Capturing Software Evolution and Change through Code Repository Smells

摘要

In the last years we have seen the rise and the fall of many version control systems. These systems collect a large amount of data spanning from the path of the files involved in changes to the exact text changed in every file. This data can be exploited to produce an overview about how the system changed over time and evolved. We have developed a tool, called VCS-Analyzer, to use this information, both for data retrieval and analysis tasks. Currently, VCS-Analyzer implements six different analyses: two based on source code for the computation of metrics and the detection of code smells, and four original analysis based on repositories metadata, which are based on the concepts of Repository Metrics and Code Repository Smells. In this paper, we describe one smell and two metrics we have defined for source code repositories analysis.