Integrating Retrieval-Augmented Generation for Enhanced Code Reuse: A Comprehensive Framework for Efficient Software Development

With the rapid development of ubiquitous computing, the demand for efficient software development is growing stronger. Code reuse is an effective means to enhance software development efficiency, significantly reducing the development cycle. Although existing large language models for code generation can assist developers in quickly writing code, they primarily aid in generating generic code and are unable to produce specific business code, making code reuse indispensable. In this paper, We propose a retrieval-augmented generation-based framework to streamline code reuse. This involves extracting reusable code units, generating natural language summaries, and matching queries with these summaries using search and reranking techniques. We conducted extensive experiments from both code search and code generation perspectives, the experimental results demonstrate that our approach significantly improves the efficiency and quality of code reuse.