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

object work02 {
  def main(args: Array[String]): Unit = {
    val conf = new SparkConf().setAppName("localtion").setMaster("local[*]")
    val sc=new SparkContext(conf)
    //獲取資料
    val user1:RDD[String]=sc.textFile("E://aaa//testdata.txt",1)
    //1.統計Spark行數
    val result=user1.map((lines:String)=>if(lines.contains("Spark"))
      ("Spark",1) else (" ",0)).filter(_.equals(("Spark",1)))
    val result1=result.reduceByKey(_+_)
    //result1.foreach(println)
    //2.單詞最多的哪一行的單詞數
    var i:Int=0
    val result2=user1.map((lines:String)=>{
      val str=lines.split(" ").toList
      i=i+1
      (str,i)
    })
    val result3=result2.map(temp=>(temp._1.length,temp._2)).sortByKey(false).take(1)
    //result3.foreach(println)
  //3.統計資料中包含“a”的行數和包含“b”的行數
  val result4=user1.map((lines:String)=>if(lines.contains("a"))
    ("a",1) else (" ",0)).filter(_.equals(("a",1)))
    val result5=result4.reduceByKey(_+_)
    result5.foreach(println)

    val result6=user1.map((lines:String)=>if(lines.contains("b"))
      ("b",1) else (" ",0)).filter(_.equals(("b",1)))
    val result7=result6.reduceByKey(_+_)
    result7.foreach(println)
  }

}