Re: aggregation and states



Date view Thread view Subject view Author view Attachment view

From: Les Munday (baldrick@ureach.com)
Date: Thu 30 May 2002 - 02:08:42 BST


I don't see why there would be any constraints, as with 
inherited classes. Especially in the case of aggregation where 
a component may exist outside of the context of its parent. 
(Wheel -<> Car for example.)

I would expect to see event communication between a parent 
aggregation class and it's parts. (Using the car example the 
car probably wants to know the state of each of its wheels in 
order to calculate the actual state of the car, but there 
doesn't need to be a direct correlation between states.)

Les.

---- On Wed, 29 May 2002, Jonathan N. D. Whittle 
(jonathw@ptolemy.arc.nasa.gov) wrote:

> Is there any work out there on how aggregations constrain 
behavioral states? 
> E.g., if classes A1 and A2 are part of class B, do the states 
of class B 
> necessarily contain those of A1 and A2 (e.g., by using 
orthogonality in a 
> statechart for B)? I know of work in this direction for 
inheritance of course, 
> but I haven't seen anything for aggregations.
> 
> Thanks,
> 
> Jon
> 
> 
> 
> 
> To remove yourself from this list please mail puml-list-
request@cs.york.ac.uk
> with a message containing the word "unsubscribe".
> 
> 
> 


________________________________________________
Get your own "800" number
Voicemail, fax, email, and a lot more
http://www.ureach.com/reg/tag

Date view Thread view Subject view Author view Attachment view