[
https://issues.apache.org/jira/browse/FLINK-5414?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15873024#comment-15873024
]
ASF GitHub Bot commented on FLINK-5414:
---------------------------------------
Github user wuchong commented on the issue:
https://github.com/apache/flink/pull/3338
Hi @fhueske , yes, Calcite forces an Calc after each aggregate that only
renames fields, because we rename every aggregates in Table API which is not
necessary. I changed the logic of getting projections on aggregates to only
rename the duplicate aggregates. And that works good, no more Calc appended.
Hi @haohui , the ArrayRelDataType is still NOT NULL. I reverted [that
line](https://github.com/apache/flink/blob/master/flink-libraries/flink-table/src/main/scala/org/apache/flink/table/calcite/FlinkTypeFactory.scala#L121)
which is not need to be changed in this PR.
Cheers,
Jark Wu
> Bump up Calcite version to 1.11
> -------------------------------
>
> Key: FLINK-5414
> URL: https://issues.apache.org/jira/browse/FLINK-5414
> Project: Flink
> Issue Type: Improvement
> Components: Table API & SQL
> Reporter: Timo Walther
> Assignee: Jark Wu
>
> The upcoming Calcite release 1.11 has a lot of stability fixes and new
> features. We should update it for the Table API.
> E.g. we can hopefully merge FLINK-4864
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
