Demoen, B., de la Banda, M. J. G., Harvey, W., Marriott, K., and Stuckey, P. J. 1999. An Overview of HAL. In CP'99: Proceedings of the 5th International Conference on Principles and Practice of Constraint Programming, J. Ja#ar, Ed. Lecture Notes in Computer Science, vol. 1713. Springer Verlag, Alexandria, Virginia, USA, 174--188.  Home/Search   Document Details and Download   Summary   Related Articles   Check

.... of domains such as planning [25] semantic web [13] optimization [12] database security [3] machine learning [23] model checking [19] These developments have renovated the research efforts towards the design of models for the efficient execution of this class of languages and systems (e.g. [4, 16, 27, 5, 24]) From the beginning [9] parallelism has been identified as one of the most promising avenues to provide efficiency. Indeed, the essence of this class of systems non determinism provides a natural approach for the (automatic) exploitation of parallelism: non determinism implies the presence ....

B. Demoen et al. An Overview of HAL. In Constraint Programming, pages 174-188. 1999.

....[5] have indeed observed in the dProlog system maximal trail sizes for the PARMA scheme that are on average twice as large as with the WAM scheme. Even in the WAM it pays o to avoid trailing whenever possible and a fortiori this holds in the PARMA scheme. The PARMA scheme is used in HAL: HAL [4, 3] is a constraint logic language designed to support the construction, extension and use of constraint solvers. HAL requires type declarations and has optional mode and determinism declarations. It is compiled to Mercury [16] so as to leverage from its sophisticated compilation techniques. However, ....

B. Demoen, M. G. de la Banda, W. Harvey, K. Marriott, and P. J. Stuckey. An Overview of HAL. In Principles and Practice of Constraint Programming, pages 174-188, 1999.

....are therefore high. Hence we believe it is a useful research goal to develop a CTGC system for this language. In addition, mastering it for Mercury should be a useful stepping stone for systems such as Ciao Prolog [12] which has optional declarations and includes the impurities of Prolog) and HAL [8] (a Mercury based constraint language) The intention of the CTGC system is to discover at compile time when data is not referenced anymore, and how it can best be reused. Mulkers et al. 18] have developed an analysis for Prolog which detects when memory cells become available for reuse. This ....

B. Demoen, M. Garca de la Banda, W. Harvey, K. Marriott, and P. J. Stuckey. An overview of HAL. In Proceedings of the International Conference on Principles and Practice of Constraint Programming, pages 174-188, Virginia, USA, October 1999. Springer Verlag.

....a useful stepping stone for developing such an analysis for systems such as Ciao Prolog [9] where declarations are This work has been supported by the GOA project LP , the ESPRIT project ARGo, and the FWO Vlaanderen. optional and where one has to cope with the impurities of Prolog and HAL [6] which is a constraint language having many similarities with Mercury. Mulkers et al. 14] have developed an analysis for Prolog which detects when memory cells become available for reuse; however the lack of declarations and the impurity of Prolog make it rather infeasible to obtain acceptable ....

B. Demoen, M. Garca de la Banda, W. Harvey, K. Marriott, and P. Stuckey. An overview of HAL. In Proceedings of the International Conference on Principles and Practice of Constraint Programming, pages 174-188, Virginia, USA, October 1999. Springer Verlag.

No context found.

Demoen, B., de la Banda, M. J. G., Harvey, W., Marriott, K., and Stuckey, P. J. 1999. An Overview of HAL. In CP'99: Proceedings of the 5th International Conference on Principles and Practice of Constraint Programming, J. Ja#ar, Ed. Lecture Notes in Computer Science, vol. 1713. Springer Verlag, Alexandria, Virginia, USA, 174--188.

No context found.

B. Demoen, M. G. de la Banda, W. Harvey, K. Marriott, and P. J. Stuckey. An Overview of HAL. In Principles and Practice of Constraint Programming, pages 174--188, 1999.

No context found.

B. Demoen, M. G. de la Banda, W. Harvey, K. Marriott, and P. J. Stuckey. An Overview of HAL. In Principles and Practice of Constraint Programming, pages 174--188, 1999.

No context found.

B. Demoen, M. Garca de la Banda, W. Harvey, K. Mariott, and P. Stuckey. An overview of HAL. In J. Jaar, editor, Proceedings of the International Conference on Principles and Practice of Constraint Programming, volume 1713 of LNCS, pages 174{ 188. Springer, 1999.

No context found.

B. Demoen, M. Garca de la Banda, W. Harvey, K. Marriott, and P.J. Stuckey. An overview of HAL. In Proc. of CP'99, LNCS 1713, pages 174-188. Springer-Verlag, 1999. 21

....solvers. Indeed, it does not even fully support Herbrand constraints since it provides only a limited form of uni cation. This paper provides a high level introduction to HAL. For more detailed explanation of various aspects of HAL the interested reader is referred to our earlier publications [4, 3, 9, 2, 13]. In the next section we introduce the HAL language. In Section 3 we discuss the declarations supported by HAL, then in Section 4 we elaborate on how constraint solvers t within in the language. In Section 5 we illustrate the user extensible delay mechanism provided by HAL. Next in Section 6 we ....

B. Demoen, M. Garca de la Banda, W. Harvey, K. Marriott, and P.J. Stuckey. An overview of HAL. In J. Jaar, editor, Proceedings of the Fourth International Conference on Principles and Practices of Constraint Programming, LNCS, pages 174-188. Springer-Verlag, October 1999.

