990-32产品经理:Ultimate Guide to the Different Types of Software Testing in 2024 2024年不同类型软件测试终极指南

Software testing is an essential element of the Software Development Life Cycle (SDLC) that ensures the quality, reliability, and functionality of software applications. And with software testing engineers earning salaries in excess of $95k, there’s no denying that it’s a valuable skill to have.
All the same, if you’re asking the question, what are the different types of software testing, we’ve got you covered. In this article, we’ll dive into the different types of software testing you need to know in 2024, including definitions and usage.
软件测试是确保软件应用程序的质量、可靠性和功能性的软件开发生命周期（SDLC）的基本要素。软件测试工程师的薪水超过95k美元，不可否认，这是一项有价值的技能。
尽管如此，如果你想知道软件测试有哪些不同的类型，我们已经为你准备好了。在本文中，我们将深入探讨2024年你需要了解的不同类型的软件测试，包括定义和使用。

The Core Principles of Software Testing 软件测试的核心原则

Testing is context-dependent: Different applications have different requirements and constraints, so your test should be tailored to the context.
Absence of errors fallacy无错误谬误: The absence of errors does not guarantee the absence of defects in a software application. Software testing should be designed to cover identified and potential bugs.
Early testing: This helps detect errors early, reducing the cost and effort required to fix them later.
Pesticide paradox农药悖论: Repeating the same tests multiple times can result in finding the same errors while missing new issues. Tests should be revised regularly to include new cases.
Defect clustering缺陷集群: A small number of modules in a software application are often responsible for the majority of defects, so testing should focus on identifying and testing these thoroughly.
Exhaustive tests are impossible: Due to the often limitless number of test scenarios, testing every combination of inputs and conditions is impossible.

测试是依赖于上下文的：不同的应用程序有不同的需求和约束，所以你的测试应该根据上下文进行调整。
无错误谬误：无错误并不能保证软件应用程序中没有缺陷。软件测试的设计应涵盖已识别的和潜在的错误。
早期测试：这有助于及早发现错误，减少以后修复错误所需的成本和工作量。
农药悖论：多次重复相同的测试可能会导致发现相同的错误，而忽略了新的问题。应定期修订测试，以纳入新病例。
缺陷集群：软件应用程序中的一小部分模块通常会导致大部分的缺陷，所以测试应该集中于彻底地识别和测试这些模块。
穷举测试是不可能的：由于测试场景的数量通常是无限的，测试输入和条件的每一个组合是不可能的。

Manual vs Automated Testing 手动测试与自动化测试

When it comes to software testing and types of testing, the most fundamental distinction is between manual and automated testing.
Manual testing: Human testers execute tests on a software application to identify errors. The tester follows a predetermined set of test cases to ensure the application performs as expected. This is a labor-intensive process and can be slow.
Automated testing: Uses software testing tools to execute tests on software applications. With this, tests are pre-programmed, allowing automated testing tools to compare actual results with expected results.
Generally speaking, automated testing is faster, more efficient, and more reliable than manual testing, but it requires expertise in the programming language and software tools used.
Let’s look at some more differences between manual and automated testing.
说到软件测试和测试类型，最根本的区别是手工测试和自动化测试。
手工测试：人工测试人员在软件应用程序上执行测试，以识别错误。测试人员遵循一组预先确定的测试用例，以确保应用程序按预期执行。这是一个劳动密集型的过程，可能会很慢。
自动化测试：使用软件测试工具对软件应用程序执行测试。了这一点，测试被预先编程，允许自动化测试工具将实际结果与预期结果进行比较。
一般来说，自动化测试比手工测试更快、更有效、更可靠，但它需要程序设计语言和软件工具方面的专业知识。
让我们来看看手工测试和自动化测试之间的更多区别。

The Different Types Of Software Testing 不同类型的软件测试

Software testing can be broadly divided into two types based on the techniques used and the level of knowledge about the software application being tested.
These are known as functional and non-functional testing, and within each category, there are several sub-categories of software testing.
根据所使用的技术和对被测试软件应用程序的知识水平，软件测试可以大致分为两种类型。
这些测试被称为功能测试和非功能测试，在每一个类别中，又有几个子类别的软件测试。

