Petra Hofstedt

Integration of declarative and constraint programming

Combining a set of existing constraint solvers into an integrated system of cooperating solvers is a useful and economic principle to solve hybrid constraint problems. In this paper we show that this approach can also be used to integrate different language paradigms into a unified framework. Furthermore, we study the syntactic, semantic and operational impacts of this idea for the amalgamation of declarative and constraint programming.

Better Communication for Tighter Cooperation

We propose a general scheme for the cooperation of different constraint solvers. A uniform interface for constraint solvers allows to formally specify information exchange between them and it enables the development of an open and very flexible combination mechanism. This mechanism allows the definition of a wide range of different cooperation strategies according to the current requirements such that our overall system forms a general framework for cooperating constraint solvers.

Turtle: A Constraint Imperative Programming Language

Ideally, in constraint programs, the solutions of problems are obtained by specifying their desired properties, whereas in imperative programs, the steps which lead to a solution must be defined explicitly, rather than being derived automatically. This paper describes the design and implementation of the programming language TURTLE, which integrates declarative constraints and imperative language elements in order to combine their advantages and to form a more flexible programming paradigm suitable for solving a wide range of problems.

Cooperating Constraint Solvers

We propose a general scheme for the cooperation of different constraint solvers. On top of a uniform interface for constraint we stepwise develop reduction systems which describe the behaviour of an overall combined system. The modularity of our definitions of reduction relations at different levels allows the definition of cooperation strategies for the solvers according to the current requirements such that our overall system forms a general framework for cooperating solvers.

A Flexible Meta-solver Framework for Constraint Solver Collaboration

The solving of multi-domain constraint problems with the help of collaborating solvers has seen extended interest in recent years. We describe the implementation (Meta-S) and extension of a previously proposed theoretical framework of cooperating constraint solvers. Meta-S allows the dynamic integration of arbitrary external (stand-alone) solvers to enable the collaborative processing of constraints. The modular structure allows the easy experimentation with different cooperation strategies. This is further amplified by an integrated strategy description language which supports the adaption of general strategies for problem specific needs to reduce the incurred cooperation overhead.

A General Approach for Building Constraint Languages

This paper describes a general approach for the integration of arbitrary declarative languages and constraint systems. The main idea is to consider declarative programs together with the language evaluation mechanisms as constraint solvers and to integrate them into an overall system of cooperating solvers. Exemplarily,w e present the integration of a logic language with a constraint system,and the extension of a functional logic language with constraints. The approach allows to build constraint languages according to current requirements and thus it enables comfortable modelling and solving of many problems.

Solution strategies for multi-domain constraint logic programs

We integrated a logic programming language into Meta-S, a flexible and extendable constraint solver cooperation system, by treating the language evaluation mechanism resolution as constraint solver. This new approach easily yields a CLP language with support for solver cooperation that fits nicely into our cooperation framework. Applying the strategy definition framework of Meta-S we define classical search strategies and more sophisticated ones and discuss their effects on an efficient evaluation of multi-domain constraint logic programs by illustrating examples.

Spatial Inference - Learning vs. Constraint Solving

We present a comparison of two new approaches for solving constraints occurring in spatial inference. In contrast to qualitative spatial reasoning we use a metric description, where relations between pairs of objects are represented by parameterized homogenous transformation matrices with numerical (nonlinear) constraints. We employ interval arithmetics based constraint solving and methods of machine learning in combination with a new algorithm for generating depictions for spatial inference.

TURTLE++ – a CIP-Library for c++

This article introduces the turtle++ library which combines constraint-based and imperative paradigms and enables in this way constraint imperative programming (CIP) with c++. Integrating CIP into c++ allows to exploit the powerful expressiveness of the CIP paradigm within a language widely used and accepted in practice. We discuss the main concepts and implementation and illustrate programming with turtle++ by means of typical examples.