From Coverage Computation to Fault Localization: A Generic Framework for Domain-Specific Languages

To test a system efficiently, we need to know how good are the defined test cases and to localize detected faults in the system. Measuring test coverage can address both concerns as it is a popular metric for test quality evaluation and, at the same time, is the foundation of advanced fault localization techniques. However, for Domain-Specific Languages (DSLs), coverage metrics and associated tools are usually manually defined for each DSL representing costly, error-prone, and non-reusable work. To address this problem, we propose a generic coverage computation and fault localization framework for DSLs. Considering a test suite executed on a model conforming to a DSL, we compute a coverage matrix based on three ingredients: the DSL specification, the coverage rules, and the model’s execution trace. Using the test execution result and the computed coverage matrix, the framework calculates the suspiciousness-based ranking of the model’s elements based on existing spectrum-based techniques to help the user in localizing the model’s faults. We provide a tool atop the Eclipse GEMOC Studio and evaluate our approach using four different DSLs, with 297 test cases for 21 models in total. Results show that we can successfully create meaningful coverage matrices for all investigated DSLs and models. The applied fault localization techniques are capable of identifying the defects injected in the models based on the provided coverage measurements, thus demonstrating the usefulness of the automatically computed measurement