Closures and Modules Within Linear Logic Concurrent Constraint Programming

Abstract

Abstract. There are two somewhat contradictory ways of looking at modules in a given programming language. On the one hand, module systems are largely independent of the particulars of programming languages. On the other hand, the module constructs may interfere with the programming constructs, and may be redundant with the other scope mechanisms of a specific programming language, such as closures for instance. There is therefore a need to unify the programming concepts that are similar, and retain a minimum number of essential constructs to avoid arbitrary programming choices. In this paper, we realize this aim in the framework of linear logic concurrent constraint programming (LCC) languages. We first show how declarations and closures can be internalized as agents in a variant of LCC for which we provide precise operational and logical semantics in linear logic. Then, we show how a complete module system can be represented within LCC, and prove for it a general code protection property. Finally we study the instanciation of this scheme to the implementation of a safe module system for constraint logic programs, and conclude on the generality of this approach to programming languages with logical variables.

Abstract interpretation of temporal concurrent constraint programs

“…We take inspiration on the structural operational semantics (SOS) for linear ccp in (Fages et al 2001;Haemmerlé et al 2007) to define the behavior of processes. We consider transitions between configurations of the form x; P ; c where c is a constraint representing the current store, P a process and x is a set of distinct variables representing the bound (local) variables of c and P .…”

Semantics-preserving translations between linear concurrent constraint programming and constraint handling rules

“…values in functional languages or tokens in LCC. The adaptations of functional concepts in LCC languages have been initiated with the embedding of closures and modules [13].…”

“…Transient asks wake up at most one time. Persistent asks are introduced [13] to replace declarations. The agent ∀x(c ⇒ a) can wake up as many times as c is entailed.…”

“…Here is the LCC version for dining philosophers [8,20]. Compared to the CHR version, the following code is reentrant: the variable K identifies tokens and let several diners to be run in parallel (a banquet [20]) and separation results from the LCC module theory prove that tables cannot steal cutlery from each other [13].…”

“…The following transformation from pure λ-terms to LCC is proved correct and complete with respect to the call-byvalue semantics of the λ-calculus [13]. Every function, aka λ-value, is represented by a variable K. The constraint apply(K, X, V ) represents that V should code the result of the application of the function (coded by) K to the λ-term (coded by) X.…”