holdenk commented on PR #46143: URL: https://github.com/apache/spark/pull/46143#issuecomment-3483065675
While starting to get the context back @cloud-fan I think the cost based mechanism is the way to instead of just blacklisting UDFS. For example: `sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/CallMethodViaReflection.scala` is fairly "UDF like" but is not its self a UDF. Similarly, someone might write their own custom expression; regexes can also be fairly expensive, and so on. I can start with making this feature configurable while we discuss what we don't want to double eval. -- 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]
