[
https://issues.apache.org/jira/browse/DERBY-3613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12588600#action_12588600
]
Knut Anders Hatlen commented on DERBY-3613:
-------------------------------------------
The first change (two rows --> one row) happened in this commit:
------------------------------------------------------------------------
r267239 | bandaram | 2005-09-02 20:07:39 +0200 (Fri, 02 Sep 2005) | 6 lines
DERBY-504: Disable pushing down DISTINCT when number of columns from the parent
query and the subquery don't match. Can lead to wrong results.
Thanks to Rick for persistant reviews and running the tests.
Submitted by Knut Anders Hatlen ([EMAIL PROTECTED])
------------------------------------------------------------------------
The old behaviour (one row --> two rows) was reintroduced in this commit:
------------------------------------------------------------------------
r516454 | abrown | 2007-03-09 17:37:20 +0100 (Fri, 09 Mar 2007) | 8 lines
DERBY-681: Remove the "wrap group by's in a subselect" rewrite in the parser.
This patch preserves the having clause through bind and optimize phases and
then, during the final rewrite for aggregates in the GroupByNode, it transforms
the having clause to a valid restriction. See text file attached to the Jira
for more information.
Contributed by Manish Khettry ([EMAIL PROTECTED])
------------------------------------------------------------------------
> SELECT DISTINCT field FROM TABLE_NAME GROUP BY field, field2
> -------------------------------------------------------------
>
> Key: DERBY-3613
> URL: https://issues.apache.org/jira/browse/DERBY-3613
> Project: Derby
> Issue Type: Bug
> Affects Versions: 10.3.2.1
> Environment: Windows XP
> Reporter: Artur Kuś
>
> Query 'SELECT DISTINCT field FROM TABLE_NAME GROUP BY field, field2' not
> work ok.
> Distinct is ignored !!!
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
