File "railgun"

Since the recent version of Dogelog Player supports cyclic terms, we could let the Jini out of the bottle, and provide the experimental library(edge/railgun) to model delayed goals with nothing else than the ISO core standard Prolog and Alain Colmerauers rational trees. The results were a Lean CLP of ca. 100 lines of code.

Meanwhile the library is in steady iteration with major milestones. The current state is that it provides the basic CLP(FD) vocabulary as found in SWI-Prolog library(clpfd) by Markus Triska. But it still differs considerably since its constraint solver is not propagation based. Instead a static variable ordering technique is deployed before labeling.

The following railgun predicates are provided:

T in D:

The predicate delays the goal in D on the term T.

L ins D:

The predicate delays the goal in D on the elements L.

S #= T:

The predicate delays the goal S =:= T on the terms S and T.

S #\= T:

The predicate delays the goal S =\= T on the terms S and T.

S #< T:

The predicate delays the goal S < T on the terms S and T.

S #=< T:

The predicate delays the goal S =< T on the terms S and T.

S #> T:

The predicate delays the goal S > T on the terms S and T.

S #>= T:

The predicate delays the goal S >= T on the terms S and T.

all_different(L):

The predicate delays the goal that all elements of L are different.

label(L):

The predicate resolves the elements of L, left to right..

labeling(O, L):

The predicate resolves the elements of L, statically reordered, accordingly to the option list O. The available labeling options are:

zero: Reorder by degree only.
constr: Reorder by degree and number of constraints.
plausi: Reorder by degree and discounted number of constraints and consistency tests.

Use Privacy (c) 2005-2026 XLOG Technologies AG