相關閱讀:
Step1:創建項目
直接通過Spring 官方提供的 Spring Initializr 創建或者直接使用 IDEA 創建皆可。
Step2: 配置 Kafka
通過 application.yml 配置文件配置 Kafka 基本信息
server:
port: 9090
spring:
kafka:
consumer:
bootstrap-servers: localhost:9092
# 配置消費者消息offset是否自動重置(消費者重連會能夠接收最開始的消息)
auto-offset-reset: earliest
producer:
bootstrap-servers: localhost:9092
# 發送的對象信息變爲json格式
value-serializer: org.springframework.kafka.support.serializer.JsonSerializer
kafka:
topic:
my-topic: my-topic
my-topic2: my-topic2
Kafka 額外配置類:
package cn.javaguide.springbootkafka01sendobjects.config;
import org.apache.kafka.clients.admin.NewTopic;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.kafka.support.converter.RecordMessageConverter;
import org.springframework.kafka.support.converter.StringJsonMessageConverter;
@Configuration
public class KafkaConfig {
@Value("${kafka.topic.my-topic}")
String myTopic;
@Value("${kafka.topic.my-topic2}")
String myTopic2;
/**
* JSON消息轉換器
*/
@Bean
public RecordMessageConverter jsonConverter() {
return new StringJsonMessageConverter();
}
/**
* 通過注入一個 NewTopic 類型的 Bean 來創建 topic,如果 topic 已存在,則會忽略。
*/
@Bean
public NewTopic myTopic() {
return new NewTopic(myTopic, 2, (short) 1);
}
@Bean
public NewTopic myTopic2() {
return new NewTopic(myTopic2, 1, (short) 1);
}
}
當我們到了這一步之後,你就可以試着運行項目了,運行成功後你會發現 Spring Boot 會爲你創建兩個topic:
- my-topic: partition 數爲 2, replica 數爲 1
- my-topic2:partition 數爲 1, replica 數爲 1
“通過上一節說的:
”kafka-topics --describe --zookeeper zoo1:2181
命令查看或者直接通過IDEA 提供的 Kafka 可視化管理插件-Kafkalytic 來查看
Step3:創建要發送的消息實體類
package cn.javaguide.springbootkafka01sendobjects.entity;
public class Book {
private Long id;
private String name;
public Book() {
}
public Book(Long id, String name) {
this.id = id;
this.name = name;
}
省略 getter/setter以及 toString方法
}
Step4:創建發送消息的生產者
“這一步內容比較長,會一步一步優化生產者的代碼。
”
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.kafka.core.KafkaTemplate;
import org.springframework.stereotype.Service;
@Service
public class BookProducerService {
private static final Logger logger = LoggerFactory.getLogger(BookProducerService.class);
private final KafkaTemplate<String, Object> kafkaTemplate;
public BookProducerService(KafkaTemplate<String, Object> kafkaTemplate) {
this.kafkaTemplate = kafkaTemplate;
}
public void sendMessage(String topic, Object o) {
kafkaTemplate.send(topic, o);
}
}
我們使用Kafka 提供的 KafkaTemplate
調用 send()
方法出入要發往的topic和消息內容即可很方便的完成消息的發送:
kafkaTemplate.send(topic, o);
如果我們想要知道消息發送的結果的話,sendMessage
方法這樣寫:
public void sendMessage(String topic, Object o) {
try {
SendResult<String, Object> sendResult = kafkaTemplate.send(topic, o).get();
if (sendResult.getRecordMetadata() != null) {
logger.info("生產者成功發送消息到" + sendResult.getProducerRecord().topic() + "-> " + sendResult.getProducerRecord().value().toString());
}
} catch (InterruptedException | ExecutionException e) {
e.printStackTrace();
}
}
但是這種屬於同步的發送方式並不推薦,沒有利用到 Future
對象的特性。
KafkaTemplate
調用 send()
方法實際上返回的是ListenableFuture
對象。
send()
方法源碼如下:
@Override
public ListenableFuture<SendResult<K, V>> send(String topic, @Nullable V data) {
ProducerRecord<K, V> producerRecord = new ProducerRecord<>(topic, data);
return doSend(producerRecord);
}
ListenableFuture
是Spring提供了繼承自Future
的接口。
ListenableFuture
方法源碼如下:
public interface ListenableFuture<T> extends Future<T> {
void addCallback(ListenableFutureCallback<? super T> var1);
void addCallback(SuccessCallback<? super T> var1, FailureCallback var2);
default CompletableFuture<T> completable() {
CompletableFuture<T> completable = new DelegatingCompletableFuture(this);
this.addCallback(completable::complete, completable::completeExceptionally);
return completable;
}
}
繼續優化sendMessage
方法
public void sendMessage(String topic, Object o) {
ListenableFuture<SendResult<String, Object>> future = kafkaTemplate.send(topic, o);
future.addCallback(new ListenableFutureCallback<SendResult<String, Object>>() {
@Override
public void onSuccess(SendResult<String, Object> sendResult) {
logger.info("生產者成功發送消息到" + topic + "-> " + sendResult.getProducerRecord().value().toString());
}
@Override
public void onFailure(Throwable throwable) {
logger.error("生產者發送消息:{} 失敗,原因:{}", o.toString(), throwable.getMessage());
}
});
}
使用lambda表達式再繼續優化:
public void sendMessage(String topic, Object o) {
ListenableFuture<SendResult<String, Object>> future = kafkaTemplate.send(topic, o);
future.addCallback(result -> logger.info("生產者成功發送消息到topic:{} partition:{}的消息", result.getRecordMetadata().topic(), result.getRecordMetadata().partition()),
ex -> logger.error("生產者發送消失敗,原因:{}", ex.getMessage()));
}
再來簡單研究一下 send(String topic, @Nullable V data)
方法。
我們使用send(String topic, @Nullable V data)
方法的時候實際會new 一個ProducerRecord
對象發送,
@Override
public ListenableFuture<SendResult<K, V>> send(String topic, @Nullable V data) {
ProducerRecord<K, V> producerRecord = new ProducerRecord<>(topic, data);
return doSend(producerRecord);
}
ProducerRecord
類中有多個構造方法:
public ProducerRecord(String topic, V value) {
this(topic, null, null, null, value, null);
}
public ProducerRecord(String topic, Integer partition, Long timestamp, K key, V
......
