[gecode-users] Dynamic creation of IntVarArgs

Max chaosangel at gmx.de
Tue Dec 2 13:03:31 CET 2008


Yes, of course i can do this (and have to do this),
but my tree is an abstracted form that should not know about linear 
constraints or not.
An resize function would be much easier.

Thanks for the post linear expression hint, i will have a look at.

Max

Christian Schulte wrote:
> That's the oldest trick from the schoolbook: when you create the tree,
> maintain in each node a counter that tells you the number of nodes (or
> leaves, or the nodes corresponding to a linear expression or whatever you
> need) in the subtree.
>
> As suggested earlier, this trick is used also in post linear expressions in
> mini model.
>
> Christian
>
> --
> Christian Schulte, www.ict.kth.se/~cschulte/
>
>
> -----Original Message-----
> From: users-bounces at gecode.org [mailto:users-bounces at gecode.org] On Behalf
> Of Max
> Sent: Tuesday, December 02, 2008 12:52 PM
> To: users at gecode.org
> Subject: [gecode-users] Dynamic creation of IntVarArgs
>
> Hello everybody,
>
> i wanted to create an IntVarArgs array for using linear constraints.
> I need this array in an recursive function where the size is calculated 
> at the end of the recursion(recursion size == array size).
>
> Now i found out that a can not dynamically create an IntVarArgs or an 
> IntArgs.
> This, seems reasonable, because i would have to destroy it somewhere, 
> and i don't know if gecode copies the reference it gets in the linear 
> call(i don't think so and it would be better not to do so).
>
> BUT:
> Why can i not create an IntVarArgs array without size and initializing 
> it with a size at a later time.
> Without that, i would have to recurse my tree two times, one time for 
> evaluating the size of the array, and one time to fill it. This is bad. 
> Do i have overseen something ? Why is this feature not supported.
> Any ideas how to solve my problem?
>
> Thanks in advance.
>
> Max
>
> _______________________________________________
> Gecode users mailing list
> users at gecode.org
> https://www.gecode.org/mailman/listinfo/gecode-users
>
>   





More information about the gecode-users mailing list