(6)FlinkSQL将kafka数据写入到mysql方式一

这里不展开zookeeper、kafka安装配置
(1)首先需要启动zookeeper和kafka
(2)定义一个kafka生产者
package com.producers;

import com.alibaba.fastjson.JSONObject;
import com.pojo.Event;
import com.pojo.WaterSensor;
import org.apache.kafka.clients.producer.KafkaProducer;
import org.apache.kafka.clients.producer.ProducerRecord;
import org.apache.kafka.clients.producer.RecordMetadata;
import org.apache.kafka.common.serialization.StringSerializer;

import java.util.Properties;
import java.util.Random;

/**
 * Created by lj on 2022-07-09.
 */
public class Kafaka_Producer {
    public final static String bootstrapServers = "127.0.0.1:9092";

    public static void main(String[] args) {
        Properties props = new Properties();
        //设置Kafka服务器地址
        props.put("bootstrap.servers", bootstrapServers);
        //设置数据key的序列化处理类
        props.put("key.serializer", "org.apache.kafka.common.serialization.StringSerializer");
        //设置数据value的序列化处理类
        props.put("value.serializer", "org.apache.kafka.common.serialization.StringSerializer");
        KafkaProducer<String, String> producer = new KafkaProducer<>(props);

        try {
            int i = 0;
            Random r=new Random();   //不传入种子
            String[] lang = {"flink","spark","hadoop","hive","hbase","impala","presto","superset","nbi"};

            while(true) {
                Thread.sleep(2000);
                WaterSensor waterSensor = new WaterSensor(lang[r.nextInt(lang.length)],i,i);
                i++;

                String msg = JSONObject.toJSONString(waterSensor);
                System.out.println(msg);
                RecordMetadata recordMetadata = producer.send(new ProducerRecord<>("kafka_data_waterSensor", null, null,  msg)).get();
//                System.out.println("recordMetadata: {"+ recordMetadata +"}");
            }

        } catch (Exception e) {
            System.out.println(e.getMessage());
        }
    }
}
(3)定义一个消息对象
package com.pojo;

import java.io.Serializable;

/**
 * Created by lj on 2022-07-05.
 */
public class WaterSensor implements Serializable {
    private String id;
    private long ts;
    private int vc;

    public WaterSensor(){

    }

    public WaterSensor(String id,long ts,int vc){
        this.id = id;
        this.ts = ts;
        this.vc = vc;
    }

    public int getVc() {
        return vc;
    }

    public void setVc(int vc) {
        this.vc = vc;
    }

    public String getId() {
        return id;
    }

    public void setId(String id) {
        this.id = id;
    }

    public long getTs() {
        return ts;
    }

    public void setTs(long ts) {
        this.ts = ts;
    }
}
(4)从kafka接入数据,并写入到mysql
    public static void main(String[] args) throws Exception {

        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        env.setParallelism(1);
        StreamTableEnvironment tableEnv = StreamTableEnvironment.create(env);

        //读取kafka的数据
        Properties properties = new Properties();
        properties.setProperty("bootstrap.servers","127.0.0.1:9092");
        properties.setProperty("group.id", "consumer-group");
        properties.setProperty("key.deserializer", "org.apache.kafka.common.serialization.StringDeserializer");
        properties.setProperty("value.deserializer", "org.apache.kafka.common.serialization.StringDeserializer");
        properties.setProperty("auto.offset.reset", "latest");

        DataStreamSource<String> streamSource = env.addSource(
                new FlinkKafkaConsumer<String>(
                        "kafka_waterSensor",
                        new SimpleStringSchema(),
                        properties)
        );

        SingleOutputStreamOperator<WaterSensor> waterDS = streamSource.map(new MapFunction<String, WaterSensor>() {
            @Override
            public WaterSensor map(String s) throws Exception {
                JSONObject json  = (JSONObject)JSONObject.parse(s);
                return new WaterSensor(json.getString("id"),json.getLong("ts"),json.getInteger("vc"));
            }
        });

        // 将流转化为表
        Table table = tableEnv.fromDataStream(waterDS,
                $("id"),
                $("ts"),
                $("vc"),
                $("pt").proctime());

        tableEnv.createTemporaryView("EventTable", table);


        tableEnv.executeSql("CREATE TABLE flinksink (" +
                "componentname STRING," +
                "componentcount BIGINT NOT NULL," +
                "componentsum BIGINT" +
                ") WITH (" +
                "'connector.type' = 'jdbc'," +
                "'connector.url' = 'jdbc:mysql://localhost:3306/testdb?characterEncoding=UTF-8&useUnicode=true&useSSL=false&tinyInt1isBit=false&allowPublicKeyRetrieval=true&serverTimezone=Asia/Shanghai'," +
                "'connector.table' = 'flinksink'," +
                "'connector.driver' =  'com.mysql.cj.jdbc.Driver'," +
                "'connector.username' = 'root'," +
                "'connector.password' = 'root'," +
                "'connector.write.flush.max-rows'='3'\r\n" +
                ")"
        );
        Table mysql_user = tableEnv.from("flinksink");
        mysql_user.printSchema();

        Table result = tableEnv.sqlQuery(
                "SELECT " +
                        "id as componentname, " +                //window_start, window_end,
                        "COUNT(ts) as componentcount ,SUM(ts) as componentsum " +
                        "FROM TABLE( " +
                        "TUMBLE( TABLE EventTable , " +
                        "DESCRIPTOR(pt), " +
                        "INTERVAL '10' SECOND)) " +
                        "GROUP BY id , window_start, window_end"
        );

        //方式一:写入数据库
//        result.executeInsert("flinksink").print(); //;.insertInto("flinksink");

        //方式二:写入数据库
        tableEnv.createTemporaryView("ResultTable", result);
        tableEnv.executeSql("insert into flinksink SELECT * FROM ResultTable").print();

//        tableEnv.toAppendStream(result, Row.class).print("toAppendStream");           //追加模式
        env.execute();

    }
(5)效果演示

 

posted @ 2022-08-06 16:34  NBI大数据可视化分析  阅读(285)  评论(0编辑  收藏  举报