Table of Contents
Some people thought that avoiding testing stage of the software development was a good idea. They thought it would save thier budget. How wrong they were…
The painful truth is – penny wise and pound foolish. Missing at least one bug can cause serious problems, that are pulling one another and leading to the fatal snowcrash. Of course, it’s the most pessimistic scenario, but you should be ready for such consequences.
We, in Altamira, consider testing as important as development itself. In this article, we’ll uncover the hidden truth about software testing.
What is software testing?
The general meaning of the software testing process is executing the program in order to check whether it works as intended or to identify bugs that occure in the process of execution.
Purposes of software testing:
- To find out if the final product meets the requirements mentioned in specifications.
- To find the bugs or defects and consequently fix them.
- Understand the time needed to perform certain functions.
- To check if the responses are correct to various kinds of inputs.
What are the types of software testing?
Generally, specialists distinguish 4 main types of software testing. They are unit testing, component interface testing, integration, and system testing. As you might guess, the processes of each type vary and can be performed by different people. We dare these definitions to mean nothing to you, so, we’ll explain.
Unit testing QA means that the code is not tested as a whole project, it’s divided into units, which are actually the parts of the code responsible for specific functions of the app. These tests are usually written by the developers as they work with code and know exactly how each of functions should perform. Moreover, testing should be made from different perspectives to ensure there are no corner cases or backdoors. The key point of unit testing is to reduce the cost of the development as tests are performed before passing the code to a dedicated testing team.
Unit testing can also include static code analysis, data-flow and metric analysis, peer code reviews, etc. It also may require writing automated tests.
Integration testing is performed to know whether all interconnections between components are established in accordance to specifications. This kind of test is required before joining the separate components into one complete system to make sure they do not have conflicts with each other.
Component interface testing is made to verify the expected outputs of each component, or, in plain language, to verify the functionality and usability of this component. Such tests are usually executed by the developers themselves.
System testing is performed to make sure that software is running according to specifications.
What are the stages of software testing in Altamira?
Software development is a process that must be clearly planned and documented, otherwise you never get the worthy result. We always organize the development process in a most efficient way to get the result that totally meets our clients’ expectation. That’s why we have the well-established timeline so the client is always aware of what is going on.
Testing is the mandatory stage in our dev circle and moreover, it’s provided after each milestone, not only when the product is ready.
So, how it works in Altamira?
Since the programming process is divided into multiple stages, each of them is composed of tasks, performed by developers. After the completion of each task, they are directed to testers who make various tests.
If any bug or defect is found, the task is given back to the developer to be fixed. Upon fixing comes the second check. And only when there are no issues, the task is considered as a complete one.
The golden rule of our QA team is to always test the APIs given by a client to ensure they will interact with other components properly.
Tests are also executed upon the completion of each milestone and before the last stage of product delivery. Such practice allows us to deliver the projects of a high-quality.
Dos and don’ts of software testing
As any process, software testing has its own bad and good practices.
Don’ts of software testing:
- Postpone testing to the last minute eg.: to the milestone delivery. This practice can cause such problems as lack of time for bug fixing or the incorrect work of the software.
- Rely on a third-party API, provided by a client. Since APIs tend to change, this might cause the conflicts between components working with API.
- Always consider the bug found by the client as a bug. Sometimes such “bug” is just the possible irrational user behavior that is not an actual defect, but can be used by hacker as a back door, and cause security breach.
Dos of software testing:
- Moving the deadlines if any new feature appears. This allows to change the architecture if needed, provide refactoring and testing.
- Keep all the specifications data in one place. Adding the possibility to review versions is a plus. It may greatly reduce time on testing stage.
Conclusion
After all, software testing is the crucial and fundamental part of the development process. If well planned and tested adequately, this process will end up with delivering the bug-free product. Just like all our products, if you know what we mean.