Introduction to Automated Test Framework

2021-12-03 11:32:10
Chen Qi
Original 1161
Summary : Automation testing has increasingly become the "trend" for testers because of its cost-saving advantages, improving efficiency, and reducing manual intervention. Practitioners increasingly clearly understand that realizing the automation framework is one of the critical factors to the success of software automation projects. This article will takeyou into the field of the automated testing framework, how the mechanical testing framework will give you a competitive advantage in the automated testing framework.

Automation testing has increasingly become the "trend" for testers because of its cost-saving advantages, improving efficiency, and reducing manual intervention. Practitioners increasingly clearly understand that realizing the automation framework is one of the critical factors to the success of software automation projects. This article will take you into the field of the automated testing framework, how the mechanical testing framework will give you a competitive advantage in the automated testing framework. 

Advantages of automated testing

The popularity and advantages of automated tests do not mean a panacea. Customized projects, short cycles, test ease of use, and other situations are not suitable for automated testing. But there is no denying that if carefully planning and executing a mechanical testing framework indeed bring the following benefits to software development and testing companies. 

  1. Minimum time-maximum earnings

Building any viable automated test framework and automated scripts maximizes the time spent writing and running tests, allowing for maximum output in a short period. In addition, with an excellent automated testing framework, manpower can be freed from these common problems such as synchronization, error management, local configuration, report generation, interpretation, etc

2. Reusable and readable automated code

When you use the code mentioned in the existing component library, Big can rest that it is still readable and reusable for later times. All related tasks (such as reporting, synchronization, and troubleshooting) will become more accessible.

3. Resource optimization 

Some companies do not benefit from automated implementation as conceived, as the benefits from automated tests depend on the flexibility to adopt it. If automated systems are flexible and compatible with different teams engaged in various components, they can benefit from resource optimization and knowledge sharing.

Definition of the automated testing framework

In many other industries, "automation" refers to the automatic processing of processes through intelligent algorithms and techniques that run with little or no human intervention. In the software industry, automated testing refers to various testing of software applications through automation tools (either authorized or open-source versions). From a technical perspective, the automated testing framework is a customized set of interactive components that can perform scripting testing and comprehensively report the test results.

To successfully build an automated test framework requires considering the advice of software quality assurance experts who can help observe and help control the entire test process and improve the accuracy of the test results. A carefully designed automated testing framework allows testers to perform automated tests practically, simplistically.

The type of the automated test framework

1. Tool-centric framework

Commercial and open-source automation tools help to generate reporting, testing suites, and system infrastructure for distributed testing execution in a test environment. For example, the Selenium Automation framework, which has the main component, WebDriver—— plug-based Web browser, controls and manipulates DOM models of applications in the Web browser. The Selenium Test Automation framework also has practical encoding libraries and record playback tools.

Serenity is another essential framework dedicated to tools, an accelerator built around the Selenium Web driver. In this case, to accelerate the test automation implementation process, the community combines specific components in a public entity.

Tool-specific frameworks such as TestComplete, Ranorex HP QTP are challenging to make explicit judgments because they are all pre-built by a deployed infrastructure with action simulators, reports, and script IDE. 

2. Application-centric framework

This type of framework is customized to automate a particular application project. An application-centric framework supports the need for test automation for a targeted application, driven by components built from open-source libraries. This framework can create a friendly test environment around SUT to run some basic features, including deploying developed applications, running applications, test case execution, direct test results reporting, and efficiently encoding wrapper controls. The application-centric framework also has a component to support testing runs across various cloud environments on different operating systems and browsers.

3. Keyword drive framework

Keyword keyword-driven framework can mainly attract novice developers and testers with little experience in coding, which can be either a tool-centric or project-centric framework. The keyword-driven framework involves unskilled coding employees in writing and understanding automated scripts. The keyword set used for encoding (e. g., login, NavigateToPage, Click, Type Text) is installed in the code library as a keyword repository, and a spreadsheet based on the provided keyword reference will be passed the keyword interpreter to perform the test.


Main components of the ideal automated testing framework

If you want to implement a robust and high-performance test automation framework (whether open-source or commercial), certain components must be considered at their core. Some frameworks may include all the details, and others may consist of only a few, so don t have to involve all the points mentioned below in each framework.

1. Test Base Components

a)unit testing 

The unit test base component can be a fundamental part of any automated testing framework. It can be used for the following situations:

  • Define the test method being used through specific forms like @test or [test]
  • Perform assertions that affect the final results of the automated tests
  • Run a straightforward test
Running tests from the command line, IDE, dedicated tools, or a CI (continuous integration) system, the unit test library ensures that the unit test runs.
The unit test library supports almost all programming languages, such as Java's JUnit and TestNG. NUnit and MSTest of NET, and unit test of Python (formerly PyUnit).

b)Integration and end-to-end testing

While performing integration and end-to-end test automation, it is also very appropriate to implement the functions provided by existing test libraries. API level testing driven by the application UI requires components that eliminate the unnecessary coding burden, making it easier for interaction with the application-tested application. Therefore, testers are not troubled by the coding efforts of connecting to applications, sending requests, and receiving results.

Several essential test components of this type are Selenium (available in the main languages), protractor (JavaScript-specific), and Karate DSL (Java-specific API-level integration tests).

c)Behavior-driven development(BDD)

Components specifically for BDD create executable specifications in the form of executable code. In this section, the tester can convert the different features and scenarios of the predicted behavior into code. Although unlike other testing tools, which can directly interact with the tested application, they can support the BDD process to create activity documents consistent with the scope and intent of the automated test. Typical examples of BDD components are cucumber (main languages), Jasmine (JavaScript), SpecFlow (for.NET).

