flink讀取數據源的四種方式

object SourceDemo {
  def main(args: Array[String]): Unit = {
    val env = StreamExecutionEnvironment.getExecutionEnvironment

    //數據來源
    //1.從文件讀取
    val inpath="D:\\programs\\sparkPrograms\\FlinkProgarm\\src\\main\\resources\\hello.txt"
    val stream1 =  env.readTextFile(inpath)

    //2.從socket流中讀取
    val stream2 =env.socketTextStream("hadoop01",7777)

    //3.從kafka中讀取
    val properties=new Properties()
    properties.setProperty("bootstrap.servers","hadoop01:9092")
    properties.setProperty("group.id","cosumer-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")
    val stream3 = env.addSource(new FlinkKafkaConsumer011[String]("sensor", new
        SimpleStringSchema(), properties))
    //    stream3.print("stream3")
    
    //4.自定義Source
    val stream4: DataStream[SensorReading] = env.addSource(new MySensorSource())
    stream4.print("Stream4")
    
    env.execute()
  }

  case class SensorReading(id: String, timestamp: Long, temperature: Double)

  class MySensorSource extends SourceFunction[SensorReading]{
    //flag 表示數據源是否正常運行
    var running:Boolean = true

    override def run(sourceContext: SourceFunction.SourceContext[SensorReading]): Unit = {
      //初始化一個隨機數發生器
      val rand = new Random()

      var curTemp=1.to(10).map(
        //rand.nextGaussian() 正太分佈隨機數 範圍在正負2seigama之間
        i=>("sensor_"+i,65+rand.nextGaussian()*20)
      )

      while (running){
        //更新溫度值
        curTemp = curTemp.map(
          t=>(t._1,t._2+rand.nextGaussian())
        )
        //獲取當前時間戳
        val curTime = System.currentTimeMillis()

        curTemp.foreach(
          //使用collect方法 將數據一條一條發送出去
          t=>sourceContext.collect(SensorReading(t._1,curTime,t._2))
        )
        Thread.sleep(100)
      }
    }

    override def cancel(): Unit = {
      running=false
    }
  }
}