[news.eclipse.modeling.mdt.uml2] Re: Problems with DerivedUnion during d

[Kenn Hussey <Kenn.Hussey@xxxxxxxxx>]

Timothy,

Something must not be right about the way the precondition feature is 
defined. Both it and the postcondition features should be 
non-containment subsets of the ownedRule containment feature (which, in 
UML2, is inherited from NamedElement).

Regarding the item provider implementation, I would expect only the 
ownedRule feature to be referenced by the getChildrenFeatures(...) 
method. Is that the case?

If you're still having trouble, perhaps you could send me your Ecore and 
generator models and I'll take a closer look.

Kenn

Timothy Marc wrote:
> Kenn,
>
> it's me again concerning the duplicate entry problem. As mentioned, the 
> XMI is correct and when i load the model prgramatically, all stuff is 
> correct implemented... so one question: is the default generated editor 
> from emf capable to handle multiple subset containment references?
>
> To me it seems, as there are problems with subsets, when two (or more) 
> references from one source element point to a target element. In case of 
> Operation->Constraint, there are 2 references (namely preconditios and 
> postconditions) where the asso ends are from the same type. It seems as 
> if the editor can not distinguish between a preconditon or 
> postcondition... the XMI is:
>
> <?xml version="1.0" encoding="UTF-8"?>
> <testing:TestingModel xmi:version="2.0" 
> xmlns:xmi="http://www.omg.org/XMI"; 
> xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"; 
> xmlns:testing="http://www.fokus.fraunhofer.de/testing/testing.ecore"; 
> xmi:id="_vbeNAItuEd6Wtdf04WDKwQ">
>   <packagedElement xsi:type="testing:Class" 
> xmi:id="_T2weUYtvEd6uo5hY0fhr5A">
>     <ownedOperation xmi:id="_Uv00UItvEd6uo5hY0fhr5A" 
> postcondition="_VXUSIItvEd6uo5hY0fhr5A">
>       <ownedRule xsi:type="testing:Constraint" 
> xmi:id="_VXUSIItvEd6uo5hY0fhr5A"/> //inserted as precondition!!!!!
>     </ownedOperation>
>   </packagedElement>
> </testing:TestingModel>
>
> /Timothy
>
> Timothy Marc schrieb:
> > Kenn,
> >
> > hmm... i've compared the item providers of UML2 and my model 
> > (OperationItemProvider). They look exactly like the same, so i really 
> > don't know where to apply the leverage in order to erase this 
> > misbehavior...
> >
> > All this providing stuff is still a closed book to me.
> >
> > Timothy
> >
> > Kenn Hussey schrieb:
> > > Timothy,
> > >
> > > Yeah, we used to automatically make derived properties transient (a 
> > > long time ago) but we changed it because in some cases it may be 
> > > desirable to serialize derived properties (e.g. if a value is more 
> > > compact than the values it is derived from).
> > >
> > > Regarding the duplicate display issue, I'd suggest looking at how the 
> > > item providers are defined for your features. I suspect that 
> > > providers for two different features are displaying the value in 
> > > question. In this case you probably don't want the values for pre- 
> > > and post-conditions to appear in the content tree (since they will be 
> > > coerced to non-containment) but rather in the Properties view.
> > >
> > > Kenn
> > >
> > >
> > > Timothy Marc wrote:
> > > > Kenn,
> > > >
> > > > god damn, i was aware of all other derived attributes or subset, but 
> > > > not for the unions... is there any rationale, why derived 
> > > > associations or attributes are not generally set to transient=true? 
> > > > Because of that they are already set to volatile=true i allways 
> > > > forget to set them manually to transient. But, it works now. Thanks!
> > > >
> > > > Another point: The generated editor behaves weird for the following 
> > > > snippet. I defined the following (trying to approximate the 
> > > > ownedRule semtanics of the superstructure):
> > > >
> > > > #####
> > > > Element {abstract}
> > > >     +/ownedElement * {union, read-only, containment}
