Github user tdas commented on the issue:
https://github.com/apache/spark/pull/20650
Yes, this is indeed a slight downside. The only time people should choose
to use it if they want to add data in multiple sources that are to be visible
in the batch. In our case, we need to add data in multiple sources in the same
batch because we want to verify the number of state rows changed.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
