[
https://issues.apache.org/jira/browse/FLINK-6250?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15978933#comment-15978933
]
ASF GitHub Bot commented on FLINK-6250:
---------------------------------------
Github user stefanobortoli commented on the issue:
https://github.com/apache/flink/pull/3732
@fhueske, I agree with you about the risk of temporary DIST() ingestion.
Perhaps we could meanwhile just work on the "ProcessFunction + Code
generation" keeping the DIST function for test purposes tests. My concern is
that the code my change again and all the work would just be wasted. To be
honest, the code generation is quite new to me, and I will have to learn to
work on that. Meanwhile, I have almost completed a version that relies on
current code generation, nesting the distinct logic. As it is almost done, I
will share this one as well and then if necessary move to the code generation.
what do you think?
> Distinct procTime with Rows boundaries
> --------------------------------------
>
> Key: FLINK-6250
> URL: https://issues.apache.org/jira/browse/FLINK-6250
> Project: Flink
> Issue Type: Sub-task
> Components: Table API & SQL
> Reporter: radu
> Assignee: Stefano Bortoli
>
> Support proctime with rows boundaries
> Q1.1. `SELECT SUM( DISTINCT b) OVER (ORDER BY procTime() ROWS BETWEEN 2
> PRECEDING AND CURRENT ROW) FROM stream1`
> Q1.1. `SELECT COUNT(b), SUM( DISTINCT b) OVER (ORDER BY procTime() ROWS
> BETWEEN 2 PRECEDING AND CURRENT ROW) FROM stream1`
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
