Last updated on November 5th, 2012 (Spring MVC 3.2 RC1)

In previous blog posts I introduced the Servlet 3 based async capability in Spring MVC 3.2 and used the spring-mvc-showcase and the Spring AMQP stocks sample to demonstrate it. This post presents a chat sample where the external events are not AMQP messages but rather HTTP POST requests with chat messages. In the second part of the post, I'll switch to a distributed chat where the events are Redis notifications.

Chat is not a common requirement for web applications. However it is a good example of a requirement that can only be met with real-time notifications. It is more sensitive to time delays than email or status alerts and it is not that uncommon to chat in a browser with a friend, or with a colleague during a webinar, or with a live person on a shopping site. You can imagine other types of online collaboration.

The Sample

The spring-mvc-chat sample is available on Github. Although not the focus of this blog post, the client side uses Thymeleaf, knockout.js, and jQuery. Thymeleaf is an excellent alternative to JSPs that enables clean HTML templates with support for previews allowing a designer to double-click an HTML template and view it unlike a JSP that requires a Servlet container. knockout.js is a client-side MVC framework that's very handy for attaching behavior to HTML elements. To get an idea about it quickly, follow one of its excellent tutorials. jQuery is used for DOM scripting and Ajax requests.

ChatController

The ChatController exposes operations to get and post chat message. Here is the method to get messages:

@RequestMapping(method=RequestMethod.GET)
@ResponseBody
public DeferredResult<List<String>> getMessages(@RequestParam int messageIndex) {

  final DeferredResult<List<String>> deferredResult = new DeferredResult<List<String>>(null, Collections.emptyList());
  this.chatRequests.put(deferredResult, messageIndex);

  deferredResult.onCompletion(new Runnable() {
    @Override
    public void run() {
      chatRequests.remove(deferredResult);
    }
  });

  List<String> messages = this.chatRepository.getMessages(messageIndex);
  if (!messages.isEmpty()) {
    deferredResult.setResult(messages);
  }

  return deferredResult;
}

A new DeferredResult is created and saved in a Map from where it will be removed by the registered onCompletion callback when the async request completes. Then the method checks for new messages using an in-memory ChatRepository. If new messages are found, the DeferredResult is set immediately. Otherwise it will be set later, when a new message arrives.

Below is the method that saves chat messages and updates all saved DeferredResult instances:

@RequestMapping(method=RequestMethod.POST)
@ResponseBody
public void postMessage(@RequestParam String message) {

  this.chatRepository.addMessage(message);

  // Update all chat requests as part of the POST request
  // See Redis branch for a more sophisticated, non-blocking approach

  for (Entry<DeferredResult<List<String>>, Integer> entry : this.chatRequests.entrySet()) {
    List<String> messages = this.chatRepository.getMessages(entry.getValue());
    entry.getKey().setResult(messages);
  }
}

A Distributed Chat

The above chat sample uses simple, in-memory persistence and works only when deployed to a single server. The redis branch uses a Redis-backed ChatRepository. Redis is a simple key-value store that is easy to use in Java with the help of the Spring Redis project.

The RedisChatRepository uses the Spring Redis RedisTemplate to look up and save chat messages. Feel free to take a look at the code.

The controller method that saves new chat messages is now a single line:

@RequestMapping(method=RequestMethod.POST)
@ResponseBody
public void postMessage(@RequestParam String message) {
  this.chatRepository.addMessage(message);
}

Receiving new messages is also very simple. It involves implementing the Spring Redis MessageListener interface, which can be done directly in the controller:

@Controller
@RequestMapping("/mvc/chat")
public class ChatController implements MessageListener {

  // ...

  public void onMessage(Message message, byte[] pattern) {
    for (Entry<DeferredResult<List<String>>, Integer> entry : this.chatRequests.entrySet()) {
      List<String> messages = this.chatRepository.getMessages(entry.getValue());
      entry.getKey().setResult(messages);
    }
  }

}

The Redist version of the chat will work in a cluster. A message can be posted on any server and all other servers will receive Redis notifications. The Spring Redis project makes it really simple to receive those notifications in message-driven POJO style.

Chat messages can be posted from non-Java, Redis clients too. For example, connect with the Redis command-line shell, type the below commands, and the chat messages will be deliverd to all subscribed servers and connected browsers:

redis 127.0.0.1:6379> RPUSH chat:archive "hello from the redis cli"
redis 127.0.0.1:6379> PUBLISH chat "a new chat message is available"

This concludes the blog posts covering the Spring MVC 3.2, Servlet 3 based, async support. Thanks for reading!