[
https://issues.apache.org/jira/browse/FLINK-12174?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dawid Wysakowicz closed FLINK-12174.
------------------------------------
Resolution: Fixed
Fix Version/s: 1.9.0
Implemented in 9d0951ecf2466e7b27e41b2fa894a0979527f07a
> Introduce FlinkAggregateExtractProjectRule and remove extractFieldReferences
> ----------------------------------------------------------------------------
>
> Key: FLINK-12174
> URL: https://issues.apache.org/jira/browse/FLINK-12174
> Project: Flink
> Issue Type: Improvement
> Components: Table SQL / API
> Reporter: Hequn Cheng
> Assignee: Hequn Cheng
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.9.0
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> Currently, when parsing Table API expression, an inner project will be added
> to projects fields for {{Aggregate}}s. As the code show below:
> {code:java}
> if (!extracted.getAggregations.isEmpty) {
> val projectFields = extractFieldReferences(expressionsWithResolvedCalls)
> wrap(
> operationTreeBuilder.project(extracted.getProjections,
> operationTreeBuilder.aggregate(emptyList[Expression],
> extracted.getAggregations,
> operationTreeBuilder.project(projectFields, operationTree)
> )
> )
> )
> }
> {code}
> This optimization is not very suited to added here, instead, we can use a
> rule to achieve this. The `extractFieldReferences` method can also be removed
> if we use a rule which also makes the Expression parsing logic more clear.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
