Micro-service automatic testing method based on Redis database

Abstract

The invention relates to a micro-service automatic testing method based on a Redis database. The method comprises the steps of A, setting a Mock annotation for an AOP method needing to be called by aremote service, wherein the Mock annotation class comprises a parameter key used for obtaining data in a Redis database and a remote service name corresponding to the AOP method, and the remote service name is used for marking whether the AOP method is allowed to be tested by Mock or not; B, setting a Remote annotation class associated with the class signature, wherein the Remote annotation classis used for marking a method containing remote calling in the class; and C, setting interception classes used for intercepting all the remote calling AOP methods, and returning corresponding differentresults according to the judgment whether the called AOP method allows the Mock test or not. According to the invention, the micro-service can be subjected to dynamic Mock test, the test flexibilityis greatly improved, and the cost and difficulty of developing a test program are remarkably reduced.

Description

technical field [0001] The invention relates to a testing method of an application program component, in particular to a microservice automatic testing method based on a Redis database. Background technique [0002] In the field of Internet distributed technology, the architecture model of microservices has gradually occupied the mainstream position. After the application module is split into microservices, the dependencies and collaboration network between services become intricate. How to improve the testability of microservices with external dependencies will directly affect the cycle and quality of software delivery. [0003] In the existing microservice testing technology, the way of developing interface baffle program is often used to solve the external dependency problem. But using the baffle procedure itself has significant limitations. First of all, the baffle program itself is not flexible enough. The result returned by calling the baffle program is a static resu...

AI Technical Summary

Problems solved by technology

But using the baffle program itself has significant limitations

First of all, the bezel program itself is not flexible enough. The result returned by calling the bezel program is a static result, and it is impossible to return a specific answer according to different inputs.

Second, when the interface of the microservice under test changes, the bezel program needs to be modified and upgraded in a targeted manner. In the early stage of development, the frequency of interface changes is very high

Third, the baffle program needs to be deployed separately and takes up hardware resources

Images