(State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing 210023, P.R. China)
Abstract：Refactoring is an important practice in software development and software maintenance. Many development environments provide automated tools to assist in refactoring. Since existing refactoring engines cannot guarantee the correctness of a single refactoring process, regression testing techniques are often employed to test the code after refactoring. However, large scale regression testing is too big for small refactoring activity. Re-running regression test cases is extremely time-consuming each time when we applying refactoring. Even if time is not bounded, regression test cases associated with the program may be inadequate. To secure the process of refactoring, we have to know the consequence of refactoring and hence make the test plan. In this paper, we propose a call graph based change impact analysis technique to recommend test points to developers. This technique recommends methods to developers to help them select regression test cases. If there is no unit test associated with a method, it can also help developers decide if it is necessary to generate a new test case. This technique makes the refactoring process more transparent.Keywords: Recommend；Impact analysis；Refactoring；Regression testing

References
[1] M. Fowler and K. Beck, Refactoring: improving the design of existing code. 1999.
[2] B. Daniel, D. Dig, K. Garcia, and D. Marinov, “Automated testing of refactoring engines,” in Proceedings of the the 6th joint meeting of the European software engineering conference and the ACM SIGSOFT symposium on The foundations of software engineering - ESEC-FSE  ’07, 2007, p. 185.
[3] G. Soares, D. Cavalcanti, and R. Gheyi, “SAFEREFACTOR - Tool for Checking Refactoring Safety,” Source. 2008.
[4] G. Soares, R. Gheyi, and T. Massoni, “Automated Behavioral Testing of Refactoring Engines,” IEEE Transactions on Software Engineering, pp. 1–19, 2012.
[5] S. Max, E. Torbjörn, and de M. Oege, “Sound and extensible renaming for java,” ACM SIGPLAN Notices, vol. 43, no. 10, p. 277, Oct. 2008.
[6] M. Kim, T. Zimmermann, and N. Nagappan, “A field study of refactoring challenges and benefits,” in Proceedings of the ACM SIGSOFT 20th International Symposium on the Foundations of Software Engineering - FSE  ’12, 2012, p. 1.
[7] S. Bohner, “Software change impact analysis,” 1996.
[8] C. Pacheco and M. Ernst, “Randoop: feedback-directed random testing for Java,” OOPSLA, 2007.
[9] H. Do, S. Elbaum, and G. Rothermel, “Supporting controlled experimentation with testing techniques: An infrastructure and its potential impact,” Empirical Software Engineering, 2005.
[10] N. Rachatasumrit and M. Kim, “An empirical investigation into the impact of refactoring on regression testing,” in 2012 28th IEEE International Conference on Software Maintenance (ICSM), 2012, pp. 357–366.

作者簡介：
呂軍，男，南京大學計算機科學與技術系碩士生，主要研究領域為程序分析、重構、測試等。