No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Visualizing explanations to exhibit dynamic structure in constraint problems
Abstract
no abstract
... This approach is implemented in the Grace tool [18] for finite domains visualization, and in OPL studio [2] for search tree and constraint propagation visualization. Some of the tools in this category rely on the enrichment of the constraint solver to help improve the visualization, whether by grouping constraints [15] or by keeping explanations of domain reductions [13]. − dynamic visualization and control tools which allow interaction with a CLP process through different visualizations. ...
CLPGUI is a generic graphical user interface for visualizing and con- trolling the execution of constraint logic programs. CLPGUI has been designed to be used in different contexts: initially for teaching purposes, then for debugging complex programs of real-world scale, and recently for developing end-user interfaces. The challenge of developing a tool which is generic w.r.t. both the constraint logic programming system and the visualizers, is addressed by a client-server architecture for connecting a CLP process to a Java-based GUI process, and by a non-intrusive tracing and control method based on annotations in the CLP program. Arbitrary constraints and goals can be posted incrementally from the GUI in an interactive manner, and arbitrary states can be recomputed. We describe several generic 2D and 3D viewers of the variables and of the search tree, and argue that the D representation is best-suited to apprehend the shape of large search trees. We also illustrate the use of CLPGUI for developing application-oriented end-user interfaces on two placement problems, one in virtual reality.
... Looking at the way variables and constraints interact together through explanations [36] leads to consider a more intricate usage of explanations within constraint programming languages. Therefore, one can design new programming gimmicks and new approaches. ...
La programmation par contraintes est un sujet de recherche qui tire profit de nombreuses autres disciplines : mathématiques discrètes, analyse numérique, intelligence artificielle, recherche opérationnelle et calcul formel. Elle a prouvé son intérêt et son efficacité dans de nombreux domaines : optimisation combinatoire, ordonnancement, finance, simulation et synthèse de composants, diagnostic de panne, biologie moléculaire, ou encore problèmes géométriques. Néanmoins, un certain nombre de limitations et de difficultés ont été identifiées dans le domaine : conception d'algorithmes génériques et stables, traitement des problèmes dynamiques, accessibilité des concepts et des outils, ...
Dans ce document, nous plaidons pour l'utilisation de la notion d'explication au sein de la programmation par contraintes. Notre but est double : non seulement présenter un tableau général des explications (définition, génération et utilisations) mais aussi montrer comment leur utilisation permet de contribuer à lever certains des problèmes ouverts en programmation par contraintes. Nous présentons aussi une démarche générale de résolution de problème dans un environnement expliqué. Enfin, nous montrons comment ce nouveau sujet semble promis à un bel avenir.
Debugging tools are essential to help tune constraint solving programs and each platform has its environment tools. However,
at present, these tools are specific and have to be redesigned and re-implemented for each constraint solver whereas much
could be factorized. This article sets the foundations to enable debugging tools to be defined almost independently from finite
domain solvers, and conversely, tracers to be built independently from these tools. We propose a generic trace schema based on a generic observational semantics which formalizes relevant aspects of constraint programming and solving. We illustrate the genericity of the schema on three
representative families of finite domain solvers: CLP (Gnu-Prolog), CSP (Choco) and explanation based CP (PaLM). Debugging
tools can use the generic trace and do not have to take into account all the details of the solvers. We experimented several
combinations of new debugging tools with the above mentioned solvers. We show that they all find in the generic trace the
whole information they require.
The 3 years OADymPPaC project ended may 2004.
The expected impacts of the project were of three kinds: new results in constraint program development and debugging, and
visualization of dynamic phenomena; improvement of the industrial and academic platforms: constraint solvers and their tools
for constraint program development and modules of visualization, enhancement of several platforms (GNU-Prolog, PaLM, Choco,
CHIP); contribution to the teaching of the programming with constraints.
All results (reports and systems) may be found at the project URL:
http://contraintes.inria.fr/OADymPPaC
The Prolog programmer's needs have always been the focus for guiding the development of the SWI-Prolog system. This article ac- companies an invited talk about how the SWI-Prolog environment helps the Prolog programmer solve common problems. It describes the central parts of the graphical development environment as well as the command line tools which we see as vital to the success of the system. We hope this comprehensive overview of particularly useful features will both in- spire other Prolog developers, and help SWI-Prolog users to make more productive use of the system.
ResearchGate has not been able to resolve any references for this publication.