Jan

This sounds strange to me. I think you can change the business object
(the non-Peer objects) and add methods to them or fields that need not
be stored in the database. The moment you have the need for
fields/properties that you want to store in the database, you should
change your scheme and recreate the Peer objects (the bussiness objects
are not touched by regeneration, so changes there remain.

Yes, you can change both the extended business object and even the extended Peer. Only the Base* classes are overwritten by Torque when you generate the model. So, yes, you could just manually change the Torque schema and regenerate the model.


So I am wondering: if you add those properties to your objects, do you
also adapt the peer classes?

No, you don't *have* to do anything to the Peer classes, unless you want to extend their function.


Ok, i can imagine that if you add fields according to some kind of
naming convention and then regenerate this into the xml-schema. But then
again, is this really a big thing. You have to remember the properties
you add and add them to the xml scheme.

Sure, this is simple for simple changes. But what if you refactor your model, making significant changes, such as breaking out a set of properties into a one-to-many relationship? When you deploy the changed app, you need to change all the existing data in the database. That makes it more complex.


Eric


--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to