In this post I'll introduce Puma4j, a new open-source project I recently published that I hope will make writing JUnit 5 tests a bit easier.
Check the project repository on GitHub.
The Project
In testing microservices, we often need to simulate the behavior of external integrations, mocking HTTP requests/responses and messages arriving in topics or queues. For this, we need to build a lot of payloads in our tests.
Instead of creating test data in our code which sometimes leaves our tests a little "dirty," or creating the same boilerplate code all the time to read and parse resource files, why not leverage JUnit 5 extensions to make a simple and standardized way to do this and also have some fun coding new stuff?
That's why Puma4j was created: with it, you can load and convert resources with just a couple of annotations.
Take a look at the usage example below:
Installation with Gradle
implementation "io.github.vitorsalgado.puma4j:puma4j-junit5-extension:4.0.2"Usage Example
Consider a scenario where you have the following files in your test resources directory:
- model.json
- test.txt
- data.yml
@UsePuma4j
class YourJUnit5Test {
static class Model {
public String hello;
}
@Res("data.yml")
private static Model ymlData;
@Res("test.txt")
private String txt;
@Test
void test(@Res("model.json") Model model) {
assertEquals("world", model.hello);
assertEquals("world", ymlData.hello);
assertEquals("hello world", txt);
}
}Following the example above, to use Puma4j we need to annotate our test class with @UsePuma4j and then, all fields and method parameters into which you want resources injected need to be annotated with @Res(RESOURCE_NAME_AND_EXTENSION).
Puma4j will automatically load the resources and convert them based on field or parameter type. For example, you can load the raw file content as a String or byte[].
Supported File Types
Puma4j can convert the following resources into Java object instances:
- JSON
- YAML
- Properties
The content of other file types, like .txt, can be loaded into a String or byte array.
Custom Converters
You can use a custom converter for your resources. Implement the Unmarshaller<O> interface and provide the class reference to the annotation @Use(YourCustomConverter.class). The annotation @Use can be applied to the class, field, and method parameters.
Switching JSON Converters
By default, Puma4j uses Jackson Object Mapper to convert JSON resources into Java objects. To switch to Gson, use the annotation @UseGson. The annotation can be applied to class, field, and method parameters. You can also use @UseJackson to force a specific resource to use Object Mapper.
Take a look at the project documentation and source code on GitHub.