基于Spark下的查询数据库中信息Demo

import java.sql.DriverManager

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

object JDBCRDDDemo extends App {
  val conf = new SparkConf().setAppName("JDBCRDDDemo").setMaster("local[2]")
  val sc = new SparkContext(conf)

  val connection = ()=>{
    //获取驱动
    Class.forName("com.mysql.jdbc.Driver").newInstance()
    //建立连接
    DriverManager.getConnection("jdbc:mysql:///video","root","123456")
  }
 /* 源码class JdbcRDD[T](sc : org.apache.spark.SparkContext,
 getConnection : scala.Function0[java.sql.Connection],
  sql : scala.Predef.String,
  lowerBound : scala.Long,
   upperBound : scala.Long,
   numPartitions : scala.Int,
   mapRow : scala.Function1[java.sql.ResultSet, T] = {
   /* compiled code */ })
   (implicit evidence$1 : scala.reflect.ClassTag[T]) extends org.apache.spark.rdd.RDD[T] with org.apache.spark.Logging {
    override def getPartitions : scala.Array[org.apache.spark.Partition] = { /* compiled code */ }
    override def compute(thePart : org.apache.spark.Partition, context : org.apache.spark.TaskContext) : scala.Iterator[T] = { /* compiled code */ }
  }*/
  val jdbcRdd = new JdbcRDD(sc,
    connection,
  "select * from user where id >= ? and id <= ?",
    10,
    20,
    2,
    r =>{
      val id = r.getInt(1)
      val email = r.getString(2)
      (id,email)
    })

  val jrdd = jdbcRdd.collect()
  println(jrdd.toBuffer)
  sc.stop()
}