Section: New Results

Certification of Languages, Tools and Systems

• A. Tafat and C. Marché developed a certified VC generator using Why3. The challenge was to formalize the operational semantics of an imperative language, and a corresponding weakest precondition calculus, without the possibility to use Coq advanced features such as dependent types nor higher-order functions. The classical issues with local bindings, names and substitutions were solved by identifying appropriate lemmas. It was shown that Why3 can offer a very significantly higher amount of proof automation compared to Coq [36]

• A. Charguéraud, together with the other members of the JsCert team have developed this year the first complete formalization of the semantics of the JavaScript programming language. This project is joint work with Philippa Gardner, Sergio Maffeis, Gareth Smith, Daniele Filaretti and Daiva Naudziuniene from Imperial College, and Alan Schmitt and Martin Bodin from Inria Rennes – Bretagne Atlantique (see http://jscert.org ).

  The formalization consists of a set of inductive rules translating the prose from the ECMAScript Language Specification, version 5. These rules can be used to formally reason about program behaviors or to establish the correctness of program transformations. In addition to the inductive rules, a reference interpreter has been proved correct. This interpreter may be used to run actual JavaScript program following the rules of the formal semantics. It has been used in particular to validate the formal semantics against official JavaScript test suites.

  The formalization of JavaScript has been published at POPL 2014 [24] . A key ingredient in this formalization is the use of the pretty-big-step semantics. This technique allows for representing evaluation rules in big-step style without suffering from a duplication of several premises across different rules. The pretty-big-step technique is described in a paper published by A. Charguéraud at ESOP 2013 [27] .

• É. Contejean, together with V. Benzaken and their PhD student S. Dumbrava, have proposed a Coq formalization of the relational data model which underlies relational database systems [21] . Proposing such a formalization is the first, essential step, that will allow to prove that existing systems conform to their specifications and to verify both production implementations of database systems and database-backed applications. More precisely, they present and formalize the data definition part of the model including integrity constraints, attributes, tuples, relations, schemas and integrity constraints (including the so-called Armstrong's system and the chase). They model two different query language formalisms: relational algebra and conjunctive queries. The former is the basis of the SQL commercial query language and the latter underlies graphical languages, such as Microsoft Access or Query By Example (QBE). They also present logical query optimization and prove the main “database theorems”: algebraic equivalences, the homomorphism theorem and conjunctive query minimization.