Evaluation of code quality and hotspot prioritization using repository mining

Abstract

Software development is characterized by continuous changes and improvements to meet evolving requirements and address emerging issues. Software repositories contain historical and valuable information about the overall development of software systems. To proactively manage code quality and address potential challenges, this thesis presents a service-based tool that employs repository mining techniques, using the Python Framework PyDriller, to assess code quality and prioritize hotspots in GitHub repositories. This thesis proposes a tool, named CodeInspector, the proposed tool aims to assist software engineers and researchers in identifying critical areas in their codebases that require immediate attention and understand the impact of individual commits on the overall maintainability of the software system. We achieve this by analyzing complexity and churn metrics and employing the Delta Maintainability Model (DMM) to assess code changes. The thesis emphasizes the importance of code quality in software development, equipping developers and researchers with a powerful tool that empowers them to create and maintain high-quality software systems.