[teiid-issues] [JBoss JIRA] (TEIID-5783) OUTER JOIN yields wrong results when used with GROUP BY, CASE and COUNT DISTINCT

[Dmitrii Pogorelov (Jira)]

    [ https://issues.jboss.org/browse/TEIID-5783?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13751394#comment-13751394 ] 

Dmitrii Pogorelov commented on TEIID-5783:
------------------------------------------

Thx a lot [~shawkins] for the fix, I've just checked these changes locally for left join, right join and full join - everything works perfectly!

> OUTER JOIN yields wrong results when used with GROUP BY, CASE and COUNT DISTINCT
> --------------------------------------------------------------------------------
>
>                 Key: TEIID-5783
>                 URL: https://issues.jboss.org/browse/TEIID-5783
>             Project: Teiid
>          Issue Type: Bug
>          Components: Query Engine
>    Affects Versions: 12.0
>         Environment: teiid-12.0.0 on WildFly Full 14.0.1.Final (WildFly Core 6.0.2.Final)
>            Reporter: Dmitrii Pogorelov
>            Assignee: Steven Hawkins
>            Priority: Blocker
>             Fix For: 12.3, 12.1.2, 12.2.2
>
>         Attachments: test1.csv, test2.csv
>
>
> Some queries that involve an OUTER JOIN between objects from different data sources, a GROUP BY with a CASE statement and a COUNT DISTINCT yield inconsistent results on a SELECT expression which is identical to the GROUP BY expression with CASE.
> * Reproduced in combination of <File> LEFT JOIN <MySQL> and <File> LEFT JOIN <MS SQL> and the same for the other OUTER JOIN variants.
> * screenshot for the different queries are attached
> * sample data for reproduction is attached
> The same problem happens with LEFT OUTER JOIN, RIGHT OUTER JOIN and FULL OUTER JOIN.



--
This message was sent by Atlassian Jira
(v7.12.1#712002)