Adaptive random testing approach based on coverage for object oriented software
1. Informatization Department(Center), Jiangsu University, Zhenjiang, Jiangsu 212013, China; 2. School of Computer Science and Communication Engineering, Jiangsu University, Zhenjiang, Jiangsu 212013, China
Abstract：To improve the effectiveness of selecting test cases and the failure detection effectiveness in current adaptive random testing (ART) for object oriented software (OOS), an adaptive random testing approach was proposed based on coverage for object oriented software. In the proposed approach, the code coverage information was considered when the candidate test cases were generated. The information of code coverage and the test case dissimilarity were combined to select the executed test cases, and the failure detection effectiveness was improved. The overall framework of coverage dynamic ranking monitoring technology, the pretreatment module, the program instrumentation module and the ordering module were proposed. The dynamic test case generation for object oriented software based on coverage (DTCGOOSCov) algorithm was also described. The proposed approach was compared with the traditional adaptive random testing method and the pure random testing method. The comparison experiments were conducted on six class libraries under the tests of ID3Manage, RabbitAndFox, SATM, SchoolManagement, WaveletLibrary and WindShieldWiper. The results show that the adaptive random testing method based on coverage for OOS outperforms the other two methods on Fm and Em metrics.
2019, Vol. 40 
