摘要： 针对软件规范输入域与被测软件输入域之间边界不一致引起的测试用例失效问题,提出一种基于时序逻辑语言XYZ/E软件规范说明的测试用例自动生成方法。该方法采用XYZ/E描述,首先确保规范与被测软件的语义一致,其次,利用XYZ/E规范的逻辑程序属性,实现了对测试用例中输出期望的Prolog自动推导求值。最后,通过一个实例验证了该方法的可行性。与传统方法相比,该方法既保证了软件测试用例的有效性,又减少了测试用例输出的计算工作量。

关键词: 测试用例生成, 软件规范, XYZ/E, Prolog, 边界值测试

Abstract: To solve the test case ineffective problem due to inconsistency between the input domain boundaries of the specification and the software under test,a test case generation method based on software specification,which is described in temporal logic language XYZ/E,is proposed.The method,which describes software specification in XYZ/E,can assure semantic consistency between specification and software.On the other hand,the output expectation of test cases can be automatically evaluated from the specification as a Prolog program.The method can assure test case effectiveness and reduce the effort on computing test case outputs.Furthermore,whole procedure of the method is illustrated in a case study.

中图分类号: 

• TP311.52