Quantified, Typed Events for Improved Separation of Concerns

Implicit invocation and aspect-oriented languages provide related but distinct mechanisms for separation of concerns. Implicit invocation languages have explicitly announced events, which runs registered observer methods. Aspect-oriented languages have implicitly announced events, called “join points,” which run methodlike but more powerful advice. A limitation of implicit invocation languages is their inability to refer to a large set of events succinctly. They also lack the expressive power of aspect-oriented advice, and require code to manage event registration and announcement. Aspect-oriented languages also have several limitations, including the potential for fragile dependence on syntactic structure that may hurt maintainability, limits in the set of join points and the reflective contextual information that they make available. Quantified, typed events solve all these problems. They extend implicit invocation languages with a key idea from aspectoriented languages: the ability to quantify over events (join points). Programmers declare named event types that contain information about the names and types of event arguments (exposed context). An event type declaratively identifies an expression as an event. This event type can then be used to quantify over all such events. Event types reduce the coupling between the observers and the set of events, and similarly between the advising and advised code.