Github user cloud-fan commented on the issue:
https://github.com/apache/spark/pull/20579
After some more thoughts, we should try our best to not introduce behavior
change to existing data sources. How about we only add this check for
file-based data sources(all of them are built-in)? We can have a follow-up JIRA
to add this check for data source v2.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
