You'd add to CM a place to look for other other models (either a classpath approach or explicitly identify the other models). CM wouldn't need to edit the other models, but in the relationship inspector allow referencing the other entities in the model group. At least that's a first approach. I didn't use model groups extensively and my memory is a bit rusty. I can try to find more info on them later if that helps.
Thanks, mrg On Thu, Nov 3, 2011 at 9:33 AM, Andrus Adamchik <[email protected]> wrote: > What does that mean in practical terms? How do we define that "group" in the > Modeler? > > On Nov 3, 2011, at 4:25 PM, Michael Gentry wrote: > >> EOModeler had model groups where you could reference external models. >> That's probably the approach Cayenne Modeler would need to take. >> >> On Thu, Nov 3, 2011 at 9:12 AM, Andrus Adamchik <[email protected]> >> wrote: >>> This is one of the unsolved things. The relationships between DataMaps are >>> supported in runtime (all DataMaps are loaded in a single namespace, so >>> entities can reference each other freely). >>> >>> During modeling I am not sure how we can do that. We just had this >>> discussion in my company recently, but no solution yet. If there are ideas >>> how that can be implemented, let's discuss. >>> >>> Andrus >>> >>> >>> On Nov 3, 2011, at 1:45 PM, Hugi Thordarson wrote: >>> >>>> Hi all! >>>> >>>> Most of our projects share a set of common DB tables (users, permissions >>>> etc). The code and cayenne.xml for these entities resides in a standalone >>>> project, works great. >>>> However, I'm wondering if I can create relationships from entities in our >>>> client projects to entities in the common project? And if so, do I have to >>>> do it in code, or is there any way for CayenneModeler to pick up the >>>> entities in the common project, so I can model the relationships in the >>>> GUI? >>>> >>>> Cheers, >>>> - hugi >>>> >>>> >>>> PS: I started a huge thread regarding outer joins, and I haven't had the >>>> time to respond yet. Preparing a massive thesis on that one ;-) >>> >>> >> > >
