Constructing Rule-Based Solvers for Intentionally-Defined Constraints

Developing constraint solvers which are key requisites of constraint programming languages is time consuming and difficult. In this paper, we propose a generic algorithm that symbolically constructs rulebased solvers from the intensional definition of the constraint. Unlike the well-established “generate and test” approach, our symbolic construction approach is capable of generating recursive rules from a recursive constraint definition. Combining the two approaches gives better filtering capabilities than either of the approaches acting alone.