【大数据课程】高途课程实践-Day03：Scala实现商品实时销售统计

〇、概述

1、实现内容

使用Scala编写代码，通过Flink的Source、Sink以及时间语义实现实时销量展示

2、过程

（1）导包并下载依赖

（2）创建数据源数据表并写⼊数据

（3）在Mysql数据库中创建统计结果表

（4）编写Flink计算代码

a.参考ShopMysqlSource.scala 代码，进⾏Flink Source 编写，从Mysql读取数据

b.参考GaotuShopFlinkStat.scala代码，进⾏统计逻辑的编写，进⾏FlinkSQL 查询

c.参考ShopStatMysqlSink.scala 代码，进⾏FlinkSink 编写，存⼊数据到Mysql

一、导包并下载依赖

 

二、创建数据源数据表并写⼊数据

参考执⾏GenerateOrders.scala 代码

package com.gaotu.flink
import org.apache.commons.lang.time.FastDateFormat

import java.sql.DriverManager
import java.sql.Connection
import java.util.concurrent.TimeUnit
import scala.util.Random

/**
 * 向Mysql数据库中添加数据
 * CREATE TABLE `g_orders` (
      `id` int(11) NOT NULL AUTO_INCREMENT COMMENT '交易id',
      `stock_name` varchar(20)DEFAULT NULL COMMENT '商品名称',
      `user_id` int(11) DEFAULT NULL COMMENT '用户id',
      `deal_status` varchar(20) DEFAULT NULL COMMENT '交易状态',
      `stock_cnt` int(11) DEFAULT NULL COMMENT '商品数量',
      `deal_amount` double DEFAULT NULL COMMENT '订单金额',
      `oper_time` bigint  COMMENT '处理时间',
      PRIMARY KEY (`id`)
    ) ENGINE=InnoDB AUTO_INCREMENT=120 DEFAULT CHARSET=utf8;
 *
 */
object GenerateOrders {

  def main(args: Array[String]): Unit = {
    val driver = "com.mysql.cj.jdbc.Driver"
    val url = "jdbc:mysql://localhost/plato"
    val username = "root"
    val password = "root"
    var connection:Connection = null

    //商品数组
    val stocks = Array("HUAWEI Mate40","Apple iphone13","Apple MacBook Pro 14"
      ,"ThinkBook 14p","RedmiBook Pro14","飞鹤星飞帆幼儿奶粉","爱他美 幼儿奶粉"
      ,"李宁运动男卫裤","小米踏步机椭圆机","欧莱雅面膜","御泥坊面膜","欧莱雅男士套装","金六福白酒"
      ,"牛栏山42度","茅台飞天")
    val amount = Array(6569.00,6099.00,14999.00,6799.00,4899.00,275,392,199,1299.00,599,399,389,469,175,1399.00)

    try {
      Class.forName(driver)
      connection = DriverManager.getConnection(url, username, password)
      val ps = connection.createStatement()
      for (i <- 1 to 10000){
        val formater = FastDateFormat.getInstance("yyyy-MM-dd HH:mm:ss")
        val item = scala.util.Random.nextInt(14)
        val deal_time = formater.format(System.currentTimeMillis())
        val stock_name = stocks(item)
        val user_id = Random.nextInt(100)
        val deal_status = "有效"
        val stock_cnt = Random.nextInt(20)
        val deal_amount = amount(item)
        val oper_time = System.currentTimeMillis()
        val query = s"insert into g_orders(deal_time,stock_name,user_id,deal_status,stock_cnt,deal_amount,oper_time) values('$deal_time', '$stock_name', $user_id, '$deal_status', $stock_cnt, $deal_amount, $oper_time)"
        ps.addBatch(query)
        println(query)

        TimeUnit.SECONDS.sleep(1)
      }
      ps.executeBatch()
    } catch {
      case e => e.printStackTrace
    }
    connection.close()
  }
}

三、在Mysql数据库中创建统计结果表

-- 统计结果表 
 CREATE TABLE `g_orders_stat` ( 
 `stock_name` varchar(20)DEFAULT NULL COMMENT '商品名称', 
 `order_cnt` int(11) DEFAULT NULL COMMENT '商品数量', 
 `stock_sales_cnt` int(11) DEFAULT NULL COMMENT '商品数量', 
 `stock_sales_amt` double DEFAULT NULL COMMENT '商品数量' 
 ) ENGINE=InnoDB AUTO_INCREMENT=120 DEFAULT CHARSET=utf8;

四、编写Flink计算代码

1、参考ShopMysqlSource.scala 代码，进⾏Flink Source 编写，从Mysql读取数据

package com.gaotu.flink

import org.apache.flink.streaming.api.functions.source.RichSourceFunction
import org.apache.flink.streaming.api.functions.source.SourceFunction.SourceContext
import java.sql.{Connection, DriverManager, PreparedStatement}
import java.util.concurrent.TimeUnit

