Enhancing Automated Program Repair with Solution Design

TL;DR

This paper introduces DRCodePilot, a novel approach that leverages solution design rationales from issue logs to significantly improve automated program repair using GPT-4, demonstrating substantial performance gains on a new benchmark.

Contribution

The paper proposes DRCodePilot, integrating design rationales into GPT-4-based APR and implementing a feedback framework, advancing the effectiveness of automated bug fixing.

Findings

DRCodePilot achieves 4.7x higher full-match ratio than GPT-4 alone.

Incorporating design rationales improves repair accuracy across multiple GPT models.

The approach enhances CodeBLEU scores, indicating better code quality.

Abstract

Automatic Program Repair (APR) endeavors to autonomously rectify issues within specific projects, which generally encompasses three categories of tasks: bug resolution, new feature development, and feature enhancement. Despite extensive research proposing various methodologies, their efficacy in addressing real issues remains unsatisfactory. It's worth noting that, typically, engineers have design rationales (DR) on solution-planed solutions and a set of underlying reasons-before they start patching code. In open-source projects, these DRs are frequently captured in issue logs through project management tools like Jira. This raises a compelling question: How can we leverage DR scattered across the issue logs to efficiently enhance APR? To investigate this premise, we introduce DRCodePilot, an approach designed to augment GPT-4-Turbo's APR capabilities by incorporating DR into the prompt instruction. Furthermore, given GPT-4's constraints in fully grasping the broader project context and occasional shortcomings in generating precise identifiers, we have devised a feedback-based self-reflective framework, in which we prompt GPT-4 to reconsider and refine its outputs by referencing a provided patch and suggested identifiers. We have established a benchmark comprising 938 issue-patch pairs sourced from two open-source repositories hosted on GitHub and Jira. Our experimental results are impressive: DRCodePilot achieves a full-match ratio that is a remarkable 4.7x higher than when GPT-4 is utilized directly. Additionally, the CodeBLEU scores also exhibit promising enhancements. Moreover, our findings reveal that the standalone application of DR can yield promising increase in the full-match ratio across CodeLlama, GPT-3.5, and GPT-4 within our benchmark suite. We believe that our DRCodePilot initiative heralds a novel human-in-the-loop avenue for advancing the field of APR.