Automatic program instrumentation in generation of test data using genetic algorithm for multiple paths coverage

Software testing is an important step during software development. By doing automation of software testing we can increase the robustness of software and decrease the cost of development. The automation of test data generation is done using genetic algorithm. GA analyses the fitness based on the path coverage. This paper presents a novel method for test data generation with multiple paths coverage for instrumented programs. The dynamic instrumentation method is used for analyzing the source code which has been incorporated in java using AspectJ. This reduces the effort of man power required for debugging programs through inserting probes. The experimental results show that the Path Reuse Method (PRM) given in this paper reduces the execution time and no of generations. The PRM automates the test input generation for testing aspectual behavior of object-oriented programs.