| With the continuous popularization of information technology,people cannot do without the assistance of computer software in their daily lives.And software testing is an important means to ensure software quality,and perfect software testing can detect most of the software failures.But with the increasing scale and complexity of computer software nowadays,it is a very time-consuming and labor-intensive activity to conduct a comprehensive test on it.A lot of experience and practice of predecessors have shown that combined testing is an effective testing method.It can detect the factors and factors in the software system by generating fewer test cases under the premise of ensuring the failure detection rate.The impact of the existing interaction on the software system.Under the premise of the given coverage requirements of the software system under test,generating a set of test cases as small as possible for the specific system under test is the current research direction of combinatorial testing.Based on this,this article first briefly describes the relevant research status and basic concepts in the field of combined testing,and then classifies and specifically introduces the current combined test case generation technology and research status used in academia and industry,and conducts a comparative study.The advantages and disadvantages of each method are summarized.With its good scalability,the IPO algorithm can deal with complex and changeable systems under test,thereby obtaining a better set of test cases.But at the same time,it also has its own inherent shortcomings,such as the problem of initial matrix construction,parameter expansion and value selection;in addition,because the IPO algorithm is also one of the traditional fixedstrength combination testing methods,it also has a fixed-strength combination test.The common problem contained in the method is that once the combined coverage strength is determined,it cannot be changed,which leads to problems such as redundant test case sets or poor error detection capabilities.Regarding the above-mentioned problems,this paper analyzes and summarizes many combination testing algorithms nowadays,and proposes an IPO＿I algorithm based on the original IPO algorithm idea.While inheriting the advantages of strong scalability of the original algorithm,the algorithm is based on the interaction relationship existing in the actual system under test,and provides a basis for the initial matrix construction of the original algorithm,the expansion of uncovered parameters and the value step,so that the algorithm can obtain A better set of test cases.Finally,this paper implements a test tool based on this method,and proves the effectiveness of the IPO＿I algorithm in the experiment. |