}
如果我們想在發送的時候帶上timestamp(時間戳)、key等信息的話,sendMessage()
方法可以這樣寫:
public void sendMessage(String topic, Object o) {
// 分區編號最好爲 null,交給 kafka 自己去分配
ProducerRecord<String, Object> producerRecord = new ProducerRecord<>(topic, null, System.currentTimeMillis(), String.valueOf(o.hashCode()), o);
ListenableFuture<SendResult<String, Object>> future = kafkaTemplate.send(producerRecord);
future.addCallback(result -> logger.info("生產者成功發送消息到topic:{} partition:{}的消息", result.getRecordMetadata().topic(), result.getRecordMetadata().partition()),
ex -> logger.error("生產者發送消失敗,原因:{}", ex.getMessage()));
}
Step5:創建消費消息的消費者
通過在方法上使用 @KafkaListener
註解監聽消息,當有消息的時候就會通過 poll 下來消費。
import cn.javaguide.springbootkafka01sendobjects.entity.Book;
import com.fasterxml.jackson.core.JsonProcessingException;
import com.fasterxml.jackson.databind.ObjectMapper;
import org.apache.kafka.clients.consumer.ConsumerRecord;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.kafka.annotation.KafkaListener;
import org.springframework.stereotype.Service;
@Service
public class BookConsumerService {
@Value("${kafka.topic.my-topic}")
private String myTopic;
@Value("${kafka.topic.my-topic2}")
private String myTopic2;
private final Logger logger = LoggerFactory.getLogger(BookProducerService.class);
private final ObjectMapper objectMapper = new ObjectMapper();
@KafkaListener(topics = {"${kafka.topic.my-topic}"}, groupId = "group1")
public void consumeMessage(ConsumerRecord<String, String> bookConsumerRecord) {
try {
Book book = objectMapper.readValue(bookConsumerRecord.value(), Book.class);
logger.info("消費者消費topic:{} partition:{}的消息 -> {}", bookConsumerRecord.topic(), bookConsumerRecord.partition(), book.toString());
} catch (JsonProcessingException e) {
e.printStackTrace();
}
}
@KafkaListener(topics = {"${kafka.topic.my-topic2}"}, groupId = "group2")
public void consumeMessage2(Book book) {
logger.info("消費者消費{}的消息 -> {}", myTopic2, book.toString());
}
}
Step6:創建一個 Rest Controller
import cn.javaguide.springbootkafka01sendobjects.entity.Book;
import cn.javaguide.springbootkafka01sendobjects.service.BookProducerService;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.web.bind.annotation.PostMapping;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RequestParam;
import org.springframework.web.bind.annotation.RestController;
import java.util.concurrent.atomic.AtomicLong;
@RestController
@RequestMapping(value = "/book")
public class BookController {
@Value("${kafka.topic.my-topic}")
String myTopic;
@Value("${kafka.topic.my-topic2}")
String myTopic2;
private final BookProducerService producer;
private AtomicLong atomicLong = new AtomicLong();
BookController(BookProducerService producer) {
this.producer = producer;
}
@PostMapping
public void sendMessageToKafkaTopic(@RequestParam("name") String name) {
this.producer.sendMessage(myTopic, new Book(atomicLong.addAndGet(1), name));
this.producer.sendMessage(myTopic2, new Book(atomicLong.addAndGet(1), name));
}
}
Step7:測試
輸入命令:
curl -X POST -F 'name=Java' http://localhost:9090/book
控制檯打印出的效果如下:
my-topic 有2個partition(分區) 當你嘗試發送多條消息的時候,你會發現消息會被比較均勻地發送到每個 partion 中。