Language-independent quantification and weaving for feature composition

Abstract

Based on a general model of feature composition, we present a composition language that enables programmers by means of quantification and weaving to formulate extensions to programs written in different languages. We explore the design space of composition languages that rely on quantification and weaving and discuss our choices. We outline a tool that extends an existing infrastructure for feature composition and discuss results of three initial case studies. We found that, due to its language independence, our approach is less powerful than aspect-oriented languages but still usable for many implementation problems. © 2009 Springer Berlin Heidelberg.