Translation API Test Best Practices-Organize your tests

Organize your tests

Applicable level: Beginner

At the bottom, a test step is used to verify a single operation. Combining several test steps to test cases allows you to verify the functionality that is separated from one another, which is required by the application. Next, several test cases can form a test suite to verify the full functionality of one deliverable, which is what the user wants. Finally, by combining several test suites into a single test project (TestProject), you can verify the functionality of a complete product.

Word Engineering (Project) and Suite (suite) can be used interchangeably in some cases, but they all have the same meaning, and include multiple levels of testing in various ranges. How do you understand this sentence? Can not be understood as a traditional test level, the traditional Test level for component testing, integration testing, system testing, acceptance testing and so on. Here to combine the content of the previous paragraph, the level refers to the test steps, test cases, test suites, test engineering, this level is the scope of the test step, the scope of the test case, the scope of the test suite, test the scope of the project.

Small Building Blocks

From childhood parents to educate us to do things to start from childhood, eat not a fat man, greedy not what good end. Therefore, API testing is best to start with a single test step call, and then combine them into a large user scenario, and the benefit is that it can help you pinpoint defects quickly and accurately. It also allows you to step by step to understand the details of the program under test, functional logic. The test is aut,application under test.

API is usually not a serious document, there is a lack of documentation and maintenance, no document is really not. So the test cases you want to organize build should be like stacked wood, starting with a very small part, and then using the knowledge you learned to accumulate other large parts.

When you run the test case and find a flaw, the developer will often find a way to quickly identify the damaged part of the minimum range. So, if you have a long description of the problem with the single function you're testing, it's going to take a lot of time to figure out the problem and even the clues, no matter how the process is automated.

In addition, the rational and effective organization of your test cases into manageable logical blocks will make your test cases more flexible and more maintainable.

You will probably choose a small piece of functionality to test. If you have a set of thousands of test cases, but one of your applications has only one (serious) bug fixed, you will probably need to quickly pick up the test cases that are related to the feature or problem. Almost all modern test frameworks allow you to create a set of test cases or sets of test cases, and then select one or more test cases to run from the group that you create.

How small this feature piece (block) is, depends on the product you are testing. If you are a house-building person, the smallest department may be doors, windows, walls and so on. The next big thing, you might consider building a bedroom out, a kitchen or a hallway, and so on. A test project or project is like building a cottage, two floors, no difference. (Really?) ）

If you are the supplier of a house-building person, the smallest part may be the hinge of a door, the handle, the bathroom sink, the kitchen sink, the faucet and so on. Larger parts may be outside doors, interior doors, kitchen cabinets and so on. A test project or project can also be likened to a full kitchen or a full bathroom.

(not to be continued ...) ）

Translation API Test Best Practices-Organize your tests