Spark core使用JDBC连接MySQL

添加依赖

<dependency>
    <groupId>mysql</groupId>
    <artifactId>mysql-connector-java</artifactId>
    <version>5.1.27</version>
</dependency>

读取MySQL数据

    val conf: SparkConf = new SparkConf().setAppName(this.getClass.getName).setMaster("local[*]")
    val sc = new SparkContext(conf)
    // 连接数据库四要素
    val driver = "com.mysql.jdbc.Driver"
    val url = "jdbc:mysql://localhost:3306/student"
    val userName = "root"
    val passWord = "root"

    // 创建JdbcRDD
    /*
    sc: SparkContext,  spark程序的入口，上下文对象
    getConnection: () => Connection,    数据库连接
    sql: String,    要执行的sql
    lowerBound: Long,   sql语句的参数1
    upperBound: Long,   sql语句的参数2
    numPartitions: Int, 分区数量
    mapRow: (ResultSet) 处理结果集
     */
    val rdd: JdbcRDD[(Int, String)] = new JdbcRDD(
      sc,
      () => {
        // 注册驱动
        Class.forName(driver)
        // 创建连接
        DriverManager.getConnection(url, userName, passWord)
      },
      "select * from stuinfo where id>=? and id<=?",
      1000,
      2000,
      1,
      // 处理结果集
      rs => {
        (rs.getInt(1), rs.getString(2))
      }
    )

    // 打印结果
    rdd.foreach(println)
    sc.stop()

插入数据到MySQL

	val conf: SparkConf = new SparkConf().setAppName(this.getClass.getName).setMaster("local[*]")
    val sc = new SparkContext(conf)
    val rdd: RDD[(String, String)] = sc.makeRDD(List(("Tom", "男"), ("Jerry", "男"),
      ("柯南", "男"), ("毛利兰", "女")
    ))

    // 连接数据库四要素
    val driver = "com.mysql.jdbc.Driver"
    val url = "jdbc:mysql://localhost:3306/student"
    val userName = "root"
    val passWord = "root"

    // 向mysql插入数据
    // 每一条数据都要执行连接关闭操作，十分消耗性能
    /*rdd.foreach {
      case (name, gender) => {
        // 注册驱动
        Class.forName(driver)
        // 创建连接
        val conn: Connection = DriverManager.getConnection(url, userName, passWord)
        val sql = "insert into stuinfo(sname,gender) values(?,?)"
        val ps: PreparedStatement = conn.prepareStatement(sql)
        // 赋值
        ps.setString(1, name)
        ps.setString(2, gender)
        // 执行sql
        ps.executeUpdate()

        // 关闭连接
        ps.close()
        conn.close()
      }
     }*/

    // 将一个分区内的数据进行操作，可以减少连接关闭操作的次数，节约资源
    rdd.foreachPartition(
      datas => {
        // 注册驱动
        Class.forName(driver)
        // 创建连接
        val conn: Connection = DriverManager.getConnection(url, userName, passWord)
        val sql = "insert into stuinfo(sname,gender) values(?,?)"
        val ps: PreparedStatement = conn.prepareStatement(sql)
        // 遍历迭代器中的元素
        datas.foreach {
          case (name, gender) => {
            // 赋值
            ps.setString(1, name)
            ps.setString(2, gender)
            // 执行
            ps.executeUpdate()
          }
        }
        // 关闭连接
        ps.close()
        conn.close()
      })
    sc.stop()