[great-users] negative application condition

Feng Shi fengshi at isis.vanderbilt.edu
Thu Feb 9 10:38:36 CST 2006


Hi, Arda,

 

Would you please export the transformation model MI2SI.mga and test model myUML.mga into XML(.xme) file and send me these two files? I cannot directly open the mga file on my machine.

 

Thanks,

Feng

 

-----Original Message-----
From: great-users-bounces at list.isis.vanderbilt.edu [mailto:great-users-bounces at list.isis.vanderbilt.edu] On Behalf Of Arda Göknil
Sent: Thursday, February 09, 2006 10:28 AM
To: great-users
Subject: RE: [great-users] negative application condition

 

Hi,

In my pattern, I expected the pattern to match with  ClassA, ClassB and FeatureA classes in my input model named myUML.mga. I try to decompose the rule into two rules. In the first rule, I define the pattern which specifies that classA aggregates another class (classB) and has a feature. In the second rule, I negated the association between the feature of the ClassA and the ClassB but nothing changed.


Aditya Agrawal <aditya.agrawal at gmail.com> wrote:

	Hi Feng,
	
	In the example, A, F and C are classes and the arrows are association
	classes. 
	
	The pattern cardinality of the classes are set to 1 and the pattern
	cardinality of the associations between the classes have been shown on the
	figure. 
	
	In the previous mail I attached a zip containing Arda's original example
	with the two cases I mentions. I have also exported the UMT file to xme.
	
	Thanks,
	Adi
	
	-----Original Message-----
	From: great-users-bounces at list.isis.vanderbilt.edu
	[mailto:great-users-bounces at list.isis.vanderbilt.edu] On Behalf Of Feng Shi
	Sent: Monday, February 06, 2006 12:00 PM
	To: great-users
	Subject: RE: [great-users] negative application condition
	
	Hi Adi,
	
	In the first case mentioned in you email:
	
	Is the "F" association class between ClassA and ClassC and its cardinality
	set to 0?
	
	If yes, currently the user needs to set the cardinality of edge A->F and
	F->C to 0(not -1).
	
	Thanks,
	Feng
	
	-----Original Message-----
	From: great-users-bounces at list.isis.vanderbilt.edu
	[mailto:great-users-bounces at list.isis.vanderbilt.edu] On Behalf Of Aditya
	Agrawal
	Sent: Friday, February 03, 2006 1:02 PM
	To: great-users
	Subject: RE: [great-users] negative application condition
	
	Hi Arda and Feng,
	
	I ran some experiments and realized the following.
	
	1. The pattern arda drew does not work. However, it is not because the class
	called feature is negated. Because that were the case then for your example
	the output would show classE but it doesn't.
	
	R-->A--1->F<-1-->C
	| ^ 
	|-----0-----|
	
	This pattern yields the wrong result.
	
	
	2. A pattern of the form 
	
	R-->A--1->B
	|
	|--0->C
	
	This pattern yields the right result.
	
	
	>From my understanding the A-->C edge in the first case is handled in a
	different part of the pattern matcher than the other edges and I guess we
	may not be checking the 0 cardinality correctly there.
	
	Thanks,
	Adi
	
	
	________________________________________
	From: great-users-bounces at list.isis.vanderbilt.edu
	[mailto:great-users-bounces at list.isis.vanderbilt.edu] On Behalf Of Feng Shi
	Sent: Friday, February 03, 2006 11:47 AM
	To: great-users
	Subject: RE: [great-users] negative application condition
	
	Hi, Arda,
	 
	I am not sure want you're meaning the "feature", is it another Class? I
	could not open your transformation model and would you please send its xme
	file to me?
	 
