基于scala使用flink将kafka数据写入mysql示例

使用Flink消费Kafka中ChangeRecord主题的数据，统计每三分钟各设备状态为“预警”且未处理的数据总数。将结果存入MySQL的shtd_industry.threemin_warning_state_agg表（追加写入），表结构如下，同时备份到Hbase一份，表结构同MySQL表的。请在将任务启动命令截图，启动且数据进入后按照设备id倒序排序查询threemin_warning_state_agg表进行截图，第一次截图后等待三分钟再次查询并截图,将结果截图粘贴至对应报告中。

---

连接kafka

val kafkaSource=KafkaSource.builder()
      .setTopics("ChangeRecord")
      .setBootstrapServers("bigdata1:9092")
      .setValueOnlyDeserializer(new SimpleStringSchema())
      .setStartingOffsets(OffsetsInitializer.earliest())
      .build()

设置flink流处理环境

val env:StreamExecutionEnvironment=StreamExecutionEnvironment.getExecutionEnvironment
    env.setParallelism(1)

从kafka源创建数据流

val dataStream = env.fromSource(kafkaSource, WatermarkStrategy.noWatermarks(), "03")

• formSource ： 从外部数据源创建一个数据流
• kafkaSource ： kafka地址及配置
• WatermarkStrategy.noWatermarks() ： 不使用水印，用处理时间来处理数据
• 03 ： 船体给fromSource方法的第二个参数，表示消费组ID

指定kafka数据 并显示

 val transsformDataStream=dataStream
      .map(line => {
        val data = line.split(",")
        (data(1).toInt, data(3),data(6).toInt)
      })
      .filter(_._2 == "预警")
      .filter(_._3%2==0)
      .keyBy(_._1)
      .window(SlidingProcessingTimeWindows.of(Time.minutes(1),Time.seconds(1)))
      .min(0)
      .map(x=>SensorReading(x._1,x._3,new SimpleDateFormat("yyyy-MM-dd HH:mm:ss").format(new Date )))
    //                       获取当前系统时间
transsformDataStream.print()
transsformDataStream.addSink(new JDBCSink)

写入mysql

transsformDataStream.addSink(new JDBCSink)

执行flink程序

env.execute()

创建样例类

case class SensorReading(change_machine_id:Int,totalwarning:Int,window_end_time:String)

创建与MySQL连接方法的类

class JDBCSink() extends RichSinkFunction[SensorReading]{
  //定义sql连接、预编译器
  var conn: sql.Connection = _
  var insertStmt: sql.PreparedStatement = _
  var updateStmt: sql.PreparedStatement = _
  //_是一个占位符，表示该变量尚未被初始化

  override def open(parameters: Configuration): Unit = {
    super.open(parameters)
    conn=DriverManager.getConnection("jdbc:mysql://localhost:3306/databasename", "root", "密码")
    insertStmt = conn.prepareStatement("INSERT INTO threemin_warning_state_agg (change_machine_id, totalwarning, window_end_time) VALUES (?,?,?)")
    //updateStmt = conn.prepareStatement("UPDATE threemin_warning_state_agg SET totalwarning = ?,window_end_time=? WHERE change_machine_id = ?")
  }

//调用连接，执行sql
  override def invoke(value: SensorReading, context: SinkFunction.Context): Unit = {
    //执行插入语句
      insertStmt.setInt(1,value.change_machine_id)
      insertStmt.setInt(2,value.totalwarning)
      insertStmt.setString(3,value.window_end_time)
      insertStmt.execute()
  }

  //关闭时做清理工作
  override def close(): Unit = {
    insertStmt.close()
    updateStmt.close()
    conn.close()
  }
}

完整代码

import com.ibm.icu.text.SimpleDateFormat
import flink.g1.MyRedisMapper
import org.apache.flink.api.common.eventtime.WatermarkStrategy
import org.apache.flink.api.common.serialization.SimpleStringSchema
import org.apache.flink.api.scala.createTypeInformation
import org.apache.flink.configuration.Configuration
import org.apache.flink.connector.kafka.source.KafkaSource
import org.apache.flink.connector.kafka.source.enumerator.initializer.OffsetsInitializer
import org.apache.flink.streaming.api.functions.sink.{RichSinkFunction, SinkFunction}
import org.apache.flink.streaming.api.scala.StreamExecutionEnvironment
import org.apache.flink.streaming.api.windowing.assigners.SlidingProcessingTimeWindows
import org.apache.flink.streaming.api.windowing.time.Time

import java.sql. DriverManager
import java.sql
import java.util.Date

case class SensorReading(change_machine_id:Int,totalwarning:Int,window_end_time:String)

object g3 {
  def main(args: Array[String]): Unit = {
    val kafkaSource=KafkaSource.builder()
      .setTopics("ChangeRecord")
      .setBootstrapServers("bigdata1:9092")
      .setValueOnlyDeserializer(new SimpleStringSchema())
      .setStartingOffsets(OffsetsInitializer.earliest())
      .build()

    val env:StreamExecutionEnvironment=StreamExecutionEnvironment.getExecutionEnvironment
    env.setParallelism(1)

    // 实时读取Kafka数据,不设置水印:watermark,目的是使用处理时间
    //不设置水印，不生成时间戳 时间戳表示事件时间 不生成则使用处理时间
    val dataStream = env.fromSource(kafkaSource, WatermarkStrategy.noWatermarks(), "03")

    val transsformDataStream=dataStream
      .filter( _ != "==> /data_log/2024-01-23@20:34-changerecord.csv <==")
      .map(line => {
        val data = line.split(",")
        (data(1).toInt, data(3),data(6).toInt)
      })
      .filter(_._2 == "预警")
      .filter(_._3%2==0)
      .keyBy(_._1)
      .window(SlidingProcessingTimeWindows.of(Time.minutes(1),Time.seconds(1)))
      .min(0)
      .map(x=>SensorReading(x._1,x._3,new SimpleDateFormat("yyyy-MM-dd HH:mm:ss").format(new Date )))
    //                       获取当前系统时间
    transsformDataStream.print()
    transsformDataStream.addSink(new JDBCSink)
    //执行Flink程序
    env.execute()

  }
}

class JDBCSink() extends RichSinkFunction[SensorReading]{
  //定义sql连接、预编译器
  //_是一个占位符，表示该变量尚未被初始化
  var conn: sql.Connection = _
  var insertStmt: sql.PreparedStatement = _
  var updateStmt: sql.PreparedStatement = _

  override def open(parameters: Configuration): Unit = {
    super.open(parameters)
    conn=DriverManager.getConnection("jdbc:mysql://10.2.60.156:3306/shtd_industry", "root", "123456")
    insertStmt = conn.prepareStatement("INSERT INTO threemin_warning_state_agg (change_machine_id, totalwarning, window_end_time) VALUES (?,?,?)")
    updateStmt = conn.prepareStatement("UPDATE threemin_warning_state_agg SET totalwarning = ?,window_end_time=? WHERE change_machine_id = ?")
  }

//调用连接，执行sql
  override def invoke(value: SensorReading, context: SinkFunction.Context): Unit = {
    //执行插入语句
      insertStmt.setInt(1,value.change_machine_id)
      insertStmt.setInt(2,value.totalwarning)
      insertStmt.setString(3,value.window_end_time)
      insertStmt.execute()
  }

  //关闭时做清理工作
  override def close(): Unit = {
    insertStmt.close()
    updateStmt.close()
    conn.close()
  }
}