一种针对交互式软件的测试用例生成模型研究与实现

测试用例的自动生成是软件测试的一个重要方面，测试用例生成的复杂度与质量直接影响测试的效率和覆盖面，已有的针对软件功能测试的自动测试理论和方法在针对CAD和图像处理这类人机交互复杂、系统状态繁多的系统时，因复杂度大或者测试效率较低而难以在实际中应用．本文提出的一种针对复杂人机交互软件的测试模型一基于功能对象的测试用例生成模型，将用户使用被测试系统的操作流程看成是若干个处理过程的组合，采用面向对象的思想将用户的操作流程进行分解和封装，构建出用以生成测试流程的功能对象，并依照一定约束条件对功能对象的实例进行重组，派生出测试用例．本文提出的测试用例生成模型，已在三维数字化设计系统的专用测试平台CADTesting中成功实现，实践证明，该模型与现有的一些模型相比，实现复杂度低且测试用例的质量较高．

Research on Model of Automated Test Case Generation for Complicated Interactive Software

The automated test case generation is an important aspect of the software testing. The quality of test case directly determines the efficiency and overlay of the testing. While faced with systems such as CAD system or Image Processing system which involve complicated human-machine interactions, it is difficult to generate test case with the existing automated test theory and models. This paper proposed a software testing model which analyzes the user's operating stream and encapsulate it into functional objects with object-oriented theory. Then recombine the functional objects according to certain restrictions to generate test cases. The testing model has been successfully applied to theCAD Testing a special testing platform for CAD system. The experiment results show that the complexity of this new model is lower than the existing test models and the quality of the test case generated is higher.