An Expressive Aspect Composition Language for UML State Diagrams

摘要

The goal of aspect-oriented software development is to maintain a clear separation of concerns throughout the software lifecycle. Concerns that are separated, however, must be composed at some point. The hypothesis in this paper is that existing aspect-oriented modeling composition methods are not expressive enough for composing state-dependent behavioral models. The paper presents a new aspect composition language, ,SDMATA, for UML state diagrams. SDMATA supports a richer form of model composition than previous approaches to aspect-oriented modeling. Firstly, pointcuts are given as patterns which allows for sequence pointcuts, loop pointcuts, etc. Secondly, SDMATA supports rich forms of composition including parallel composition and alternative composition. The language is applied to the use case slice technique of Jacobson and Ng. The findings are that it is possible to maintain the separation of state-dependent models during software design and that expressive model composition methods are necessary to do this in practice.