Font Size: a A A

Research On Test Case Generation Algorithm Based On Sequence Diagram And Construction Of Automated Test Platform

Posted on:2019-04-17Degree:MasterType:Thesis
Country:ChinaCandidate:Y M LiFull Text:PDF
GTID:2428330545452132Subject:Software engineering
Abstract/Summary:PDF Full Text Request
With the construction and development of modern information society,various application software and system software have played an irreplaceable role in various fields.While various software systems have been applied to all walks of life,its existence has also become more and more diverse,complicated,and large-scale.The quality of software is the focus that people have always paid attention to since the birth of software.In today's modernization,people's attention and requirements for software quality have also increased.Software testing as an important means of ensuring software quality has also attracted more and more scholars to invest in software testing.Automated testing has also evolved from simple recording playback to keyword-driven,data-driven.The more in-depth research on the traditional automated testing process,the transition from more manual testing to automated testing,is of great significance for improving test automation and testing efficiency.In the traditional automated testing process,automated processes related to test cases are performed only by automated use cases.At present,most of the test cases in enterprise applications are still manually designed and rely heavily on the testing experience of testers.Test cases are the basis of testing.Increasing test case coverage and reducing test case sets are very important for improving the efficiency of software automated testing.The purpose of this paper is to develop a solution to the automatic generation of test cases based on UML timing diagrams by studying the automatic generation of test cases.The purpose is to increase the test case coverage and reduce the purpose of test case sets.analysis.Based on this,the algorithm is applied to the test case generation module of the automated test system.This paper takes test case generation as the core,starting from the semi-formal definition of UML sequence diagram,studying the rule,conversion process and naming convention of UML sequence diagram,and getting the basic data model generated by the test case.Then compare the existing test case generation algorithm,comprehensive analysis of the applicable scene,complexity and algorithm efficiency,select the appropriate algorithm to generate test cases,and study the improvement method of the selected algorithm.Secondly,the characteristics of the system and the characteristics of the system-related roles are analyzed.Based on the system use cases and the use case descriptions,the detailed requirements of the system's functionality and non-functionality are described.The automated test system mainly includes a test case automatic generation module,a test case resolution module,a test script execution module,a rights management module,and a test report generation module.Then,according to the design of the technical scheme of this paper,genetic ant colony algorithm is improved and the feasibility of the algorithm is proved by experiments.In the outline design section,the system is outlined using a system architecture,functional module partitioning,a conceptual model of the database,and a physical model.With the help of the flow chart and class diagram of each module,the detailed design and implementation details of the system are described.Finally,the automated testing process of the automated test system was demonstrated in connection with the triangle program,and the future development of the project is forecasted.After the system is implemented,the software test automation is improved,and the software quality problems caused by incomplete design coverage,low test execution,and low test quality are reduced,which indirectly saves the time and effort of the software testers in the test work.
Keywords/Search Tags:UML sequence diagram, Automatically generate test cases, Combination of Genetic algorithm and Ant colony, Automated test
PDF Full Text Request
Related items