MuString: A Mutation Testing Tool for Reducing Execution Time of Mutants Using String Matching Algorithm

Abstract

Today, testing is the most challenging and dominating activity used by industry, therefore, improvement in its effectiveness, both with respect to the time and resources, is taken as a major factor by many researchers. Software testing allows programmers to determine the quality of the software. Mutation testing is a branch of software testing that does more than this. It helps determine whether the test cases that have been created, effectively detect all the possible faults in the software. This allows the development of better test sets, thus, ensuring that maximum software quality is achieved. Mutation testing works by inserting faults in the program. Various mutation operators are used to create these faulty programs. These programs are called mutants. The mutants depict software faults that may be caused by programmers while writing the software. Test cases are then executed on these mutants to determine if they have been killed or not. Test sets that kill all the mutants are considered to be good as they successfully detect all the possible program faults. This dissertation targets this issue of time, effort that is needed for generating and executing large amount of mutants. A new tool MuString is developed that generate and execute mutants using String Matching Algorithm that reduces total execution time of mutants compared to other previous algorithms. This technique takes the program as an input and then applies changes to that program using String Matching Algorithm. Complete methodology and its effectiveness have been demonstrated with the help of suitable examples.