Bachelor Thesis

In my bachelor's thesis, I designed a framework for a holistic approach to managing code smells. This includes the phases of prevention, detection, preparation & interpretation, refactoring, and finally monitoring & learnings.

Based on this framework, I selected various tools for code smell management and evaluated them using a criteria catalog. The candidates examined were SonarQube, CodeScene, and Embold. The research showed that these tools cover different focal points of code smell management. While SonarQube has clear strengths in detection and integration into workflows and processes, CodeScene convinces with additional data from Git commit histories in aspects such as preparation and interpretation.

Through my research, I gained the following insights and skills:

• Deep understanding of code smells and their impact on code quality
• Practical application of code smell tools and their integration into processes / CI pipelines
• Understanding of technical debt and its management
• Knowledge in academic research and writing

Visualization of development hotspots

CodeScene provides the ability to visualize development hotspots in a project. These hotspots reveal which parts of the code are particularly complex and frequently changed. This helps identify potential problem areas in the code and work on improving them in a targeted manner.