Testers, how to prepare test data

In daily testing work , Will you encounter similar problems ？

1、 For example, the page data is not enough , The page turning function cannot be tested

2、 Some functions of the page have not found the data that meets the requirements for the time being

3、 When doing data analysis , It takes a lot of data , However, the amount of data in the existing environment can not meet the demand

......

How to solve the problem of data ？

1）、 be based on GUI Construct test data

benefits ：

1、 It's not just data , It's essentially an end-to-end test

2、 There are not too many skill requirements , Familiar with the page , It's OK to operate on the page

shortcoming ：

1、 Creating data is inefficient

2、 Creating data depends on too much （ Depends on the background interface 、 Test environment and development support ）

3、 be based on GUI The method of creating test data is not suitable to be encapsulated as a test data tool  

4、 The cost of making data is high , Poor stability , Too dependent

2） adopt api call （python Of faker library ,jmeter Methods such as ）

advantage ：

1、 The data generated is reliable

2、 Efficient

3、 The script that constructs the data can be changed into an interface case

4、 Can skip the front desk , Independent of the front end

shortcoming ：

1、 The cost of learning

2、 You need to sort out the interfaces

3、 There are individual businesses , You need to use some parameters other than the interface （ Some data may be transferred to the interface after front-end processing ）, It's very troublesome to deal with

3) Through the database (sql) Generate test data   

advantage ：

1、 High efficiency

shortcoming

1、 It's very difficult to organize the relationship of database , Sort out all the corresponding information of a business sql It's not easy.

4） Based on the existing online business data, after desensitization, it is imported into the test environment

When it comes to big data testing , A lot of data is often needed to validate logic , At this time, we can consider desensitizing the online data , Synchronize to the database of the test environment , At the same time, it can avoid that the data constructed by itself is not comprehensive , Cause missing test .

5) Comprehensive use API And database

be based on GUI Operation generating test data is the most primitive method , But it's inefficient , And introduce unnecessary

rely on ; adopt API Call and database operation to generate test data is the current mainstream practice , adopt API transfer

This method has the characteristics of high data accuracy but low creation efficiency , But through the database way has the creation efficiency to be high

But the maintenance complexity is also high .