Français Anglais
Accueil Annuaire Plan du site
Accueil > Production scientifique > Résultat majeur
Production scientifique
Résultat majeur : MODEL TRANSFORMATION AS CONSERVATIVE THEORY-TRANSFORMATION
MODEL TRANSFORMATION AS CONSERVATIVE THEORY-TRANSFORMATION
30 octobre 2020

We present a new technique to construct tool support for domain-specific languages (DSLs) inside the interactive theorem prover environment Isabelle. Our approach is based on modeling the DSL formally in higher-order logic (HOL).
Model transformations play a central role in model-driven software development. Hence, logical unsafe model transformation can result in erroneous systems. Still, most model transformations are written in languages that do not provide built-in safeness guarantees. We present a new technique to construct tool support for domain-specific languages (DSLs) inside the interactive theorem prover environment Isabelle. Our approach is based on modeling the DSL formally in higher-order logic (HOL), modeling the API of Isabelle inside it, and defining the transformation between these two. Reflection via the powerful code generators yields code that can be integrated as extension into Isabelle and its user interface. Moreover, we use code generation to produce tactic code which is bound to appropriate command-level syntax. Our approach ensures the logical safeness (conservativity) of the theorem prover extension and, thus, provides a certified tool for the DSL in all aspects: the deductive capacities of theorem prover, code generation, and IDE support. We demonstrate our approach by extending Isabelle/HOL with support for UML/OCL and, more generally, providing support for a formal object-oriented modeling method.

This work was authored by Achim Brucker, Frederic Tuong and Burkhart Wolff.

Activités de recherche
  [aucun]

Equipe
  [aucun]

Contact
  ° WOLFF Burkhart
Résultats majeurs
BEST STUDENT PAPER AWARD (ML) AT ECML 2019
20 septembre 2019
Guillaume Doquet (A&O), Best Student Paper Award (category Machine Learning) at ECML 2019.

BEST PAPER AWARD - HPCS 2019 - ON SERVER-SIDE FILE ACCESS PATTERN MATCHING
17 juillet 2019
Francieli Zanon Boito¹ , Ramon Nou², Laércio Lima Pilla³, Jean Luca Bez⁴, Jean-François Méhaut¹, T

BEST FULL PAPER AWARD EDM 2019 - EDUCATIONAL DATA MINING
05 juillet 2019
DAS3H: Modeling Student Learning and Forgetting for Optimally Scheduling Distributed Practice of Ski

BEST PAPER AWARD - CODIT2019 - STOCHASTIC DUAL DYNAMIC INTEGER PROGRAMMING FOR A MULTI-ECHELON LOT-SIZING PROBLEM WITH REMANUFACTURING AND LOST SALES
14 mai 2019
Franco Quezada, Céline Gicquel and Safia Kedad-Sidhoum

BEST PAPER AWARD ICONS 2019: RESONANCE THINKING AND INDUCTIVE MACHINE LEARNING
06 mai 2019
Yves Kodratoff & Marta Franova