A meta level dynamic approach to visualize impact analysis for regression testing

Abstract

Fixing software bugs, extending base applications with new functionalities, as well as adapting to changing environments are among the reasons for software evolution. To facilitate such a process and to help maintainers make informed decision, there is a need to be able to estimate and determine the impacts of evolution to the overall software system. While there are already quite a number of useful tools, termed impact analysis tools, developed either as research prototypes or commercial products that addresses such issues, much of which is static based and adopts traditional text based impact reporting. To address some of these issues, we propose a new change impact analysis, called JRegres, that can dynamically generate trails for impact analysis as well as support impact visualization. JRegres serves as a research vehicle to investigate the hypothesis that suggests impact visualization is useful for supporting regression testing.