Github user chenghao-intel commented on the pull request:
https://github.com/apache/spark/pull/9429#issuecomment-153540121
After checking with Hive like:
```
hive> select sum(a-b) as ab from mytable group by b with rollup;
FAILED: SemanticException [Error 10210]: Grouping sets aggregations (with
rollups or cubes) are not allowed if aggregation function parameters overlap
with the aggregation functions columns
```
Hive actually doesn't support the overlap with the aggregation functions
columns. Probably we can have a simple fixing based on the current master
branch if we need to support that.
And after double checking, the master branch will be optimized for
expression constant folding while with the `Expand` operator, it means better
performance than re-implemented based on UDTF, so I am a little struggling
which is the better approach for the implementation.
---
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]
