Flink自定义数据源（scala实现）

import org.apache.flink.streaming.api.scala._

case class SensorReading(id: String, timestamp: Long, temperature: Double)
//样例类对象
object Sensor {
  def main(args: Array[String]): Unit = {
    val env = StreamExecutionEnvironment.getExecutionEnvironment
    env.setParallelism(1)
    val stream1 = env.fromCollection(List(
     三个字段分别是传感器 id，时间戳，温度
      SensorReading("sensor_1", 1547718199, 35.80018327300259),
      SensorReading("sensor_6", 1547718201, 15.402984393403084),
      SensorReading("sensor_7", 1547718202, 6.720945201171228),
      SensorReading("sensor_10", 1547718205, 38.101067604893444)
    ))
    stream1.print("stream1:")

    env.execute()

  }

}

import java.util.Random

import org.apache.flink.streaming.api.functions.source.SourceFunction
import org.apache.flink.streaming.api.scala._
//实现类
object demo01 {

  def main(args: Array[String]): Unit = {

    val env = StreamExecutionEnvironment.getExecutionEnvironment

    env.setParallelism(1)

    val dataStream: DataStream[SensorReading] = env.addSource(new SensorSource())

    dataStream.print()

    env.execute("  job")
  }
}

//自定义数据源  继承样例对象
class SensorSource() extends SourceFunction[SensorReading]{

  // 定义一个flag：表示数据源是否还在正常运行
  var running:Boolean = true

  override def run(ctx: SourceFunction.SourceContext[SensorReading]): Unit = {

    // 创建一个随机数发生器，对温度的定义
    val random = new Random()

    // 随机初始换生成10个传感器的温度数据，之后在它基础随机波动生成流数据
    var curTemp = 1.to(10).map(
      i=>("sensor_" + i, 60 + random.nextGaussian() * 20)
    )

    // 无限循环生成流数据，除非被cancel
    while (running){
    
      curTemp = curTemp.map(

        // 更新温度值
        t => (t._1, t._2 + random.nextGaussian())
      )

      // 获取当前的时间戳
      val curTime = System.currentTimeMillis()

      // 包装成SensorReading，输出
      curTemp.foreach(
        t => ctx.collect( SensorReading(t._1, curTime, t._2) )
      )

      // 间隔时间
      Thread.sleep(1000)

    }
  }

  override def cancel(): Unit = running  = false

}