2. Test data management

The most considerable difficulty in software test automation and test creation is how to utilize test data management systems. As the number of automated tests increases, there is always a problem of ensuring that certain test data required to perform a specific test is available or created when executed. There is no foolproof solution for this situation, requiring reliable test data management methods to make automation work successfully. So, the automated test framework employed needs to be fully configured to provide essential remedies to enter or create and clear the test data to be performed. The solution to this problem is to use appropriate simulation tools to make the data more accessible, clearer, and easier to understand.

3. mock、stubs and virtual assets

During the practice and exploration of automated testing:

● You want to isolate the module from the connected component experienced in unit testing.
● Processing the tedious and critical dependencies common in the integration of modern applications, or end-to-end testing

In these cases, it feels necessary to create mock, stubs, and virtual assets to reflect the behavioral patterns of connected components, and may also find processing phony and stub a broad, large-volume task. In any case, testers will realize that it is essential to choose helpful virtualization tools while developing an automated testing framework.

The universal mechanism for implementing the mode

In addition to the automation as mentioned above framework components, there are several valuable mechanisms to help create, use, and maintain automated tests, for example:

Wrapper methods: Creating a custom wrapper makes handling errors with Selenium WebDriver components easier. Creating a custom wrapper for Selenium API calls enables better timeouts, exception handling, and fault reports. After that, people who have made automated tests can reuse them to avoid focusing on complex processes and valuable testing.

Abstraction methods: Abstract mechanisms imply improved readability and concealment of redundant details. For example, the purpose of using a page object when creating Selenium WebDriver tests is to expose user input operations on a web page, including entering credentials or clicking somewhere on the page, and the goal is to achieve advanced testing methods bypassing beyond or bypassing the needs of exploring specific elements of the page. This method applies to many similar applications and automated tests.

Test results report

When choosing a library or mechanism to report test results to the automated framework, you need to take the target audience to view this report into account. In this regard, the following considerations should be followed:

● The information generated by unit testing frameworks such as Junit and TestNG is mainly for receiving systems like CI (continuous integration) servers, which are eventually interpreted and presented in the XML format available to other software.

● When using languages that most people can understand as tools for reporting, you may need to consider using business tools compatible with the unit testing framework, such as the UFT-Pro for Junit, NUnit, and TestNG.

● An alternative is to use third-party libraries such as ExtentReports that create test results reports in an easy-to-understand format, including visual interpretation through pie charts, graphics, or images.

Auxiliary platform

In addition to the automated testing framework itself, some auxiliary platforms associated with continuous integration will also be beneficial to automated testing. For example, CI can perform deliverables deployment, test scheduling; the source code platform can manage code for the product and test script; dependency manager is generally a dependency management tool for a language library, such as universal java types developed with nexus management employees; Zentao ZTF contains features such as continuous integration, unit testing, and version library management.


CI terrace

To perform application testing faster, more coherently, the continuous integration platform helps you regularly build software and run a variety of tests for new versions. This method gives developers and testers regular feedback and rapid response to application quality when developing and deploying new features and updating existing capabilities. Current prominent, continuous integration platforms are TeamCity, CircleCI, Jenkins, etc.

Source code management

As with manual testing, automated testing involves writing and storing versions of the source code. Each development company has a select source code and version control system to save and protect the source code. Automated testing requires a sound source code management system, which is convenient when processing production code. Typical source management examples are Git and Subversion.

Create a dependency manager

The primary purpose of the Dependency Manager is to assist in the collection, managing existing dependencies and libraries that have already been applied in the functions of automated software solutions. Some tools (such as Maven and Gradle) act as dependency managers and help build tools. The build tool is designed to help you develop automated software and run tests from the source code and support libraries. Tools for this aspect include Ant, npm, and NuGet.

Zentao ZTF

ZTF supports seamless integration with Zentao, binding Zentao use cases and automated test scripts one by one. Execution results can be submitted to Zentao as test results, and executing failed scripts can also automatically create a bug. The ZTF automated testing framework enables continuous integration capabilities with Jenkins. After the user initiated the task, the user automatically executed the test script through ZTF and returned the unit test results to Zentao. The two cooperation opened the continuous integration closed loop.

Framework construction and implementation process

Here are several ways to plan for automated test solutions:

● Explore the practical applicability of automation from a customer perspective, check its good appearance from all views, and test it on underused technology. In contrast, automated testing is unnecessary if the efforts of mechanical development greatly exceed the expected advantage.

● It is crucial to keep a close eye on the system-tested techniques to find the most suitable automated testing tools that can perfectly simulate the user behavior.

● A phase-based implementation is recommended, where each stage has a priority for delivering automated test scripts while adding framework functionality for the expected script execution.

● Before starting the software automated test, to ensure proper implementation of automated decisions, the postimplementation of ROI, proof of concept, time of the manual regression or smoke test run, and the number of running cycles for each version must first be estimated.


In today's fast-paced, brutal software development ecosystem, the automated testing framework plays an integral role in maintaining the software testing cycle's speed, efficiency, and clarity. As AI is introduced into software testing, organizations considering an automated testing framework must dive into the final framework's design before they get into this field. Good framework design and strategies for using components will lay the foundation for the final automated testing framework.

The best way to build a mature, sophisticated, and flexible automated testing framework architecture is to start small, frequently test and review, and gradually build extended versions to a higher level. It is easy to prepare numerous automated tests early on and see the working framework faster to avoid conflicts or damage during the computerized testing phase later.

The guidelines outlined above help software testers and companies learn from projects that successfully perform automated tests.


Author bio

Chen Qi, a senior agile test consultant, as a team member of ZenTao, a well-known domestic project management software, is mainly responsible for the open-source automated test management framework-the development of ZTF. With more than ten years of practical experience in the agile process, he is now committed to the practice and research in test automation and DevOps.

Write a Comment
Comment will be posted after it is reviewed.