[
https://issues.apache.org/jira/browse/JDO-730?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14058996#comment-14058996
]
Craig L Russell commented on JDO-730:
-------------------------------------
Using a variable to identify the second table should result in the additional
table in the FROM clause. The condition should then appear in the WHERE clause
of the query. Putting the condition into the JOIN clause is simply SQL
shorthand. The optimizer should not generate a different execution plan for the
two queries.
If this doesn't make sense, please post the SQL that you would like to see and
its execution plan and then the generated SQL with its execution plan so we can
understand your concern.
> Creating the innerjoin without relationship between objects
> -----------------------------------------------------------
>
> Key: JDO-730
> URL: https://issues.apache.org/jira/browse/JDO-730
> Project: JDO
> Issue Type: Wish
> Components: api
> Affects Versions: JDO 3.1-rc1
> Reporter: RaviRaja
> Assignee: Michael Bouschen
> Fix For: JDO 3.1
>
>
> Cannot make innerjoin like this without relationship:
> select a.x, b.y from Table1 as a INNER JOIN Table 2 as b ON a.z=b.w
> Note: a.z and b.w are not primary and foreign keys
--
This message was sent by Atlassian JIRA
(v6.2#6252)
