A Test Case Prioritization Method Based on UML Model

Abstract

The invention relates to a test case prioritization method based on a UML model. The method comprises the following steps: considering class complexity and an inter-method dependence relation as important factors; first analyzing a CK (Chidamber and Kemerer) metric unit reflecting the class complexity according to a UML class diagram in combination with a UML sequence diagram, and then generating a class failure rate according to the CK metric unit and a software failure relation formula; generating an inter-migration data dependence and control dependence relation through a finite state automata generated by the UML sequence diagram; finally generating the importance degree of the migration by comprehensively considering the dependence relation of the migration and a class failure rate corresponding to migration, and regarding the uncovered and unmigrated total importance degree in test cases as the priority to order the test cases. By use of the method disclosed by the invention, the limitation that the traditional test case ordering technology cannot test software based on a software model due to excessive dependence on program codes is overcome, and the method is more suitable for the testing of the software developed based on a model drive mode.

Description

technical field [0001] The invention is an optimized processing method for a test case set, aiming at enabling the test cases processed by the technology to efficiently detect errors when detecting software developed by using model drive, and achieve the effect of reducing the test cost. Background technique [0002] Software testing plays an extremely important role in the software development process, but software testing is also a high-cost process in software development. Testers either manually design test cases selectively, or automatically generate test cases for a software through software testing tools, and use the test cases to test the software. Designing test cases by hand is highly subjective and difficult to guarantee test quality. Therefore, automatic generation of test cases by test case generation tools is the current development trend of software testing. But for the automatic generation of test cases, the biggest problem is how to select an excellent subs...

AI Technical Summary

Problems solved by technology

[0004] However, these two technologies still have certain limitations. For example, in the process of reducing the test case set, TCR technology may delete some test cases that could detect software vulnerabilities as redundant test cases, thus making the simplified test The use case set has a decrease in error detection ability and loses the ability to detect some potential vulnerabilities

The traditional TCP technology focuses on sorting the test cases according to a certain sorting strategy or according to the relationship between the software code and the error so that the test cases covering the error-prone area are prioritized, ignoring the relationship between the software model and the error. This makes the traditional TCP technology unable to achieve good results when applied to model-based software testing

Images