This paper presents a type-based information flow analysis for a call-by-value λ-calculus equipped with references, exceptions and let-polymorphism, which we refer to as Core ML. The type system is constraint-based and has decidable type inference. Its noninterference proof is reasonably light-weight, thanks to the use of a number of orthogonal techniques. First, a syntactic segregation between values and expressions allows a lighter formulation of the type system. Second, noninterference is reduced to subject reduction for a nonstandard language extension. Lastly, a semi-syntactic approach to type soundness allows dealing with constraint-based polymorphism separately.

This paper shows how to systematically extend an arbitrary type system with dependency information, and how soundness and non-interference proofs for the new system may rely upon, rather than duplicate, the soundness proof of the original system. This allows enriching virtually any of the type systems known today with information flow analysis, while requiring only a minimal proof effort.Our approach is based on an untyped operational semantics for a labelled calculus akin to core ML. Thus, it is simple, and should be applicable to other computing paradigms, such as object or process calculi.The paper also discusses access control, and shows it may be viewed as entirely independent of information flow control. Letting the two mechanisms coexist, without interacting, yields a simple and expressive type system, which allows, in particular, "selective" declassification.

We offer a solution to the type inference problem for an extension of Hindley and Milner's type system with generalized algebraic data types. Our approach is in two strata. The bottom stratum is a core language that marries type inference in the style of Hindley and Milner with type checking for generalized algebraic data types. This results in an extremely simple specification, where case constructs must carry an explicit type annotation and type conversions must be made explicit. The top stratum consists of (two variants of) an independent shape inference algorithm. This algorithm accepts a source term that contains some explicit type information, propagates this information in a local, predictable way, and produces a new source term that carries more explicit type information. It can be viewed as a preprocessor that helps produce some of the type annotations required by the bottom stratum. It is proven sound in the sense that it never inserts annotations that could contradict the type derivation that the programmer has in mind.

No abstract

This paper studies type inference for a functional, ML-style language with subtyping, and focuses on the issue of simplifying inferred constraint sets. We propose a powerful notion of entailment between constraint sets, as well as an algorithm to check it, which we prove to be sound. The algorithm, although very powerful in practice, is not complete. We also introduce two new typing rules which allow simplifying constraint sets. These rules give very good practical results.