Change Impact Analysis Based Regression Testing of Web Services

Reducing the effort required to make changes in web services is one of the primary goals in web service projects maintenance and evolution. Normally, functional and non-functional testing of a web service is performed by testing the operations specified in its WSDL. The regression testing is performed by identifying the changes made thereafter to the web service code and the WSDL. In this thesis, we present a tool-supported approach to perform efficient regression testing of web services. By representing a web service as a directed graph of WSDL elements, we identify and gathers the changed portions of the graph and use this information to reduce regression testing efforts. Specifically, we identify, categorize, and capture the web service testing needs in two different ways, namely, Operationalized Regression Testing of Web Service (ORTWS) and Parameterized Regression Testing of Web Service (PRTWS). Both of the approach can be combined to reduce the regression testing efforts in the web service project. The proposed approach is prototyped as a tool, named as Automatic Web Service Change Management (AWSCM), which helps in selecting the relevant test cases to construct reduced test suite from the old test suite. We present few case studies on different web service projects to demonstrate the applicability of the proposed tool. The reduction in the effort for regression testing of web service is also estimated.