Github user mgaido91 commented on the issue:
https://github.com/apache/spark/pull/22008
Actually the SQL cartesian product already uses an optimized (but unsafe
version) of the cartesian RDD. So this rule is not needed at all. We might
still want tough to expose the optimized version in the RDD API too. I am
closing this PR and eventually create a new one for the RDD API. Thanks.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