> > > >     +/ownedRule 0..* {subsets ownedElement, containment}
> > > >
> > > > NamedElemet extends Element
> > > >
> > > > BehavioralFeature extends NamedElement
> > > >
> > > > Operation extends BehavioralFeature
> > > >     +precondition 0..* {subsets ownedRule, containment}
> > > >     +postcondition 0..* {subsets ownedRule, containment} #####
> > > >
> > > > If i add an 'ownedRule' to the operation, everything is fine. But 
> > > > any adding of either pre- or postcondition has the effect, that 2 
> > > > instances of the same constraint is displayed. Well, the XMI looks 
> > > > okay, IMHO:
> > > >
> > > > <packagedElement xsi:type="testing:Class" 
> > > > xmi:id="_cUg_8IXsEd6Lu8REkryexw">
> > > >     <ownedOperation xmi:id="_fkrwMIXsEd6Lu8REkryexw" 
> > > > postcondition="_gJMLcIXsEd6Lu8REkryexw">
> > > >       <ownedRule xmi:id="_gJMLcIXsEd6Lu8REkryexw"/>
> > > >     </ownedOperation>
> > > >   </packagedElement>
> > > >
> > > > The postcondition belongs both to the ownedRule containment 
> > > > reference, and to the postcondition reference of its context 
> > > > operation. In the editor, the constraint is shown twice instead of 
> > > > just being a single element. If i add a constraint via the ownedRule 
> > > > reference, only a single element is displayed.
> > > >
> > > > The java code for Operation::postcondition is:
> > > >
> > > > public EList<Constraint> getPostconditions() {
> > > >     if (postconditions == null) {
> > > >         postconditions = new 
> > > > SubsetSupersetEObjectResolvingEList<Constraint>(Constraint.class, 
> > > > this, TestingPackage.OPERATION__POSTCONDITION, 
> > > > POSTCONDITION_ESUPERSETS, null);
> > > >     }
> > > >     return postconditions;
> > > > }
> > > >
> > > > and for Element::ownedRule it is
> > > >
> > > > public EList<Constraint> getOwnedRules() {
> > > >     if (ownedRules == null) {
> > > >         ownedRules = new 
> > > > EObjectContainmentWithInverseEList<Constraint>(Constraint.class, 
> > > > this, TestingPackage.MODEL_ELEMENT__OWNED_RULE, 
> > > > TestingPackage.CONSTRAINT__CONTEXT);
> > > >     }
> > > >     return ownedRules;
> > > > }
> > > >
> > > > Again some perfect ideas?
> > > >
> > > > Thanks in advance
> > > > Timothy
> > > >
> > > > Kenn Hussey schrieb:
> > > > > Timothy,
> > > > >
> > > > > I wouldn't have expected to see an 'ownedElement' element in the 
> > > > > serialization, since derived unions are, by definition read-only 
> > > > > (hence the reason why the addition is failing upon 
> > > > > deserialization). Did you make that feature transient in the 
> > > > > metamodel?
> > > > >
> > > > > Kenn
> > > > >
> > > > > Timothy Marc wrote:
> > > > > > Hey all,
> > > > > >
> > > > > > i'm using the UML2 genmodel to generate a uml-like metamodel with 
> > > > > > subsets and unions etc. I've rebuild the Element -> ownedComment 
> > > > > > relationship of the UML 2.2 Superstructure (Figure 7.3). Instead 
> > > > > > of Comment, i've sed the term Annotation. The model looks like
> > > > > >
> > > > > > Element {abstract}
> > > > > >     +/owner 0..1 {union, read-only}
> > > > > >     +/ownedElement * {union, read-only}
> > > > > >
> > > > > > TestingModel extends Element
> > > > > >     + annotation * {subsets ownedElement}
> > > > > >
> > > > > > Annotation extends Element
> > > > > >
> > > > > >
> > > > > > When i define a simple model like the model mentioned in the 
> > > > > > following XMI snippet
> > > > > >
> > > > > > <?xml version="1.0" encoding="UTF-8"?>
> > > > > > <union:TestingModel xmi:version="2.0" 
> > > > > > xmlns:xmi="http://www.omg.org/XMI"; 
> > > > > > xmlns:union="http:///union.ecore"; ownedElement="//@annotation.0">
> > > > > >   <annotation/>
> > > > > > </union:TestingModel>
> > > > > >
> > > > > > i ran allways in an exception during deserialization. I've tried 
> > > > > > both the editor and the programmatically deserialization. Both 
