parthchandra commented on PR #6784: URL: https://github.com/apache/datafusion-comet/pull/6784#issuecomment-6156457471
> If we think the common use case here is broadcasted tables (i.e., small input) then why not start with adding a config that enables SparkToColumnar for this data source, a la `COMET_CONVERT_FROM_PARQUET_ENABLED`? That avoids maintaining what looks to be complicated logic like what [#6784 (comment)](https://github.com/apache/datafusion-comet/pull/6784#discussion_r4253018828) alludes to. > > edit: I guess that's similar to what #6786 is? Exactly. #6786 is essentially what you are suggesting. It is in fact more generic. Any unsupported feature (not just a txt file) that can cause the build side to fall back can be covered by inserting a `SparkToColumnar` before the join. #6786 tries to be conservative though ( to avoid breaking DPP) -- 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]
