[teiid-issues] [JBoss JIRA] (TEIID-3525) dependent join handling of multi-way joins

[Steven Hawkins (JIRA)]

     [ https://issues.jboss.org/browse/TEIID-3525?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Steven Hawkins resolved TEIID-3525.
-----------------------------------
    Resolution: Done


Updated the handling in cases without access patterns to detect when duplicate columns are being used and to ensure that only 1 will be utilized.

> dependent join handling of multi-way joins
> ------------------------------------------
>
>                 Key: TEIID-3525
>                 URL: https://issues.jboss.org/browse/TEIID-3525
>             Project: Teiid
>          Issue Type: Quality Risk
>          Components: Query Engine
>            Reporter: Steven Hawkins
>            Assignee: Steven Hawkins
>             Fix For: 8.12
>
>
> In scenarios where the same column is used multiple times in a multi-way dependent join, it's affect may be duplicated in source predicates.
> For example:
> select pm1.g1.e1, pm1.g1.e2, pm2.g1.e2 FROM pm1.g2, pm1.g1, /*+ makedep */ pm2.g1 where pm1.g1.e1 = pm2.g1.e1 and pm1.g2.e1 = pm2.g1.e1
> Should have a source query of the form:
> SELECT g_0.e1 AS c_0, g_0.e2 AS c_1 FROM pm2.g1 AS g_0 WHERE g_0.e1 IN (<dependent values>) ORDER BY c_0
> Rather than having multiple dependent values predicates.



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