sparkStreaming的wordCount

1．需求：使用netcat工具向9999端口不断的发送数据，通过SparkStreaming读取端口数据并统计不同单词出现的次数
2．添加依赖

<dependency>
    <groupId>org.apache.spark</groupId>
    <artifactId>spark-streaming_2.11</artifactId>
    <version>2.1.1</version>
</dependency>

3．编写代码：这种wordCount是无状态的，只计算当前批次

```cpp
package com.atguigu

import org.apache.spark.streaming.dstream.{DStream, ReceiverInputDStream}
import org.apache.spark.streaming.{Seconds, StreamingContext}
import org.apache.spark.SparkConf

object StreamWordCount {

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

    //1.初始化Spark配置信息
    val sparkConf = new SparkConf().setMaster("local[*]").setAppName("StreamWordCount")

    //2.初始化SparkStreamingContext
    val ssc = new StreamingContext(sparkConf, Seconds(5))

    //3.通过监控端口创建DStream，读进来的数据为一行行
    val lineStreams = ssc.socketTextStream("hadoop102", 9999)

    //将每一行数据做切分，形成一个个单词
    val wordStreams = lineStreams.flatMap(_.split(" "))

    //将单词映射成元组（word,1）
    val wordAndOneStreams = wordStreams.map((_, 1))

    //将相同的单词次数做统计
    val wordAndCountStreams = wordAndOneStreams.reduceByKey(_+_)

    //打印
    wordAndCountStreams.print()

    //启动SparkStreamingContext
    ssc.start()
    ssc.awaitTermination()
  }
}

4．编写代码：有状态的wordCount

object WordCountV2 {

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

    val conf = new SparkConf().setMaster("local[*]").setAppName("sparkStreaming1217")
    val steamingContex = new StreamingContext(conf, Duration(5000))
    val dstream1: DStream[String] = steamingContex.socketTextStream("hadoop-101",999)
  
    val dstream2: DStream[String] = dstream1.flatMap(line => {
      val fields = line.split(" ")
      fields
    })
    steamingContex.sparkContext.setLogLevel("error")
    
    val wordsAndOneDstream: DStream[(String, Int)] = dstream2.map(t => (t, 1))

    val stateDstream: DStream[(String, Int)] = wordsAndOneDstream.updateStateByKey {
      //seq 当前批次相同key的values形成的序列
      //buffer 历时累加数据
      case (seq, buffer) => {
        var sum = buffer.getOrElse(0) + seq.sum
        //相当于把最新的sum给buffer
        Option(sum)
      }
    }
    stateDstream.print()
    
    //可以理解为启动采集器
    steamingContex.start()
    
    //等待采集器终止
    steamingContex.awaitTermination() //防止main方法结束程序结束  
    //同时因为要和采集器绑定所以while(true) 方法无效


  }

}