Re: [atl_discussion] How to target a UML profile in ATL? (bis)

Hi all,
> I feel that it's more than urgent to come up with a clarifying
> example. I'll prepare the little website today :-)
A preliminary website is available at
http://www.fots.ua.ac.be/motmot/docs/profileTutorial.php

It could be restructured in the upcoming days/weeks but the early
birds can already have a look.

I hope it helps understanding the relation between UML profiles and
MOF based model transformation tools. Feel free to contact me if you
have any questions.

Kind regards,
-- Pieter Van Gorp
http://www.fots.ua.ac.be/~pvgorp/research/
http://motmot.sourceforge.net/

---------- Forwarded message ----------
From: Pieter Van Gorp <pietervangorp@gmail.com>
Date: Thu, 24 Mar 2005 10:08:04 +0100
Subject: Re: [atl_discussion] How to target a UML profile in ATL? (bis)
To: Frédéric Jouault <f.jouault@gmail.com>
Cc: Patrice Boisieau <pboisieau@yahoo.com>

Dear Patrice and Frédéric,
I feel that it's more than urgent to come up with a clarifying
example. I'll prepare the little website today :-)

Kind regards,
-- Pieter.

On Wed, 23 Mar 2005 19:01:59 +0100, Frédéric Jouault
<f.jouault@gmail.com> wrote:
> > Thank you to both of you, Frédéric and Pieter, you
> > give me nice ideas for dealing with my problem.
>
> You are welcome.
>
>
> > However, if there are some solutions for DEFINING
> > and/or ACCESSING stereotypes in an ATL transformation
> > (e.g. by setting the profile as an input model as
> > suggested by Frédéric), it seems there is no easy
> > solution for APPLYING these same stereotypes on output
> > model elements.
> >
> > Indeed, if I have well understood the UML standard,
> > the application mechanism of a stereotype is not
> > defined at the M2 level (i.e. in the UML metamodel).
> > Therefore it is not possible to create an "applies"
> > relation between a stereotype and a UML element at M1
> > level such that it conforms to the UML metamodel.
> >
> > Hence UML modelers have their specific approach for
> > applying a stereotype to a model element. And when we
> > want to export a stereotyped model, stereotypes are
> > lost or cannot be retrieved inside another tool.
>
> As I said before, I am not a UML expert. Mainly because I believe in
> the multiple DSLs approach rather than in the unified language
> approach.
>
> However, it is nice that you are trying ATL on UML problems.
>
> I you find a working solution, please let us know.
>
> > In conclusion, I think that this problem is not
> > ATL-specific but UML-specific.
>
> I am inclined to agree with you.
>
> Regards,
>
> Frédéric Jouault
>

           
Received on Wed Mar 30 08:07:57 2005