Execution base path evolution generation method based on statistical analysis

Abstract

The invention discloses an executable base path evolution generation method based on statistical analysis, in order to make the generated test path can be applied to any path test technology, so as to improve the efficiency of software testing. First, the statistical analysis method is used to design and analyze the model to determine the control flow graph, DD-graph and mutex sides of the tested program. Then, a variable length genetic algorithm based on statistical analysis is proposed. The method is used for designing the executable path generation model and a basic path detection model that evolves to generate an executable base path. The prior method is to generate the test path and detect the executable path separately. The invention designs a two-part content in one way. The basic path of the evolution is itself an executable path and overcomes the loop problem in the control flow graph , and can greatly improve the efficiency of the test path generation.

Description

technical field [0001] The invention relates to the field of computer software testing, and designs a new evolutionary method for automatically generating executable basic paths based on statistical analysis technology. This method is different from the original method in that the generated test paths are executable independent paths satisfying linear independence, that is, executable basic paths. Background technique [0002] Software testing is an important part of the software life cycle. Structural testing is the main method of software testing. Path testing is the most effective coverage criterion in structural testing. It needs to execute the test path of the program under test. Therefore, automatic test path generation is the key task of software testing, and it is an effective method to simplify the software testing process, which can reduce the cost of software testing and improve the testing efficiency. [0003] Basic path testing is one of the most effective test...

AI Technical Summary

Problems solved by technology

J. Yan et al. proposed a finite set F method for generating executable paths, F satisfies the basic path coverage criterion, but this method needs to execute all paths when detecting executable paths, which is very time-consuming

Z. Zhonglin et al. and D. Qingfeng et al. used cyclomatic complexity to generate linearly independent path sets. Since the variables contained in decision nodes have data dependencies, most of the generated basic path sets are non-executable paths. For this reason, the baseline method combined with dependencies Avoid choosing non-executable paths, but the method does not handle looping constructs in the program

Ahmed S. Ghiduk proposed a variable-length genetic algorithm to generate a test path set. The length of the chromosome increases gradually with the number of iterations to represent paths of different lengths. The path set generated by this method contains all test paths. Contains many non-executable paths, in order to delete non-executable paths, this method uses non-executable path detection model to detect all generated test paths, this process is very time-consuming

Images