[
https://issues.apache.org/jira/browse/FLINK-2138?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14605774#comment-14605774
]
ASF GitHub Bot commented on FLINK-2138:
---------------------------------------
Github user gaborhermann commented on the pull request:
https://github.com/apache/flink/pull/872#issuecomment-116736041
Sorry for not making myself clear.
I would actually go for
4. Only the Scala function (both in the streaming and batch API)
I don't understand how changing from partitioner implementation to function
implementation in the batch API would mess up determining the compatibility of
the partitioning. By compatibility I mean the type of the key must be the same
as the input of the partitioner.
I suppose there was another reason (that I do not understand) for choosing
the partitioner implementation for the Scala batch API, so if (4) is not an
option, I would go for (2) (only partitioner, sync with batch API).
> PartitionCustom for streaming
> -----------------------------
>
> Key: FLINK-2138
> URL: https://issues.apache.org/jira/browse/FLINK-2138
> Project: Flink
> Issue Type: New Feature
> Components: Streaming
> Affects Versions: 0.9
> Reporter: Márton Balassi
> Assignee: Gábor Hermann
> Priority: Minor
>
> The batch API has support for custom partitioning, this should be added for
> streaming with a similar signature.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
