[teiid-issues] [JBoss JIRA] (TEIID-3573) Infinispan-dsl-cache translator: Operator <> incorrectly handles NULL values

Steven Hawkins (JIRA) issues at jboss.org
Mon Jul 27 22:00:04 EDT 2015


    [ https://issues.jboss.org/browse/TEIID-3573?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13093265#comment-13093265 ] 

Steven Hawkins commented on TEIID-3573:
---------------------------------------

> The issue with changing SupportNot = False, is that the pushed down criteria is no longer done and end up doing table scans.

Yes, that is expected.

> The combination of intNum <> 23 and intNum is Null criteria is an issue in both cases.

I'm not sure what you mean.

> But in the first case where supportNot = TRUE, <null> values are returned in the result set when only specifying intNum <> 23.

Since the backend will likely perform a full scan for a negated predicate, there isn't as much overhead as you would expect for turning off support.

It would be more work, impose additional capability limitations, and require extension metadata to pursue the fix that was done for ldap (where the predicate is pushed and evaluated in the engine), so I'm ok with the quick fix for now.


> Infinispan-dsl-cache translator: Operator <> incorrectly handles NULL values
> ----------------------------------------------------------------------------
>
>                 Key: TEIID-3573
>                 URL: https://issues.jboss.org/browse/TEIID-3573
>             Project: Teiid
>          Issue Type: Bug
>          Components: Misc. Connectors
>    Affects Versions: 8.7.1
>            Reporter: Filip Elias
>            Assignee: Van Halbert
>
> Operator '<>' returns true for NULL <> 1
> Example:
> {code}
> select intKey,intNum from smallA where intNum<>1
> {code} 
> It returns also rows which have NULL in column intNum
> I believe that NULL <> 1 is not true in SQL.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)


More information about the teiid-issues mailing list