[
https://issues.apache.org/jira/browse/CALCITE-7242?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18032660#comment-18032660
]
weihua zhang commented on CALCITE-7242:
---------------------------------------
For example, implementations like this one
(https://github.com/apache/impala/blob/master/java/calcite-planner/src/main/java/org/apache/impala/calcite/rules/ExtractLiteralAgg.java)
have strict limitations and lack universality, so enhancements can be made
based on this foundation.
> Implement a rule to get rid of the LITERAL_AGG into other database can handle
> -----------------------------------------------------------------------------
>
> Key: CALCITE-7242
> URL: https://issues.apache.org/jira/browse/CALCITE-7242
> Project: Calcite
> Issue Type: Improvement
> Components: core
> Reporter: weihua zhang
> Priority: Major
>
> Calcite can create an Aggregate RelNode with a single LITERAL_AGG
> AggregateCall to check for existence in a subquery. In this case, the
> Aggregate will do a group by, and just spit out a literal value for each row.
> The LITERAL_AGG function does not exist in other databases. In order to
> create a RelNode structure that other databases supports, this Aggregate is
> turned into an Aggregate that still does the "group by" on the relevant
> groups, but places a Project RelNode on top that returns the literal value.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
