On the Use of Software Visualization to Analyze Library Dependency Evolution: an Exploratory Study
Rodrigo Magnavita, Renato Lima Novais, Thiago Souto Mendes, Manoel Mendonça · 2016
Software evolution produces a large amount of data. Software engineers may benefit from these data while carrying out daily activities. The challenge is to understand all the data and recovery valuable information. One of these activities is to maintain library dependencies of a project. This is a complex task, since it is usual a system and its library dependencies evolving separately. It is necessary to be careful when updating the libraries of a project. The aintainers need to know the history about a system ́s past upgrade decisions and which dependencies were adopted at the same time. Software Evolution Visualization (SEV) can be a promising approach to this end. In this paper, we present an exploratory study that uses a SEV tool, called EVOWAVE, to analyze Library Dependency Evolution. EVOWAVE is able to visualize different types of data generated in software evolution using both overview-based and detail based approaches. In the study, we performed an analysis of FindBug’s library dependency history. The results highlighted the benefits of the SEV tool on analyzing a large amount of data that encompasses more than 10 years.