Github user cloud-fan commented on the issue:
https://github.com/apache/spark/pull/21698
For `zip`, it's hard to define what result is "corrected", given the fact
that RDD is unordered. I think sample should be similar.
`repartition` is special because we may change the number of outputs if
this bug is triggered, which is obviously wrong result.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
