[
https://issues.apache.org/jira/browse/HDFS-5224?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13792125#comment-13792125
]
Chris Nauroth commented on HDFS-5224:
-------------------------------------
No worries on the delay, and thanks for the feedback.
bq. I think we should make it so the PCD path is qualified at creation time,
else you might create a PCD with a relative path, change working directories,
and the PCD now refers to a different path. I think the easiest way of doing
this is a factory method in DFS.
This is actually consistent with all of the other DFS methods that accept a
{{Path}}. If you create a relative {{Path}} instance, then call
{{DistributedFileSystem#setWorkingDirectory}}, and then pass the {{Path}} to a
method, the method internally resolves against the new working directory, not
the working directory that was in effect at {{Path}} creation time. Thus,
adding a factory method like this on DFS would create an inconsistency in the
API. If a caller doesn't like this behavior for their
{{PathBasedCacheDirective}} instances, then they can create it using the result
of {{DistributedFileSystem#makeQualified}}, passing the relative {{Path}}, and
this would be consistent with the rest of the DFS API. Let me know what you
think.
> Refactor PathBasedCache* methods to use a Path rather than a String
> -------------------------------------------------------------------
>
> Key: HDFS-5224
> URL: https://issues.apache.org/jira/browse/HDFS-5224
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: datanode, namenode
> Affects Versions: HDFS-4949
> Reporter: Andrew Wang
> Assignee: Chris Nauroth
> Attachments: HDFS-5224.1.patch, HDFS-5224.2.patch
>
>
> As discussed in HDFS-5213, we should refactor PathBasedCacheDirective and
> related methods in DistributedFileSystem to use a Path to represent paths to
> cache, rather than a String.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
