COMPASS: composition-centric mapping of aspectual requirements to architecture

  • Authors:

  • Ruzanna Chitchyan;Mónica Pinto;Awais Rashid;Lidia Fuentes

  • Affiliations:

  • Computing Department, Lancaster University, Lancaster, UK;Dept. Lenguajes y Ciencias de la Computación, University of Málaga, Málaga, Spain;Computing Department, Lancaster University, Lancaster, UK;Dept. Lenguajes y Ciencias de la Computación, University of Málaga, Málaga, Spain

  • Venue:
  • Transactions on aspect-oriented software development IV
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Currently there are several approaches available for aspect-oriented requirements engineering and architecture design. However, the relationship between aspectual requirements and architectural aspects is poorly understood. This is because aspect-oriented requirements engineering approaches normally extend existing requirements engineering techniques. Although this provides backward compatibility, the composition semantics of the aspect-oriented extension are limited by those of the approaches being extended. Consequently, there is limited or no knowledge about how requirements-level aspects and their compositions map on to architecture-level aspects and architectural composition. In this paper, we present COMPASS, an approach that offers a systematic means to derive an aspect-oriented architecture from a given aspect-oriented requirements specification. COMPASS is centred on an aspect-oriented requirements description language (RDL) that enriches the usual informal natural language requirements with additional compositional information derived from the semantics of the natural language descriptions themselves. COMPASS also offers an aspect-oriented architecture description language (AO-ADL) that uses components and connectors as the basic structural elements (similar to traditional ADLs) with aspects treated as specific types of components. Lastly, COMPASS provides a set of concrete mapping guidelines, derived from a detailed case study, based on mapping patterns of compositions and dependencies in the RDL to patterns of compositions and dependencies in the AO-ADL. The mapping patterns are supported via a structural mapping of the RDL and AO-ADL meta-models.