Leslie De Koninck

User-definable rule priorities for CHR

This paper introduces CHRrp: Constraint Handling Rules with user-definable rule priorities. CHRrp offers flexible execution control which is lacking in CHR. A formal operational semantics for the extended language is given and is shown to be an instance of the theoretical operational semantics of CHR. It is discussed how the CHR rp semantics influences confluence results. A translation scheme for CHRrp programs with static rule priorities into (regular) CHR is presented. The translation is proven correct and bench-mark results are given. CHRrp is related to priority systems in other constraint programming and rule based languages.

As time goes by: Constraint Handling Rules: A survey of CHR research from 1998 to 2007

Constraint Handling Rules (CHR) is a high-level programming language based on multiheaded multiset rewrite rules. Originally designed for writing user-defined constraint solvers, it is now recognized as an elegant general purpose language. Constraint Handling Rules related research has surged during the decade following the previous survey by Fruhwirth ( J. Logic Programming, Special Issue on Constraint Logic Programming , 1998, vol. 37, nos. 1–3, pp. 95–138). Covering more than 180 publications, this new survey provides an overview of recent results in a wide range of research areas, from semantics and analysis to systems, extensions, and applications.

Optimizing compilation of CHR with rule priorities

Constraint Handling Rules were recently extended with user-definable rule priorities. This paper shows how this extended language can be efficiently compiled into the underlying host language. It extends previous work by supporting rules with dynamic priorities and by introducing various optimizations. The effects of the optimizations are empirically evaluated and the new compiler is compared with the state-of-the-art K.U. Leuven CHR system.

The correspondence between the logical algorithms language and CHR

This paper investigates the relationship between the Logical Algorithms language (LA) of Ganzinger and McAllester and Constraint Handling Rules (CHR). We present a translation scheme from LA to CHRrp: CHR with rule priorities and show that the meta-complexity theorem for LA can be applied to a subset of CHRrp via inverse translation. This result is compared with previous work. Inspired by the high-level implementation proposal of Ganzinger and McAllester, we demonstrate how LA programs can be compiled into CHR rules that interact with a scheduler written in CHR. This forms the first actual implementation of LA. Our implementation achieves the complexity required for the metacomplexity theorem to hold and can execute a subset of CHRrp with strong complexity bounds.

A Flexible Search Framework for CHR

This paper introduces a framework for the specification of tree search strategies in CHR with disjunction (CHR *** ). We support the specification of common search strategies such as depth-first, breadth-first and best-first, as well as constrained optimization by means of branch & bound search. The framework is given as an extension of CHR with rule priorities (CHRrp) in which each branch of the search tree is assigned a branch priority . This approach leads to a uniform solution to execution control in CHR.

Logical algorithms meets chr: A meta-complexity result for constraint handling rules with rule priorities

This paper investigates the relationship between the Logical Algorithms (LA) language of Ganzinger and McAllester and Constraint Handling Rules (CHR). We present a translation schema from LA to CHRrp: CHR with rule priorities, and show that the meta-complexity theorem for LA can be applied to a subset of CHRrp via inverse translation. Inspired by the high-level implementation proposal for Logical Algorithm by Ganzinger and McAllester and based on a new scheduling algorithm, we propose an alternative implementation for CHRrp that gives strong complexity guarantees and results in a new and accurate meta-complexity theorem for CHRrp. It is furthermore shown that the translation from Logical Algorithms to CHRrp combined with the new CHRrp implementation satisfies the required complexity for the Logical Algorithms meta-complexity result to hold.

Constraints in Non-Boolean Contexts

In high-level constraint modelling languages, constraints can occur in non-Boolean contexts: implicitly, in the form of partial functions, or more explicitly, in the form of constraints on local variables in non-Boolean expressions. Specifications using these facilities are often more succinct. However, these specifications are typically executed on solvers that only support questions of the form of existentially quantified conjunctions of constraints. We show how we can translate expressions with constraints appearing in non-Boolean contexts into conjunctions of ordinary constraints. The translation is clearly structured into constrained type elimination, local variable lifting and partial function elimination. We explain our approach in the context of the modelling language Zinc. An implementation of it is an integral part of our Zinc compiler.

Execution Control for CHR

This is a summary of the PhD of the author [1], which deals with the topic of execution control for Constraint Handling Rules.

Demand-Driven Normalisation for ACD Term Rewriting

ACD Term Rewriting (ACDTR) is term rewriting modulo associativity, commutativity, and a limited form of distributivity called conjunctive context. Previous work presented an implementation for ACDTR based on bottom-up eager normalisation, extended to support the conjunctive context. This paper investigates the possibility of using a demand-driven normalisation strategy for ACDTR. Again, dealing with the conjunctive context proves to be challenging. The alternative normalisation strategy is compared with the current form of eager normalisation and potential further improvements on the strategy are investigated.