I agree that EOModel support would likely always be a second-class
citizen in the Cayenne Modeler, and the EOModel users would have to
limit which parts of the Cayenne modeler that they used as not all
Cayenne features would translate to an EOModel.
However, I don't think the amount of work would really be all that
great. The only missing functionality is already on the TODO list
for the standard Cayenne modeler (prototypes, user properties, uml
diagramming, certain kinds of inheritance). I also don't think the
divergence is going to be significant. At some point, I think you
will see the ability to save an equivalent EOModel right from the
Cayenne modeler. But a dedicated EOModel modeler makes more sense
because of the useability issues above.
I haven't looked at the Entity Modeler yet, but may do so down the road.
On 3/12/07, Mike Schrag <mschra..dimension.com> wrote:
> > Hmm. There may be more work involved than I originally considered.
> > You'd need to somehow preserve EOPrototypes as the Cayenne modeler
> > doesn't yet support them. User properties would also need to be
> > preserved, although a feature to support user properties has been the
> > topic of serious discussion on the cayenne lists recently, and will
> > probably be implemented by someone (maybe me) in the near future.
> Yeah, I looked at this option before working on Entity Modeler and
> just decided it was going to be fair amount of work, had a high
> probability of making both products confusing, and would be
> potentially problematic as both products moved forward in
> progressively divergent directions.
This archive was generated by hypermail 2.0.0 : Mon Mar 12 2007 - 14:52:33 EDT