Source: oose Innovative Informatik eG ( Axel Scheithauer)
UML22-21asked for some clarifications about uniqueness. In its resolution following sentence was added to the specification:
NOTE. Even when all ends of the AssociationClass have isUnique=true, it is possible to have several instances associating the same set of instances of the end Classes.
I'm afraid, this makes AssociationClasses obsolete, because they can now be replaced with normal Classes without loosing expressive power.
It is correct, that the uniqueness of an AssociationClass is independent of the uniqueness of its member ends, and that adding a 'unique' property is out of the scope of an RTF. However, it would be possible to simply define that an AssociationClass with only unique member ends is itself unique. Should the modeler require another semantics, she can use a normal Class.
Without this, defining unique links is much more cumbersome.
Reported: UML 2.5 — Mon, 4 Mar 2019 14:11 GMT
Updated: Mon, 2 Jan 2023 09:44 GMT