【摘要】 软件测试是软件工程领域中重要组成部分。随着软件规模的不断扩大,测试工作的复杂性也不断升高,而改善、提高软件易测试性则是降低测试复杂性的有效手段。为达到这一目标,首先要能对软件的易测试性进行准确的度量。这种度量结果除了作为软件度量的一个量化指标外,还应能为改善易测试性提供指导、为测试的设计提供有效帮助。本文对现有软件易测试性分析方法进行归类,将已有技术大致归并为基于复杂度分析、基于信息论分析、基于PIE技术分析和基于UML类图分析四种基本类型。简要介绍了每一类方法,对这些方法的特点与不足进行分析比较,并探讨今后的研究方向。更多还原

【Abstract】 Software testing is an important part in the domain of software engineering. With the enlargement of soft- ware scale, software testing becomes more complex. The improvement of software testability is an efficient method to reduce the testing complexity. The precise measurement of software testability is the first step to achieve the above goal. Besides as a quantified guideline, the measurement result should be direction for the improvement of software testability and be helpful to testing design. In this paper, the existing analysis methods of software testability are classified roughly as analysis based on software complexity, analysis based on information theory, analysis based on PIE technology and analysis based on UML class graph. This paper briefly introduces some representative methods of each class, compares the merits and demerits of these methods. Finally, future research directions are discussed.更多还原