[
https://issues.apache.org/jira/browse/MAPREDUCE-5021?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13584539#comment-13584539
]
Chris Nauroth commented on MAPREDUCE-5021:
------------------------------------------
{quote}
The method isn't exactly suitable for the feature described here, but it might
be an opportunity for refactoring and reuse.
{quote}
What I meant by this comment is that calling
{{FileUtil#createJarWithClassPath}} won't help build this feature, but the
method contains code that might be reusable if refactored out. The method
contains code very similar to what you just posted:
{code}
// Append all jars that match the wildcard
Path globPath = new Path(classPathEntry).suffix("{.jar,.JAR}");
FileStatus[] wildcardJars = FileContext.getLocalFSFileContext().util()
.globStatus(globPath);
if (wildcardJars != null) {
for (FileStatus wildcardJar: wildcardJars) {
classPathEntryList.add(wildcardJar.getPath().toUri().toURL()
.toExternalForm());
}
}
{code}
> Add an addDirectoryToClassPath method DistributedCache
> ------------------------------------------------------
>
> Key: MAPREDUCE-5021
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-5021
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Components: client, distributed-cache
> Affects Versions: 2.0.3-alpha
> Reporter: Sandy Ryza
>
> As adding a directory of jars to the class path is a common use for the
> distributed cache it would be easier on API consumers if they were able to
> call a method that would add all the the files in a directory for them.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
