Research On Defect Location Technology Of Adaptive Combination Test Input Domain Based On Error Location Table

Combinatorial testing aims to detect potential failures caused by the interaction of various factors affecting a software system.When performing composite testing,the usual steps are to first generate a set of test cases to cover each interaction that could cause a failure,and then identify the interaction that caused the failure after the failure is detected.Although this method of execution is straightforward,studies have shown that it is not ideal in practice.This is because in the test case generation stage,testers hope that the test case set can not only comprehensively detect the interaction between some key parameters in the component,but also generate as few test case sets as possible;People want to be able to determine the root cause of a software failure before all test cases are generated and executed,because early identification of the failure-causing interaction can guide the remaining test case generation,avoiding the generation of many unnecessary and invalid test cases.Aiming at the above goals,this paper conducts research from two aspects of test case generation and defect location technology.The specific research results are as follows:First,the paper proposes a method for generating test cases with variable intensity based on the in-parameter-order(IPO)algorithm.Using 20 Boolean reduction expressions and their variants in the Boolean reduction set as a single-level system and 5 programs in the Siemens program set as a mixed level system,the two strategies in the IPO algorithm are given specific definitions,and then the optimal solution is obtained through analysis and comparison.Strategy combination,the IPO algorithm under this strategy combination is applied to the generation of variable strength coverage table,and it has better effect than other variable strength coverage table generation algorithms in the number of required test case sets and execution time.The results confirm that the variable strength overlay table is superior to the fixed strength overlay table in terms of error detection ability.Second,the paper proposes an adaptive defect localization method based on the error localization table ELA,which allows the generation and recognition processes to interact with each other.Therefore,both the generation and recognition phases will be done more effectively and efficiently.Using 20 Boolean reduction expressions in the Boolean reduction set and expressions with safe values ??in their variants as experimental objects,a series of empirical studies were carried out.The evaluation metrics are used as evaluation metrics,and the results show that this method can locate the fault-causing interaction faster than traditional non-adaptive methods,while requiring less overall cost than adaptive methods based on the FIC algorithm.