[
https://issues.apache.org/jira/browse/FLINK-8689?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16449240#comment-16449240
]
ASF GitHub Bot commented on FLINK-8689:
---------------------------------------
Github user walterddr commented on a diff in the pull request:
https://github.com/apache/flink/pull/5555#discussion_r183222267
--- Diff:
flink-libraries/flink-table/src/main/scala/org/apache/flink/table/codegen/AggregationCodeGenerator.scala
---
@@ -151,8 +157,15 @@ class AggregationCodeGenerator(
}
}
- // initialize and create data views
- addReusableDataViews()
+ // get distinct filter of acc fields for each aggregate functions
+ val distinctAccType = s"${classOf[DistinctAccumulator[_, _]].getName}"
+ val isDistinctAggs = distinctAggs.map(_.nonEmpty)
--- End diff --
Hmm I tried to generate MapViewSpec in codegen but
`AggregationCodeGenerator.generateAggregations` call signature seems to miss
the typeinfo for accumulator (`accTypes` only shows the accumulator type, not
the `PojoField` info).
I can certainly changed to a more simple: `DistinctAccumulator[Object,
Long]` but I think there's pros and cons, also making a `MapTypeInfo` for plain
`Object` seems funny to me. How about we leave it to the future optimization.
> Add runtime support of distinct filter using MapView
> -----------------------------------------------------
>
> Key: FLINK-8689
> URL: https://issues.apache.org/jira/browse/FLINK-8689
> Project: Flink
> Issue Type: Sub-task
> Reporter: Rong Rong
> Assignee: Rong Rong
> Priority: Major
>
> This ticket should cover distinct aggregate function support to codegen for
> *AggregateCall*, where *isDistinct* fields is set to true.
> This can be verified using the following SQL, which is not currently
> producing correct results.
> {code:java}
> SELECT
> a,
> SUM(b) OVER (PARTITION BY a ORDER BY proctime ROWS BETWEEN 5 PRECEDING AND
> CURRENT ROW)
> FROM
> MyTable{code}
>
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
