Koper

Koper is a MQ-based and event-driven distributed programming framework. [中文README]

Koper provides a simplified distributed listener and data listener model,which can help you build your async application in a simple way.

[Wiki]
[Quick Start] [快速指南]
[User Guide] [用户指南]

Concept

Core architecture: Message Architeture， Event Driven Architecture（EDA）
Core concept: producer, consumer, message, Message Queue(MQ), topic, subscribe
Core component: MessageSender, MessageListener, DataEventListener.
High-level concept: consumer group, message partition

Feature

Simplified MQ-based listener model and API.
Simplified Data Event model and API.
MQ provider independent.
- Koper supports Kafka by default, but also supports other message queue as provider, e.g. RabbitMQ, RocketMQ.
High performance and throughput.
High scalability based on MQ.
High-level feature: time spot recording, message tracking.

Programming Demo

1. Listener Model

Member signup example. When a member signs up, a message is sent to MQ by messageSender.

   messageSender.send("koper.memberSignup", "Signed up successfully! " + member.getPhoneNo());

On the other hand, a consumer subscribes the topic and handle the message, e.g. send a SMS to notify member.

@Component
@Listen(topic = "koper.memberSignup")
public class MemberSignupListener  {

   @Autowired
   private SmsService smsService;
   
   public void onMessage(String msg) {
       smsService.sendSms(msg);
   }
   
}

2. Data Event & Data Listener Model

Order example.

Koper support EDA programming. The Data Event machanism can intercept method invocation of object(such as DAO) and send it to MQ as data event.

orderDao.insertOrder( order);
orderDao.updateOrder( order);

DataListener responds to the event.

@Component
@DataListener(dataObject = "koper.message.demo.dataevent.dao.impl.OrderDaoImpl")
public class OrderListener {
   // data event: onInsertOrder
   public void onInsertOrder(Order order) {
       System.out.println("orderNo : " + order.getOrderNo());
       System.out.println("create time : " + order.getCreatedTime());
       // do some other operations
   }
   //data event: onUpdateOrder
   public void onUpdateOrder(Order order) {
       System.out.println("orderNo : " + order.getOrderNo());
       // do some other operations such as cache refresh
   }
  //data event: exception on updateOrder
   public void onUpdateOrder_X(Order order, DataEvent event) {
      String ex = event.getException();
      System.out.println("onUpdateOrder exception :" +ex);
   }
}

What can we do with Koper?

Koper is designed to support distributed asynchornous programming based on Message Queue and Event Driven Ahchitecture. It's suitable for scenarios such as Web Application, Business Monitor, Data statistics or Big Data etc.

In a large-scale application, the system architecture and event driven architecture look like something below.

Typical use cases

Asynchronous Business Process
Distributed Data Update or Cache Update
Data Log
Business Monitor and Alarm

Refer to Async Scenarios and examples for more demos.

Contribute

1 Fix bug or enhance Koper

Feel free to Fork the source code and commit bug fix or enhancement. And then make a pull request to us.

2 Contribute other MQ provider

Koper provides a Kafka provider implementation by default, but also provides high extensibility. You can implement other MQ provider easily, such as RabbitMQ, RocketMQ, ActiveMQ etc.

Refer to Developer Guide for details.

liuzchao/koper