基于软件历史仓库和抽象语法树的相似缺陷识别方法

软件开发过程中,软件开发人员常常通过搜索软件历史仓库(historical software repository, HSR),再经复制/粘贴以实现软件复用。HSR中会保存被复用的代码的缺陷及修复信息,辅助处理相似缺陷。基于此,提出一种基于HSR挖掘的相似缺陷识别方法。首先,基于变更日志的分析,从HSR中提取出已知缺陷的模块,建立bug模块库。然后,采用基于抽象语法树(abstract syntax tree, AST)的相似代码检测方法,识别待测试软件与bug模块库中相似的代码,并借助HSR中保存的相应缺陷及修复信息,完成待测试软件中可能包含潜在缺陷的模块的识别。同时,为提高相似代码的识别精度,优化基于AST的代码特征度量。在18个C程序、164对克隆代码上进行实验,结果表明所提方法能够识别出全部相似代码且性能优于已有工具。在人工构建的bug模块库上验证了代码相似性对相似缺陷识别的作用。最后,在8个真实的大型C项目上进行验证,平均缺陷召回率达到94%,表明挖掘HSR可以有效地为跨项目传播的相似代码提供缺陷理解支持。

New multiple attribute decision-making model with triangular fuzzy numbers based on connection numbers

To study the fuzzy multiple attribute decision-making problem that the attribute weights and the attribute values are both indicated in the forms of triangular fuzzy numbers,a method to solve fuzzy multi-attribute decision-making problem denoted by triangular fuzzy numbers based on connection number is presented.By applying the uncertainty systems theory in the set-pair-analysis,the median,and the interval of values prescribed by supremum and infimum of the triangular fuzzy numbers,the triangular fuzzy numbers are transformed into connection numbers,and the program of making is given.The median and the supremum and infimum of the triangular fuzzy numbers are used in the model,it is more objective.A practical example indicates the method is simple and credible.