Graph-based detection of library API imitations

It has been a common practice nowadays to employ third-party libraries in software projects. Software libraries encapsulate a large number of useful, well-tested and robust functions, so that they can help improve programmers' productivity and program quality. To interact with libraries, programmers only need to invoke Application Programming Interfaces (APIs) exported from libraries. However, programmers do not always use libraries as effectively as expected in their application development. One commonly observed phenomenon is that some library behaviors are re-implemented by client code. Such re-implementation, or imitation, is not just a waste of resource and energy, but its failure to abstract away similar code also tends to make software error-prone. In this paper, we propose a novel approach based on trace subsumption relation of data dependency graphs to detect imitations of library APIs for achieving better software maintainability. Furthermore, we have implemented a prototype of this approach and applied it to ten large real-world open-source projects. The experiments show 313 imitations of explicitly imported libraries with high precision average of 82%, and 116 imitations of static libraries with precision average of 75%.