[great-users] Terminating & Aspects

[Sandeep Neema]

Matt/Tivadar,

Sorry for responding without reading all the way thru. Matt, I stand
corrected - I didn't realize that GME has changed its ways in how it
stores position :-), tells you its been a long time since I did any real
work :-)

 

Thanks,

Sandeep

 

--

Sandeep Neema

Research Scientist,

Institute of Software Integrated Systems, Vanderbilt University

Phone: 615-343-9996

Email: sandeep.k.neema at vanderbilt.edu

  _____  

From: great-users-bounces at list.isis.vanderbilt.edu
[mailto:great-users-bounces at list.isis.vanderbilt.edu] On Behalf Of
Sandeep Neema
Sent: Wednesday, May 25, 2005 10:27 AM
To: great-users
Subject: RE: [great-users] Terminating & Aspects

 

Matt/Tivadar,

I believe that it is possible to access the position information for a
specific attribute in a specific aspect, and I have done that in the
past. Unfortunately I was not able to locate a code snippet that I could
send over to you. As far as I recall you could define position as a
"registry" attribute in your UML class diagram (generated from
MetaGME2UML - you will have to hand modify the generated UML class
diagram). If you have multiple aspects then GME stores position as a
string in the registry node somewhat like this
"Aspect1(100,34),Aspect2(150,95),...", you can look up the registry node
in your GME model to see how specifically it is stored. Using UDM DS-API
you should be able to access this string, from that point on it should
be manageable -

 

-- Sandeep

 

 

--

Sandeep Neema

Research Scientist,

Institute of Software Integrated Systems, Vanderbilt University

Phone: 615-343-9996

Email: sandeep.k.neema at vanderbilt.edu

  _____  

From: great-users-bounces at list.isis.vanderbilt.edu
[mailto:great-users-bounces at list.isis.vanderbilt.edu] On Behalf Of
Matthew J. Emerson
Sent: Wednesday, May 25, 2005 10:02 AM
To: great-users; udm-users
Subject: RE: [great-users] Terminating & Aspects

 

In response to 2, here's how it's done in BON2:

        // Obtaining the position of an FCO in the Aspect 'FOO'

        BON::Point pt = BON::FCOExRegistryNode(

                fco->getRegistry())->getLocation("FOO");

I know that it's possible to work with registry entries in UDM, but I
really doubt you can get the position information for a specific aspect.
Look how it shows up in the registry:

 

You can see that this is the registry for an object named RootFolder,
which appears in two aspects: Transform and Template. It has a position
attribute for each aspect in which it appears. But, the position part is
nested. IIRC, UDM doesn't allow access to or creation of nested registry
entries.

I'm adding this to the UDM mailing list, since I think it's really a UDM
question.

-Matt

> -----Original Message-----

> From: great-users-bounces at list.isis.vanderbilt.edu
[mailto:great-users-

> bounces at list.isis.vanderbilt.edu] On Behalf Of Tivadar Szemethy

> Sent: Wednesday, May 25, 2005 9:18 AM

> To: great-users

> Subject: [great-users] Terminating & Aspects

> 

> Hi,

> 

> I have 2 questions:

> 

> 1) How do I terminate the transformation nicely from an
AttributeMapping /

> Guard box ?

> 

> I figured out that the MessageBox(...) call works well and I can warn
the

> user if something is wrong with the input model. What if I want to
quit

> right after that ?

> 

> 2) How do I work with aspects ?

> 

> How do I figure out which aspect is used by the Object.position()

> attribute in the first place ? It looks arbitrary to me.

> The problem is that I have an input model with multiple aspects (I
have to

> define multiple ordering for certain objects and the meta uses
multiple

> aspects to achieve this). I'd like to extract the position within each

> aspect from a GReAT rule. How can I do that ?

> 

> Thanks,

> Tivadar

> _______________________________________________

> great-users mailing list

> great-users at list.isis.vanderbilt.edu

> http://list.isis.vanderbilt.edu/mailman/listinfo/great-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://list.isis.vanderbilt.edu/pipermail/great-users/attachments/20050525/00013c31/attachment.htm