Functional Testing

This type of testing evaluates software application functionality or behavior based on the requirements and specifications of the application.
Functional testing can be performed at various stages of the SDLC to ensure that the application meets the functional requirements and works as expected. Let’s take a look at the various types of functional testing.
这种类型的测试基于应用程序的需求和规范来评估软件应用程序的功能或行为。
可以在SDLC的各个阶段进行功能测试，以确保应用程序满足功能需求并按预期工作。让我们来看看各种类型的功能测试。

Unit Testing

This type of functional testing is conducted at the module or component level. The goal of unit testing is to test individual code modules(or units) to ensure that they are functioning correctly.
Often this is done by software developers during the development phase, and it can be performed manually or with automated testing tools.
Some use cases of unit testing include verifying the correctness of algorithms, testing boundary conditions, and ensuring that the code is working as expected.
The following code snippet is an example of a unit test for an asynchronous JavaScript function that fetches data from the popular JSONPlaceholder API.
This test uses MochaJS and checks whether the function correctly returns the name “Leanne Graham” when given an input of 1. This is the expected result since this is the first name in the users list from the JSONPlaceholder API.
这种类型的功能测试是在模块或组件级别进行的。单元测试的目标是测试单个代码模块（或单元），以确保它们正常工作。
常这是由软件开发人员在开发阶段完成的，它可以手动执行，也可以使用自动化测试工具。
单元测试的一些用例包括验证算法的正确性，测试边界条件，以及确保代码按预期工作。
下面的代码片段是一个异步JavaScript函数的单元测试示例，该函数从流行的JSONPlaceholder API获取数据。
该测试使用MochaJS并检查当输入为1时函数是否正确返回名称“Leanne Graham”。这是预期的结果，因为这是来自JSO Placeholder API的用户列表中的第一个名称。

White Box Testing

This involves testing the internal workings of the software application, including the code, structure, and design.
Testers with intermediate/advanced programming knowledge perform it. Typically it’s done during the development phase. The objective of white box testing is to ensure that the code functions correctly and identify any code errors.
白盒测试
这包括测试软件应用程序的内部工作，包括代码、结构和设计。
中/高级编程知识的测试人员来执行。通常是在开发阶段完成的。白盒测试的目的是确保代码正确运行并识别任何代码错误。

Integration Testing

This type of functional testing examines the interaction between different modules or components of the software application. The goal here is to ensure components are working correctly when integrated and it can be performed at different levels.
集成测试
这种类型的功能测试检查软件应用程序的不同模块或组件之间的交互。这里的目标是确保组件在集成时能够正确工作，并且可以在不同级别上执行。

Incremental Testing

This involves integrating components one at a time and testing them as they are integrated. This helps to identify bugs early in the development process as you can tell if a component works correctly before integrating it with other components. There are two ways to go about this.
Top-Down: Higher-level components are tested first while gradually integrating lower-level modules or components. This helps identify major defects early in the development process and safeguards against critical/major functionalities before moving to lower-level components.
Bottom-Up: Test lower-level components first and then proceed to integrate the higher-level components. This way, you identify minor defects early in the development process before moving to higher-level components.
增量测试
这涉及到一次集成一个组件，并在集成时对其进行测试。这有助于在开发过程的早期识别bug，因为在将组件与其他组件集成之前，您可以判断组件是否正常工作。有两种方法可以解决这个问题。
自上而下：首先测试较高级的组件，同时逐渐集成较低级别的模块或组件。这有助于在开发过程的早期识别主要缺陷，并在转移到较低级别的组件之前防止关键/主要功能。
自下而上：首先测试较低级别的组件，然后继续集成较高级别的组件。这样，您就可以在转移到更高级别的组件之前，在开发过程的早期识别出小的缺陷。

Non-incremental Testing

With this approach, you integrate all of the components at once and test them together. This is less efficient than incremental testing because it becomes difficult to identify and debug errors when all the components are integrated at once.
The following example of non-incremental integration testing uses JavaScript, the Mocha.js testing framework, and the Chai assertion library.
This tests an application created with Express.js by ensuring that the root URL (/) returns a 200 response code. The second test checks if the page title is correct.
非增量测试
使用这种方法，您可以一次集成所有组件并一起测试它们。这比增量测试效率低，因为当所有组件同时集成时，识别和调试错误变得很困难。
下面的非增量集成测试示例使用了JavaScript、Mocha.js测试框架和Chai断言库。
这将通过确保根URL（‘/’）返回200个响应代码来测试用Express.js创建的应用程序。第二个测试检查页面标题是否正确。