Github user cloud-fan commented on the issue:
https://github.com/apache/spark/pull/13680
> `(4-bit nullbits)[0110] [offset0] [-offset3] [-offset3] [offset3]`
Why negative the `offset3`? The old format negative it to indicate
nullability, but we have null bits now.
> it fills non-zero into a field that have `null`
This is fine. We zero out the null field instead of leaving some random
bytes there because we wanna make sure equality is deterministic. The new
rule(using next offset for null element) is also deterministic.
---
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]