[ 
https://issues.apache.org/jira/browse/HIVE-4302?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13624589#comment-13624589
 ] 

Phabricator commented on HIVE-4302:
-----------------------------------

hbutani has commented on the revision "HIVE-4302 [jira] Fix how RowSchema and 
RowResolver are set on ReduceSinkOp that precedes PTFOp".

  Will add the test

INLINE COMMENTS
  ql/src/java/org/apache/hadoop/hive/ql/parse/SemanticAnalyzer.java:10543 No 
this is not about the having clause for windowing.
  This has to do with the GroupBy RR that can precede window processing. For 
Having expressions a Aggregation expression is added twice to the RR. This 
logic is to preserve this. The comment on line 10566 tries to explain what is 
going on.

REVISION DETAIL
  https://reviews.facebook.net/D10005

BRANCH
  HIVE-4302

ARCANIST PROJECT
  hive

To: JIRA, ashutoshc, hbutani

                
> Fix how RowSchema and RowResolver are set on ReduceSinkOp that precedes PTFOp
> -----------------------------------------------------------------------------
>
>                 Key: HIVE-4302
>                 URL: https://issues.apache.org/jira/browse/HIVE-4302
>             Project: Hive
>          Issue Type: Bug
>          Components: PTF-Windowing
>            Reporter: Harish Butani
>            Assignee: Harish Butani
>         Attachments: HIVE-4302.D10005.1.patch
>
>
> Currently the RowSchema and RowResolver for the ReduceSinkOp just point to 
> the 'input' Op's structures. This causes issues when input Op's structures 
> are changed during Optimization. See Jira 2340 of a problem.

--
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

Reply via email to