Automating third-party library migrations

Abstract

Manual migration between various third-party libraries is a problem for software developers. Developers usually need to study the application programming interfaces of both libraries, as well as read their documentation to find suitable comparisons between the replacement and the replaced methods. In this article, I will present a new approach (MIG) to machine learning that recommends mappings between the methods of two API libraries. My model learns from manually found data of implemented migrations, extracts a set of functions related to the similarity of the method signature and text documentation. I evaluated the model using 8 popular migrations compiled from 57,447 open source Java projects. The results show that the model can recommend appropriate library API mappings with an average accuracy rate of 87%.   This study examines the problem of recommending method comparisons when migrating between third-party libraries. A new approach is described that recommends the comparison of methods between two unknown libraries using features extracted from the lexical similarity between method names and textual similarity of method documentation. I evaluated the result by checking how this approach and three other most commonly used approaches recommend a comparison of migration methods for 8 popular libraries. I have shown that the proposed approach shows much better accuracy and performance than the other 3 methods. Qualitative and quantitative analysis of the results shows an increase in accuracy by 39.51% in comparison with other well-known approaches.