spark mr 文件统计

lyzx1,19
lyzx2,20
lyzx3,21
lyzx4,22
lyzx5,23
lyzx6,24
lyzx7,25
托塔天王
lyzx7,25,哈哈

package com.zxwa.live.process.test

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

object ScalaTs {


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

    val sparkContext = new SparkContext(new SparkConf().setAppName("ProductSalesStat").setMaster("local[*]"))
    val rdd = sparkContext.textFile("E:\\Data\\LIVE-DATA-SPARK\\src\\main\\resources\\people.txt")
    rdd.map(line => line.split(","))
      .map(rt => 
        if (rt.length == 1) rt(0)
        else if (rt.length == 2) (rt(0), rt(1))
        else (rt(0), rt(1), rt(2))
      )
      .map {
        case (one: String) => "one:" + one
        case (name: String, age: String) => ("name:" + name, "age:" + age)
        case _ => ("_name", "_age", "_")
      }
      .foreach(println)
  }
}