Software test method and system

Abstract

The present invention provides a software test method, including: obtaining a first configuration file, wherein the first configuration file is set with instrumentations and the corresponding instrumentation codes for generating test scenarios; according to the first configuration file, inserting the corresponding instrumentation codes at the instrumentations in the testing software and generating the instrumented testing software; running the instrumented testing software, and when the instrumentations in the instrumented testing software are triggered, executing the instrumentation codes corresponding to the triggered instrumentations. The software test method can simulate a variety of abnormal test scenarios and save the time cost of testing. The present invention also provides a software test system.

Description

technical field [0001] The invention relates to the technical field of software testing, in particular to a software testing method and system. Background technique [0002] In order to ensure the stability of the software before the software is released, it is necessary to test the software. In software testing, in order to ensure software quality, it is necessary to perform coverage tests on various scenarios that the software may face, including normal and abnormal scenarios, especially some abnormal scenarios, such as: insufficient machine memory, insufficient disk space, local file Read and write errors and abnormal data, etc. In traditional technology, two schemes are usually adopted to implement software testing, namely, function-based black-box testing method and code-based white-box testing method. [0003] The function-based black-box testing method, if you want to cover various abnormal scenarios, for example, for the abnormal scenario where the machine memory i...

AI Technical Summary

Problems solved by technology

Therefore, the black-box testing method based on function is difficult to implement, and requires a certain test time cost

Moreover, for some abnormal branches inside the program, such as the abnormal return of a function, the function-based black-box testing method is more difficult to simulate

[0004] Although the code-based white-box testing method can set breakpoints when the program is running, and simulate such abnormal scenarios by manually modifying the program runtime data or temporarily modifying the code, the test time cost of this method is also very high, especially For the simulation of some specific scenarios involving many code module modifications, for example, the specific scenario where all new function calls return NULL, the time cost of this testing method will be even higher

[0005] Therefore, traditional software testing methods have the technical problems that abnormal scenarios are difficult to simulate and the cost of testing time is high.

Images