The introduction of unrestricted overloading in languages with type systems based on implicit parametric polymorphism generally destroys the principal type property: namely that the type of every expression can uniformly be represented by a single type expression over some set of type variables. As a consequence, type inference in the presence of unrestricted overloading can become a NP-complete problem. In this paper we define the concept of parametric overloading as a restricted form of overloading which is easily combined with parametric polymorphism. Parametric overloading preserves the principal type property, thereby allowing the design of efficient type inference algorithms. We present sound type deduction systems, both for predefined and programmer defined overloading. Finally we state that parametric overloading can be resolved either statically, at compile time, or dynamically, during program execution.
CITATION STYLE
Kaes, S. (1988). Parametric overloading in polymorphic programming languages. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 300 LNCS, pp. 131–144). Springer Verlag. https://doi.org/10.1007/3-540-19027-9_9
Mendeley helps you to discover research relevant for your work.