SEAL: Integrating Program Analysis and Repository Mining

Software projects are complex technical and organizational systems involving large numbers of artifacts developers. To understand tame software complexity, a wide variety program analysis techniques have been developed for bug detection, comprehension, verification, more. At the same time, repository mining aim at obtaining insights into inner socio-technical workings larger scale. While both successful on their own, they largely isolated, which leaves considerable potential synergies untapped. We present SEAL, first integrated approach that combines low-level with high-level information. SEAL maps information, mined from development history project, onto intermediate representation, making it available state-of-the-art analysis. SEAL’s allows us to efficiently address engineering problems span multiple levels abstraction, data flow demonstrate its merits practicality, we use determine code changes modify central parts given how authors interact (indirectly) each other through code, putting static analysis’ results context improves expressiveness interpretability.