Typing Context-Dependent Behavioural Variation

TL;DR

This paper introduces ContextML, a core language with static analysis and type system guarantees, enabling predictable context-dependent behavioral adaptation in programs.

Contribution

It presents a formal core language with COP primitives and a type system that ensures reliable context-based program dispatching.

Findings

Type system guarantees successful dispatch at runtime

Formal semantics for context-dependent behavior

Predictability in context-oriented programming

Abstract

Context Oriented Programming (COP) concerns the ability of programs to adapt to changes in their running environment. A number of programming languages endowed with COP constructs and features have been developed. However, some foundational issues remain unclear. This paper proposes adopting static analysis techniques to reason on and predict how programs adapt their behaviour. We introduce a core functional language, ContextML, equipped with COP primitives for manipulating contexts and for programming behavioural variations. In particular, we specify the dispatching mechanism, used to select the program fragments to be executed in the current active context. Besides the dynamic semantics we present an annotated type system. It guarantees that the well-typed programs adapt to any context, i.e. the dispatching mechanism always succeeds at run-time.