[
https://issues.apache.org/jira/browse/SPARK-50221?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17896749#comment-17896749
]
Daniel commented on SPARK-50221:
--------------------------------
Looking at this more, the proposed semantics would not be consistent with how
aggregation works with pipe operators, where the GROUP BY expressions arrive
out of the operator followed by the aggregate functions. I will close this.
As an aside, we might want to consider returning errors if any of the aggregate
functions contain column references that are not aggregated, since the pipe
operator returns the grouping expressions and it is always possible to use a
projection ({{{}|> SELECT{}}}) afterwards based on the grouping/aggregate
functions.
> GROUP BY ALL support
> --------------------
>
> Key: SPARK-50221
> URL: https://issues.apache.org/jira/browse/SPARK-50221
> Project: Spark
> Issue Type: Sub-task
> Components: SQL
> Affects Versions: 4.0.0
> Reporter: Daniel
> Priority: Major
> Labels: pull-request-available
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
