flink kafka消费pojo类型数据实战详解

Posted 2022-09-28 bitcarmanlee

1.pojo类数据接口

在实际生产环境中，我们一般会将数据封装成一个pojo类(或者其他rpc框架通过IDL生成一个java类)，这样能方便我们后续的数据传输与解析。该pojo类就相当于标准数据接口，可以在任何地方被引用或者使用。下面我们来看看，怎么通过kafka的producer来生产这些pojo类数据，又怎么通过flink或者kafka的consumer来消费这些数据。

2.kafka producer生产数据

2.1 定义pojo类

首先，我们定义一个pojo类

public class User 
    public String name;
    public int age;

    public User() 

    public User(String name, int age) 
        this.name = name;
        this.age = age;
    

    public String getName() 
        return name;
    

    public void setName(String name) 
        this.name = name;
    

    public int getAge() 
        return age;
    

    public void setAge(int age) 
        this.age = age;
    

    @Override
    public String toString() 
        return "user" +
                "name='" + name + '\\'' +
                ", age=" + age +
                "";
    

我们定义了一个User类，该类有两个字段name与age。需要注意的是，该类必须要包含有默认的构造函数，否则后续代码使用过程中会出现问题，具体可以参考如下链接

https://stackoverflow.com/questions/7625783/jsonmappingexception-no-suitable-constructor-found-for-type-simple-type-class

2.2 定义序列化类

import org.apache.kafka.common.serialization.Serializer;
import org.codehaus.jackson.map.ObjectMapper;

import java.io.IOException;
import java.util.Map;

/**
 * author: wanglei
 * create: 2022-09-26
 */
public class UserSerializer implements Serializer<User> 

    @Override
    public void configure(Map configs, boolean isKey) 
    

    @Override
    public byte[] serialize(String topic, User user) 
        if (user == null) return null;
        ObjectMapper objectMapper = new ObjectMapper();
        try 
            return objectMapper.writeValueAsString(user).getBytes();
         catch (IOException e) 
            e.printStackTrace();
        
        return null;
    

    @Override
    public void close() 

    

上面类的作用，是将User对象序列化的过程，方便后面的数据传输。上面使用了ObjectMapper类进行序列化，需要引入如下依赖

        <dependency>
            <groupId>org.codehaus.jackson</groupId>
            <artifactId>jackson-mapper-asl</artifactId>
            <version>1.9.13</version>
        </dependency>

2.3 实现producer

import edu.bit.leilei.serialize.User;
import edu.bit.leilei.serialize.UserSerializer;
import org.apache.kafka.clients.producer.KafkaProducer;
import org.apache.kafka.clients.producer.ProducerConfig;
import org.apache.kafka.clients.producer.ProducerRecord;
import org.apache.kafka.common.serialization.StringSerializer;

import java.util.Properties;

/**
 * author: wanglei
 * create: 2022-09-26
 */
public class PojoProducer 

    public static void main(String[] args) 
        Properties properties = new Properties();
        properties.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG, "localhost:9092");
        properties.put(ProducerConfig.KEY_SERIALIZER_CLASS_CONFIG, StringSerializer.class.getName());
        properties.put(ProducerConfig.VALUE_SERIALIZER_CLASS_CONFIG, UserSerializer.class.getName());
        KafkaProducer<String, User> producer = new KafkaProducer<String, User>(properties);

        String topic = "pojotest";
        for(int i=0; i<5; i++) 
            User user = new User("my name-" + i, i);
            ProducerRecord<String, User> record = new ProducerRecord<String, User>(topic, "key-"+i, user);
            producer.send(record);
        

        producer.close();
    

上面的代码，生成了5个User对象。将代码先运行2次，总共生成了10个user对象。再将里面的一行稍作修改

            User user = new User("my name-" + i, i*10);

再运行一次，这样topic里面总共就有了15个对象。

3.kafka consumer消费数据

3.1 编写反序列化类

consumer要消费数据，首先需要做的是对数据进行反序列化。我们先来编写反序列化类代码

import org.apache.kafka.common.serialization.Deserializer;
import org.codehaus.jackson.map.ObjectMapper;

import java.io.IOException;
import java.util.Map;

/**
 * author: wanglei
 * create: 2022-09-26
 */
public class UserDeserializer implements Deserializer<User> 
    @Override
    public void configure(Map<String, ?> configs, boolean isKey) 
    

    @Override
    public User deserialize(String topic, byte[] data) 
        ObjectMapper objectMapper = new ObjectMapper();
        try 
            return objectMapper.readValue(data, User.class);
         catch (IOException e) 
            e.printStackTrace();
        
        return null;
    

    @Override
    public void close() 

    

上面同样用到的是ObjectMapper对byte[]数据进行反序列化。

3.2 kafka consumer代码

import edu.bit.leilei.serialize.User;
import edu.bit.leilei.serialize.UserDeserializer;
import org.apache.kafka.clients.consumer.ConsumerConfig;
import org.apache.kafka.clients.consumer.ConsumerRecord;
import org.apache.kafka.clients.consumer.ConsumerRecords;
import org.apache.kafka.clients.consumer.KafkaConsumer;
import org.apache.kafka.common.serialization.StringDeserializer;

import java.util.Arrays;
import java.util.Properties;

