A general framework for the integration of crosscutting concerns in BIP -

Abstract

Most computer systems almost certainly harbor undetected errors because of the gap between requirements and implementation. In this thesis, we define a method that combines Aspect Oriented Programming (AOP) and Component-based Systems (CBSs). AOP is a programming paradigm aiming at supporting the separation of concerns during the development of monolithic systems. We use the Behavior Interaction Priority (BIP) framework which is a component-based framework with formal operational semantics. We distinguish two types of aspects: Local and Global. Local aspects are used to model concerns to refine components. Global aspects are used to model concerns at the architecture-level, and hence refine communications (synchronization and data transfer) between components. We formalize local and global aspects as well as their composition and integration into a BIP system. Our combination of AOP and CBS yields a CBS framework with formal semantics and rigorous transformation primitives. Our method is fully implemented and tested on non-trivial examples.