Automated Testing-Understanding

In fact, I am currently in the phase of automated testing, but in order not to continue confused, I decided to extricate myself from confusion. the following are some of the things I have summarized for the seeking direction during the period of confusion.

Since automation is required, we should not be blind. We should at least understand the reasons for automation and fundamentally solve the problem. I think if we can open the door to automation testing in the following ways, automation is not that out of reach.

First, let's talk about automated testing?

Automated testing is a process of converting human-Driven Testing behavior into machine execution. Generally, after a test case is designed and passed the review, the tester executes the test step by step according to the procedure described in the test case to compare the actual results with the expected results. In this process, in order to save manpower, time or hardware resources and improve testing efficiency, the concept of automated testing is introduced.

This is an official definition. I think it is concise to simulate human operation steps and execute operations instead of people to complete our testing process. The main purpose is to complete testing tasks in batches and improve work efficiency.

Second, what can automated testing do?

After learning about automated testing, we already have a macro understanding of automation, but I think this is what we need to do, from the above understanding, what automation can do is to complete the test work instead of humans. But what work can be done? Is it impossible for all work to be completed, after all, it is not time for computers to replace the human brain. in my opinion, from a simple test case, I think what automation can do is to simulate the execution process of the test case and simulate the driving data of each execution in the test case. however, if we explain at a micro level what automation can do, it can be automated from test management to test execution, or even to the output of the final test result, what we can talk about here is limited according to everyone's programming ability. If someone with development ability can do automation, he can design an automation process, from management to report output, this is the so-called automated testing framework. I think the biggest advantage of automation is that you can develop your own automated testing framework based on your technical characteristics and the characteristics of each project.

Third, what knowledge is required to learn automation well?

In fact, when it comes to automation, it is not difficult to think of programming capabilities. Without the foundation of programming languages, learning automation can only be regarded as a result, because you cannot feel the process of designing automation tests by yourself, therefore, the program must be mastered. Based on the current automation, the language to be mastered is VB, Ruby, Java, and so on. You can choose one of them. currently, automation on the market is generally used for web, so you must have knowledge about html. we should also talk about the basic knowledge of databases. Automation is mainly about data-driven testing, so database applications are also necessary.

Fourth, What are the limitations of automation?

Speaking of limitations, I also realized that, in view of the increasing number of Web architecture projects and the focus of automation on the web, more applications are Web Services, for C/S structure programs, it is not impossible to do, but the support is getting fewer and fewer, so resources are scarce. after all, qtp can still be done, that is, VB can. It can be automated by calling objects in windows, but it is too limited. I will not talk about it here. I will know it once I use it, I have also studied vbs for a while. let's talk about the limitations of automated testing from another perspective. Automated Testing is suitable for projects with stable requirements, this will cause the test scripts built in the early stage to follow the demand changes at any time. Once the changes are large and the project is unstable, a large amount of maintenance work on the automated test scripts will greatly reduce the automation efficiency, in the early stage of automation, all aspects of the evaluation must be done. I am currently engaged in learning automation mainly for portal websites, because this requirement is relatively stable, so it is more suitable.

Fifth, is to assess whether your project is suitable for automation?

In this regard, we have already talked about how to assess whether the project is suitable for automation. In fact, it is still complicated. First, we must communicate with the demand side in detail, to ensure the accuracy and stability of requirements, we also need to understand the development model of the project. For example, agile development is not suitable for automated testing, because the requirements are agile.

Sixth, how to choose an automated testing tool?

Qtp is the most widely used tool on the market, but it is not affordable for general companies because it is a tool for charging fees. Therefore, we are a civilian people like us, choose open-source. After all, open-source technologies are becoming more and more mature. It cannot be said that they can compete with qtp. At least some are open-source technologies, and qtp may not be able to do so. here are some examples of Ruby + watir, selenium, and vbs.