/**
 * author: wanglei
 * create: 2022-09-26
 */
public class PojoConsumer 
    public static void main(String[] args) 
        String topic = "pojotest";
        String groupId = "group_leilei";

        Properties props = new Properties();
        props.put(ConsumerConfig.BOOTSTRAP_SERVERS_CONFIG, "localhost:9092");
        props.put(ConsumerConfig.KEY_DESERIALIZER_CLASS_CONFIG, StringDeserializer.class.getName());
        props.put(ConsumerConfig.VALUE_DESERIALIZER_CLASS_CONFIG, UserDeserializer.class.getName());
        props.put(ConsumerConfig.GROUP_ID_CONFIG, groupId);
        props.put("auto.commit.interval.ms", "1000");
        props.put("session.timeout.ms", "30000");
        props.put("enable.auto.commit", "true");//设置为自动提交
        props.put("auto.offset.reset", "earliest");

        KafkaConsumer<String, User> consumer = new KafkaConsumer<String, User>(props);
        consumer.subscribe(Arrays.asList(topic));

        while(true) 
            ConsumerRecords<String, User> records = consumer.poll(1L);
            for(ConsumerRecord record : records) 
                System.out.printf("patition = %d , offset = %d, key = %s, value = %s%n",
                        record.partition(), record.offset(), record.key(), record.value());
            
        
    

上面的代码中，指定了反序列化类为UserDeserializer，并且从topic的最早位置开始进行消费。

最后输出的结果为

patition = 0 , offset = 0, key = key-0, value = username='my name-0', age=0
patition = 0 , offset = 1, key = key-1, value = username='my name-1', age=1
patition = 0 , offset = 2, key = key-2, value = username='my name-2', age=2
patition = 0 , offset = 3, key = key-3, value = username='my name-3', age=3
patition = 0 , offset = 4, key = key-4, value = username='my name-4', age=4
patition = 0 , offset = 5, key = key-0, value = username='my name-0', age=0
patition = 0 , offset = 6, key = key-1, value = username='my name-1', age=1
patition = 0 , offset = 7, key = key-2, value = username='my name-2', age=2
patition = 0 , offset = 8, key = key-3, value = username='my name-3', age=3
patition = 0 , offset = 9, key = key-4, value = username='my name-4', age=4
patition = 0 , offset = 10, key = key-0, value = username='my name-0', age=0
patition = 0 , offset = 11, key = key-1, value = username='my name-1', age=10
patition = 0 , offset = 12, key = key-2, value = username='my name-2', age=20
patition = 0 , offset = 13, key = key-3, value = username='my name-3', age=30
patition = 0 , offset = 14, key = key-4, value = username='my name-4', age=40

4.flink消费pojo类型数据

我们尝试使用flink来消费pojo类的数据。

4.1 反序列化类

同样的，首先也需要编写flink中反序列化的schema类。

import org.apache.flink.api.common.serialization.DeserializationSchema;
import org.apache.flink.api.common.typeinfo.TypeInformation;
import org.codehaus.jackson.map.ObjectMapper;

import java.io.IOException;

/**
 * author: wanglei
 * create: 2022-09-27
 */
public class UserFlinkDeserializer implements DeserializationSchema<User> 
    @Override
    public User deserialize(byte[] message) throws IOException 
        ObjectMapper objectMapper = new ObjectMapper();
        try 
            return objectMapper.readValue(message, User.class);
         catch (IOException e) 
            e.printStackTrace();
        
        return null;
    

    @Override
    public boolean isEndOfStream(User nextElement) 
        return false;
    

    @Override
    public TypeInformation<User> getProducedType() 
        return TypeInformation.of(User.class);
    

2.编写flink相关代码

import edu.bit.leilei.serialize.User;
import edu.bit.leilei.serialize.UserFlinkDeserializer;
import org.apache.flink.streaming.api.datastream.DataStream;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.streaming.connectors.kafka.FlinkKafkaConsumer09;

import java.util.Properties;

/**
 * author: wanglei
 * create: 2022-09-27
 */
public class StreamPojo 

    public static void main(String[] args) throws Exception 
        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        env.setParallelism(1);
        String topic = "pojotest";
        Properties properties = new Properties();
        properties.setProperty("bootstrap.servers", "localhost:9092");
        properties.put("group.id", topic);

        FlinkKafkaConsumer09<User> myConsumer = new FlinkKafkaConsumer09<User>(topic, new UserFlinkDeserializer(), properties);
        myConsumer.setStartFromEarliest();
        DataStream<User> stream = env.addSource(myConsumer);

        stream.print();
        env.execute();
    

代码运行以后得到的输出结果为

username='my name-0', age=0
username='my name-1', age=1
username='my name-2', age=2
username='my name-3', age=3
username='my name-4', age=4
username='my name-0', age=0
username='my name-1', age=1
username='my name-2', age=2
username='my name-3', age=3
username='my name-4', age=4
username='my name-0', age=0
username='my name-1', age=10
username='my name-2', age=20
username='my name-3', age=30
username='my name-4', age=40