Method and device for generating software test data

Abstract

The invention discloses a method and device for generating software test data, comprising: automatically analyzing the interface document of each interface of the server, generating and saving interface request data of a standard structure; parameterizing each of the interface request data; A business process including multiple associated interfaces is encapsulated into independent data generation rules, and each data generation rule is defined through a label; the data generation rule corresponding to the label is invoked through the label to generate the required business data. The method in the present invention can make the operation visualized and the data sample controllable. In a software system with complex business logic, deep path and high coupling, the tester does not need to understand the complex pre-business logic of the current test object, but only needs to The required test data can be generated by entering the corresponding tags and attributes, and the data link is complete, with high accuracy and high automatic generation efficiency.

Description

Technical field [0001] The present invention relates to preparation and maintenance of data, and more particularly to a software test data generation method and apparatus. Background technique [0002] Now most products, especially Internet products, are now using databases to save data. Each function of the product also relies on the corresponding business data, so in software testing, the preparation and maintenance of test data is a very important aspect. Especially for software systems with complex service logic, high path, high coupling, when the test object is in the end of the business, it is time-consuming time-consuming. Similarly, the preparation and management of test data is also the premise of continuous integration and automation tests. Otherwise, due to business restrictions, it will cause automation cases that cannot be repeatedly executed, and it has lost its meaning. [0003] The way to test data generation can be roughly divided into: generate directly using SQ...

AI Technical Summary

Problems solved by technology

All of these have their advantages and disadvantages. SQL generation requires testers to fully understand the business code and data structure, which is equivalent to reimplementing the business method with SQL, and the operation cost is high; the Docker container cannot provide enough sample data, because the data of the container itself needs to be prepared , the production traffic playback needs to keep the local test and production environment exactly the same, and the test data generation is mostly to serve the test of new requirements

The method of calling the business interface generation is the most reliable way, but because the business interface is independent, when the business scenario path is deep, a business scenario will cover a large number of interfaces, making manual maintenance difficult, and interface input parameters also need to be manually modified. Data generation efficiency is no different from manual operation

Images