A self-replication algorithm to flexibly match execution traces

Abstract

Stateful aspects react to the history of a computation. Stateful aspect developers define program execution patterns of interest to which aspects react. Various stateful aspect languages have been proposed, each with non-customizable semantics for matching a join point trace. For instance, most languages allow multiple matches of a sequence when the associated context information is different. Obtaining a different matching semantics requires ad hoc modifications of the aspects, if at all possible. In order to allow flexible customization of the matching semantics of a given aspect, this paper presents a self-replication algorithm called MatcherCells. Through the composition of simple reaction rules, MatcherCells makes it possible to express a wide range of matching semantics, per aspect. In addition, we present an initial implementation of our proposal.