Hi,
The problem I encounter was already asked several months ago , therefore I'm just quoting the old mail: - - - snip - - - Say I'm trying to reverse engineer package a.b.c, which may have a number of subpackages. I followed the steps given in the tutorial. For the examples below, assume that a.b.c.I1 is an interface and a.b.c.C1 a class. If a class a.b.c.d.C2 extends a.b.c.C1, the model captures this but the diagram has an extra <<data type>> C1 to which the generalization relationship is established. Is it possible to have the tool connect a.b.c.d.C2 to a.b.c.C1 (across package boundaries) directly? If class a.b.c.d.C2 implements a.b.c.I1, *the model* does not capture this relationship at all- I must manually add the relationship to my model/diagram. Say a.b.c.d.C2 has a nested class a.b.c.d.C2.C3. The model does capture this relationship, but there is no UML connector between the two classes to indicate the nesting. - - - snip - - - Has anyone found any solutions for these problems, especially for "a.b.c.d.C2 extends a.b.c.C1"? I would not like to redraw all relationships by hand for my projects. IMHO that "datatype" should only exist if the real class cannot be found in the project scope. In the trackers for bugs and features I couldn't find anything related. Does it make sense to create a new bug report? I appreciate any help or ideas. Thanks, Sigmar
_______________________________________________ Topcased-users mailing list [email protected] http://lists.gforge.enseeiht.fr/mailman/listinfo/topcased-users
