[
https://issues.apache.org/jira/browse/HADOOP-1795?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12523687
]
Hadoop QA commented on HADOOP-1795:
-----------------------------------
+1
http://issues.apache.org/jira/secure/attachment/12364765/HADOOP-1795.patch
applied and successfully tested against trunk revision r570881.
Test results:
http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/646/testReport/
Console output:
http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/646/console
> Task.moveTaskOutputs is escaping special characters in output filenames
> -----------------------------------------------------------------------
>
> Key: HADOOP-1795
> URL: https://issues.apache.org/jira/browse/HADOOP-1795
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.14.0
> Reporter: Frédéric Bertin
> Priority: Critical
> Attachments: HADOOP-1795.patch
>
>
> after a migration from 0.10.1 to 0.14.0, jobs can't generate output files
> with special characters in their name, just like '[' or ']' for example,
> because they are escaped during the {{Task.moveTaskOutputs}} process.
> For example, if you try to generate an output file named {{/foo/bar[0]}}, it
> ends up being named {{/foo/bar%5B0%5B}}.
> The culprit is {{Task.getFinalPath()}}, when it does
> {{relativePath.toString()}}, where I think it should do
> {{relativePath.getPath()}}.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
