依托开源软件资源的移植型自动修复技术研究

随着软件行业不断走向成熟，软件资源极大丰富，从开源软件资源中挖掘补丁并通过移植的方式实现软件缺陷修复的机会大大增加。项目拟借鉴器官移植的思想，重点研究移植型代码块配型技术、基于依存度的移植代码块剪切方法、移植代码块在目标软件中的适应性改造技术、移植代码块的干扰效应分析和抑制技术等等，研制基于开源软件资源的移植型软件缺陷自动修复工具，并通过麒麟操作系统缺陷和CVE、NVD公开漏洞的自动修复试验对移植型自动修复方法进行演示验证。

The classes and amount of software repositories become rich and abundant with the maturing of software industry. The probability to repair software bugs through transplantation of patches mining from open-source software repositories are largely increased. This project intends to adopt the idea of the organ transplantation, and focuses on the matching technique of transplant code, the extraction method of the transplant code based on the degree of dependence, the adaptation technique for the transplant code in the target software, and analysis of interference effect and inhibition of transplant code, and etc. An automatic repair tool through transplantation from open-source software repositories will be developed. The effectiveness of these achievements will be illustrated by transplant experiments of automated repairing of bugs in the Kylin operating system and public vulnerabilities in CVE or NVD.