Combining Model-Based and Capture-Replay Testing Techniques of Graphical User Interfaces: An Industrial Approach

Abstract

Various approaches for the automated test case generation and execution in the area of graphical user interface testing (GUI) have emerged in recent years. Two notable trends are model-based testing as well as capture and replay technique. In this contribution we present an approach which allows a meaningful combination of both in order to be able to derive fully executable test cases from usage-models and to keep them highly maintainable in a SCRUM process. We first define a set of re-useable test steps by means of a capture and replay tool (TAISpy). Next, we show how it is possible to directly integrate these test steps into the usage models of a GUI and subsequently to generate executable test cases from them without further refinement or transformation to other formats. General idea, implemented architecture, tool chain and, finally, a case study are presented.