A Dedicated Language for Context Composition and Execution of True Black-Box Model Transformations

摘要

Model-Driven Engineering (MDE) automates development activities by employing model transformations. Thereby, a plethora of model transformation approaches with individual capabilities have been developed. In certain cases, complex and automated MDE activities require the interaction of various, potentially heterogeneous, model transformations. This can be achieved by a loosely coupled and highly cohesive composition of model transformations implemented in different model transformation languages. However, existing approaches either do not support context composition, using other model transformations as additional context, or they violate the important black-box principle because they require adapting model transformations for context composition. In this paper, we present a dedicated model transformation composition framework (MoTCoF) that does not require the adaptation of model transformations and, thus, treats model transformations as true black-boxes. We illustrate our approach with an application example taken from an industrial case study.