Software Composition 

Abstract: Traditional compiler development is non-modular. Although syntax extension and DSL embedding is making its way back in modern language design and implementation, componentisation in compiler construction is still an overlooked matter. Neverlang is a language development framework that emphasises modularity and code reuse. Neverlang makes extension, restriction and feature sharing easier, by letting developers define language components in distinct, independent units, that can be compiled independently and shared across different language implementations, even in their compiled form. The semantics of the implemented languages can be specified using any JVM-supported language. In this paper we will present the architecture and implementation of Neverlang 2, by the help of an example inspired by mobile devices and context-dependent behaviour. The Neverlang framework is already being employed successfully in real-world environments.

Abstract: Component-Based Software Engineering (CBSE) has now emerged as a discipline for system development. An important issue is to fill the gap between high-level models (needed for analysis) and implementation. This paper describes a component model with explicit symbolic protocols based on Symbolic Transition Systems (STSs), and its implementation in Java. This implementation relies on controllers that encapsulate protocols and channels devoted to (possibly remote) communications between components.

Abstract: Context-Aware Composition allows to automatically select optimal variants of algorithms, data-structures, and schedules at runtime using generalized dynamic Dispatch Tables. These tables grow exponentially with the number of significant context attributes. To make Context-Aware Composition scale, we suggest four alternative implementations to Dispatch Tables, all well-known in the field of machine learning: Decision Trees, Decision Diagrams, Naive Bayes and Support Vector Machines classifiers. We assess their decision overhead and memory consumption theoretically and practically in a number of experiments on different hardware platforms. Decision Diagrams turn out to be more compact compared to Dispatch Tables, almost as accurate, and faster in decision making. Using Decision Diagrams in Context-Aware Composition leads to a better scalability, i.e., Context-Aware Composition can be applied at more program points and regard more context attributes than before.

Abstract: In a component-based system, connectors are used to compose components. Connectors should have a semantics that makes them simple to construct and use. At the same time, their semantics should be rich enough to endow them with desirable properties such as genericity, compositionality and reusability. For connector construction, compositionality would be particularly useful, since it would facilitate systematic construction. In this paper we describe a hierarchical approach to connector definition and construction that allows connectors to be defined and constructed from sub-connectors. These composite connectors are indeed generic, compositional and reusable. They behave like design patterns, and provide powerful composition connectors.