| --- | --- |
| Bugzilla Link | 562862 |
| Status | NEW |
| Importance | P3 normal |
| Reported | May 06, 2020 03:59 EDT |
| Modified | May 06, 2020 04:00 EDT |
| See also | 562833 |
| Reporter | Ed Willink |
Description
Following on from Bug 562833, support for UML 2.5.1 looks fairly easy since very little changes.
a) add the MOF/XMI/UML 20161101 family namespaces
b) UMLR-92: UML/OCL spec mismatch-Constraint.context vs Constraint.constrainedElement
This is a pedantry of no consequence for Eclipse UML2. The new wording is vaguer but accommodates the prevailing Eclipse OCL practice but also allows something that Eclipse OCL doesn't. Until the OCL integration with UML is clear, this is just largely ignored words.
c) UMLR-685: StateMachine Vertex needs to be made a kind of RedefinableElement instead of State
this seems to move RedefinableElement up the inheritance hierarchy from State to Vertex. Nominally trivial but seems to cause many text changes and affects property names.
d) UMLR-696: The behavior of an OpaqueExpression should be allowed to have input parameters
Looks like a relaxation on a constraint.
UML 2.5.1 should be 99.99% upward compatible. Only slight 'change' might be the renaming of redefinedState/redefinedVertex that could be fixed with a legacy synonym.
| --- | --- | | Bugzilla Link | 562862 | | Status | NEW | | Importance | P3 normal | | Reported | May 06, 2020 03:59 EDT | | Modified | May 06, 2020 04:00 EDT | | See also | 562833 | | Reporter | Ed Willink |
Description
Following on from Bug 562833, support for UML 2.5.1 looks fairly easy since very little changes.
a) add the MOF/XMI/UML 20161101 family namespaces
b) UMLR-92: UML/OCL spec mismatch-Constraint.context vs Constraint.constrainedElement
This is a pedantry of no consequence for Eclipse UML2. The new wording is vaguer but accommodates the prevailing Eclipse OCL practice but also allows something that Eclipse OCL doesn't. Until the OCL integration with UML is clear, this is just largely ignored words.
c) UMLR-685: StateMachine Vertex needs to be made a kind of RedefinableElement instead of State
this seems to move RedefinableElement up the inheritance hierarchy from State to Vertex. Nominally trivial but seems to cause many text changes and affects property names.
d) UMLR-696: The behavior of an OpaqueExpression should be allowed to have input parameters
Looks like a relaxation on a constraint.
UML 2.5.1 should be 99.99% upward compatible. Only slight 'change' might be the renaming of redefinedState/redefinedVertex that could be fixed with a legacy synonym.