Java整合Kafka实现生产及消费

时间：09-07来源：作者：点击数：26

前提条件

搭建Kafka环境，参考Kafka集群环境搭建及使用
Java环境：JDK1.8
Maven版本：apache-maven-3.6.3
开发工具：IntelliJ IDEA项目环境
创建maven项目。
pom.xml文件中引入kafka依赖。

代码语言：xml

<dependencies>
        <dependency>
            <groupId>org.apache.kafka</groupId>
            <artifactId>kafka\_2.11</artifactId>
            <version>2.1.0</version>
        </dependency>
</dependencies>

创建Topic创建topic命名为testtopic并指定2个分区。

代码语言：bash

./kafka-topics.sh --bootstrap-server 127.0.0.1:9092 --create --topic testtopic --partitions 2

生产消息

代码语言：java

public class Producer {
    public static void main(String[] args) throws ExecutionException, InterruptedException {
        // 生产参数配置
        Properties properties = new Properties();
        properties.setProperty(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG, "localhost:9092");
        properties.setProperty(ProducerConfig.KEY_SERIALIZER_CLASS_CONFIG, StringSerializer.class.getName());
        properties.setProperty(ProducerConfig.VALUE_SERIALIZER_CLASS_CONFIG, StringSerializer.class.getName());
        KafkaProducer<String, String> kafkaProducer = new KafkaProducer<String, String>(properties);
        int i=0;
        while (true) {
            //生产消息
            Future<RecordMetadata> future = kafkaProducer.send(new ProducerRecord<String, String>("testtopic", "key"+i, "value"+i));
            //获取生产的数据信息
            RecordMetadata recordMetadata = future.get();
            System.out.println("time:"+recordMetadata.timestamp()+" key:"+i+" value:"+i+" partition:"+recordMetadata.partition()+" offset:"+recordMetadata.offset());
            Thread.sleep(1000);
            i+=1;
        }
    }
}

生产者参数配置

代码语言：java

// ACK机制，默认为1 (0,1,-1)
properties.setProperty(ProducerConfig.ACKS_CONFIG, "");
// Socket发送消息缓冲区大小，默认为128K，设置为-1代表操作系统的默认值
properties.setProperty(ProducerConfig.SEND_BUFFER_CONFIG, ""); 
// Socket接收消息缓冲区大小，默认为32K，设置为-1代表操作系统的默认值
properties.setProperty(ProducerConfig.RECEIVE_BUFFER_CONFIG, ""); 
// 生产者客户端发送消息的最大值，默认1M
properties.setProperty(ProducerConfig.MAX_REQUEST_SIZE_CONFIG, ""); 
// 发送消息异常时重试次数，默认为0
properties.setProperty(ProducerConfig.RETRIES_CONFIG, "");   
// 重试间隔时间，默认100
properties.setProperty(ProducerConfig.RETRY_BACKOFF_MS_CONFIG, "");    
// 生产消息自定义分区策略类
properties.setProperty(ProducerConfig.PARTITIONER_CLASS_CONFIG, "");
// 开启幂等 ,默认true
properties.setProperty(ProducerConfig.ENABLE_IDEMPOTENCE_CONFIG, "");

更多配置信息查看ProducerConfig类

生产自定义分区策略

创建分区策略类，实现org.apache.kafka.clients.producer.Partitioner接口，编写具体策略。

代码语言：java

public class PartitionPolicy implements Partitioner {

    private final ConcurrentMap<String, AtomicInteger> topicCounterMap = new ConcurrentHashMap();

    @Override
    public int partition(String topic, Object key, byte[] keyBytes, Object value, byte[] valueBytes, Cluster cluster) {
        List<PartitionInfo> partitions = cluster.partitionsForTopic(topic);
        int numPartitions = partitions.size();
        if (keyBytes == null) {
            int nextValue = this.nextValue(topic);
            List<PartitionInfo> availablePartitions = cluster.availablePartitionsForTopic(topic);
            if (availablePartitions.size() > 0) {
                int part = Utils.toPositive(nextValue) % availablePartitions.size();
                return ((PartitionInfo)availablePartitions.get(part)).partition();
            } else {
                return Utils.toPositive(nextValue) % numPartitions;
            }
        } else {
            return Utils.toPositive(Utils.murmur2(keyBytes)) % numPartitions;
        }
    }


    private int nextValue(String topic) {
        AtomicInteger counter = (AtomicInteger)this.topicCounterMap.get(topic);
        if (null == counter) {
            counter = new AtomicInteger(ThreadLocalRandom.current().nextInt());
            AtomicInteger currentCounter = (AtomicInteger)this.topicCounterMap.putIfAbsent(topic, counter);
            if (currentCounter != null) {
                counter = currentCounter;
            }
        }

        return counter.getAndIncrement();
    }

    @Override
    public void close() {

    }

    @Override
    public void configure(Map<String, ?> map) {

    }
}

参数配置。

代码语言：java

properties.setProperty(ProducerConfig.PARTITIONER_CLASS_CONFIG, PartitionPolicy.class.getName());

