Re: [Repast-developer] interesting copy question

[Nick C. <sr...@ea...>]

I think Mike's suggestion here is a good one. But a question, under what 
conditions will there be duplicate ties. Is it something we can protect 
against when ties are created, a switch to all multiplex networks, or an 
exception is thrown?

nick

Skye Bender-deMoll wrote:
> Nick, Tom,
> 
> So I'm doing some network stuff where I want to make changes to the
> network (removing duplicate ties as preperation for stats methods)
> without makeing changes to the actual node "Agents" of the model. 
> Unfortunatly, methods like ArrayList.clone() only return shallow copies,
> so any changes to the copied nodes will change the origial nodes.  Now
> in my case, I can probably come up with a workaround:
> 
> 1) modify the stat function to check the condition (very awkward, but
> probably the least expensive)
> 2) since I am only interested in edges (and not the rest of nodes
> properties), I can make a method which constructs a new network,
> identical to the old one, but constructed from default nodes and edges,
> and perform operations on it. 
> 
> However, there may be other cases in which people want to do things to
> agents, without changeing the "real" agents.  I was thinking that we
> should add copy methods to the existing node structures, and add
> getCopy() to the Node interface, but this would break all existing
> nodes, and still not insure that all relevent variables would be
> copied.  Tom suggested creating a "copyable" interface and having the
> abstract classes implement it...
> 
> Nick, do you know of any other way of getting deep copies of objects? 
> (clone() is protected)  This would still be a problem for networks, as a
> deep copy of a node would necessarily have to return the whole network,
> as each edge contains nodes, which contain edges....
> 
> Something to think about anyway!
> 
> (in the meantime, I just won't let people work with multiplex networks!)
> -skye
> 
> _______________________________________________
> Repast-developer mailing list
> Rep...@li...
> http://lists.sourceforge.net/lists/listinfo/repast-developer
> 
> 


-- 
Nick Collier
Social Science Research Computing
University of Chicago
http://repast.sourceforge.net