> > > > > > fail with the following exception trace. It seems, that somewhere 
> > > > > > the DerivedEObjectEList.add() method will be called.
> > > > > >
> > > > > > org.eclipse.emf.ecore.xmi.IllegalValueException: Value 
> > > > > > 'union.impl.AnnotationImpl@352d87' is not legal. 
> > > > > > (platform:/resource/f/My.union, -1, -1)
> > > > > >     at 
> > > > > > org.eclipse.emf.ecore.xmi.impl.XMLHandler.setFeatureValue(XMLHandler.java:2648) 
> > > > > >
> > > > > >     at 
> > > > > > org.eclipse.emf.ecore.xmi.impl.XMLHandler.handleForwardReferences(XMLHandler.java:1135) 
> > > > > >
> > > > > >     at 
> > > > > > org.eclipse.emf.ecore.xmi.impl.XMLHandler.endDocument(XMLHandler.java:1214) 
> > > > > >
> > > > > >     at 
> > > > > > com.sun.org.apache.xerces.internal.parsers.AbstractSAXParser.endDocument(Unknown 
> > > > > > Source)
> > > > > >     at 
> > > > > > com.sun.org.apache.xerces.internal.impl.XMLDocumentFragmentScannerImpl.scanDocument(Unknown 
> > > > > > Source)
> > > > > >     at 
> > > > > > com.sun.org.apache.xerces.internal.parsers.XML11Configuration.parse(Unknown 
> > > > > > Source)
> > > > > >     at 
> > > > > > com.sun.org.apache.xerces.internal.parsers.XML11Configuration.parse(Unknown 
> > > > > > Source)
> > > > > >     at 
> > > > > > com.sun.org.apache.xerces.internal.parsers.XMLParser.parse(Unknown 
> > > > > > Source)
> > > > > >     at 
> > > > > > com.sun.org.apache.xerces.internal.parsers.AbstractSAXParser.parse(Unknown 
> > > > > > Source)
> > > > > >     at 
> > > > > > com.sun.org.apache.xerces.internal.jaxp.SAXParserImpl$JAXPSAXParser.parse(Unknown 
> > > > > > Source)
> > > > > >     at javax.xml.parsers.SAXParser.parse(Unknown Source)
> > > > > >     at 
> > > > > > org.eclipse.emf.ecore.xmi.impl.XMLLoadImpl.load(XMLLoadImpl.java:181)
> > > > > >     at 
> > > > > > org.eclipse.emf.ecore.xmi.impl.XMLResourceImpl.doLoad(XMLResourceImpl.java:180) 
> > > > > >
> > > > > >     at 
> > > > > > org.eclipse.emf.ecore.resource.impl.ResourceImpl.load(ResourceImpl.java:1445) 
> > > > > >
> > > > > >     at 
> > > > > > org.eclipse.emf.ecore.resource.impl.ResourceImpl.load(ResourceImpl.java:1241) 
> > > > > >
> > > > > >     at 
> > > > > > org.eclipse.emf.ecore.resource.impl.ResourceSetImpl.demandLoad(ResourceSetImpl.java:255) 
> > > > > >
> > > > > >     at 
> > > > > > org.eclipse.emf.ecore.resource.impl.ResourceSetImpl.demandLoadHelper(ResourceSetImpl.java:270) 
> > > > > >
> > > > > >     at 
> > > > > > org.eclipse.emf.ecore.resource.impl.ResourceSetImpl.getResource(ResourceSetImpl.java:397) 
> > > > > >
> > > > > >     at 
> > > > > > union.presentation.UnionEditor.createModel(UnionEditor.java:950)
> > > > > >     at 
> > > > > > union.presentation.UnionEditor.createPages(UnionEditor.java:1007)
> > > > > >     at 
> > > > > > org.eclipse.ui.part.MultiPageEditorPart.createPartControl(MultiPageEditorPart.java:310) 
> > > > > >
> > > > > >     at ....
> > > > > >
> > > > > >
> > > > > > Caused by: java.lang.UnsupportedOperationException
> > > > > >     at 
> > > > > > org.eclipse.uml2.common.util.DerivedEObjectEList$DerivedListIterator.add(DerivedEObjectEList.java:278) 
> > > > > >
> > > > > >     at java.util.AbstractSequentialList.add(Unknown Source)
> > > > > >     at 
> > > > > > org.eclipse.uml2.common.util.DerivedEObjectEList.addUnique(DerivedEObjectEList.java:636) 
> > > > > >
> > > > > >     at 
> > > > > > org.eclipse.emf.ecore.xmi.impl.XMLHelperImpl.setValue(XMLHelperImpl.java:1192) 
> > > > > >
> > > > > >     at 
> > > > > > org.eclipse.emf.ecore.xmi.impl.XMLHandler.setFeatureValue(XMLHandler.java:2643) 
> > > > > >
> > > > > >     ... 58 more
> > > > > >
> > > > > >
> > > > > > For serialization and deserialization i'm using an 
> > > > > > XMIBResource(Factory) created by the genmodel automatically. 
> > > > > > Please let me know, what is going wrong in this really simple 
> > > > > > example.
> > > > > >
> > > > > > Thanks in advance
> > > > > > Timothy