Generating Testing Sequences for UML Statecharts  

作　　者：MIAO Li ZHANG Dafang 

机构地区：[1]College of Computer and Communication, Hunan University, Changsha 410082, China

出　　处：《Chinese Journal of Electronics》2006年第2期225-230,共6页电子学报（英文版）

基　　金：This work is supported by the National Natural Science Foundation of China （No.60473031, No.60273070）.

摘　　要：UML Statechart has become a popular notion for specifying complex reactive systems, especially for software systems. Due to the complex and informal semantics of Statecharts, testing Statecharts is a challenging problem. Compared with classical Statecharts, Statecharts can receive and generate events in one step, but the generated events for the UML Statechart semantics in one step may not be the input of the next. Instead, these events may be entered into an event queue in a random order until they will be dispatched in the future. The nondeterminism of the event queue induces the nondeterminism of execution paths of the UML Statecharts. The nondeterminism is the source of one of difficulties when we test and verify a UML Statechart model. In this paper, we address the problem of generating executable testing sequences from UML Statecharts, first, give a formal description of UML Statechart semantics, which is the base of further analysis; second, present an improved algorithm to transform an UML Statechart to a CSA （Compact semantic automata）, which can describe the behavior of the UML Statechart completely and explicitly. There exist many methods for generating testing sequences from compact semantic automata since the CSAs can be treated as finite state machines.

关 键 词：STATECHART Software testing Generate testing sequences Specification-based testing. 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]