Similarity-based regression test case prioritization

Abstract

With the continuous evolution of software systems, test suites often grow very large. Rerunning all test cases may be impractical in regression testing under limited resources. Coverage-based test case prioritization techniques have been proposed to improve the effectiveness of regression testing. The original test suite often contains some test cases which are designed for exercising production features or exceptional behaviors, rather than for code coverage. Therefore, coverage-based prioritization techniques do not always generate satisfactory results. In this context, we propose a global similarity-based regression test case prioritization approach. The approach reschedules the execution order of test cases based on the distances between pair-wise test cases. We designed and conducted empirical studies on four C programs to validate the effectiveness of our proposed approach. Moreover, we also empirically compared the effects of six similarity measures on the global similarity-based test case prioritization approach. Experimental results illustrate that the global similarity-based regression test case prioritization approach using Euclidean distance is the most effective. This study aims at providing practical guidelines for picking the appropriate similarity measures.