Scala入门案例---Wordcount

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

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

    if (args.size != 2) {
      println(
        """
          |Usage:clw.spark.day01.ScalaWordCount
          |<srcPath>,<desPath>
        """.stripMargin)
      sys.exit(-1)
    }

    val Array(srcPath, desPath) = args

    //创建Spark配置,设置应用的名字
    val sparkConf = new SparkConf()
      .setAppName("ScalaWordCount")
      .setMaster("local[*]")
    //创建Spark的执行入口
    val sc = new SparkContext(sparkConf)

    //指定以后从哪里读取数据,创建RDD
    val data: RDD[String] = sc.textFile(srcPath)

    //切分压平
    val words: RDD[String] = data.flatMap(_.split(","))

    //将单词和1组装
    val wordAndOne: RDD[(String, Int)] = words.map((_, 1))

    //按key进行聚合
    val result: RDD[(String, Int)] = wordAndOne.reduceByKey(_ + _)

    //排序
    val resultSort: RDD[(String, Int)] = result.sortBy(-_._2)

    //将数据存入指定文件
    result.saveAsTextFile(desPath)


    //释放资源
    sc.stop()
  }

}