Keeping the software architecture of a system and its implementation consistent can be tough. The larger and more complex a software system is, the more likely software architecture erosion occurs. This effect can lead to a decrease of quality with respect to adaptability, maintainability, or reusability. Refactorings can help to reverse software architecture erosion through systematically applying them to resolve architecture violations. However, it can be difficult in complex systems to manually resolve all violations in an efficient way due to the complex interdependencies between them. In this paper, we propose a new approach to the automatic recommendation of refactorings to resolve architecture violations based on a meta-heuristic search for an efficient set of refactorings. The approach is applied to resolve architectural dependency violations using the "move class" refactoring. © 2014 Springer International Publishing Switzerland.
CITATION STYLE
Herold, S., & Mair, M. (2014). Recommending refactorings to re-establish architectural consistency. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8627 LNCS, pp. 390–397). Springer Verlag. https://doi.org/10.1007/978-3-319-09970-5_33
Mendeley helps you to discover research relevant for your work.