spark sql 之 RDD与DataFrame互相转化

一、RDD转DataFrame

　　方法一：通过 case class 创建 DataFrames

　　

import org.apache.spark.SparkConf
import org.apache.spark.SparkContext
import org.apache.spark.sql.sqlContext


object TestDataFrame {
  
  def main(args: Array[String]): Unit = {
    
    /**
     * 1、初始化 spark config
     */
    val conf = new SparkConf().setAppName("TestDataFrame").setMaster("local");    
    /**
     * 2、初始化spark context
     */
    val sc = new SparkContext(conf);
    
    /**
     * 3、初始化spark sql context
     */
    val ssc = new sqlContext(sc);
    
    /**
     * 4、做spark sql 的df获取工作
     */
    val PeopleRDD = sc.textFile("F:\\input.txt").map(line => People(line.split(" ")(0),line.split(" ")(1).trim.toInt))
    
    import ssc.implicits._
    
    var df = PeopleRDD.toDF
    
    //将DataFrame注册成临时的一张表，这张表相当于临时注册到内存中，是逻辑上的表，不会物化到磁盘  这种方式用的比较多
    df.registerTempTable("peopel")
    
    var df2 =ssc.sql("select * from peopel where age > 23")show()
    
    /**
     * 5、spark context 结束工作
     */
    sc.stop();
    
  }
}
case class People(var name:String ,var age : Int)

　　方法二：通过 structType创建 DataFrames

import org.apache.spark.SparkConf
import org.apache.spark.SparkContext
import org.apache.spark.sql.sqlContext
import org.apache.spark.sql.DataFrame
import org.apache.spark.sql.Row
import org.apache.spark.sql.types.{StructType,StructField,StringType,IntegerType}

object TestDataFrame2{
  def test2(): Unit = {
    /**
     * 1、初始化 spark config
     */
    val conf = new SparkConf().setAppName("TestDataFrame").setMaster("local");    
    /**
     * 2、初始化spark context
     */
    val sc = new SparkContext(conf);
    
    /**
     * 3、初始化spark sql context
     */
    val ssc = new sqlContext(sc);
    
    /**
     * 4、做spark sql 的df获取工作
     */
    val peopleRDD = sc.textFile("F:\\input.txt")map(line => 
      Row(line.split(" ")(0),line.split(" ")(1).trim().toInt))
   
    // 创建 StructType 来定义结构
    val structType : StructType = StructType(
        StructField("name",StringType,true)::
        StructField("age",IntegerType,true) ::Nil       
    );
    
    val df : DataFrame = ssc.createDataFrame(peopleRDD, structType);
    df.registerTempTable("peopel");
    
    ssc.sql("select * from peopel").show();
    
     /**
     * 5、spark context 结束工作
     */
    sc.stop();
  }  
}

　　方法三:通过json创建 DataFream

import org.apache.spark.SparkConf
import org.apache.spark.SparkContext
import org.apache.spark.sql.sqlContext
import org.apache.spark.sql.DataFrame
import org.apache.spark.sql.Row
import org.apache.spark.sql.types.{StructType,StructField,StringType,IntegerType}
import org.apache.spark.sql.DataFrame

object TestDataFrame2{
  def test3() : Unit={
    /**
     * 1、初始化 spark config
     */
    val conf = new SparkConf().setAppName("TestDataFrame").setMaster("local");    
    /**
     * 2、初始化spark context
     */
    val sc = new SparkContext(conf);
    
    /**
     * 3、初始化spark sql context
     */
    val ssc = new sqlContext(sc);
    
    /**
     * 4、做spark sql 的df获取工作
     */
    val df :DataFrame = ssc.read.json("F:\\json.json")
    df.registerTempTable("people")
    ssc.sql("select * from people").show();
    
     /**
     * 5、spark context 结束工作
     */
    sc.stop();
  }
}

二、RDD转DataFrame

df.rdd