根据配置文件一步步去查看：

spring-consumer.xml

DefaultKafkaConsumerFactory就是根据参数配置生产consumer，略过。

监听器类，实现了监听类一些列接口的某一个。我这里使用了手动提交的AcknowledgingMessageListener 接口，以此为例。

首先他又继承了一个接口

public interface AcknowledgingMessageListener extends GenericAcknowledgingMessageListener> {

}

继承的接口

public interface GenericAcknowledgingMessageListener extends KafkaDataListener {

void onMessage(T var1, Acknowledgment var2);

}

我的实现类，重写发放就好，略。。。继续下一步

public class KafkaConsumerListener implements AcknowledgingMessageListener {

@Override

public void onMessage(ConsumerRecord stringStringConsumerRecord, Acknowledgment acknowledgment) {

System.out.printf("offset= %d, key= %s, value= %s,topic= %s,partition= %s\n",

stringStringConsumerRecord.offset(),

stringStringConsumerRecord.key(),

stringStringConsumerRecord.value(),

stringStringConsumerRecord.topic(),

stringStringConsumerRecord.partition());

acknowledgment.acknowledge();

}

}

topic1

topic2

org.springframework.kafka.listener.config.ContainerProperties

关注的第一个构造方法，我可以传入多个topic，也就是说我可以通过传入一个list来同时订阅多个主题的消息

public ContainerProperties(String... topics) {

this.ackMode = AckMode.BATCH;

this.pollTimeout = 1000L;

this.shutdownTimeout = 10000L;

this.syncCommits = true;

this.ackOnError = true;

Assert.notEmpty(topics, "An array of topicPartitions must be provided");

this.topics = (String[])Arrays.asList(topics).toArray(new String[topics.length]);

this.topicPattern = null;

this.topicPartitions = null;

}

关注的第二个构造方法，我可以传入一个正则表达式对象来匹配主题

public ContainerProperties(Pattern topicPattern) {

this.ackMode = AckMode.BATCH;

this.pollTimeout = 1000L;

this.shutdownTimeout = 10000L;

this.syncCommits = true;

this.ackOnError = true;

this.topics = null;

this.topicPattern = topicPattern;

this.topicPartitions = null;

}

关注的第三个构造方法，我可以传入一个或多个set，对应的键值对为主题和分区，关注特定主题特定分区的消息

public ContainerProperties(TopicPartitionInitialOffset... topicPartitions) {

this.ackMode = AckMode.BATCH;

this.pollTimeout = 1000L;

this.shutdownTimeout = 10000L;

this.syncCommits = true;

this.ackOnError = true;

this.topics = null;

this.topicPattern = null;

Assert.notEmpty(topicPartitions, "An array of topicPartitions must be provided");

this.topicPartitions = (TopicPartitionInitialOffset[])(new LinkedHashSet(Arrays.asList(topicPartitions))).toArray(new TopicPartitionInitialOffset[topicPartitions.length]);

}

关注点私有属性

/**

* 确认模式(自动确认属性为false时使用)

*

*

1.RECORD逐条确认: 每条消息被发送给监听者后确认

*

2.BATCH批量确认: 当批量消息记录被消费者接收到并传送给监听器时确认

*

3.TIME超时确认:当超过设置的超时时间毫秒数时确认(should be greater than

* {@code #setPollTimeout(long) pollTimeout}.

*

4.COUNT计数确认: 当接收到指定数量之后确认

*

5.MANUAL手动确认：由监听器负责确认(AcknowledgingMessageListener)

*/

private AckMode ackMode;

//手动提交次数

private int ackCount;

//手动提交时间

private long ackTime;

//消息监听器，必须是 MessageListener或者AcknowledgingMessageListener两者中的一个

private Object messageListener;

//阻止消费者等待记录的最长时间。

private volatile long pollTimeout;

//线程执行器：轮询消费者

private AsyncListenableTaskExecutor consumerTaskExecutor;

//错误回调，当监听器抛出异常时

private GenericErrorHandler> errorHandler;

//停止容器超时时间

private long shutdownTimeout;

//用户定义的消费者再平衡监听器实现类

private ConsumerRebalanceListener consumerRebalanceListener;

//提交回调，默认记录日志。

private OffsetCommitCallback commitCallback;

private boolean syncCommits;

private boolean ackOnError;

private Long idleEventInterval;

private String groupId;

private PlatformTransactionManager transactionManager;

org.springframework.kafka.listener.KafkaMessageListenerContainer

关注点，构造函数

public KafkaMessageListenerContainer(ConsumerFactory consumerFactory, ContainerProperties containerProperties) {

this(consumerFactory, containerProperties, (TopicPartitionInitialOffset[])null);

}

略。。。。看不懂了。。