[
https://issues.apache.org/jira/browse/CALCITE-6164?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17796497#comment-17796497
]
Julian Hyde commented on CALCITE-6164:
--------------------------------------
Not a goal of this test. In fact we should have pushed back on the two major
methods, not related to CSV file parsing, that were added to that test in
CALCITE-4260.
> add testPushDownProjectAggregateWithAvg on CsvTest
> --------------------------------------------------
>
> Key: CALCITE-6164
> URL: https://issues.apache.org/jira/browse/CALCITE-6164
> Project: Calcite
> Issue Type: Test
> Components: tests
> Affects Versions: 1.36.0
> Environment: linux+java
> Reporter: Caican Cai
> Priority: Trivial
> Labels: pull-request-available
> Fix For: 1.37.0
>
>
> The avg operator in the agg operator is actually different from the
> aggregation operators such as max and count, because avg will be split into
> count/avg. I think the avg operator can be formed into a separate test
> I have observed that some other open source projects display avg tests
> separately, e.g: spark sql
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
