Uma abordagem para criação de máquinas de transformações de KDM para PSM
Armijo, Guisella Clara Angulo
MetadataShow full item record
Business needs compel companies to modernize their systems, but changes in legacy systems are complex and costly because the legacy code has programming logic, design decisions, user requirements, and business rules that are difficult to extract. In this context, OMG created Architectural Modernization (ADM), making it possible to model all the artifacts of the legacy system as models and establishing transformations between the different levels of abstraction. ADM is a type of software reengineering that employs standard models throughout the process and handles the entire system architecture. The main metamodel is the Knowledge Discovery Metamodel (KDM), which is an independent language and platform model capable of representing various aspects of a software system. Although a significant number of researches can be found in the reverse engineering phase of ADM, little can be found in relation to the forward engineering; especially in the generation of platform-specific models (PSM) from the KDM. This phase is essential because it belongs to the final part of the ADM horseshoe cycle, completing the entire automated reengineering process. However, the lack of research and the lack of support of available tools hinder ADM adoption in practice. In order to contribute to the advanced engineering phase of ADM, in this project a tool called RUTE-K2J was developed, which is a transformation engine to generate a Java Model from a KDM model. In addition, from this practical experience, a process was developed to support modernization engineers both in (i) creating KDM transformation mechanisms for any other PSM and (ii) in the evolution of RUTE-K2J towards a more stable transformation mechanism and complete. The tool RUTE-K2J was evaluated with a test strategy that considered typical software scenarios in order to validate the correctness of the transformation rules that make up the engine.