Eclipse Community Forums - RDF feed
https://www.eclipse.org/forums/
Eclipse Community ForumsGMF contraints
https://www.eclipse.org/forums/index.php/mv/msg/70539/220685/#msg_220685
Originally posted by: tobk42.gmx.de
Hello,
I'm not sure for which version(s) of GMF this applies, but when I
started modeling with GMF, declaring a java constraint resulted in a
method-to-be-implemented of the name given in the GMFmap, which was a
really nice thing, as one knew exactly which method was supposed to
implement which constraint [1]. However, for some time now, instead of
methods having that name, a number of adapters is created, which each
have a numeric identifier, and the only way to look up which adapter
handles which constraint one has to find the constraint with the same
identifier in the plugin XML.
And even worse: When inserting a new contraint, that new constraint is
not given a new number, but the number depends on the position in the
list (which, when using groups, is not always at the end of the list),
and existing constraints will be given a new number. If one has a high
number of contraints (in my case: more than fifty) re-assigning the
constraints is extremely labarous and error prone task.
Just wanted to tell this. It would be great if this could be improved in
the next version of GMF. I think simply using the names (the "body")
given in the constraint in the gmfMap instead of a numeric identifier
would be sufficient and shouldn't be much of a problem.
Regards,
Tobias
PS.: Apart from this, GMF rocks! ;-)
[1] (Of course it would have been even better if the Error-Message and
Description would have appeared in the generated method's javadoc, but a
carefully chosen method name was fine, too).]]>2009-03-06T12:42:44-00:00Re: GMF contraints
https://www.eclipse.org/forums/index.php/mv/msg/70539/220694/#msg_220694
http://wiki.eclipse.org/GMF_2.1_New_and_Noteworthy#Inject_Ja va_Expression_Bo
dy
> Hello,
>
> I'm not sure for which version(s) of GMF this applies, but when I
> started modeling with GMF, declaring a java constraint resulted in a
> method-to-be-implemented of the name given in the GMFmap, which was a
> really nice thing, as one knew exactly which method was supposed to
> implement which constraint [1]. However, for some time now, instead of
> methods having that name, a number of adapters is created, which each
> have a numeric identifier, and the only way to look up which adapter
> handles which constraint one has to find the constraint with the same
> identifier in the plugin XML.
>
> And even worse: When inserting a new contraint, that new constraint is
> not given a new number, but the number depends on the position in the
> list (which, when using groups, is not always at the end of the list),
> and existing constraints will be given a new number. If one has a high
> number of contraints (in my case: more than fifty) re-assigning the
> constraints is extremely labarous and error prone task.
>
> Just wanted to tell this. It would be great if this could be improved in
> the next version of GMF. I think simply using the names (the "body")
> given in the constraint in the gmfMap instead of a numeric identifier
> would be sufficient and shouldn't be much of a problem.
>
> Regards,
> Tobias
>
> PS.: Apart from this, GMF rocks! ;-)
>
> [1] (Of course it would have been even better if the Error-Message and
> Description would have appeared in the generated method's javadoc, but a
> carefully chosen method name was fine, too).]]>Richard Gronback2009-03-06T12:54:46-00:00