A model refactoring does not only need to ensure behavior preservation. First of all, it needs to ensure that specific well-formedness constraints of the modeling language under consideration are preserved (consistency preservation). The consistency of model refactorings can be ensured by runtime checks. However, this means that not the developer of the refactorings but the user is confronted with the problem. In this paper we present an approach to statically check for consistency preservation of rule-based refactorings at design time. Thereby, refactoring rules describe which side-effects may take place on the model to be refactored. We formalize rule-based refactorings using graph transformation and consistency using graph constraints. We extend a verification technique capable of proving statically that refactoring rule applications preserve consistency. By automatically computing meaningful counterexamples, this technique allows for the iterative development of refactoring rules guaranteeing consistency preservation. We demonstrate the approach for common Java refactorings applied to a fine grained EMF model and can show that bugs that were present in refactorings of former Eclipse versions could have been avoided using our approach. © 2011 Springer-Verlag.
CITATION STYLE
Becker, B., Lambers, L., Dyck, J., Birth, S., & Giese, H. (2011). Iterative development of consistency-preserving rule-based refactorings. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6707 LNCS, pp. 123–137). https://doi.org/10.1007/978-3-642-21732-6_9
Mendeley helps you to discover research relevant for your work.