Model-based development promises to increase productivity and quality through domain-specific modeling languages. When a modeling language and its metamodel evolve, existing models must be migrated. An approach to model migration faces two main challenges. First, it must automate migration as far as possible to reduce migration effort. Second, it must lead to a correct migration preserving the meaning of all models. In this thesis, we present the approach COPE that addresses both challenges. COPE defines model migration as a sequence of small, modular operations that are coupled with metamodel adaptations. To reduce migration effort, recurring coupled operations can be reused. To foster correctness, COPE records the operations when the metamodel is evolved. An evaluation on six real-life case studies showed that reusing over 95% of the operations highly automates migration, and that recording small operations fosters correct migrations.
«
Model-based development promises to increase productivity and quality through domain-specific modeling languages. When a modeling language and its metamodel evolve, existing models must be migrated. An approach to model migration faces two main challenges. First, it must automate migration as far as possible to reduce migration effort. Second, it must lead to a correct migration preserving the meaning of all models. In this thesis, we present the approach COPE that addresses both challenges. COP...
»