Method for testing multi-parameter software

Abstract

The invention relates to a method for testing multi-parameter software. The method comprises the following steps of: 1) generating combined test data of which the intensity is t; 2) recognizing a non-sensitive element of the combined test data, wherein the non-sensitive element indicates that the variation of the value of the element does not affect the intensity t of the combined test data; 3) instantiating the non-sensitive element so as to achieve a possibly high coverage rate of a value combination of a t+m tuple in the combined test data after instantiation, wherein m is a natural number, can be any natural number ranging from 0 to k-t and is preferably 1; 4) taking the combined test data which increases the coverage rate of the value combination of the t+m tuple as optimized combined test data; and 5) testing software to be tested by using the optimized combined test data. On the premise of not reducing the original test intensity or increasing any test cost, the method increases the coverage rate of the test data on a high-dimension tuple, so that higher error detection ratio can be achieved with the same test cost.

Description

technical field

[0001] The invention relates to combined testing technology in the field of software engineering, in particular to a software testing method for optimizing used combined testing data. Background technique

[0002] With the rapid development of information technology, software has been widely used in various industries such as industry, service industry, finance, national defense, and people's daily life. In order to ensure the reliability of the software, it is of great significance to discover as many potential errors in the system as possible through software testing technology before the software is released.

[0003] Hidden bugs in multiparameter software are likely to be triggered by the interaction of more than two parameters. Therefore, if only the possible values ​​of each parameter are tested individually, software errors caused by the interaction of multiple parameters are likely to be missed. The ideal test plan is to exhaustively test the value ...

Images