ABCE: A Novel Framework for Improved Branch Coverage Analysis

Abstract

Abstract The software testing practices generate test cases manually, that affects both the effectiveness and efficiency. In aerospace and safety critical domains, software quality assurance is strict to rules and regulations such as DO-178B standard. To resolve these issues, concolic testing generates test suite that can attain high coverage using an automated technique based on Branch Coverage. In this paper, we propose a framework to compute branch coverage percentage for test case generation. To achieve an increase in branch coverage, we transform the input Java program, P , into its transformed version, P t , using JPCT. Then we use JCUTE to generate test cases and find branch coverage percentage. Our experimental studies on branch coverage percentage consists of two steps: the first observation is made without using Java program code transformer, the second observation is made by using Java program code transformer. The Java program code transformer adds additional conditional statements for each decision that causes variation in branch coverage. We quantify this variation in the form of branch coverage percentage. This approach resolves some of the bottleneck issues associated with traditional concolic testers. The experimental studies show that our approach achieves 20.146% of average increase in branch coverage. This increase in branch coverage percentage is achieved in an average computation time of 33837.6 milli seconds.