| With the scale of software system increasing,software development is becoming more and more complex.Historically,large software failure occurred frequently,how to improve the quality of the software becomes a new research focus.As the policy of design and guiding for software development,software architecture has played an irreplaceable role in the development of software.Research shows that we can improve the quality of the software by means of testing and formal.While the software architecture-level testing can detect the structure of software in the early stage of the development cycle,which is unmatched by traditional testing.The current research on software architecture testing starts just now,only to find the effective testing techniques,which can apply to the software architecture,the application can make software development successful.This paper is mainly to study on the shortest path search of C2 software architecture and we design and implement a genetic algorithm,in which the length of chromosome is variable.In the experiment,the feasible solution can be indicated with binary code,then use the improved fitness proportional selection to copy the adaptability of the individual.According to the preset cutting and splicing probability,operate randomly selected individuals to generate new individuals.The crossover operation in the traditional genetic algorithm is replaced by cutting and splicing operation.Apply the mutation operation in the process,and then decode the final population as the solution.Compare the experimental results of two kinds of algorithm,explain that genetic algorithm used to generate testing paths of C2 architecture is feasible.Finally,analysis the coverage of testing paths,and pointe out the next step of the research project.The purpose of this paper is to design and implementation a testing tool to generate testing paths for C2 software architecture,so that testing paths can be used in integration testing. |
PDF Full Text Request