Abstract-Abstractions are used in model checking to tackle problems like state space explosion or modeling of IO. The application of these abstractions in real software development processes, however, lacks engineering support. This is one reason why model checking is not widely used in practice yet and testing is still state of the art in falsification. We show how user-defined abstractions can be integrated into a Java PathFinder setting with tools like AspectJ or Javassist and discuss implications of remaining weaknesses of these tools. We believe that a principled engineering approach to designing and implementing abstractions will improve the applicability of model checking in practice.

The implementation of unit tests with mock objects and stubs often involves substantial manual work. Stubbed methods return simple default values, therefore variations of these values require separate test cases. The integration of mock objects often requires more infrastructure code and design decisions exposing private data. We present an approach that both improves the expressiveness of test cases using non-deterministic choice and reduces design modifications using dynamic aspect-oriented programming techniques.Non-deterministic choice facilitates local definitions of multiple executions without parameterization or generation of tests. It also eases modelling naturally non-deterministic program features like IO or multi-threading in integration tests. Dynamic AOP facilitates powerful design adaptations without exposing test features, keeping the scope of these adaptations local to each test. We also combine nondeterminism and dynamic aspects in a new approach to testing multi-threaded programs using co-routines.

Abstract. Dynamic aspect-oriented programming has been widely used for the development of dynamic analyses to abstract over low-level program instrumentation. Due to particular feature requirements in different analysis domains like debugging or testing, many different aspect languages were developed from scratch or by extensive compiler or interpreter extensions. We introduce another level of abstraction in form of a meta-aspect protocol to separate the host language from the analysis domain. A language expert can use this protocol to tailor an analysisspecific aspect language, based on which a domain expert can develop a particular analysis. Our design enables a flexible specification of the join point model, configurability of aspect deployment and scoping, and extensibility of pointcut and advice language. We present the application of our design to different dynamic analysis domains.