Application - Petstore Jakarta EE 9

• Author : Antonio Goncalves
• Level : Intermediate
• Technologies : Jakarta EE 9 (Jakarta Persistence 3.0, CDI 2.0, Jakarta Validation 2.0, Jakarta Enterprise Beans 4.0, Faces 3.0, Jakarta Rest 2.0), Java SE 8 (because that's the minimum required by Jakarta EE 9), Twitter Bootstrap (Bootstrap 3.x, JQuery 2.x, PrimeFaces 6.x)
• Application Servers : GlassFish 6
• Summary : A Petstore-like application using Jakarta EE 9

Download the code from GitHub

Purpose of this application

Do you remember the good old Java Petstore ? It was a sample application created by Sun for its Java BluePrints program. The Java Petstore was designed to illustrate how J2EE (and then Java EE) could be used to develop an eCommerce web application. Yes, the point of the Petstore is to sell pets online. The Petstore had a huge momentum and we started to see plenty of Petstore-like applications flourish. The idea was to build an application with a certain technology. Let's face it, the J2EE version was far too complex using plenty of (today outdated) design patterns. When I wrote my Java EE 5 book back in 2006, I decided to write a Petstore-like application but much simpler. But again, it's out-dated today.

What you have here is another Petstore-like application but using Jakarta EE 9 and all its goodies (CDI, Enterprise Beans Lite, REST interface). It is based on the Petstore I developed for my Java EE 5 book (sorry, it's written in French). I've updated it based on my Java EE 6 book, and now I'm updating it again so it uses some new features of Jakarta EE 9 described on my Java EE 7 book. The goals of this sample is to :

• use Jakarta EE 9 and just Jakarta EE 9 : no external framework or dependency (except web frameworks or logging APIs)
• make it simple : no complex business algorithm, the point is to bring Java EE 8 technologies together to create an eCommerce website

If you want to use a different web interface, external frameworks, add some sexy alternative JVM language... feel free to fork the code. But the goal of this EE 9 Petstore is to remain simple and to stick to Jakarta EE 9.

The only external framework used are Arquillian, Twitter Bootstrap and PrimeFaces. Arquillian is used for integration testing. Using Maven profile, you can test services, injection, persistence... against different application servers. Twitter Bootstrap and PrimeFaces bring a bit of beauty to the web interface.

Compile and package

Being Maven centric, you can compile and package it without tests using mvn clean compile -Dmaven.test.skip=true, mvn clean package -Dmaven.test.skip=true or mvn clean install -Dmaven.test.skip=true. Once you have your war file, you can deploy it.

Test with Arquillian

Launching tests under WildFly is straight forward. You only have to launch WidlFly and execute the tests using the Maven profile :

mvn clean test -Parquillian-wildfly-remote

Or if you prefer the managed mode :

mvn clean test -Parquillian-wildfly-managed

Execute the sample

Once deployed go to the following URL and start buying some pets: http://localhost:8080/applicationPetstore.

The admin REST interface allows you to create/update/remove items in the catalog, orders or customers. You can run the following curl commands :

• curl -X GET http://localhost:8080/applicationPetstore/rest/categories
• curl -X GET http://localhost:8080/applicationPetstore/rest/products
• curl -X GET http://localhost:8080/applicationPetstore/rest/items
• curl -X GET http://localhost:8080/applicationPetstore/rest/countries
• curl -X GET http://localhost:8080/applicationPetstore/rest/customers

You can also get a JSON representation as follow :

• curl -X GET -H "accept: application/json" http://localhost:8080/applicationPetstore/rest/items

Check the Swagger contract on : http://localhost:8080/applicationPetstore/swagger.json

Databases

The persistence.xml defines a persistence unit called applicationPetstorePU that uses the default JBoss database :

<jta-data-source>java:jboss/datasources/ExampleDS</jta-data-source>

H2

By default, the application uses the in-memory H2 database. If you log into the WildFly Admin Console, go to http://localhost:9990/console/App.html#profile/datasources;name=ExampleDS and you will see the H2 Driver as well as the Connection URL pointing at the in-memory H2 database jdbc:h2:mem:test;DB_CLOSE_DELAY=-1;DB_CLOSE_ON_EXIT=FALSE

Postgresql

If instead a H2 in-memory database you want to use PostgreSQL, you need to do the following steps.

Install the PostgreSQL driver into Wildfly

This good article explains you how.

1. Go to $WILDFLY_HOME/modules/system/layers/base/ and create the folder org/postgresql/main
2. Copy the Postgresql JDBC driver jar file (eg. postgresql-42.1.4.jar) to the new folder $WILDFLY_HOME/modules/system/layers/base/org/postgresql/main
3. Create the file $WILDFLY_HOME/modules/system/layers/base/org/postgresql/main/module.xml with the following content:

<?xml version="1.0" encoding="UTF-8"?>
<module xmlns="urn:jboss:module:1.1" name="org.postgresql">
    <resources>
        <resource-root path="postgresql-42.1.4.jar"/>
    </resources>
    <dependencies>
        <module name="javax.api"/>
        <module name="javax.transaction.api"/>
    </dependencies>
</module>

4. Reference the module as a driver in WildFly configuration

WILDFLY_HOME/bin $ ./jboss-cli.sh
You are disconnected at the moment. Type 'connect' to connect to the server or 'help' for the list of supported commands.
[disconnected /] connect
[standalone@localhost:9990 /] /subsystem=datasources/jdbc-driver=postgresql:add(driver-name=postgresql,driver-module-name=org.postgresql, driver-class-name=org.postgresql.Driver)
{"outcome" => "success"}

Modify the default Datasource

In the Wildfly Admin Console check the default datasource ExampleDS. As you can see, it points to an in-memory H2 database. Make the following changes so it points at Postgres:

1. Attribute Tab: Change the driver to postgresql
2. Connection Tab: Change the Connection URL to jdbc:postgresql://localhost:5432/postgres
3. Security Tab: Change User name to postgres and no password

Once Postgres is up and running, you can hit the button Test Connection. It should be ok.

Startup PostgreSQL

The easiest is to use the Docker file to start Postgres

$ docker-compose -f src/main/docker/postgresql.yml up -d

Test this application on CloudBees

Third Party Tools & Frameworks

Twitter Bootstrap

When, like me, you have no web designer skills at all and your web pages look ugly, you use Twitter Bootstrap ;o)

Icons

I use:

• Font Awesome
• Silk Icons which are in Creative Commons

Arquillian

Arquillian for the integration tests.

Developpers

Some people who worked on this project :

• Antoine Sabot-Durand
• Brice Leporini
• Hervé Le Morvan

Bugs & Workaround

Licensing

This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.

antonio goncalves