hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Gang Tim Liu (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HIVE-3343) Hive: Query misaligned result for Group by followed by Join with filter and skip a group-by result
Date Wed, 08 Aug 2012 00:10:11 GMT

    [ https://issues.apache.org/jira/browse/HIVE-3343?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13430757#comment-13430757
] 

Gang Tim Liu commented on HIVE-3343:
------------------------------------

If we remove "b.k1 < 5", it will generate right result.

"explain extended" shows Select Operator has different order of columns (_col0, _col3, _col4,
_col2) from filter operator's (_col0,_col2,_col3,_col4). Here is snapshot:

Reduce Operator Tree:
        Group By Operator
          ...
          outputColumnNames: _col0, _col1
          Select Operator
            ...
            outputColumnNames: _col0, _col3, _col4, _col2
            Filter Operator
              ...
              predicate:
                  expr: (_col2 < 5.0)
                  type: boolean
              File Output Operator
                ...
                table:
                    ...
                    properties:
                      columns _col0,_col2,_col3,_col4
                      columns.types string,double,double,double
                      escape.delim \
                ...
                
> Hive: Query misaligned result for Group by followed by Join with filter and skip a group-by
result
> --------------------------------------------------------------------------------------------------
>
>                 Key: HIVE-3343
>                 URL: https://issues.apache.org/jira/browse/HIVE-3343
>             Project: Hive
>          Issue Type: Bug
>          Components: Query Processor
>            Reporter: Gang Tim Liu
>            Assignee: Gang Tim Liu
>
> This simple Hive query would generate wrong result:
> select a.key, b.k2, b.k3
> from src a
> join (
>   select key, 
>          min(key) as k,
>          min(key)+1 as k1,
>          min(key)+2 as k2,
>          min(key)+3 as k3
>   from src
>   group by key
> ) b
> on a.key=b.key and b.k1 < 5;
> 0       3.0     1.0
> 0       3.0     1.0
> 0       3.0     1.0
> 2       5.0     3.0
> The right result is:
> 0	2.0	3.0
> 0	2.0	3.0
> 0	2.0	3.0
> 2	4.0	5.0

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message