首页 | 本学科首页   官方微博 | 高级检索  
     检索      

软件配置测试的测试数据生成
引用本文:聂长海,徐宝文.软件配置测试的测试数据生成[J].东南大学学报,2004,20(1):26-30.
作者姓名:聂长海  徐宝文
作者单位:[1]东南大学计算机科学与工程系,南京210096 [2]国防科技大学计算机学院,长沙410073
基金项目:国家自然科学基金,Opening Foundation of State Key Laboratory of Software Engineering in Wuhan University,Opening Foundation of Jiangsu Key Laboratory of Computer Information Processing Technology in Soochow University
摘    要:本文提出利用多次单因素试验、正交实验设计和均匀设计等常用方法为软件的配置测试产生测试数据, 以确保待测软件可以在不同的配置上正确运行, 并分析比较了这些方法的优缺点. 在此基础上提出使用两两组合覆盖方法来改善正交试验设计方法的不足、使用多因素组合覆盖方法来降低因不完全测试而带来的风险, 并提出一种单因素覆盖方法, 该方法可以产生最少的测试数据, 在覆盖各个系统因素的同时, 尽可能多地覆盖各个因素间的两两组合. 最后, 对这些测试数据生成方法从测试数据集的规模、组合覆盖率、可用性等方面进行了比较.

关 键 词:软件测试  配置测试  测试用例  组合覆盖

Generating test cases for software configuration testing
Nie Changhai,Xu Baowen.Generating test cases for software configuration testing[J].Journal of Southeast University(English Edition),2004,20(1):26-30.
Authors:Nie Changhai  Xu Baowen
Abstract:Software configuration testing is used to test a piece of software with all kinds of hardware to ensure that it can run properly on them. This paper generates test cases for configuration testing with several common methods, such as multiple single-factor experiments, uniform design, and orthogonal experiment design used in other fields. This paper analyzes their merits and improves the orthogonal experiment design method with pairwise testing, and decrease the testing risk caused by incomplete testing with a method of multiple-factors-covering. It also presents a simple factor cover method which can cover all the factors and pairwise combinations to the greatest degree. Finally some comparisons of these methods are made on the aspects of test suite scale, coverage, and usability, etc.
Keywords:software testing  configuration testing  test case  combination cover
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号