[
https://issues.apache.org/jira/browse/HADOOP-8320?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14339708#comment-14339708
]
Vijay Bhat commented on HADOOP-8320:
------------------------------------
[~atm], [~daryn], I can take this.
Here's how I plan to refactor the code - move the checkPath logic from the
AbstractFileSystem and FileSystem classes to a "checkFileSystemPath" method in
the Path class. The AbstracFileSystem and FileSystem checkPath method will call
the common code in checkFileSystemPath (adapter pattern). This is because the
two classes aren't derived from any common ancestor.
However, I wanted your input of what I should do with the "canonicalizeUri"
method in the FileSystem class. Since the AbstractFileSystem checkPath method
should also be performing canonicalization (and it doesn't now), do you think
it makes sense to move that to the Path class or is there a more logical
location for it?
> FileSystem#checkPath and AbstractFileSystem#checkPath should share code
> -----------------------------------------------------------------------
>
> Key: HADOOP-8320
> URL: https://issues.apache.org/jira/browse/HADOOP-8320
> Project: Hadoop Common
> Issue Type: Improvement
> Components: fs
> Affects Versions: 2.0.0-alpha
> Reporter: Aaron T. Myers
> Assignee: Vijay Bhat
> Priority: Minor
>
> Per the discussion on HADOOP-8310, these two methods can be refactored to
> share some code.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
