Constraining type parameters of UML 2 templates with substitutable classifiers

14Citations
Citations of this article
8Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Generic programming is a field of computer science which consists in defining abstract and reusable representations of efficient data structures and algorithms. In popular imperative languages, it is usually supported by a template-like notation, where generic elements are represented by templates exposing formal parameters. Defining such generic artifacts may require defining constraints on the actual types that can be provided in a particular substitution. UML 2 templates support two mechanisms for expressing such constraints. Unfortunately, the UML specification provides very few details on their usage. The purpose of our article is to provide such details with regard to one of these constraining mechanisms (namely, "substitutable constraining classifiers") as well as modeling patterns inspired by practices from generic programming. © 2009 Springer Berlin Heidelberg.

Cite

CITATION STYLE

APA

Cuccuru, A., Radermacher, A., Gérard, S., & Terrier, F. (2009). Constraining type parameters of UML 2 templates with substitutable classifiers. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5795 LNCS, pp. 644–649). https://doi.org/10.1007/978-3-642-04425-0_51

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