Grey-box GUI Testing: Efficient Generation of Event Sequences

Graphical user interfaces (GUIs) encode, as event sequences, potentially unbounded ways to interact with software. During testing it becomes necessary to effectively sample the GUI’s event space. Ideally, for increasing the efficiency and effectiveness of GUI testing, one would like to sample the GUI’s event space by only generating sequences that (1) are allowed by the GUI’s structure, and (2) chain together only those events that have data dependencies between their event handlers. We propose a new model, called an eventdependency graph (EDG) of the GUI that captures data dependencies between the code of event handlers. We develop a mapping between an EDG and an existing black-box model of the GUI’s structure, called an event-flow graph (EFG). We automate the EDG construction in a tool that analyzes the bytecode of each event handler. We evaluate our “grey-box”approach using four open-source applications and compare it with the EFG approach. Our results show that using the EDG reduces the number of event sequences with respect to the EFG, while still achieving at least the same coverage. Furthermore, we are able to detect 2 new bugs in the subject applications.