....of their garbage collectors to the hProlog one in Section 2. The SICStus Prolog implementation is described in [5] About Yap one can nd implementation details in [8] hProlog is a successor of dProlog [10] and is available from the rst author. hProlog is meant to become a back end to HAL [9]. We will use the following terminology: the heap is the area in which the WAM allocates structured terms elsewhere the term global stack is often used for this area; the local stack is the stack with environments or stack frames in more traditional terminology; the choice point stack contains ....

B. Demoen, M. Garca de la Banda, W. Harvey, K. Mariott, and P. Stuckey. An overview of HAL. In J. Jaar, editor, Proceedings of the International Conference on Principles and Practice of Constraint Programming, volume 1713 of LNCS, pages 174-188. Springer, 1999.

....set. Transitions serve to simplify constraints and detect satis ability and unsatis ability. CHRs have been used extensively (see e.g. 4] Ecient implementations are already available for the languages SICStus Prolog and Eclipse Prolog, and will soon appear for others such as Java [5] and HAL [2] In this paper we discuss how to improve the compilation of CHRs by using additional information derived either from declarations provided by the user or from the analysis of the constraint handling rules themselves. The major improvements we discuss over previous papers [4] on CHR compilation ....

....uent CHR programs, there are applications where a predictable order of rule applications is important. Hence, their textual order is used to resolve rule applicability con icts in favor of earlier rules. In this paper we focus on the implementation of CHRs in a programming language, such as HAL [2], which requires programmers to provide type, mode and determinism information. A simple example of a HAL CHR program to compute the greatest common divisor of two positive numbers a and b (using the goal gcd(a) gcd(b) is given below. module gcd. L1) import int. L2) export ....

[Article contains additional citation context not shown here]

B. Demoen, M. Garca de la Banda, W. Harvey, K. Marriott, and P.J. Stuckey. An overview of HAL. In Proceedings of the Fourth International Conference on Principles and Practices of Constraint Programming, pages 174-188, 1999.

....solving combinatorial optimization and constraint satisfaction problems. Thus, constraint programmers would like to be able to easily experiment with di erent constraint solvers and to readily develop new problem speci c constraint solvers. The new constraint logic programming (CLP) language HAL [3] has been speci cally designed to allow the user to easily experiment with di erent constraint solvers over the same domain, to support extension of solvers and construction of hybrid solvers, and to call procedures (in particular, solvers) written in other languages with little overhead. In ....

....nition of propagators. Finally, HAL provides global variables which allow ecient implementation of a persistent constraint store. They behave in a similar manner to C s static variables and are only visible within the module in which they are de ned. The initial design of HAL was described in [3]. The current paper extends this in ve main ways. First, we describe how the addition of type classes provides a natural way of specifying a constraint solver s capabilities and, therefore, support for plug and play with solvers. Second, we give a more detailed description of how HAL supports ....

[Article contains additional citation context not shown here]

B. Demoen, M. Garca de la Banda, W. Harvey, K. Marriott, and P.J. Stuckey. An overview of HAL. In Procs. of PPCP'99, LNCS, pages 174-188, 1999.

....require reordering of clause body literals during mode checking. We also sketch the mode checking algorithms used in the HAL compiler. 1 Introduction While traditional logic and constraint logic programming (CLP) languages are untyped and unmoded, recent languages such as Mercury [13] and HAL [4] require type, mode and determinism declarations for (exported) predicates and functions. This information allows the generation of ecient target code (e.g. mode information provides a substantial speed improvement [3] improves robustness and facilitates ecient integration with foreign language ....

B. Demoen, M. Garca de la Banda, W. Harvey, K. Marriott, and P.J. Stuckey. An overview of HAL. In Proc. of CP99, LNCS. Springer-Verlag, October 1999.

....require reordering of clause body literals during mode checking. We also sketch the mode checking algorithms used in the HAL compiler. 1 Introduction While traditional logic and constraint logic programming (CLP) languages are untyped and unmoded, recent languages such as Mercury [14] and HAL [5] require type, mode and determinism declarations for (exported) predicates and functions. This information allows the generation of ecient target code (e.g. mode information provides a substantial speed improvement [4] improves robustness and facilitates ecient integration with foreign language ....

B. Demoen, M. Garca de la Banda, W. Harvey, K. Marriott, and P.J. Stuckey. An overview of HAL. In Proc. of CP99, LNCS. Springer-Verlag, October 1999.

....Here we investigate whether it is possible to have Mercury like eciency, yet still allow true logical variables. We describe our experience with HAL, a new constraint logic programming language speci cally designed to support the construction of and experimentation with constraint solvers [2]. Features of HAL include Mercury like declarations, a well de ned solver interface, dynamic scheduling and global variables. HAL is compiled to Mercury so as to leverage from Mercury s compilation techniques. Unlike Mercury, HAL includes a built in Herbrand constraint solver which provides full ....

....This would explain why our performance improvements are more uniform (and larger) across all benchmarks, regardless of size. 2 The HAL Language In this section we provide a brief overview of the HAL language, concentrating on its support for Herbrand constraints. For more details see [2]. The basic HAL syntax follows the standard CLP syntax, with variables, rules and predicates de ned as usual. The module system in HAL is similar to that of Mercury. The base language supports integer, oat, string, atom and term data types. However, this support is limited to assignment, testing ....

B. Demoen, M. Garca de la Banda, W. Harvey, K. Marriott, and P. Stuckey. An overview of HAL. In Procs. of CP99, to appear, 1999.

No context found.

B. Demoen, M. G. de la Banda, W. Harvey, K. Marriott, and P. J. Stuckey. An Overview of HAL. In Principles and Practice of Constraint Programming, pages 174--188, 1999.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC