A Test Automation Language for Behavioral Models

Model-based testers design tests in terms of models, such as paths in graphs. Abstract tests cannot be run directly because they use names and events that exist in the model, but not the implementation. Testers usually solve this mapping problem by hand. Model elements often appear in many abstract tests, so testers write the same redundant code many times. This is time-consuming, labor-intensive, and error-prone. This paper presents a language to automate the creation of mappings from abstract tests to concrete tests. Three issues are addressed: (1) creating mappings and generating test values, (2) transforming graphs and using coverage criteria to generate test paths, and (3) solving constraints and generating concrete test. The paper also presents results from an empirical comparison of testers using the mapping language with manual mapping on 17 open source and example programs. We found that the automated test generation method took a fraction of the time the manual method took, and the manual tests contained numerous errors in which concrete tests did not match their abstract tests.