Github user cloud-fan commented on the issue:
https://github.com/apache/spark/pull/17293
My main concern is about how to validate it. This is one pain point of
Spark: we don't control the storage layer and we can't trust any constraints
like nullable, primary key, etc.
I think it's a safe choice to always treat input data as nullable, and I
think it's all about performance, validating also have performance penalty, and
it may eliminate the benefit of non-nullable optimization.
BTW another reason is Spark doesn't work well if the nullable information
is wrong, we may return wrong result which is very hard to debug.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
