Flink--Source(集合，文件)

从集合、文件中读取数据

基于Flink快速，灵巧，实时性高的特点以温度传感器的数据采集为场景做练习。

------集合------

第一步:定义传感器样例类

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

第二步:创建执行环境（流式数据）

val env = StreamExecutionEnvironment.getExecutionEnvironment

第三步:定义数据源集合

val sensorList = List (
      SensorReading("Sensor_1", 1547718199, 36.0),
      SensorReading("Sensor_2", 1547718257, 35.2),
      SensorReading("Sensor_3", 1547718358, 34.8),
      SensorReading("Sensor_4", 1547718455, 32.8),
      SensorReading("Sensor_5", 1547718587, 32.5),
      SensorReading("Sensor_6", 1547718698, 31.3),
      SensorReading("Sensor_7", 1547718785, 37.9)
      )

第四步:从集合中读取数据

val sensorData: DataStream[SensorReading] = env.fromCollection(sensorList)

//也可以使用fromelements方法
val sensorData: DataStream[SensorReading] = env.fromElements(
        SensorReading("Sensor_1", 1547718199, 36.0),
        SensorReading("Sensor_2", 1547718257, 35.2),
        SensorReading("Sensor_3", 1547718358, 34.8),
        SensorReading("Sensor_4", 1547718455, 32.8),
        SensorReading("Sensor_5", 1547718587, 32.5),
        SensorReading("Sensor_6", 1547718698, 31.3),
        SensorReading("Sensor_7", 1547718785, 37.9))

fromCollection参数为包含数据的一个seq.
fromElements参数直接为数据。这些数据可以是不同类型的，可用于测试，但是不常用

第五步:打印

sensorData.print()

第六步:执行

env.execute("apitest")

结果：
2> SensorReading(Sensor_1,1547718199,36.0)
3> SensorReading(Sensor_2,1547718257,35.2)
2> SensorReading(Sensor_5,1547718587,32.5)
3> SensorReading(Sensor_6,1547718698,31.3)
1> SensorReading(Sensor_4,1547718455,32.8)
4> SensorReading(Sensor_3,1547718358,34.8)
4> SensorReading(Sensor_7,1547718785,37.9)

虽然代码中只对数据进行了读取和打印，依然需要进行隐式转换
因为里面设计了typeinformation的隐式转换

因为数据源为有界流数据处理完会直接结束

------文件------

第一步:创建执行环境（流式数据）
第二步:创建数据源文件
source下创建txt文件
“Sensor_1”, 1547718199, 36.0
“Sensor_2”, 1547718257, 35.2
“Sensor_3”, 1547718358, 34.8
“Sensor_4”, 1547718455, 32.8
“Sensor_5”, 1547718587, 32.5
“Sensor_6”, 1547718698, 31.3
“Sensor_7”, 1547718785, 37.9
第三步:读取文件

val inputpath = "F:\\idea1\\flinktest\\src\\main\\resources\\sensorfile.txt"
val sensorData2: DataStream[String] = env.readTextFile(inputpath)

第四步:打印
第五步:执行
完整代码

package com.erke.apitest

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

//定义样例类
case class SensorReading(id:String,timeStamp:Long,temperature:Double)

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

    //创建执行环境
    val env: StreamExecutionEnvironment = StreamExecutionEnvironment.getExecutionEnvironment

    //定义数据源集合
    val sensorList = List (
      SensorReading("Sensor_1", 1547718199, 36.0),
      SensorReading("Sensor_2", 1547718257, 35.2),
      SensorReading("Sensor_3", 1547718358, 34.8),
      SensorReading("Sensor_4", 1547718455, 32.8),
      SensorReading("Sensor_5", 1547718587, 32.5),
      SensorReading("Sensor_6", 1547718698, 31.3),
      SensorReading("Sensor_7", 1547718785, 37.9)
      )

    //读取集合中数据
    val sensorData: DataStream[SensorReading] = env.fromCollection(sensorList)

    //读取文件中数据
    val inputpath = "F:\\idea1\\flinktest\\src\\main\\resources\\sensorfile.txt"
    val sensorData2: DataStream[String] = env.readTextFile(inputpath)

    //打印
    sensorData.print()
    sensorData2.print()

    //执行
    env.execute("apitest")
  }
}