Calculating Metrics from Large C++ Programs

In this work we present a new method called compiler wrapping for extracting information from the source code of large software systems written in the C++ language. This new method can be used without having to modify the analyzed source code in any way. With the extracted information we can calculate different object oriented metrics and characterize the analyzed system. For source code analysis and metrics calculation we employed the Columbus reverse engineering framework. To demonstrate the operability of our new approach we tested it on the open source internet suite Mozilla and found it very effective in obtaining the desired metrics.