Computing ripple effect for web applications. (c2007)

Abstract

The number of internet web applications is rapidly increasing in a variety of fields and not much work has been done for ensuring their quality, especially after modification. Modifying any part of a web application may affect other parts. If the stability of a web application is poor, then the impact of modification will be costly in terms of maintenance and testing. Ripple effect is a measure of the structural stability of source code upon changing a part of the code. It provides an assessment of how much a local modification in the web application may affect other parts. No previous work has been published on computing the ripple effect for web application. In this thesis, we propose, for the first time, a technique for computing ripple effect in web applications. This technique is based on direct-change impact analysis and dependence analysis for web applications developed in the .Net environment. Dependences include control dependences, data dependences, call dependences, semantic dependence, event-based dependences, functional dependences, and class dependences. Also, a complexity metric is proposed to be included in computing the ripple effect. A case study is used to show the applicability of our technique.