An Approach for Domain Reduction with Data Dependence in Mutation Testing

As a testing strategy to evaluate the completeness of test cases, mutation testing has been identified as a "faultoriented" technique for unit testing, which is mainly used to generate complete test cases. The path-oriented technique of test data generation is a highly efficient technique which implements test data generation by building and solving constraint systems. Most of path-oriented generation techniques only take control dependence among statements into consideration, which is to build constraint system by analyzing control flow graph. However, it neglects the influence of data dependence among statements on constraint system. Therefore, this paper improved test data generation technique of domain reduction and proposed a new domain reduction method with data dependence. It added detecting of equivalent mutants and solved influences on constraint systems caused by multiple conditional branch statement. Experimental results showed that this method improved success rate and execution efficiency of test data generation in a significant extent.