[
https://issues.apache.org/jira/browse/FLINK-6105?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16090343#comment-16090343
]
ASF GitHub Bot commented on FLINK-6105:
---------------------------------------
Github user StephanEwen commented on the issue:
https://github.com/apache/flink/pull/4316
We could make this change. I have not seen a lot of use of
`InterruptedIOException`, probably because it is a bit of a strange class, with
its public mutable int field.
I am +/- 0 on this. Do you have a concrete case where this change would
lead to a benefit?
> Properly handle InterruptedException in HadoopInputFormatBase
> -------------------------------------------------------------
>
> Key: FLINK-6105
> URL: https://issues.apache.org/jira/browse/FLINK-6105
> Project: Flink
> Issue Type: Bug
> Components: DataStream API
> Reporter: Ted Yu
> Assignee: mingleizhang
>
> When catching InterruptedException, we should throw InterruptedIOException
> instead of IOException.
> The following example is from HadoopInputFormatBase :
> {code}
> try {
> splits = this.mapreduceInputFormat.getSplits(jobContext);
> } catch (InterruptedException e) {
> throw new IOException("Could not get Splits.", e);
> }
> {code}
> There may be other places where IOE is thrown.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
