Hi all,

I've been struggling here for a bit. I'm trying to use the saveOrUpdate to save/update an object where I don't want to use the primary key (id which is generated) but rather a different column (name for example). Avoiding a lot of background discussion, the situation I'm in I have no control over. I'm given a file of serialized objects of many differeny types generated at a remote site and I need to persist them locally. Obviously, the generated IDs of the objects are not going to match - they are from two SQL servers that have no communication to each other. So saveOrUpdate will always just create new rows. What I want to do is update if the 'name' is the same or save (create) if the name is unique.

I tried changing the @Id annotation to the name colum on my dev machine and saveOrUpdate worked perfectly. Sadly, I don't have control over that in production.

I've seen some good articles on merge, where I set the new object's ID to the local ID and then merge, but it poses some problems as I would have to know the object type. I do - but I'd love to keep this as generic as possible. There is a lof of code to change.

After all that rambling, here is my question - Is there an annotation I can put on the entity for saveOrUpdate to use in place of the ID? I can't use a hibernate XML file - everything is in code.




References:
Cameron McKenzie's 'Hibernate Made Easy book
http://stackoverflow.com/questions/2857 ... veorupdate

Just found:
viewtopic.php?f=1&t=994931&start=0


Which is helpful, but I still have to imagine there's an annotation for the Entity that would solve this problem.

@Entity
public class Something {

@TellHibernateToUseThisForSaveOrUpdate
String name;

@Id
@GeneratedValue
long id;

...
}


Thanks,
Mike