A 3-part type inference engine

8Citations
Citations of this article
5Readers
Mendeley users who have this article in their library.

This article is free to access.

Abstract

Extending a subtyping-constraint-based type inference frame- work with conditional constraints and rows yields a powerful type inference engine. We illustrate this claim by proposing solutions to three delicate type inference problems: “accurate" pattern matchings, record concatenation, and “dynamic" messages. Until now, known solutions required significantly different techniques; our theoretical contribution is in using only a single (and simple) set of tools. On the practical side, this allows all three problems to benefit from a common set of constraint simplification techniques, leading to efficient solutions.

Cite

CITATION STYLE

APA

Pottier, F. (2000). A 3-part type inference engine. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1782, pp. 320–335). Springer Verlag. https://doi.org/10.1007/3-540-46425-5_21

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free