大数据系列篇-SPARK-SQL读写数据源

大数据系列篇-SPARK-SQL读写数据源

package com.test

import org.apache.spark.SparkConf
import org.apache.spark.sql.{SaveMode, SparkSession}

//练习read-write数据源
object SparkSqlRw {

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

    val sparkConf = new SparkConf().setAppName("练习SparkSqlRw").setMaster("local[*]")
    val spark = SparkSession.builder().config(sparkConf).getOrCreate()

    //读json1
    val df = spark.read.json("data/user.json")
    //用VIEW显示
    df.createOrReplaceTempView("user")
    spark.sql("SELECT * FROM user").show()
    //或读json2
    spark.sql("SELECT * FROM json.`data/user.json`").show()

    //写json
    df.write.format("json").mode(SaveMode.Overwrite).save("data/output/json")

    //默认写parquet
    df.write.mode(SaveMode.Overwrite).save("data/output/parquet")

    //读parquet
    val df2 = spark.read.load("data/user.parquet")
    val df3 = spark.read.parquet("data/user.parquet")
    //显示
    df2.show()
    df3.show()
    //用VIEW显示
    df2.createOrReplaceTempView("user2")
    spark.sql("SELECT age age2,userName userName2 FROM user2").show()
    df3.createOrReplaceTempView("user3")
    spark.sql("SELECT age as age3,userName as userName3 FROM user3").show()

    //jdbc
    //    val df1 = spark.read.format("jdbc")
    //      .option("url", "")
    //      .option("driver", "")
    //      .option("user", "")
    //      .option("password", "")
    //      .option("dbtable", "")
    //      .load()

    //    df1.write.format("jdbc")
    //      .option("url", "")
    //      .option("driver", "")
    //      .option("user", "")
    //      .option("password", "")
    //      .option("dbtable", "")
    //      .mode(SaveMode.Append)
    //      .save()

    spark.close()
  }
}