Re: navigable vs. exposed by reference



Date view Thread view Subject view Author view Attachment view

From: SainTiss (saintiss@arklinux.org)
Date: Tue 27 Jan 2004 - 09:55:45 GMT


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Steffen,

If I understand correctly, the document names "navigating across non-navigable 
associations" as a compliance point for OCL tools.
However, I'd like to know how associations should be handled which *are* 
navigable, but just not exposed by a reference? I was wondering if some 
standard says something about that?

the MOF standard does say the following (3.9.3.1)

<quote>
<expr>"."<identifier>
In this context (i.e., the definition of the MOF Model), the "<identifier>" is 
interpreted differently. In the MOF Model, the interfaces for navigating 
Associations are specified using References rather than AssociationEnds. Thus 
in the MOF version of OCL, link navigation is expressed using the name of a 
Reference for the "<expr>" object as the "<identifier>". However, the overall 
meaning is analogous to the UML case.
</quote>

So I guess this means an OCL WFR on a metamodel can't use navigable 
associations which aren't exposed by a reference, right?
If that is the case, then how is one to know which associations are not 
exposed by a reference? I can't find anything on that in e.g. the UML 
standard. Surely the intention can't be to just "try out"?

Thanks in advance,

Hans


On Monday 26 January 2004 16:32, Steffen Zschaler wrote:
> Hi Hans,
>
> This is in fact a compliance point for OCL 2.0. See ptc/2003-10-14, p. 1.
>
> Steffen
>
> SainTiss schrieb:
> >-----BEGIN PGP SIGNED MESSAGE-----
> >Hash: SHA1
> >
> >Hi,
> >
> >if I understand correctly, there is a difference in the UML metamodel
> > between a navigable associationEnd, and one which is exposed by a
> > reference. Does this difference exist at the modeling level already, or
> > only at code level (e.g. JMI interfaces)?
> >I looked at the UML standard, but couldn't find anything about it... As
> > far as I can see, all associations in the UML metamodel can be navigated
> > bidirectionally and so I assume that they are also exposed by a reference
> > at both ends.
> >
> >Is this correct? If not, are OCL constraints allowed to use non-exposed
> >association-ends, and is this documented somewhere?
> >
> >Thanks in advance,
> >
> >Hans Schippers
> >
> >
> >- --
> >Ark Linux - Linux for the Masses (http://arklinux.org)
> >
> >Capitalism:
> >You don't have any cows. The bank will not lend you money to buy cows,
> >because you don't have any cows to put up as collateral.
> >
> >Representative Democracy:
> >You have two cows. Your neighbors pick someone to tell you who gets the
> > milk.
> >
> >In a world without walls and fences, who needs windows and gates?
> >
> >Hans Schippers
> >2LIC INF
> >UA 2003-2004
> >-----BEGIN PGP SIGNATURE-----
> >Version: GnuPG v1.2.3 (GNU/Linux)
> >
> >iD8DBQFAFSvVXlnUYIbmLOQRAgFAAKCLz04Vs2bkpguZKxhm3mhLG6qAMwCfQJno
> >7CbD9pvKynuqWUy9w0Dl7Cg=
> >=lZ1h
> >-----END PGP SIGNATURE-----
> >
> >
> >
> >
> >To remove yourself from this list please mail
> > puml-list-request@cs.york.ac.uk with a message containing the word
> > "unsubscribe".

- -- 
Ark Linux - Linux for the Masses (http://arklinux.org)

Capitalism: 
You don't have any cows. The bank will not lend you money to buy cows, 
because you don't have any cows to put up as collateral.

Representative Democracy:
You have two cows. Your neighbors pick someone to tell you who gets the milk. 

In a world without walls and fences, who needs windows and gates?

Hans Schippers
2LIC INF
UA 2003-2004
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQFAFjWlXlnUYIbmLOQRAvf0AJwKf2xLXfZwMpsmdp0NgwvWalbUsgCfdHqV
LM6pIh35e5KbKLWDrYzNtwA=
=tmYy
-----END PGP SIGNATURE-----

Date view Thread view Subject view Author view Attachment view