[gme-users] RE: Multiplicity Enforcement

[Zoltan Molnar]

Hi,

There is a beta version of the GME (4.11.10) with a new metainterpreter
with the features we talked about:
- you can specify events on which the env should check whether
constraints are violated 
- you can set the priority level for all such constraints

The first option is more important, since the priority was set to 1 (the
highest) already, but the events to react was left empty (eventmask =
0x00000000).

This option setting refers to the cardinality based constraints only,
and requires input during the metainterpretation process.

Br, 
Zoli

> -----Original Message-----
> From: Steven D Baker [mailto:Steven_D_Baker at raytheon.com] 
> Sent: Friday, October 01, 2004 12:10 PM
> To: Akos Ledeczi
> Cc: Zoltan Molnar
> Subject: Re: Multiplicity Enforcement
> 
> 
> 
> 
> 
> 
> Akos and Zoltan:
> 
> Thanks for the explanation. Now that I know what's going on, 
> here's my $0.10 as a GME metamodeler.
> 
> You have the option of leaving associations/connections 
> without multiplicity constraints (i.e., multiplicity is 
> 0..*). They're like that by default. If you modify them to be 
> more restrictive, it's highly likely that you want to see 
> some kind of enforcement. I'm kind of surprised that at least 
> weak enforcement (priority > 1) isn't defaulted to on, with 
> some kind of reasonable event tie in. That said, I agree with 
> you both that a user setting would offer the most 
> flexibility. For right now, my best option to get the 
> behavior I want is write my own OCL constraints for multiplicity.
> 
> I'll keep an eye on future releases, in case you decide to 
> add some new functionality along the lines mentioned.
> 
> Thanks,
> 
> Steve
> 
> 
>