[Hibernate-issues] [Hibernate-JIRA] Commented: (HHH-1837) discussion: implicit joins are always INNER

[Bill S. (JIRA) <no...@at...>]

    [ http://opensource.atlassian.com/projects/hibernate/browse/HHH-1837?page=comments#action_23338 ] 

Bill Schneider commented on HHH-1837:
-------------------------------------

This is really a request for documentation.  I agree the HQL behavior is correct. 

The only issue is that, because the implicit-join syntax looks deceptively similar to something you could have in JSTL-EL, the SQL inner-join behavior is not immediately obvious to someone looking at this as a Java developer.  

So I think it would be helpful to just include a blurb to the effect of the above in the section in the hibernate reference on HQL implicit joins.  

> discussion: implicit joins are always INNER
> -------------------------------------------
>
>          Key: HHH-1837
>          URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-1837
>      Project: Hibernate3
>         Type: Improvement

>     Reporter: Bill Schneider

>
>
> It would be helpful to explain the consequences in the docs re. implicit joins always resulting in INNER joins.  
> This causes semantics of HQL expressions to be quite different from similar-looking JSTL expressions.  
> in particular consider (copied from comment on HHH-989) this HQL expression
> "cat.mate.name= ? OR cat.name=?"
> if cat.mate is null, this will not return any results, **even if**  the right hand cat.name=? expression matches.
> A similar condition in JSTL (<c:if test="${cat.mate.name = ... || cat.name = ...}">) would behave differently.
> The HQL behavior may be correct, but sure could be confusing at first.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://opensource.atlassian.com/projects/hibernate/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira