aokolnychyi commented on issue #2319: URL: https://github.com/apache/iceberg/issues/2319#issuecomment-799796432
@rdblue, I am afraid the changes done in Spark and whatever we have in our actions invalidate only cached data. I don't think we invalidate plans. For example, if someone creates a view that references a V2 table, Spark won't invalidate it. In our `SparkTable`, we have `refreshEagerly` flag, which is set as true whenever the caching catalog is not used. Using the caching catalog without expiry, ensures the view will always be invalidated whenever there is an operation in the current session. Operations happing in other query engines won't invalidate the Spark view, though. It looks like we should be able to handle this case by moving the caching/invalidation logic to `SparkTable`. Thoughts? ---------------------------------------------------------------- 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]
