A systematic comparison of roundtrip software engineering approaches

• Autor(a) principal: Rosca, Dionisie
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/20305
• topic: Model-driven engineering; Round-trip engineering; Forward engineering; Reverse engineering; UML modeling tools; Metamodel; Models; Code generation; Traceability; Benchmarking; Ferramentas de modelação UML; Metamodelo; Transformação de modelos; Geração de código; Rastreabilidade

Model-based software engineering contemplates several software development approaches in which models play an important role. One such approach is round-trip engineering. Very briefly, round-trip engineering is code generation from models, and models are updated whenever a code change occurs. The objective of this dissertation is to benchmark the comparative analysis of the round-trip engineering capability of the UML, Papyrus, Modelio and Visual Paradigm modeling tools. In more detailed terms, the work will focus on evaluating tools to automatically or semi-automatically support round-trip engineering processes for each selected diagram. Collaterally, this dissertation will allow us to gain insight into the current round-trip engineering landscape, establishing the state-of-the-art UML modeling tool support for this approach. Qualitative and quantitative analysis of the round-trip engineering capabilities of the tools show that the Papyrus, Modeling and Visual Paradigm tools yielded satisfactory results by applying the Reverse and Forward Engineering scenarios without changing the models and codes but applying the Round-trip engineering scenario with changes in model and code presented results with some gaps in model and code coherence. It was concluded that they arose because the semantic definition of the models was done informally. The conclusions drawn throughout the dissertation will answer the questions: How effective are current code generation tools for documenting application evolution? Where will it support the decision made? objectives and will support the recommendations of the best tools that address the round-trip engineering method.