An Extensible Feature-Oriented Approach for Fine-Grained Code Quality Analysis

Abstract

Assessing code quality is crucial for effective software maintenance and evolution. Traditional tools like SonarQube offer valuable insights at the application level but lack the granularity needed for detailed, feature-specific analysis. This paper emphasizes the importance of feature-oriented code quality analysis, often overlooked by mainstream tools due to the challenge of correlating high-level feature descriptions with low-level code implementations. To tackle this issue, we leverage existing feature location techniques to introduce a novel approach enabling granular analysis tailored to specific application features. We discuss the motivations for this approach, highlighting its potential to improve precision in enhancement and maintenance strategies. Additionally, this paper introduces a tool-based approach known as InsightMapper. We also present a study demonstrating the benefits of this method through the analysis of two case studies, featuring a recognized benchmark in the feature location domain.