Spark中使用RDD算子GroupBy做词频统计的方法

hello
world
java
world
java
java

import org.apache.spark.rdd.RDD
import org.apache.spark.{SparkConf, SparkContext}

object GroupBy {
  def main(args: Array[String]): Unit = {
  	// 创建Spark执行环境
    val sparkConf: SparkConf =
      new SparkConf().setMaster("local").setAppName("GroupBy")
    // 新建会话
    val sc = new SparkContext(sparkConf)
	// 读取本地文件到RDD
    val rdd: RDD[String] = sc.textFile("D://tmp/spark.txt")
    // 对rdd做map映射,返回(hello,1)...
    val rdd2: RDD[(String, Int)] = rdd.map(v => {
      val arr: Array[String] = v.split("\t")
      (arr(0), 1)
    })
	// 打印map映射结果
    rdd2.foreach(v=>println(v))

	// 对rdd2进行groupBy操作
    val rdd3: RDD[(String, Iterable[(String, Int)])] = rdd2.groupBy(v => v._1)
    // 遍历打印最终结果
    rdd3.map(v => (v._1, v._2.size)).foreach(v => println(v))
	//结束Spark会话
    sc.stop()
  }
}

(hello,1)
(world,1)
(java,1)
(world,1)
(java,1)
(java,1)

(hello,1)
(java,3)
(world,2)