AutoDBT: A framework for Automatic Testing of Web Database applications

Abstract

The complex functionalities and high demands of software quality make manual testing of a web application ineffective. Automatic software testing methods can help to determine if a web application is working correctly, but existing methods are unable to test whether such an application interacts correctly with a back-end database. This paper elaborates an approach, called the Automatic Database Tester (AutoDBT), that extends the functional or black-box testing of a web database application to include database updates. AutoDBT takes as input a model of the application and a set of test criteria. The model consists of a state transition diagram showing how a user navigates among pages in the application, and a data specification which captures how data flows in the application and how the database is updated. AutoDBT uses the model along with the test criteria to generate test cases for functional testing of the application. © Springer-Verlag 2004.