Software test case automatic generating method

Abstract

The invention discloses a software test case automatic generating method, belonging to the technical field of software test automation. The method comprises the following steps: 1, in the first phase, establishing a component Petri network model on the basis of a tested object, wherein three objects, namely a structured language program, requirement or scene based description and a object oriented program are taken in the first phase); 2, in the second phase, collecting and storing fore-set and post-set of places and changes in a Petri net (PN) model and related information of initialized condition and other model structures by utilizing an appropriate data structure; 3, in the third phase, analyzing and verifying the model (deadlock, activity, boundedness, accessibility and the like) by adopting a proper algorithm according to the information collected in the second phase; 4, in the fourth phase, generating a test case; 5, in the fifth phase, analyzing the test case according to different coverage rate indexes; and 6, in the sixth phase, comprehensively analyzing according to the original requirement and quality requirement of a user so as to re-generate a test case set in accordance with the requirements.

Description

technical field [0001] The invention relates to a method for automatically generating software test cases based on Petri net, which belongs to the technical field of software test automation. Background technique [0002] Software testing is an important means to ensure the correctness of the software system. It finds errors in the software by running selected test cases, and makes the quality of the software meet the requirements by correcting the errors. The key problem of software testing is to construct a reasonable and limited set of test cases, which can cover the problem space of the software under test as much as possible. The generation of software test data occupies a large proportion in the time and cost of software system development. However, the manual selection method mainly used at present has a large workload, a long test cycle, is prone to omissions, and has low utility. Therefore, if there is a way to make this selection process It would be of great value...

AI Technical Summary

Problems solved by technology

[0008] (3) The collection of structural information of the built model has not established specific rules and standards, which is not conducive to the analysis and verification of the model;

[0009] (4) There is no clear evaluation mechanism for the built model, that is, there is no way to evaluate the rationality of the model itself, which directly affects the effectiveness of test cases;

[0010] (5) There are too many steps of human participation, and more relevant professional knowledge is required to implement the corresponding method;

[0011] (6) The obtained test cases do not have a relatively complete evaluation mechanism

Under the new model, the test case coverage analysis mostly follows the traditional method, and the indicators are relatively single, which cannot fully reflect the quality of the software;

[0012] (7) The test case management mechanism is lacking, and the revised content based on the performance analysis of the test cases is not reflected in the test case collection in time, and the effect of gradual optimization cannot be achieved

Images