Reusing black box test paths for white box testing of websites

Abstract

As the numbers of web users are increasing exponentially, the software complexity is increasing exponentially and the malwares are increasing exponentially, so exhaustive and extensive testing of websites has become a necessity today. But testing of a website is not 100% exhaustive as the page explosion problem is also very usual. In this paper, we propose to reuse the basis test paths as obtained from the Page-Test-Trees (PTTs) for white box testing of websites. We traverse the same set of paths (obtained above) and test for the source code at these nodes. This saves significant amount of time required to generate test paths and hence test cases as compared to the existing approaches of white box testing. The cost and efforts are also minimized. The proposed technique ensures better website testing coverage as white box testing provides better results than black box testing. Then we validate the proposed reusability testing with two web navigational structures. The results show that doing regression testing can save several billion dollars. These test cases can be further minimized by using prioritization techniques of regression testing.