简单易懂的_kafka工具类

此kafka工具类包含

1.创建生产者，
2.创建消费者，
3.发送消息，
4.读取消息

1、引入依赖

<dependency>
    <groupId>org.apache.kafka</groupId>
    <artifactId>kafka-clients</artifactId>
    <version>1.0.1</version>
</dependency>

2、具体代码，通过加载配置文件的方式创建

public class KafkaUtils{

	//kafka地址
	public final static String server_ip = "127.0.0.1:9092";

	/**
     * 私有静态方法，创建Kafka生产者
     * @return KafkaProducer
     */
    public static KafkaProducer<String,String> createProducer(){
    	//生产者配置
		Properties properties = new Properties();
		properties.put("bootstrap.servers",server_ip);
		properties.put("acks","all");
		//消息发送重试次数
		properties.put("retries",0);
		properties.put("batch.size",0);
		//提交延迟
		properties.put("linger.ms",1);
		properties.put("buffer.memory",33554432);
		//Kafka提供的序列化和反序列化类
		properties.put("key.serializer","org.apache.kafka.common.serialization.StringSerializer");
		properties.put("value.serializer","org.apache.kafka.common.serialization.StringSerializer");
		//创建KafkaProducer; ctrl+p可以查看传什么类型的参数，
		KafkaProducer<String,String> producer = new KafkaProducer<String,String>(properties);
		return producer;
	}
	
	/**
     * 私有静态方法，创建Kafka消费者
     * @param groupid 分组名称
     * @return KafkaComsumer
     */
     public static KafkaConsumer<String,String> createConsumer(String groupid){
     	//消费者配置
		Properties properties = new Properties();
		properties.put("bootstrap.servers",server_ip);
		//分组名称
		properties.put("group.id",groupid);
		//每次拉取一百条，一条条消费，当然是具体业务状况设置
		properties.put("max.poll.records",100);
		properties.put("enable.auto.commit",true);
		properties.put("auto.offset.reset","earliest");
		//Kafka提供的序列化和反序列化类
		properties.put("key.deserializer","org.apache.kafka.common.serialization.StringDeserializer");
		properties.put("value.deserializer","org.apache.kafka.common.serialization.StringDeserializer");
		//创建KafkaConsumer; ctrl+p可以查看传什么类型的参数
		KafkaConsumer<String,String> consumer = new KafkaConsumer<String,String>(properties); 
		return consumer;
	} 

	/**
     * 传入kafka约定的topicName,json格式字符串，发送给kafka(集群)
     * @param topicName 交换机
     * @param jsonMessage 消息体
     */	
     public static void sendMsg(String topicName,String jsonMessage){
		//调用创建生产者的方法，创建生产者
		KafkaProducer<String,String> producer = createProducer();
		SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd:HH:mm:ss");
		//调用发送消息的方法
		producer.send(new ProducerRecord<String,String>(topicName,0,sdf.format(System.currentTimeMillis()),jsonMessage));
		//关闭
		producer.close();
	}

	/**
    * 传入kafka约定的topicName,groupid
    * 注意数据掉失，消费未保存，所以获取数据结合业务保存来
    * @param topicName
    * @param groupid
    */
    public static ConsumerRecords<String,String> getMsg(String topicName,String groupid){
		//调用方法，创建一个消费者
		KafkaConsumer<String,String> consumer = createConsumer(groupid);
		List<PartitionInfo> ps = consumer.partitionsFor(topicName);
		consumer.subscribe(Arrays.asList(topicName));
		ConsumerRecords<String,String> consumerRecords = consumer.poll(1000);
		consumer.commitAsync();
		//关闭
		consumer.close();
		return consumerRecords;
	}

}

后续可以采用监听的方式监听消息队列，采用注解。

@Component
@Slf4j
public class Consumer {

	// 监听主题(交换机) ；分组a
	@KafkaListener(topics =("${spring.kafka.topic}") ,groupId = "a")
	public  void  getMsg(ConsumerRecord message, Acknowledgment ack) {
        //确认收到消息
        ack.acknowledge();
    }
		
}