	Thanks,
	Feng
	 
	-----Original Message-----
	From: great-users-bounces at list.isis.vanderbilt.edu
	[mailto:great-users-bounces at list.isis.vanderbilt.edu] On Behalf Of Arda
	Göknil
	Sent: Friday, February 03, 2006 2:00 AM
	To: great-users
	Subject: RE: [great-users] negative application condition
	 
	I set the pattern cardinalities of the connection element to "0" but it
	negates the element not the association. I am sending my example. I want to
	define a pattern that defines a Class A which aggregates Class B and a
	feature but Class B should not have the feature of Class A. When I set the
	pattern cardinality of the Connection element between the Class B and the
	feature, it negates the feature, not the connection between them.
	Thanks,
	Arda
	
	Aditya Agrawal wrote:
	Hi Arda,
	 
	In your pattern specification you can change the "pattern cardinality"
	attribute even on the simple association between B and C. By simple
	association I mean an association without a class. 
	 
	You can click in the line, change the attribute called "pattern cardinality
	to '0' or '-1' " 
	 
	I hope this helps.
	 
	Thanks,
	Adi
	 
	________________________________________
	From: great-users-bounces at list.isis.vanderbilt.edu
	[mailto:great-users-bounces at list.isis.vanderbilt.edu] On Behalf Of Arda
	Göknil
	Sent: Thursday, February 02, 2006 9:15 AM
	To: A list for GReAT users to share bugs, fixes an d ideas
	Subject: RE: [great-users] negative application condition
	 
	I could not write a good example. I mean, the meta association. I imagine 
	the A, B and C classes are the meta classes and the association between them
	are the meta association, there is not a meta class named Association. The
	setting the pattern cardinality to "-1" is true if there is a Association
	entity in the meta class.
	
	Aditya Agrawal wrote: 
	Hi Arda,
	 
	As far as I remember you can set the pattern cardinality on the association
	between B and C to be "-1" instead of setting it on B. 
	 
	I don't know if things have changed in the recent past but earlier the value
	for negative condition used to be "0" not "-1".
	 
	Could Attila or Feng please comment to this?
	 
	Thanks,
	Adi
	 
	________________________________________
	From: great-users-bounces at list.isis.vanderbilt.edu
	[mailto:great-users-bounces at list.isis.vanderbilt.edu] On Behalf Of Arda
	Göknil
	Sent: Thursday, February 02, 2006 5:14 AM
	To: great-users at list.isis.vanderbilt.edu
	Subject: [great-users] negative application condition
	 
	Hi,
	I have a question about the negative application condition in GREAT. In the
	user manual, it is said that if you set the pattern cardinality to "-1", you
	negate that element. but I want to negate the element for only one
	association.
	For example, in my pattern I have three classes: classA, classB and classC.
	ClassA has a relation with ClassB and ClassC but I do not want ClassB to
	have a relation with ClassC. I want to negate ClassC for the
	association between ClassB and ClassC. But in my pattern if I set the
	pattern cardinality to -1, I can not show the whole pattern in one rule. How
	is it possible?
	Regards,
	 
	Arda Goknil
	________________________________________
	To help you stay safe and secure online, we've developed the all new Yahoo!
	Security Centre.
	_______________________________________________
	gre at-users mailing list
	great-users at list.isis.vanderbilt.edu
	http://list.isis.vanderbilt.edu/mailman/listinfo/great-users
	 
	________________________________________
	Yahoo! Photos - NEW, now offering a quality print service from just 8p a
	photo.
	_______________________________________________
	great-users mailing list
	great-users at list.isis.vanderbilt.edu
	http://list.isis.vanderbilt.edu/mailman/listinfo/great-users
	 
	________________________________________
	Yahoo! Cars NEW - sell your car and browse thousands of new and used cars
	online search now 
	________________________________________
	_______________________________________________
	great-users mailing list
	great-users at list.isis.vanderbilt.edu
	http://list.isis.vanderbilt.edu/mailman/listinfo/great-users
	
	_______________________________________________
	great-users mailing list
	great-users at list.isis.vanderbilt.edu
	http://list.isis.vanderbilt.edu/mailman/listinfo/great-users

 

  _____  

To help you stay safe and secure online, we've developed the all new Yahoo! Security Centre <http://us.rd.yahoo.com/mail/uk/taglines/default/security_centre/*http:/uk.security.yahoo.com/> .

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://list.isis.vanderbilt.edu/pipermail/great-users/attachments/20060209/d1c86a67/attachment.htm


More information about the great-users mailing list