implementation but this PR produces as well a fast implementation when the index is sorted by this field and it has low cardinality. In this case we write to disk a jump table that allows to quickly skip documents instead of manually iterating through the docs. In https://issues.apache.org/jira/browse/LUCENE-10396 it is discussed some of the use cases where this method can be used, for example computing the number of unique values for documents that match a query. On the other hand, it diverges from the sparse index approach but as this ids less intrusive, it seems appealing. Note that in order to handle backwards compatibility, I have increase the version of the codec instead of creating a new one. -- 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
