[
https://issues.apache.org/jira/browse/HIVE-838?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12756237#action_12756237
]
Raghotham Murthy commented on HIVE-838:
---------------------------------------
i guess it would be good to include this fix in 0.4.
couple of comments:
1. can you change the query with the empty partition to be a non-aggregate
query? need to make sure additional rows are not added to the result. something
like:
{code}
select * from
(select count(1) from src
union all
select 1 from srcpart where ds = '2009-08-09'
)x;
{code}
2. remove the code instead of commenting it out.
> in strict mode, no partition selected error
> -------------------------------------------
>
> Key: HIVE-838
> URL: https://issues.apache.org/jira/browse/HIVE-838
> Project: Hadoop Hive
> Issue Type: Bug
> Components: Query Processor
> Reporter: Namit Jain
> Assignee: Namit Jain
> Attachments: hive.838.1.patch
>
>
> set hive.mapred.mode=strict;
> select * from
> (select count(1) from src
> union all
> select count(1) from srcpart where ds = '2009-08-09'
> )x;
> Is it a blocker for 0.4 ?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.