[
https://issues.apache.org/jira/browse/FLINK-952?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14036025#comment-14036025 ]
Stephan Ewen commented on FLINK-952:
------------------------------------
That is true, there is actually a reason for that:
The data is treated as the type declared in the signature, because the type analysis and serializer generation happens at the "pre-flight" time. So if you have a MapFunction<A> and give it a B (subclass of A), it will treat the B as an A at runtime.
> TypeExtractor requires the argument types of the UDF to be identical to the parameter types
> -------------------------------------------------------------------------------------------
>
> Key: FLINK-952
> URL:
https://issues.apache.org/jira/browse/FLINK-952> Project: Flink
> Issue Type: Bug
> Reporter: Till Rohrmann
> Assignee: Till Rohrmann
>
> The TypeExtractor checks for each operation whether the DataSet element types are valid arguments for the UDF. However, it checks for strict equality instead of a subtype relationship. Thus the following computation would not work even though it should semantically be correct.
> DataSet[B].map(new MapFunction[A,A](){ A map(A x)}) with B being a sub class of A.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
Locked
