HeartSaVioR edited a comment on pull request #32834: URL: https://github.com/apache/spark/pull/32834#issuecomment-857460454
No, I'm referring the output of reorderKey, which will be used to partition, and compare with key in state store. Both partitioning and comparison should be the same with existing state store. My point is all about state store compatibility. If there's any chance for the output of reorder to be changed due to various ways like different Spark versions, table metadata (if that can affect the reordering), etc, state store compatibility will be broken. -- 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. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
