基于数据拓扑图的变量影响分析方法  

Variable Impact Analysis Method Based on Data Topology Graph

作　　者：潘亚飞 牟永敏 PAN Ya-fei;MU Yong-min(School of Computer,Open Computer System Laboratory,Beijing Information Science and Technology University,Beijing 100101,China)

机构地区：[1]北京信息科技大学计算机学院计算机开放系统实验室,北京100101

出　　处：《科学技术与工程》2019年第2期135-143,共9页Science Technology and Engineering

基　　金：国家自然科学基金(61370129);网络文化与数字传播北京市重点实验室开放课题(ICDD2017XX);北京市自然科学基金重点研究专题项目(Z160002)资助

摘　　要：为了解决软件中数据变量发生异常的问题,找到该异常变量在整个程序中与其余变量之间的关系以及影响范围,提出了一种基于数据拓扑图的数据影响分析方法。研究了静态分析提取变量的依赖关系包括顺序依赖、自身依赖、节点依赖和函数依赖,得到变量的依赖影响集合,生成变量的数据拓扑图。实验结果表明,该方法能全面覆盖程序中的变量并准确地生成指定变量的数据拓扑图,且数据拓扑图能够发现在程序中被异常变量影响的变量。In order to solve the problem of abnormality of data variables in software,and to find the relationship between the abnormal variables in the whole program and the remaining variables and the influence range of the variables,the data impact analysis method based on data topology graph was proposed.The dependence of static analysis on extracting variables was studied,including order dependence,self dependency,node dependence and function dependence.The dependent influence set of variables was obtained,and the data topology graph of the variables was generated.The experimental results show that the method can completely cover the variables in the program and accurately generate the data topology graph of the specified variables,and the data topology graph can find out which variables are affected by the abnormal variables in the program,and which variables are affected by the abnormal variables.

关 键 词：数据拓扑图 依赖影响集 静态分析 顺序依赖 自身依赖 节点依赖 函数依赖 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]