To be honest, I discovered Spring Boot not so long ago and I loved the simplicity of it. The project setup was done in a matter of seconds, and I was already addicted to the annotation driven development style induced by JEE.
But hell, a simple (yet polite) REST API greeting me with a kind "Hello, World!" on each of its endpoints
weighted like A FREAKING MONSTER.
So I dove deep into the Java micro-framework world, and the second (if not first) link I found was one leading to the Spark Java website.
Needless to say that I immediately felt I love with its simple and efficient implementation. And the size of the generated artifact was under 3MB! We're still talking Java here, right?
But I still felt something was wrong. Like something was missing. I read a lot of blog posts and tutorials about "How to avoid boilerplate code with Spark Java", but was never satisfied.
Slowly sinking into distress, I finally realised that if a specific mechanism was not available I just had to create it. And so did I.
spark-runner is an annotation driven library allowing you to delegate much of the already minimal Spark
configuration and verbosity.
I tried to stay agnostic here about the design, hence you won't find any reference to some Service or Repository because, hey, do whatever you want with your code.
Do not hesitate to take a look at the provided example project!
TODO
TODO
The spark-runner artifact in itself can be used as a dependency for your own projects in a simple way:
Maven:
<repositories>
<repository>
<id>fries-io</id>
<name>fries.io Maven Repository</name>
<url>http://maven.fries.io/</url>
</repository>
</repositories>
<dependencies>
<dependency>
<groupId>io.fries</groupId>
<artifactId>spark-runner</artifactId>
<version>1.0-RC1</version>
</dependency>
</dependencies>Gradle:
repositories {
maven {
url 'http://maven.fries.io/'
}
}
dependencies {
compile 'io.fries:spark-runner:1.0-RC1'
}