Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Modeling » EMF "Technology" (Ecore Tools, EMFatic, etc)  » [EMF Compare] and supertypes
[EMF Compare] and supertypes [message #666265] Tue, 19 April 2011 12:23 Go to next message
leonard.krpan is currently offline leonard.krpan
Messages: 20
Registered: March 2011
Junior Member
Hello,

as you probably all know by now Smile I have extended EMF Compare with a wrapper class which displays some extra information specific for my project and also have a possibility to pairwiese compare any number of models. However, the changes are displayed together under one root, not in one subtree per pairwise comparison.

What I am trying to do is to create a generic model which combines all the information that the compared models have in common. I've been trying to work this out for supertypes of a class, which proved to be quite hard since changes can propagate across the supertype hierarchy.

Anyways, I noticed that my custom DiffGroups also contain information about the supertypes, within the DiffGroup of the original class. I have been trying to figure out if this can somehow tell me what is the common supertype for a class across all compared models.

Here's a hierarchy example:

Model 1: E->D->C->B->A
Model 2: E->C->B->A

class D still exists in Model 2 but is not used as E's supertype anymore.

However, the class contains a subdiff group containing class D. I'm not sure about how I'm supposed to interpret this diff group?
Re: [EMF Compare] and supertypes [message #667798 is a reply to message #666265] Mon, 02 May 2011 05:39 Go to previous message
Laurent Goubet is currently offline Laurent Goubet
Messages: 1622
Registered: July 2009
Senior Member
Hi,

You should have a DiffGroup for class E with changes for its supertypes (ReferenceChange(Left/Right)Target), but that DiffGroup should not contain a DiffGroup for the class "D" changes.

"DiffGroup"s are used by EMF Compare to reproduce the class hierarchy of the compared models; and "D" is not "contained" within "E"; thus the diffgroups should reproduce that.

Are you sure that you indeed have a DiffGroup for "D" contained by the DiffGroup for "E" ? And if so, is that EMF Compare that created this DiffModel, or did you change it manually afterwards?

Laurent Goubet
Obeo
Previous Topic:Copy all non conflicting changes ... emf compare
Next Topic:[EMF compare] ignore the comparison of an attribute by its value
Goto Forum:
  


Current Time: Fri Aug 01 09:56:24 EDT 2014

Powered by FUDForum. Page generated in 0.01934 seconds