Re: [eclipse-clp-users] How to expand domain of a variable

On 04/11/11 01:38, Sergey Dymchenko wrote:
> Hello!
>
> How to expand existing domain of a variable?
> I mean, I have a variable with some domain (say, 1..10), and I want to
> allow that variable to have values from that domain or from another
> domain, like 20..30.
>
> I tried things like
> N :: [1..10], get_domain(N, D), N :: D or N :: [20..30].
> but unsurprisingly it doesn't work.
>
> Of course I can write just
> N :: [1..10] or N :: [20..30].
> but I want to add ranges to domain iteratively (in a cycle).
>
> Sergey.

Hi,

usually if you want to expand the domains of the variables it means that 
you model is not a CLP(FD) model, because in CLP(FD) domains can only 
shrink.

For this reason, ECLiPSe does not allow you to add elements to a domain, 
as you also noticed.

Said that, there are some cases in which the domain of a variable 
semantically contains a lot of elements, but you do not know them all 
(currently), and you need to add them dynamically during search. This 
may happen for example in configuration problems (see papers by 
Mailharro) or in visual search problems.

If you are interested, these are some papers on the topic:

Marco Gavanelli, Evelina Lamma, Paola Mello, and Michela Milano. Dealing 
with incomplete knowledge on CLP(FD) variable domains. ACM Transactions 
on Programming Languages and Systems, 27(2):236-263, March 2005.
http://dl.acm.org/citation.cfm?doid=1057387.1057389

Marco Alberti, Marco Gavanelli, Evelina Lamma, Paola Mello, and Michela 
Milano. A CHR-based implementation of known arc-consistency. Theory and 
Practice of Logic Programming, 5(4/5):419-440, July 2005.
http://www.ing.unife.it/docenti/MarcoGavanelli/papers/TPLP-CHR-2005.pdf

Rita Cucchiara, Marco Gavanelli, Evelina Lamma, Paola Mello, Michela 
Milano, and Massimo Piccardi. Constraint propagation and value 
acquisition: why we should do it interactively. In Thomas Dean, editor, 
Proceedings of the Sixteenth International Joint Conference on 
Artificial Intelligence, pages 468-477, Stockholm, Sweden, July 31 - 
August 6 1999.
http://ijcai.org/Past%20Proceedings/IJCAI-99-VOL-1/PDF/068.pdf

Cheers,
Marco
-- 
Marco Gavanelli, Ph.D. in Computer Science
Dept of Engineering
University of Ferrara
Tel/Fax  +39-0532-97-4833
http://www.ing.unife.it/docenti/MarcoGavanelli/