[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [emf-dev] [EMF-Compare] questions about EMF-Compare and Papyrus diagrams
- From: StÃphane Bouchet <stephane.bouchet@xxxxxxx>
- Date: Fri, 30 Sep 2011 10:47:44 +0200
- Delivered-to: email@example.com
- Organization: Obeo
- User-agent: Mozilla/5.0 (X11; Linux i686; rv:7.0) Gecko/20110923 Thunderbird/7.0
Le 29/09/2011 09:42, LORENZO Vincent a Ãcrit :
I downloaded code from your git and AddUMLAssociationImpl doesn't compile, because some methods are not implemented, so I added empty methods.
you should remove or close that plugin, it will not be used anymore by
the uml comparison, and will be removed from the repository.
I have some questions concerning papyrus diagram comparison :
- How will you manage when there is several diagrams in the model ?
The support of multiples diagrams is supported, when you select a
difference on the structureviewer, it will show the correct diagram
associated with this difference. Of course, if diagram have no
difference, you cannot displays it.
there are some screenshots and testcases for this use case.
- Concerning Papyrus diagrams, will you compare only graphical changes? That's to say, when a UML element is drawn on the first model and not drawn on the second model, there are 2 cases :
- the UML element is hidden (it is in the model)
- the UML element is deleted (removed from the model)
Will you manage these 2 kinds of changes?
Yes, both cases are supported in comparison.
We are comparing only graphical changes, so we are detecting that a
graphical node is removed, even if the semantic element is present or
not. the same applies for hidden/revealed nodes, we are detecting this
behavior and representing it with different colors.
see , and 
- when the user compares uml files, referenced by di files, will you clean the diagrams which reference obsolete elements?
there is two options for this. the first one is to let gmf/papyrus do
the cleanup itself when the diagram is displayed. the second is to use
the logical models. actually, we are using the first option.
adr;quoted-printable:BP 20773;;7 Boulevard Amp=C3=A8re;CARQUEFOU;;44481;France