flink-7 DataFrame转换算子

前言

• 通过从一个或多个 DataStream 生成新的 DataStream 的过程被称为 Transformation 操作。
• 在转换过程中，每种操作类型被定义为不同的 Operator
• Flink 程序能够将多个 Transformation 组成一个 DataFlow 的拓扑

map、filter

import org.apache.flink.streaming.api.scala.{
   DataStream, StreamExecutionEnvironment}

object map_Filter {
   
  def main(args: Array[String]): Unit = {
   
    val environment: StreamExecutionEnvironment = StreamExecutionEnvironment.getExecutionEnvironment
    import  org.apache.flink.api.scala._
    val sourceStream: DataStream[Int] = environment.fromElements(1,2,3,4,5,6)

    val mapStream: DataStream[Int] = sourceStream.map(x =>x*10)
    val resultStream: DataStream[Int] = mapStream.filter(x => x%3 ==0)
    resultStream.print()

    environment.execute()
  }

}

输出

flatMap

object flatMap_keyBy_sum {
   
  def main(args: Array[String]): Unit = {
   
    //获取程序入口类
    val environment: StreamExecutionEnvironment = StreamExecutionEnvironment.getExecutionEnvironment
    //从socket当中获取数据
    val resultDataStream: DataStream[String] = environment.socketTextStream("node01",9999)
    //导入隐式转换的包
    import org.apache.flink.api.scala._
    val value: DataStream[(String, Int)] = resultDataStream .flatMap(x => x.split(" ")).map(x => (x, 1))
    value.print()
    //执行程序
    environment.execute()
  }
}

输入

输出