[
https://issues.apache.org/jira/browse/FLINK-2576?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14803028#comment-14803028
]
ASF GitHub Bot commented on FLINK-2576:
---------------------------------------
Github user fhueske commented on the pull request:
https://github.com/apache/flink/pull/1138#issuecomment-141111654
@jkovacs and @r-pogalz, thank you very much for this PR and the detailed
description!
It's quite a bit of code so it will take some time to be reviewed. I hope
to give feedback soon.
Nonetheless, we can start a discussion about the handling of projection for
outer joins. By changing the type information to `GenericTypeInfo<TupleX>` to
support tuples with null values, a `DataSet<TupleX>` cannot be used (in a join,
groupBy, reduce, ...) as before because the runtime will use completely
different serializers and comparators. Therefore, I am more in favor of not
supporting projection for outer joins.
> Add outer joins to API and Optimizer
> ------------------------------------
>
> Key: FLINK-2576
> URL: https://issues.apache.org/jira/browse/FLINK-2576
> Project: Flink
> Issue Type: Sub-task
> Components: Java API, Optimizer, Scala API
> Reporter: Ricky Pogalz
> Priority: Minor
> Fix For: pre-apache
>
>
> Add left/right/full outer join methods to the DataSet APIs (Java, Scala) and
> to the optimizer of Flink.
> Initially, the execution strategy should be a sort-merge outer join
> (FLINK-2105) but can later be extended to hash joins for left/right outer
> joins.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
