Michael Wenz Messages: 1587 Registered: July 2009 Location: Walldorf, Germany
it seems there should be an example dealing with non-EMF domain models...
In general you should simply let the diagram editor persist and load the EMF
parts (in the non-EMF case only the diagram itself). Within this model you
can have properties that allow your independence solver to identify the
domain objects that belong to the shapes on the diagram. Therefore there
should be no need to persist the mapping information in the independence
solver. Your save logic should in addition only care about persisting your
non-EMF domain model.