IMPROVING SOFTWARE QUALITY BY SPEEDING UP MUTATION TESTING

Filimonov Artyom Alexandrovich

Mutation testing has a long history that starts back in the 1970s. The process of mutation analysis has undergone many improvements. But despite this, there are still areas for development. The number of publications is growing year after year, which shows the interest of system engineers in this topic and its relevance. Since mutation testing can be divided into several steps, most research focuses on optimizing a specific step rather than the whole process. This paper will focus on test cases and the implementation of an algorithm for their effective use in mutation testing. The algorithm being developed aims to show that selecting relevant test cases can reduce the resource cost of performing automated mutation testing. The reason for implementing such an algorithm is that testing has always been an important step in the development process, with written tests not always being able to validate the quality of the software. In order to improve the quality of testing, mutation testing is used. Conducting mutation testing is a resource-intensive process; the new approach is supposed to save the resources and make the process available to developers. The result of this work is a test case selection method based on commits and test coverage and a software tool module for mutation testing. To achieve the desired result, a systematic analysis of research work in this direction has been done. Also, test case selection algorithms have been developed, based on them the module for mutation testing tool has been implemented and tested on open source projects to confirm the applicability.

Keywords:testing quality; mutant selection; test suite reduction; test coverage; git commits