Abstract
In software testing, one of the most important issues is to validate the consistency of program and design. Usually, this kind of design information is expressed by software related documents, such as software requirements, software design, etc. Therefore, how to effectively extract the design information based on these documents is very important. In this paper, we first introduce the method of the analyzing class diagram model and method call diagram model from UML model of design document; secondly, we classify and summarize the inconsistency problems that may exist in the two types of models; thirdly, we design static and dynamic consistency algorithms for these models. Finally, we implement the CCoSaD (consistency checking of software and Design) tool and validate the effectiveness based on experiments.
Highlights
A complete software product includes the implemented software system, and the document describing the original intention, design idea and usage method of the system [1], [2]
Because this paper focuses on consistency detection of the class diagram model and method call graph model, we will introduce them separately
1) DATA STRUCTURE DESIGN Since this paper uses Java language to implement the algorithm, we will use Java class diagram to show the class diagram model, method call graph model, problem report and other data structures used in the algorithm
Summary
A complete software product includes the implemented software system, and the document describing the original intention, design idea and usage method of the system [1], [2]. The method node elements and edge elements in the method call graph model are used to explain how to extract this information from the XMI tree. (1) Method node element: From the sequence diagram model XMI tree UML:ClassfierRole, the attribute of the node name xmi.id can obtain the class name and unique ID of the method node element UML:Message. Extraction Rule 1: for the class name and method in the meta information of class file, we can get the class name, operation name and parameter type name according to rules 1, 3-6 ∼ 9 given in [1], and the unique identification of method node can be obtained according to the following format. According to the way of rule 1, the unique method node ID is used to obtain the head node and tail node from the method node set, and the edge elements < head, tail > are obtained
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
