jayzhan211 commented on PR #26228: URL: https://github.com/apache/datafusion/pull/26228#issuecomment-6244126440
How about concatenate the build batches before building either map, as the ArrayMap path already does. The hash map then indexes the concatenated batch directly, so build row i is input row i and the offset bookkeeping goes away. Inserting from the end still keeps each probe row's matches in build order. The output is the same as this PR -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
