Scala 选取数据中的某几列

原始数据：

实现代码：

object avgAge {
  def main(args:Array[String]): Unit = {
    def FILE_NAME: String = "avgAge_results"
    val conf = new SparkConf().setMaster("local[*]").setAppName("MyFirstSpark")
    val sc = new SparkContext(conf);
    val textFile = transfer(sc, "/home/saprk/data/GDYKT/2016.1-2.csv")

   val nameCounts = textFile.flatMap(line=>line.split(" ")).map(line=>(line.split(",")(0),line.split(",")(18),line.split(",")(31))) //获取相应的列
    nameCounts.saveAsTextFile(FILE_NAME+System.currentTimeMillis())
  }

运行结果：