[
https://issues.apache.org/jira/browse/HDFS-5213?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13771137#comment-13771137
]
Colin Patrick McCabe commented on HDFS-5213:
--------------------------------------------
bq. I still want to see Paths, not Strings, for the DFS methods involving cache
directives. It's non-standard and weird for the reasons I mentioned before, and
DFS has error checking for being passed a URI for a different FileSystem.
After considering this, I agree. we can have the paths normalized and
absolutized in DistributedFileSystem. I started doing this, but it got to be
too many changes for one patch. It's a little tricky how it interacts with
directive validation. Let's do this in a future frontend patch.
> separate PathBasedCacheEntry and PathBasedCacheDirectiveWithId
> --------------------------------------------------------------
>
> Key: HDFS-5213
> URL: https://issues.apache.org/jira/browse/HDFS-5213
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: datanode, namenode
> Reporter: Colin Patrick McCabe
> Assignee: Colin Patrick McCabe
> Attachments: HDFS-5213-caching.001.patch
>
>
> Since PathBasedCacheEntry is intended to be a private (implementation) class,
> return PathBasedCacheDirectiveWithId from all public APIs instead. Some
> other miscellaneous cleanups in the caching RPC stuff.
--
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
