摘要
所有基于Plan的自动程序理解工具都采用把源码段跟标准Plan库中标准Plan相比较的方法来得到源码语义信息,从而达到程序理解的目标。以程序段的特征信息为索引项,对标准Plan库进行穷尽式快速粗匹配,以克服自下而上程序理解模式搜索强度大,实际应用性不强的弱点,达到快速粗粒度匹配的目标。方法的局限是匹配过程中存在信息遗漏,需要进一步采用模糊推理机制对潜在Plan进行相似性度量。
All the automatic Plan-based instruments in program understanding acquire relevant semantic information by comparing the source codes with the standard plans in pre-constructed plan database to reach the goal Of program understanding. A rough-match method,taking characteristic symbol（CS） as the index,is proposed to match standard Plan database exhaustively. Rough-match makes bottom-up program understanding model operable, though it also leads tO information leak, which is solved by the application of fuzzy-reasoning to the potential plan set.
出处
《计算机应用与软件》
CSCD
北大核心
2007年第12期91-93,共3页
Computer Applications and Software
关键词
程序理解
特征符号
穷尽式搜索
粗匹配
Program understanding Characteristic symbol Exhaustive search Rough-match
