[
https://issues.apache.org/jira/browse/HIVE-4197?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13610242#comment-13610242
]
Ashutosh Chauhan commented on HIVE-4197:
----------------------------------------
Hmm.. One way I can think of is to sort rows ourselves. We need to add an extra
column in mapper (combination of task.id and monotonic increasing row number)
and than in PTFOperator have an implementation of ByteBasedList which uses
Priority Queue to store data in memory instead of ByteBuffer. This Priority
Queue will insert data in order of this new column we added and dropping it
soon after.
However, this seems to me is pretty heavy-handed approach. I will prefer
something simpler. Harish, this problem of stable sort is getting exposed only
on hadoop2, so I will suggest for now lets focus on other issues as we think
and come up with a better solution for it, since on hadoop1 we don't have such
a problem.
> Bring windowing support inline with SQL Standard
> ------------------------------------------------
>
> Key: HIVE-4197
> URL: https://issues.apache.org/jira/browse/HIVE-4197
> Project: Hive
> Issue Type: Bug
> Components: PTF-Windowing
> Reporter: Harish Butani
> Attachments: WindowingSpecification.pdf
>
>
> The current behavior defers from the Standard in several significant places.
> Please review attached doc; there are still a few open issues. Once we agree
> on the behavior, can proceed with fixing the implementation.
--
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
