Github user cloud-fan commented on the issue:
https://github.com/apache/spark/pull/21599
I don't think we can change behavior like this without introducing a
config. In Spark 3.0 we will add a "strict mode" and closely follow SQL
standard. But for Spark 2.4, how about we keep the "java style behavior" as it
is? We can add some document about it though.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
