Automatic detection of infeasible ALL-DU paths in the data flow test using an evolutionary approach

The infeasible path indicates that no input can execute the code anyway. The infeasible path is a kind of defect that is easily visible in software development and is also one of the most important defects in the software's white box test. And the main reason for the program's infeasible paths is the correlation between some conditional expressions.The correlation of conditional expressions plays an important role in identifying infeasible paths. In the data flow test, the pair of "def-uses" along the infeasible paths must be removed from the set of requirements for coverage by the test cases. As a result, the reduction of infeasible pairs increases the reliability of the regression test and the integration test. In this paper, we propose a new method to detect infeasible all-du paths in the data flow test using the concept of correlation tree and Particle Swarm Optimization algorithm, and we have used the concept of popularity and unpopularity in the new fitness function. we validate the performance of the proposed method through a group of experiments. To validate the proposed method, empirical analysis is performed on some of the most widely used standard programs. We use 10 subject programs written in C language for experiments. Experimental results showed that the proposed approach can detect effectively infeasible all-du paths.