交并集合

 

 

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

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

    val conf: SparkConf = new SparkConf().setAppName("My scala word count").setMaster("local")
    val sc = new SparkContext(conf)

    val rdd1: RDD[Int] = sc.makeRDD(1 to 5)
    val rdd2: RDD[Int] = sc.makeRDD(4 to 10)

    val rdd5: RDD[Int] = rdd1.intersection(rdd2)
    val rdd3: RDD[Int] = rdd1.union(rdd2)
    val rdd4: RDD[Int] = rdd1.subtract(rdd2)

    rdd3.collect().foreach(println)
    rdd4.collect().foreach(println)
    rdd5.collect().foreach(println)
  }

}