Uma abordagem para compreensão de programas baseada na localização de características em código fonte

• Ano de defesa: 2008
• Autor(a) principal: Sobreira, Victor
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; BR; Programa de Pós-graduação em Ciência da Computação; Ciências Exatas e da Terra; UFU
• Palavras-chave em Português: Manutenção de software; Compreensão de programas; Visualização de software; Localização de características; Análise dinâmica; Software - Manutenção; Software maintenance; Program comprehension; Software visualization; Feature location; Dynamic analysis; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufu.br/handle/123456789/12462

Features are an important concept for understanding software requirements. However, discovering where a desired feature is located in the source code is a challenging task because the code is generally not modularized in a feature-fashioned way, and so, features tend to be scattered over the source code. This work proposes a method and a software tool called Featincode for analyzing feature scattering through the graphical interpretation of the intersection between feature elements and source code elements. The tool collects and represents trace events of multi threaded programs for developer selected features and show some matrices that help analyzing where those features are implemented. The proposed method and tool are validated with the analysis of some features of the ArgoUML CASE tool. The conclusion is that our approach can reduce the effort for comprehending where features are implemented and which source code is specific to a feature. The method and tool can be applied to help software comprehension, maintenance and evolution activities.