By Leanminder Anonymous
Once upon a time, there was a legacy project. One can see the right architectural design patterns shaped into the application’s modules. The domain held the correct logic. The interaction with the database and system inputs were appropriately encapsulated.
However, once you start digging into the code, several features spread the logic across all layers: There were database entities in the Controllers, Transactions everywhere, incorrect use of the ORM with hardcoded SQL, cyclic dependencies, etc.
As you can imagine, bug fixing became a nightmare. The development of new features took ridiculous amounts of time. New developers Onboarding was tedious and time-consuming.
In my experience, this can be the result of several factors among which we can list:
We are knights of the code, how can we programmatically avoid this? The answer is ArchUnit.
ArchUnit is a Java library to check the architecture of your Java code using plain Java UnitTests. Using a simple syntax, we can ensure the code quality persists. Changes conflicting with architectural principles will break the build hence they will be identified early in the development process.
Enough talking, let’s see some examples:
As you can see in this example, we can identify an entity by its name or annotation. Now we can assert they reside in the right package. Note the two dots
.. are a wildcard for any value, so the following examples are valid:
But this examples will fail:
In the following example, we can ensure every class in the
controller package and annotated with
RestController, have the right suffix.
In the above examples, you can note the predicates are connecting with different logic operators (
and/or). This functionality allows you to have more specific or flexible rules as needed.
Well defined software architecture is crucial to your project. It can ensure you to maintain, evolve and extend your codebase.
Using ArchUnit you can test the code standards of your application and any violation will quickly break the build. The developer can notice the mistake at an early stage and fix it before even committing the change.
In this article, we wanted to illustrate simple yet powerful examples to programmatically ensure code standards are met.
Do you want to know more? Drop a comment and we will create a second post with further examples.
Do you want more? We invite you to subscribe to our newsletter to get the most relevan articles.
If you enjoy reading our blog, could you imagine how much fun it would be to work with us? let's do it!