This guide walks you through the process of using Spring Data Redis to publish and subscribe to messages sent with Redis.
You will build an application that uses StringRedisTemplate to publish a string message
and has a POJO subscribe for the message by using MessageListenerAdapter.
|
Note
|
It may sound strange to be using Spring Data Redis as the means to publish messages, but, as you will discover, Redis provides not only a NoSQL data store but a messaging system as well. |
Before you can build a messaging application, you need to set up the server that will handle receiving and sending messages. https://raw.githubusercontent.com/spring-guides/getting-started-macros/main/docker_compose_support.adoc
If you choose to run the Redis server yourself instead of using Spring Boot Docker Compose support, you have a few options:
- Download the server and manually run it
- Install with Homebrew, if you use a Mac
- Manually run the compose.yaml file with docker compose up
If you go with any of these alternate approaches, you should remove the spring-boot-docker-compose dependency from the Maven or Gradle build file.
You also need to add configuration to an application.properties file, as described in greater detail in the [_preparing_to_build_the_application] section.
As mentioned earlier, this guide assumes that you use Docker Compose support in Spring Boot, so additional changes to application.properties are not required at this point.
You can use this pre-initialized project and click Generate to download a ZIP file. This project is configured to fit the examples in this tutorial.
To manually initialize the project:
-
Navigate to https://start.spring.io. This service pulls in all the dependencies you need for an application and does most of the setup for you.
-
Choose either Gradle or Maven and the language you want to use. This guide assumes that you chose Java.
-
Click Dependencies and select Spring Data Redis and Docker Compose Support.
-
Click Generate.
-
Download the resulting ZIP file, which is an archive of an application that is configured with your choices.
|
Note
|
If your IDE has the Spring Initializr integration, you can complete this process from your IDE. |
In any messaging-based application, there are message publishers and messaging receivers.
To create the message receiver, implement a receiver with a method to respond to messages,
as the following example (from src/main/java/com/example/messagingredis/Receiver.java)
shows:
link:complete/src/main/java/com/example/messagingredis/Receiver.java[role=include]The Receiver is a POJO that defines a method for receiving messages. When you register
the Receiver as a message listener, you can name the message-handling method whatever
you want.
|
Note
|
For demonstration purposes, the receiver is counting the messages received. That way, it can signal when it has received a message. |
Spring Data Redis provides all the components you need to send and receive messages with Redis. Specifically, you need to configure:
-
A connection factory
-
A message listener container
-
A Redis template
You will use the Redis template to send messages, and you will register the Receiver
with the message listener container so that it will receive messages. The connection
factory drives both the template and the message listener container, letting them connect
to the Redis server.
This example uses Spring Boot’s default RedisConnectionFactory, an instance of
JedisConnectionFactory that is based on the Jedis
Redis library. The connection factory is injected into both the message listener container
and the Redis template, as the following example (from
src/main/java/com/example/messagingredis/MessagingRedisApplication.java) shows:
link:complete/src/main/java/com/example/messagingredis/MessagingRedisApplication.java[role=include]The bean defined in the listenerAdapter method is registered as a message listener in
the message listener container defined in container and will listen for messages on the
chat topic. Because the Receiver class is a POJO, it needs to be wrapped in a message
listener adapter that implements the MessageListener interface (which is required by
addMessageListener()). The message listener adapter is also configured to call the
receiveMessage() method on Receiver when a message arrives.
The connection factory and message listener container beans are all you need to listen for
messages. To send a message, you also need a Redis template. Here, it is a bean configured
as a StringRedisTemplate, an implementation of RedisTemplate that is focused on the
common use of Redis, where both keys and values are String instances.
The main() method kicks off everything by creating a Spring application context. The
application context then starts the message listener container, and the message listener
container bean starts listening for messages. The main() method then retrieves the
StringRedisTemplate bean from the application context and uses it to send a
Hello from Redis! message on the chat topic. Finally, it closes the Spring application
context, and the application ends.
You can run the main method through your IDE.
Note that, if you have cloned the project from the solution repository, your IDE may look in the wrong place for the compose.yaml file.
You can configure your IDE to look in the correct place or you could use the command line to run the application.
The ./gradlew bootRun and ./mvnw spring-boot:run commands launch the application and automatically find the compose.yaml file.
You should see the output:
yyyy-mm-ddT07:08:48.646-04:00 INFO 18338 --- [main] c.e.m.MessagingRedisApplication: Sending message...
yyyy-mm-ddT07:08:48.663-04:00 INFO 18338 --- [container-1] com.example.messagingredis.Receiver : Received <Hello from Redis!>To run the code without Spring Boot Docker Compose support, you need a version of Redis running locally.
To do this, you can use Docker Compose, but you must first make two changes to the compose.yaml file.
First, modify the ports entry in compose.yaml to be '6379:6379'.
Second, add a container_name.
The compose.yaml should now be:
services:
redis:
container_name: 'guide-redis'
image: 'redis:latest'
ports:
- '6379:6379'
You can now run docker compose up to start the Redis server.
Now you should have an external Redis server that is ready to accept requests.
You can rerun the application and see the same output using your external Redis server.
|
Note
|
No configuration is required in the application.properties file because the default values match the Redis server configuration in compose.yaml. Specifically, the properties spring.data.redis.host and spring.data.redis.port default to localhost and 6379 respectively.
|
Congratulations! You have just developed a publish-and-subscribe application with Spring and Redis.
The following guides may also be helpful: