Re: Problem with CLP

[Fergus Henderson]

On 23-Sep-2003, Jorge M. Pelizzoni <address@hidden> wrote:
> As far as I understand (I'm not a GNU Prolog user actually. I do Oz),
> you must be missing the distribute step and search itself. Constraint
> propagation has its limits and that is just what you have just observed.
> Someone out there please explain how he should complete his program in
> order to search for a definite solution set by searching iteratively in
> propagate-and-distribute steps.

Things are a little bit different in Oz than in Prolog and its CLP
derivatives, but basically your suggestion corresponds to another
poster's suggestion to use fd_labelling, which is in turn basically just a
more efficient way of doing what the original poster himself suggested:
enumerating the possible ground values for the constrained variables.

However, that does not negate the point that GNU Prolog's "yes" answer
to the original poster's original query is misleading.

-- 
Fergus Henderson <address@hidden>  |  "I have always known that the pursuit
The University of Melbourne         |  of excellence is a lethal habit"
WWW: <http://www.cs.mu.oz.au/~fjh>  |     -- the last words of T. S. Garp.