Re: Strange error message
Camille Troillard <[email protected]>
| Newsgroups | gmane.comp.web.webobjects.eof |
|---|---|
| Message-ID | <[email protected]> |
Hi Chuck, On Tuesday, Oct 8, 2002, at 20:21 Europe/Paris, Chuck Hill wrote: > What database are you using? EOF with Oracle or FrontBase usually > uses a sequence generator on (in your example) RootClass to generate > the PK for objects at or below that root. You should be able to do > the same thing. I am using FrontBase. I now realize that I have not assigned external entity names to entities like "RootClass". Do you think it could be the root of my problem? Camille > Camille Troillard wrote: > >> Chuck, that's it! >> Now I perfectly understand because my model used to work when I was >> using unique PKs. >> I guess I must rollback to the first implementation ... and find a >> way to generate PK shorter than the default implementation. >> Thanks to everybody that helped me to solve the problem!!! >> Camille >> On Tuesday, Oct 8, 2002, at 19:59 Europe/Paris, Chuck Hill wrote: >>> Ah Ha! I know the answer. EOs in an inheritance hierarchy MUST >>> MUST MUST have unique primary keys. That is the only way that EOF >>> can differentiate which class an object is. In your case >>> >>> RootClass ----> EntityA >>> ----> EntityB >>> >>> No object of EntityA, EntityB (or RootClass if not abstract) can >>> have the same PK. If EOF fetches an instance of EntityA with a PK >>> of 1000, and then fetches an instance of EntityB with a PK of 1000 >>> it will think (because the EOGlobalID is the same) that the EntityB >>> instance is actually the EntityA instance. > > > -- > > Chuck Hill [email protected] > Global Village Consulting Inc. > http://www.global-village.net > > >