Skip to main navigation Skip to search Skip to main content

An inference algorithm for strictness

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper we introduce an algorithm for detecting strictness information in typed functional programs. Our algorithm is based on a type inference system which allows to exploit the type structure of the language for the investigation of program properties. The detection of strictness information can be performed in a complete way, by reducing it to the solution of a set of constraints involving type variables. A key feature of our method is that it is compositional and there is a notion of principal type scheme. Even though the language considered in the paper is the simply typed lambda-calculus with a fixpoint constructor and arithmetic constants we can generalize our approach to polymorphic languages like Lazy ML. Although focused on strictness our method can also be applied to the investigation of other program properties, like dead code and binding time.

Original languageEnglish
Title of host publicationTyped Lambda Calculi and Applications - 3rd International Conference on Typed Lambda Calculi and Applications, TLCA 1997, Proceedings
EditorsPhilippe de Groote, J. Roger Hindley
PublisherSpringer Verlag
Pages129-146
Number of pages18
ISBN (Print)3540626883, 9783540626886
DOIs
Publication statusPublished - 1997
Externally publishedYes
Event3rd International Conference on Typed Lambda Calculi and Applications, TLCA 1997 - Nancy, France
Duration: 2 Apr 19974 Apr 1997

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1210
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference3rd International Conference on Typed Lambda Calculi and Applications, TLCA 1997
Country/TerritoryFrance
CityNancy
Period2/04/974/04/97

Fingerprint

Dive into the research topics of 'An inference algorithm for strictness'. Together they form a unique fingerprint.

Cite this