Regaining Lost Knowledge through Dynamic Analysis and Aspect Orientation

This paper describes our experiences of applying dynamic analysis solutions on an industrial legacy application written in C, with the help of Aspect Orientation (AO). We use a number of dynamic analysis techniques that can help in alleviating the problem of (1) establishing the quality of the available regression test and (2) regaining lost knowledge of the application. We also show why our aspect language for C, aspicere, is well-suited for using dynamic analysis in legacy environments. Finally, we present the case study itself, the results we have obtained and the validation thereof by the original developers and current maintainers of the application. We also mention some typical pitfalls that we encountered while dealing with legacy applications in a reengineering context.

• Type: Conference
• Year: 2006
• Where: CSMR
• Authors: Andy Zaidman, Serge Demeyer, Bram Adams, Kris De Schutter, Ghislain Hoffman, Bernard De Ruyck
• Keywords: CSMR 2006, Dynamic Analysis, Dynamic Analysis Solutions, Legacy Applications, Software Engineering