Enhancing Search-Based Techniques with Information Control Dependencies

Graduation date: 2017Software testing is a very important task during software development and it can be used to improve the quality and reliability of the software system. One potential way to reduce the cost and increase the efficiency of software testing is to generate test data automatically. Search-based approaches successfully generate unit tests for object-oriented programs, like Java. However, challenges, such as the large size of the search space, and the presence of complex predicates’ target branches, negatively affect the approaches, and, thus, cannot achieve high structural coverage for certain programs.\ud \ud The aim of this thesis is to propose enhancement techniques to improve the effectiveness of search based testing approaches and address the challenges posed by the object-oriented programs. Rather than randomly generating a sequence of method calls, our ongoing work is to focus on using static analysis to define the hidden data dependencies on predicates’ target branches (i.e., uncovered branches) and exploit method dependence relations (MDR) approach to precisely identify the method/constructors and its parameters. This method dependence information is employed to reduce the search space and used to guide the search toward regions that lead to full (or at least high) structural coverage