Hi Bob, the reverse engineering of sequence diagrams is my baby, and I'm still having great plans with it. It's a piece of a bigger picture for source code related modelling. (In short, I plan the following features: 1. tools for building sequence diagrams from code, 2. tools for building class diagrams from code, 3, a tool to select operations from the model for which a sequence diagram could be created, 4. a tool for creating a class diagram from the base classes of the classifiers of a sequence diagram, and 3. tools for creating class diagrams for associated/depending classes.)
The creation of sequence diagrams in RESequenceDiagramDialog is incomplete w.r.t. the finding of method calls in the model, but it is quite complete w.r.t. the creation of elements in a sequence diagram, so it's a good case for the interface. I tried to find the relevant functionality that is needed. Needed sequence diagram methods for the interface: 1. get the amount of ports of a classifier role 2. get the model element associated with a fig 3. get a classifier role from a model element 4. build an action and (if not existing) a classifier role for a classifier role and a message string (Most happens in buildAction(String call).) I committed a new version of RESequenceDiagramDialog where I added some TODO comments to help you identifiying where things are needed and where not. (And I resolved 2 TODOs from Tom regarding exception handling) I'm glad that this class got some attention and will help where I can. Regards, Thomas -------- Original-Nachricht -------- > Datum: Wed, 4 Jun 2008 00:04:57 +0100 > Von: "Bob Tarling" <[EMAIL PROTECTED]> > An: [email protected] > Betreff: [argouml-dev] Reverse engineering to new sequence diagrams > I don't know anything about how the the code to reverse engineer > sequence diagrams but this will need updating to work with the new > implementation before we can release that new implementation. > > Rather than just modify the existing code to work with seq2 I suggest > we establish an interface to do this. Then adapt the existing code to > use that interface to the existing sequence implementation. Once that > is done the sequence2 can be coded to implement that same interface. > > Whenever we feel confident to release seq2 we can do so without having > to change the reverse engineering subsystems at the last minute. > > Is anyone familiar with reverse engineering to sequence diagrams to > suggest what the interface ought to look like. > > We need to avoid references to Figs if possible. > > Regards > > Bob. > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] -- Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
