Tech demo of using Kotlin to create services with popular Java and Web ecosystem frameworks and tooling:
- Spring Boot and Spring MVC
- JUnit and Mockito
- React and jspm
- MQTT messaging
- Gradle and Docker
Demonstrate compatibility, concise, expressive, and safe code, and support for sophisticated features.
Caveat: I'm just learning Kotlin!
Assuming an OSX development environment. Reasonable equivalents available on Linux and Windows.
Install necessary tooling via brew:
brew cask install java
brew install kotlin
brew install gradle
brew install npm
npm install jspm@beta -g
IDEA is the recommended development environment for this project (but optional of course).
- Download IDEA CE 16+
- Install Kotlin plugin
- Import Project > From existing model > Gradle
- Specify GRADLE_HOME = /usr/local/Cellar/gradle/GRADLE_VERSION/libexec
Used for building and running distributable images (optional).
brew install docker docker-machine
docker-machine create --driver virtualbox default
eval $(docker-machine env default)
Build:
gradle assemble
Run tests and code checkers:
gradle check
Server and client can be reloaded as you make code changes. After gradle run, in a separate terminal run a watch build:
gradle assemble --continuous
The server will be reloaded by Spring Boot devtools when new built artifacts are produced by assemble.
The web client will also refresh if you install the Chrome LiveReload Extension.
gradle run
http://localhost:8002
http://localhost:8002/api/health
http://localhost:8002/api/players
http://localhost:8002/api/players/<id>
For remote debugging (will pause until debugger connected):
gradle run --debug-jvm
Of course you could just run with debugging from IDEA.
Build image at desired version:
gradle -Pversion=1.0 image
Push and run:
docker run -d -p 8002:8002 --name nebula jmailen/nebula:1.0
docker logs -f nebula
http://DOCKER_MACHINE_IP:8002/api/health
docker stop nebula
docker rm nebula