[
https://issues.apache.org/jira/browse/HIVE-2318?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13079233#comment-13079233
]
Carl Steinbach commented on HIVE-2318:
--------------------------------------
@Vaibhav: It's going to take me a while to review this patch. In the meantime I
think the biggest problem is that there are no test cases. I understand that
adding test coverage for this particular feature is hard, but without tests I'm
worried that it's going to be impossible to maintain this feature due to the
significant differences in behavior between S3 and HDFS. Can you think of a way
that we can run test queries against an S3 filesystem? Maybe we should think
about adding a MockS3FileSystem? Does anything like that already exist at
Amazon?
> Support multiple file systems
> -----------------------------
>
> Key: HIVE-2318
> URL: https://issues.apache.org/jira/browse/HIVE-2318
> Project: Hive
> Issue Type: New Feature
> Reporter: Vaibhav Aggarwal
> Assignee: Vaibhav Aggarwal
> Attachments: HIVE-2318.patch
>
>
> Currently some of the Hive tasks like MoveTask, ConditionalMergeResolver
> assume that the data is being copied or moved on the same file system.
> These operators file if the source table is in one filesystem (like HDFS) and
> destination table is in another file system (like s3).
> This patch aims at:
> 1. Support moving data between different file systems.
> 2. Add support for file systems which do not support 'move' operation like s3.
> 3. Remove redundant operations like moving data from and to the same location.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
