Github user superbobry commented on the issue:
https://github.com/apache/spark/pull/21157
Yes, that is correct. That is why I think hijacking behaviour should be
removed. It silently slows down the job and does not notify the user that a
trivial change such as making the namedtuple importable could result in a
speedup.--- --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
