Stereocode: A Tool for Automatic Identification of Method and Class Stereotypes for Software Systems

Stereocode: A Tool for Automatic Identification of Method and Class Stereotypes for Software Systems

연구 분야: Verification

학회: 2024 IEEE International Conference on Software Maintenance and Evolution (ICSME)

We present Stereocode, a static analysis tool engineered to automatically identify, and re-document software systems written in C++, C#, and/or Java with method and class stereotypes. A stereotype is a simple abstraction that encapsulates the high-level behavior of a method or a class. The tool is built around the srcML infrastructure, an XML representation of source code. Stereocode annotates the srcML input with the computed stereotypes as XML attributes to the function and class tags. We showcase Stereocode's efficiency in conducting large-scale analysis of software systems, which involves using 1050 repositories from GitHub across C++, C#, and Java. The results provide valuable insights into the distribution of stereotypes. A demo video is available at: https://youtu.be/D90xwUIPbOI.