aokolnychyi commented on pull request #35395: URL: https://github.com/apache/spark/pull/35395#issuecomment-1071474428
I agree it may be beneficial to plan the filtering subquery differently if the job planning is fast. I am open to exposing a different scan builder for runtime filtering. Then data sources may choose to reuse the same scan if they want to but Spark will be flexible. We can discuss it one more time on the PR for runtime filtering. For now, a separate scan builder seems a better option. In either case, using `SupportsRuntimeFiltering` should work well and will be backward compatible with the current API. What do you think, @cloud-fan? -- 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]
