Managing software evolution through semantic history slicing

Software change histories are results of incremental updates made by developers. As a side-effect of the software development process, version history is a surprisingly useful source of information for understanding, maintaining and reusing software. However, traditional commit-based sequential organization of version histories lacks semantic structure and thus are insufficient for many development tasks that require high-level, semantic understanding of program functionality, such as locating feature implementations and porting hot fixes. In this work, we propose to use well-organized unit tests as identifiers for corresponding software functionalities. We then present a family of automated techniques which analyze the semantics of historical changes and assist developers in many everyday practical settings. For validation, we evaluate our approaches on a benchmark of developer-annotated version history instances obtained from real-world open source software projects on GitHub.