生产到指定分区

ProducerRecord有指定分区的构造方法，设置分区号

public ProducerRecord(String topic, Integer partition, K key, V value)。

代码语言：java

Future<RecordMetadata> future = kafkaProducer.send(new ProducerRecord<String, String>("testtopic", 1, "key"+i, "value"+i));

消费消息

代码语言：java

public class Consumer {
    public static void main(String[] args) throws InterruptedException {
        Properties properties = new Properties();
        properties.setProperty(ConsumerConfig.BOOTSTRAP_SERVERS_CONFIG, "localhost:9092");
        //约定的编解码
        properties.setProperty(ConsumerConfig.KEY_DESERIALIZER_CLASS_CONFIG, StringDeserializer.class.getName());
        properties.setProperty(ConsumerConfig.VALUE_DESERIALIZER_CLASS_CONFIG, StringDeserializer.class.getName());
        properties.setProperty(ConsumerConfig.GROUP_ID_CONFIG, "test_group");
        //默认为自动提交
        properties.setProperty(ConsumerConfig.ENABLE_AUTO_COMMIT_CONFIG, "false");
        //当设置为自动提交时，默认5秒自动提交
        //properties.setProperty(ConsumerConfig.AUTO_COMMIT_INTERVAL_MS_CONFIG, "5000");
        //
        //properties.setProperty(ConsumerConfig.AUTO_OFFSET_RESET_CONFIG, "5000");
        KafkaConsumer<String, String> kafkaConsumer = new KafkaConsumer<String, String>(properties);
        //订阅topic
        kafkaConsumer.subscribe(Arrays.asList("testtopic"));
        Set<TopicPartition> assignment = kafkaConsumer.assignment();
        ConsumerRecords<String, String> records = null;
        while (assignment.size() == 0) {
            records = kafkaConsumer.poll(Duration.ofMillis(100));
            assignment = kafkaConsumer.assignment();
        }
        /*//1.根据时间戳获取 offset，设置 offset
        Map<TopicPartition, Long> offsetsForTimes=new HashMap<>();
        for (TopicPartition topicPartition : assignment) {
            offsetsForTimes.put(topicPartition,1669972273941L);
        }
        Map<TopicPartition, OffsetAndTimestamp> offsetAndTimestampMap = kafkaConsumer.offsetsForTimes(offsetsForTimes);
        offsetAndTimestampMap.forEach((tp,offsettime)->{
            kafkaConsumer.seek(tp,offsettime.offset());
        });*/
        /*//2.指定从头开始消费
        kafkaConsumer.seekToBeginning(assignment);*/
        /*//3.指定从某offset开始消费
        kafkaConsumer.seek(tp,0);*/
        while (true) {
            if (records.isEmpty()) {
                Thread.sleep(3000);
            } else {
                System.out.printf("records count:" + records.count());
                Iterator<ConsumerRecord<String, String>> iterator = records.iterator();
                while (iterator.hasNext()) {
                    ConsumerRecord<String, String> record = iterator.next();
                    System.out.println(" time:" + record.timestamp() + " key:" + record.key() + " value:" + record.value() + " partition:" + record.partition() + " offset:" + record.offset());
                }
                kafkaConsumer.commitSync();
            }
            records = kafkaConsumer.poll(Duration.ofMillis(0));
        }
    }
}

消费参数配置

代码语言：java

// 消费者必须指定一个消费组
properties.setProperty(ConsumerConfig.GROUP_ID_CONFIG, "");
// 消费者每次最多POLL的数量
properties.setProperty(ConsumerConfig.MAX_POLL_RECORDS_CONFIG, "");
// 消费者POLL的时间间隔
properties.setProperty(ConsumerConfig.MAX_POLL_INTERVAL_MS_DOC, "");
// 设置是否自动提交，默认为true
properties.setProperty(ConsumerConfig.ENABLE_AUTO_COMMIT_CONFIG, "");  
// 如果是自动提交，默认5s后提交，会发生丢失消息和重复消费情况
properties.setProperty(ConsumerConfig.AUTO_COMMIT_INTERVAL_MS_CONFIG, "");   
// 当一个新的消费组或者消费信息丢失后，在哪里开始进行消费。earliest：消费最早的消息。latest（默认）：消费最近可用的消息。none：没有找到消费组消费数据时报异常。