Hi!
I'm using Hibernate 3.1.3 and i'm having a problem with optimistic locking.
I hava a mapping with a joined sublcass.
The parent class is configured with the optimistic-locking attribute in 'all'
and the joined subclass is configured with the dynamic-update attribute in 'true'.

When I run a test case, i'm getting this exception

Caused by: org.hibernate.MappingException: optimistic-lock attribute not supported for joined-subclass mappings: com.np.iaso.mamo.model.PaqueteSuelto
at org.hibernate.persister.entity.JoinedSubclassEntityPersister.<init>(JoinedSubclassEntityPersister.java:110)
at org.hibernate.persister.PersisterFactory.createClassPersister(PersisterFactory.java:58)
at org.hibernate.impl.SessionFactoryImpl.<init>(SessionFactoryImpl.java:216)
at org.hibernate.cfg.Configuration.buildSessionFactory(Configuration.java:1176)
at com.np.iaso.


I didn't found in the documentation any info related to optimistic locking in joined subclasses. Is it supported?

Thanks a lot and sorry for my english!

Some more info.
The optimistic-lock with 'version' is it supported, but i'm not able to use this mode because other applications have access to the same database and don't know how to handle version numbers or even timestamp.

I have a similar problem: we can't (for legacy reasons) use optimistic locking in our application, but setting optimistic locking to anything but "version" throws the exception nhiggs references. The code backs this up, though the documentation doesn't comment on it, and I've not been able to track down a reason for this behavior. Does anyone know why version-based optimistic locking is required for joined-subclass?

I'm confused as well. The source explicitly disallows any optomistic lock strategy other than "version". From JoinedSubclassEntityPersister's constructor:



I can't find any documentation in the manual, HiA, the wiki, or this forum actually stating that this is an as-designed rule... even though that condition seems to have been in the class for some time (source history wasn't preserved prior to a refactoring pre-hib3).

Can anyone confirm that this constraint is still needed and maybe share some insight as to why? Christian, it might be a useful note in the inheritance modeling chapter.. as this is a limitation that will certainly affect those decisions for some folks.

Thanks in advance!

Phill

are there any news or outcomes about this problem?
i have the same problem because the reference did not mention any limitations to the optimistic-lock="dirty".

Yup. Mr. Ebersole was kind enough to fix the issue I entered:

http://opensource.atlassian.com/project ... e/HHH-2242

thx for the link.
so the optimistic locking mode "dirty" will not be supported for joined-subclasses in future?