A Tool for Detecting and Grading Code Smell in Programming Assignments

• Main Authors: Kai-Lin Yang, 楊凱霖
• Other Authors: Chien-Hung Liu
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/43yvc6

碩士 === 國立臺北科技大學 === 資訊工程系研究所 === 104 === When developing software, in addition to function properly, the upgrade software quality is also very important issue. In order to improve the quality of the student code, Import Professor Wei-Kai Chen NTUT teaching methods to eliminate bad smell in 2011, and used in Windows programming courses. The experiment confirmed that this teaching method, students can reduce the bad smell density, improve code quality. But this assistants teaching methods, the need to manually modify the courses developed 13 kinds of bad smell rules, time-consuming, and students must complete in order to correct that and other assistants work in bad smell, you can not get instant feedback. To solve this problem, this paper presents an automated detection and correcting bad smell tool called Dr.Smell, this tool can automatically detect bad smell in most courses developed, and can assist in correcting students bad ta smell. We see this tool used in 2015 on Windows programming courses, and student statistics every assignment bad smell, as well as through the questionnaire and interviews, to explore whether the student code quality improvement, tools are easy to use, and whether the tool reduce the burden of teaching assistants and the like. The results show that after the course added Dr.Smell, students work in bad smell decline in the density of the process, the display student code quality has improved. Survey data show that students generally agreed that after using Dr.Smell, can enhance the quality of code. And assistant after use Dr.Smell, these duties can significantly reduce the time of revising assigments.