DataSet转RDD

调用rdd方法即可

import org.apache.spark.rdd.RDD
import org.apache.spark.sql.SparkSession
object DS2RDD {
  def main(args: Array[String]): Unit = {
    val spark: SparkSession = SparkSession
      .builder()
      .master("local[*]")
      .appName("DS2RDD")
      .getOrCreate()
    import spark.implicits._
    val ds = Seq(User("lisi", 40), User("zs", 20)).toDS
    val rdd: RDD[User] = ds.rdd
    rdd.collect.foreach(println)
    spark.stop()
  }
}

输出

User(lisi,40)
User(zs,20)

RDD转DataSet

调用toDS方法

import org.apache.spark.sql.{Dataset, SparkSession}
// 声明一个实体类
case class User(name: String, age: Int)
object RDD2DS {
  def main(args: Array[String]): Unit = {
    val spark: SparkSession = SparkSession
      .builder()
      .master("local[*]")
      .appName("RDD2DS")
      .getOrCreate()
    import spark.implicits._
    //从集合中创建rdd
    val rdd = spark.sparkContext
      .parallelize(Array(User("lisi", 10), User("zs", 20), User("ww", 15)))
    val ds: Dataset[User] = rdd.toDS()
    ds.show()
    spark.close()
  }
}

输出

+----+---+
|name|age|
+----+---+
|lisi| 10|
|  zs| 20|
|  ww| 15|
+----+---+