A Program Model Based Regression Test Selection Technique for Object-Oriented Programs

Abstract

We propose a regression test selection technique that is based on analysis of source code of an object-oriented program. First we construct a System dependency graph model of the original program from the source code. When some modification is executed in a program, the constructed model is updated to reflect the changes. Our approach in addition to capturing control and data dependencies represents the dependencies arising from object-relations. The test cases that exercise the affected model elements in the program model are selected for regression testing. In our approach System Design Graph representation will be used for regression test selection for analyzing and comparing the code changes of original and modified program. Empirical studies carried out by us show that our technique selects on an average of 26.36. % more fault-revealing test cases compared to a Control Dependence Graph based technique while incurring about 37.34% increase in regression test suite size.