Flink初体验 -- Word Count

Flink初体验 – Word Count

object WordCount {

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

    /**
      * flink 的运行模型
      * 1.DataSource
      * 2.Transformation
      * 3.DataSink
      */
    // 获取环境，类似于spark context
    val env = StreamExecutionEnvironment.getExecutionEnvironment

    val ip = "***.***.***.***"
    // 从socket创建DataStream
    // socket  DataSource
    val socketStream = env.socketTextStream(ip, 9000)

    // Transformation  数据转换操作
    val wordsStream = socketStream.flatMap(value => value.split("\\s+"))
      .map(value => (value, 1))

    val keyValuePair1: KeyedStream[(String, Int), Tuple] = wordsStream.keyBy(0)
    val countPair: DataStream[(String, Int)] = keyValuePair1.sum(1)

    // Data Sink
    countPair.print()

    println("==================== 窗口计算 ==========================")
    // 窗口计算 批次处理（每隔5秒计算一次）
    val keyValuePair2 = wordsStream.keyBy(0).timeWindow(Time.seconds(5))
    //
    val countStream = keyValuePair2.sum(1)

    countStream.print()
    // 触发执行程序
    env.execute(" Word Count ")

  }
}