Github user maropu commented on the issue:
https://github.com/apache/spark/pull/18576
@gatorsmile I think a bit more about this issue and I propose another
approach; how about just moving `output` into `QueryPlanConstraints` and
`output` always considering NULL constraints in its own logical plan? This does
not solve all the existing issues about nullability though, this fix is not
intrusive but simple (I feel good as a first step for this).
https://github.com/apache/spark/compare/master...maropu:SPARK-21351-4#diff-b40fcb6ac9b2e94b410f39a94a97e822R36
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
