[
https://issues.apache.org/jira/browse/HADOOP-4044?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12638430#action_12638430
]
Doug Cutting commented on HADOOP-4044:
--------------------------------------
Owen, do you think this logic is unique to HDFS? If we add support for links
to another FileSystem implementation would we want to share this logic with
HDFS? If so, could we do it through a common base-class, like
LinkableFileSystem. Then FileSystems that never intend to support links could
continue to directly implement the end-user methods, while FileSystems that
support links would extend the new subclass and implement the LinkResult<T> SPI
methods. Is that consistent with your proposal?
Also, the pseudo code above will cause a stack overflow for a circular link.
Shouldn't we catch loops earlier than that? And if the loop-detection logic is
non-trivial, won't we end up with either lots of duplicated code or something
like the anonymous classes in Dhruba's patch?
> Create symbolic links in HDFS
> -----------------------------
>
> Key: HADOOP-4044
> URL: https://issues.apache.org/jira/browse/HADOOP-4044
> Project: Hadoop Core
> Issue Type: New Feature
> Components: dfs
> Reporter: dhruba borthakur
> Assignee: dhruba borthakur
> Attachments: symLink1.patch, symLink1.patch, symLink4.patch,
> symLink5.patch, symLink6.patch, symLink8.patch, symLink9.patch
>
>
> HDFS should support symbolic links. A symbolic link is a special type of file
> that contains a reference to another file or directory in the form of an
> absolute or relative path and that affects pathname resolution. Programs
> which read or write to files named by a symbolic link will behave as if
> operating directly on the target file. However, archiving utilities can
> handle symbolic links specially and manipulate them directly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
