You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Rami (JIRA)" <ji...@apache.org> on 2016/07/06 12:01:11 UTC

[jira] [Commented] (FLINK-4029) Multi-field "sum" function just like "keyBy"

    [ https://issues.apache.org/jira/browse/FLINK-4029?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15364203#comment-15364203 ] 

Rami commented on FLINK-4029:
-----------------------------

Sure, thanks for working on this:

So if we have pojo for example with those fields:
int count=1;
int bytesConsumed=32898;

Now I have a stream of those pojos. I want to be able to do:

.sum ("count","bytesConsumed")

which will sum all count values together and all bytesConsumed together. and I end up with the count of all and the sum of all the bytes consumed ( of course within the defined window)

> Multi-field "sum" function just like "keyBy"
> --------------------------------------------
>
>                 Key: FLINK-4029
>                 URL: https://issues.apache.org/jira/browse/FLINK-4029
>             Project: Flink
>          Issue Type: Improvement
>          Components: DataStream API
>            Reporter: Rami
>            Assignee: Ivan Mushketyk
>            Priority: Minor
>
> I can use keyBy as follows:
> stream.keyBy(“pojo.field1”,”pojo.field2”,…)
> Would make sense that I can use sum for example, to do its job for more than one field:
> stream.sum(“pojo.field1”,”pojo.field2”,…)



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)