低阶workcount

package tcode.chapter07
import scala.io.Source
object $17_WordCountLow {
  def main(args: Array[String]): Unit = {
    //1、读取文件
    val datas = Source.fromFile("datas/wc2.txt","utf-8").getLines().toList
    //List(hello hadoop flume kafka,kafka spark scala hadoop,hello java python hadoop,kafka flume spark spark,hello flume scala java)
    //2、切割+压平
    val words = datas.flatMap(line=> line.split(" "))
    //List(hello,hadoop,flume,kafka,kafka,spark,....)
    //3、按照单词分组
    val groupedMap = words.groupBy(x=>x)
    //Map(
    //    hello-> List(hello,hello，hello，hello，..)
    //    ...
    // )
    //
    //4、统计次数
    val result = groupedMap.map(x=>{
      //x = hello-> List(hello,hello，hello，hello，..)
      (x._1,x._2.size)
    })
    //List( (单词，总次数),(单词,总次数),... )
    result.foreach(x=>println(x))
    println("-"*100)
    //
    Source.fromFile("datas/wc.txt","utf-8").getLines().toList.flatMap(_.split(" ")).groupBy(x=>x).map(x=>(x._1,x._2.size)).foreach(println(_))
  }
}

高阶workcount

package tcode.chapter07
object $18_WordCountHight {
  def main(args: Array[String]): Unit = {
    val tupleList = List(("Hello Scala Spark World", 4), ("Hello Scala Spark", 3), ("Hello Scala", 2), ("Hello", 1))
    //1、切割+压平,给单词赋予初始次数
    val words = tupleList.flatMap(x=>{
      //x = ("Hello Scala Spark World", 4)
      //切割
      val arr = x._1.split(" ")
      //Array(Hello,Scala,Spark,Word)
      val tu = arr.map(y=>{
        //y = Hello
        (y,x._2)
      })
      tu
    })
    //List( (Hello,4),(Scala,4),(Spark,4),(Word,4),(Hello,3),(Scala,3),(Spark,3),(Hello,2),(Scala,2),(Hello,1) )
    //2、按照单词分组
    val groupedMap = words.groupBy(x=> x._1)
    //Map(
    //    Hello-> List(  (Hello,4), (Hello,3), (Hello,2), (Hello,1) )
    //    ....
    // )
    //3、统计单词个数
    val result = groupedMap.map(x=>{
      //x = Hello-> List(  (Hello,4), (Hello,3), (Hello,2), (Hello,1) )
      //val r = x._2.reduce((agg,curr)=> (agg._1, agg._2+curr._2))
      //r
      val r = x._2.map(y=>y._2).sum
      (x._1,  r )
    })
    //4、结果展示
    result.foreach(x=>println(x))
    //List((Hello,10),(Scala,9),(Spark,7),(World,4))
  }
}