How does Test-Driven Development relate to Agile?

0 Shares

During software development, developers need to deliver software that works correctly, with quality, and on time. To guarantee these points, we must test our software to find possible flaws. Test-Driven Development (TDD) has undoubtedly become one of the most popular practices among software developers.

By writing the tests beforehand, the developer ensures that a good part of your system has a test that guarantees its operation. In addition, many developers also claim that tests guide them in designing system classes.

But testing software correctly is not as simple as it may seem. Therefore, several methodologies aim to test the different components of a software solution correctly. Some of these methodologies aim to test the integration between software components, while others aim to test each component in isolation.

In this article, you will understand what means Test Driven Development and its advantages to be an agile methodology.

What is Test-Driven Development (TDD)?

As mentioned earlier, TDD is the acronym for Test-Driven Development and it means that the tests will be written even before your program that feature itself.

The idea of ​​TDD is that you work in cycles. These cycles occur in the following order:

First, write a unit test that will initially fail since the code has not yet been implemented;

Create code that satisfies this test, which means implementing the functionality in question. This first implementation should immediately satisfy the test that was written in the previous cycle;

Once the code is implemented and testing is satisfied, refactor the code to improve points such as readability. Afterward, run the test again.

The new version of the code should also pass without having to modify the test written initially.

How to apply TDD?

The practice of Teste-Driven Development brings many benefits to the development process. The first and clearest one is the external quality benefits of the product.

Automated tests, which run in a matter of seconds, are all the time for the developer. This means that we can run them all day, many times a day. Something we know is impossible with manual testing.

If something stops working, the developer is quickly notified and can fix the problem before sending the version to the client

In addition, many popular authors in the field claim that if the developer knows how to read the test code carefully, that same code can provide important information about the quality of the code being produced.

But the general idea is that if it’s difficult to write an automated test, it’s probably because the production code has some complications.

Realize then that using the Test-Driven Development practice helps the team ensure that the requirements work as expected, and also helps the developer notice code issues in their implementations.

How does Test-Driven Development (TDD) work?

Despite being considered easy, it is necessary to apply some rules in TDD development so that there is a coding standard to be followed. In summary, the flow is based on three parts:

  1. Write a test that will fail (Red)
  2. Make an effective code (Green)
  3. Eliminate Redundancy (Blue – Refactor)

The idea is that you use these cycles for unit tests. These tests aim to evaluate the behavior of specific classes and methods. You will identify what are the behaviors of the methods’ development.

Rather than implementing the code initially, you will create test code bound to those classes and methods you want to implement. Then, the test will fail (red), so that you can then implement this functionality.

With implementation, this test expectation is to pass (green). In the end, you can improve this code, avoiding duplication and unnecessary code snippets (refactor), and will test again where it should pass. So you will have a working code and already properly tested.

Practicing TDD

Whenever a developer takes a feature to do, he breaks it down into small tasks. Tasks that require writing code. Classes are created, others are modified. All these modifications have a purpose, of course. Every code written has a goal.

When practicing Teste-Driven-Development, the developer before starting to make these modifications explicit these goals. Only it does this through automated testing. Testing in code is nothing more than a snippet of code that makes it clear what a snippet of code should do.

By formalizing this objective in the form of an automated test, that test fails, of course. The developer then works to make that test pass by implementing the functionality.

Once the test passes, the developer then moves on to the next step in the cycle, critical for those who dream of producing quality code: refactoring time.

Is Teste-Driven-Development practical for a developer?

As with many agile practices, it’s hard to see the benefits when you don’t use them.

The developer will spend a good part of their time writing test code. But that doesn’t mean he’s less productive because of it.

Before moving on to arguments, it is necessary to define what it is to be productive. For many, productivity is simply the number of lines of production code that you can write each day.

The day-to-day of the developer who does manual testing is something like this: he programs the functionality and runs the application. He takes the first manual test by navigating through the application and typing in the various input data needed to run the test.

What is debugging?

Debugging is the activity where the developer runs line by line of code and sees the result. Both activities waste a lot of developer time.

At some point, the developer will encounter the problem. It will correct it, and then it will repeat the whole process: it will upload the application and do the manual test.

The developer who does manual testing repeats the same test several times a day. 

Furthermore, whenever the developer needs to test again in the future, he will do it manually, wasting time. The developer who has automated tests, however, will only run his test battery.

The developer who writes automated tests spends time on it. But he spends time smartly. Nowadays, the developer who does manual testing also spends a lot of time on testing, but unproductively.

In the medium term, that developer will spend more time testing the same functionality than what was smart and automated from the start. It’s all about thinking in the medium term.

What are the TDD benefits?

There are several benefits to writing tests before you even perform an implementation. One of the benefits is that since you know what the code needs to do in advance, you end up avoiding writing code that is too complex or doesn’t follow business prerequisites.

By writing the tests at the initial stage, it might even seem like we ended up with one more task to perform. However, in the end, your software will be less likely to fail, and you will end up developing higher-quality code.

GitScrum supports your team to better results and effective deliveries!

Set your workflow and board to guide your Agile team, assign Tasks, Subtasks and keep in charge of the whole process evolvements. Allow your Agile team to collaborate.

Reach higher levels of efficiency, productivity, and deliverability with GitScrum. Work focused on prioritizing what’s valuable and tracking your flow to overcome results.

Sign up now and make your team grow together!

Leonardo Salles

Copywriter for GitScrum.