Research On Abstract Memory Model Optimization Technology In Unit Testing

Automated test case generation is one of the key points of automated unit testing.In path-oriented automated unit testing,the abstract memory model is used to accurately extract the semantics and constraints of variables on the path,which is a prerequisite for automatically generating test cases.However,the existing abstract memory model still has shortcomings,such as incomplete type support,difficulty in handling complex operations accurately,and incomplete logging information,which fails to generate test cases for some selected paths or generates incorrect test cases,reducing test efficiency and effectiveness.This paper focuses on improving the accuracy of abstract memory model and the completeness of support.Based on the existing abstract memory model.The paper proposes an optimized abstract memory model supporting the expression of numeric,pointer,structure,array,union.And the constraint extraction algorithm and the non-numeric test case shape construction algorithm are improved,which can more accurately simulate the semantics of complex operations and extract constraints.Based on the optimized abstract memory,the existing two types of problems are solved,which provides more accurate and comprehensive information for automatic test case generation.The above abstract memory model has been implemented by the author in the Code Test System(CTS)and tested.The results show that the optimized abstract memory model proposed in this paper is more general than the original abstract memory model.Especially in complex types of programs,test cases can be generated normally,and the test coverage is significantly improved.