A Constraint-Based Framework for Test Case Generation in Method-Level Black-Box Unit Testing

張智光(Chi-Kuang Chang) ； 林迺衛(Nai-Wei Lin)

《Journal of Information Science and Engineering》 32卷2期 (2016/03) Pp. 365-387

https://doi.org/10.6688/JISE.2016.32.2.7

並列摘要

Automatic test case execution in test-driven development provides an excellent return on investment. However, test cases in test-driven development are usually designed manually. Manual acquisition of test cases is laborious, time-consuming, and error-prone. Model-based testing is a technique to automatically generate test cases from software models. Model-based test-driven development provides an opportunity to automate both test case generation and test case execution. This paper proposes and implements a constraint- based framework for automatic test case generation in method-level black-box unit testing. This framework uniformly solved the test case generation problem using constraint logic graphs and constraint logic programming. This framework effectively performs equivalence class partitioning and test coverage criteria management on constraint logic graphs, and simultaneously generates test input and expected output using constraint logic programming. This unifying constraint-based framework can serve as a nucleus for test case generation in model-based unit testing in the future, including method-level black-box, method-level white-box, and class-level unit testing.

並列關鍵字

constraint-based testing ； black-box testing ； unit testing ； method-level unit testing ； constraint satisfaction problem ； constraint logic programming