Constraint satisfaction in logic programming
Constraint satisfaction in logic programming
Implementing a Finite-domain CLP-language on Top of Prolog: a Transformational Approach
LPAR '94 Proceedings of the 5th International Conference on Logic Programming and Automated Reasoning
An Open-Ended Finite Domain Constraint Solver
PLILP '97 Proceedings of the9th International Symposium on Programming Languages: Implementations, Logics, and Programs: Including a Special Trach on Declarative Programming Languages in Education
Hi-index | 0.01 |
In this paper we describe a new implementation of the Finite Domain solver ROPE [8], called MROPE II. This new version was preceded by an implementation on top of Prolog [9] and a version using an early version of Mercury [10]. In the previous implementation Mercury was chosen for its speed, compile-time checking properties and fast reliable development. This previous experiment with Mercury was already a success, still there were some problems: for an efficient execution backtrackable destructive assignment was needed. Later releases of the Mercury system [4] contained such backtrackable destructive assignment and also impure declarations. This was all we needed for a new implementation of ROPE: MROPE II. The performance of this new system, with a very high level implementation, approaches the performance of other well known systems.