Measuring MC/DC Coverage and Boolean Fault Severity of Object-Oriented Programs Using Concolic Testing

Abstract

Fault-Based Testing targets to detect a certain set of faults in a program. There are different faults that may be found in the source code of a given program. This paper considers boolean faults only. From the literature survey, it is found that very few works have been done on targeting MC/DC (Modified Condition / Decision Coverage) score and boolean fault severity together. This motivates to present a novel technique to measure MC/DC Coverage and boolean fault severity of object-oriented programs using concolic testing. This work uses concolic testing for generating test cases for some standard Java programs. This paper proposes an algorithm that uses these test cases to measure test metrics such as MC/DC percentage and fault severity level named FEP (Fault Exposing Potential) of each Java program. FEP is measured for each predicate or boolean expression present in a program and it represents the severity level of boolean faults present in a program. This approach is validated by experimenting on twenty moderate size Java programs.