Completion work of the Latu Sensu Postgraduate Course presented to Universidade Presbiteriana Mackenzie for the title of Specialization in Systems Design and Development.
This project aims at the analysis of CSS files (Cascading Style Sheet) in open source software projects and measuring its evolution pattern, addressing aspects of their behavior. With the benefits from its implementation, CSS is a technology widely used in Web development and is present in the layer of View in an MVC architecture (Model-View-Controller). However, ensure its scalability and maintainability over time in software projects is not trivial and a lot of effort can be expended to ensure the quality of the their source code. Thus, it is crucial to identify the points that are more likely to generate failures and, from then, create strategies to work and correct these weak points. However, making this identification is not a simple or quick task. Several design information and software development are generated indirectly through committing the source code on repository, versioning source code, mailing list, e-mails, among others, thus generating a large amount of data. Focused on automated extraction of information from these sources, there is the area of Mining Software Repositories (MSR). The proposal for this project is, through the use of mining software repositories data, extract and analyze some information in CSS files, generated during the software development process, and infer relevant information about the code quality standards, more frequent changes types and the impacts that these changes can translate into other modules.
Cascading Style Sheets, Mining Software Repositories, Software Repository, Source Code.