Spark统及单词数

val rdd=sc.makeRDD(List("hello world","hello spark","hello flume","helle hadoop"))
    val rdd_r=rdd.flatMap(_.split(" ")).map(x=>(x,1)).reduceByKey((x,y)=>x+1) //理解明白x=>x是含义，将函数给拆分
    println(rdd_r.collect.mkString("\t"))