/*
 * Mysql Source
*/
class ShopMysqlSource extends RichSourceFunction[(Int, String, Int, Double, Long)] {
  val driver = "com.mysql.cj.jdbc.Driver"
  val url = "jdbc:mysql://localhost/plato"
  val username = "root"
  val password = "root"

  override def run(ctx: SourceContext[(Int, String, Int, Double, Long)]): Unit = {
    // 1. 加载MYSQL驱动
    Class.forName(driver)
    // 2. 建立MYSQL链接
    var connection: Connection = DriverManager.getConnection(url, username, password)
    var max_id = 0
    // 3. 创建PreparedStatement
    val sql = s"select id,stock_name,stock_cnt,deal_amount,oper_time from plato.g_orders where id > ${max_id} ;"
    // 4. 执行Sql查询
    val ps: PreparedStatement = connection.prepareStatement(sql)
    for(i <- 0 until 500) {
      val queryRequest = ps.executeQuery()
      // 5. 遍历结果
      while (queryRequest.next()) {
        val id = queryRequest.getInt("id")
        val stock_name = queryRequest.getString("stock_name")
        val stock_cnt = queryRequest.getInt("stock_cnt")
        val deal_amount = queryRequest.getDouble("deal_amount")
        val deal_time = queryRequest.getLong("oper_time")
        ctx.collect((id, stock_name, stock_cnt, deal_amount, deal_time))

        if(max_id < id){
          max_id = id
        }
      }
      TimeUnit.SECONDS.sleep(1)
    }
  }
  override def cancel(): Unit = {}
}

2、参考GaotuShopFlinkStat.scala代码，进⾏统计逻辑的编写，进⾏FlinkSQL 查询

package com.gaotu.flink

import org.apache.flink.api.scala._
import org.apache.flink.streaming.api.TimeCharacteristic
import org.apache.flink.streaming.api.functions.AssignerWithPeriodicWatermarks
import org.apache.flink.streaming.api.scala.{DataStream, StreamExecutionEnvironment}
import org.apache.flink.streaming.api.watermark.Watermark
import org.apache.flink.table.api.{Table, TableEnvironment}
import org.apache.flink.table.api.scala._

/**
 * 实时统计逻辑
 */
object GaotuShopFlinkStat {

  def main(args: Array[String]): Unit = {
    val env = StreamExecutionEnvironment.getExecutionEnvironment   // 1.创建流处理环境
    env.setStreamTimeCharacteristic(TimeCharacteristic.EventTime)  // 2.设置处理时间为事件
    val tableEnv = TableEnvironment.getTableEnvironment(env)       // 3.获取Table运行环境
    val orderDataStream = env.addSource(new ShopMysqlSource)       // 4.获取自定义数据源
    // 5.添加水印，允许延迟2秒
    val watermarkDataStream:DataStream[(Int, String, Int, Double, Long)] = orderDataStream.assignTimestampsAndWatermarks(
      new AssignerWithPeriodicWatermarks[(Int, String, Int, Double, Long)]{
        var currentTimestamp: Long = _
        val delayTime = 2000  // 允许延迟2秒
        override def getCurrentWatermark: Watermark = {                // 生成一个水印时间
          val watermark = new Watermark(currentTimestamp - delayTime)  // 让时间窗口延迟2秒计算
          watermark
        }
        //  从订单中获取对应的时间戳
        override def extractTimestamp(element: (Int, String, Int, Double, Long), previousElementTimestamp: Long): Long = {
          val oper_time = element._5  // 获取订单的事件时间（下单时的时间）
          currentTimestamp = Math.max(currentTimestamp, oper_time) // 当前时间与事件时间对比，获取最大值
          currentTimestamp
        }
      } )

    // 6.注册LinkSQL数据表
    tableEnv.registerDataStream("g_order", watermarkDataStream, 'id,'stock_name,'stock_cnt,'deal_amount,'oper_time,'createTime.rowtime)
    // 7.编写SQL语句进行统计
    val sql =
      """
        | select
        |     stock_name
        |     ,count(id) as order_cnt
        |     ,sum(stock_cnt) as stock_sales_cnt
        |     ,sum(deal_amount) as stock_sales_amt
        | from g_order
        | group by
        | tumble(createTime, interval '5' second ),
        | stock_name
        |""".stripMargin
    // 8.执行sql语句
    val table: Table = tableEnv.sqlQuery(sql)
    table.printSchema()

    // 9.将SQL的执行结果写入到Mysql中
    tableEnv.toAppendStream[(String, Long, Int, Double)](table).addSink(new ShopStatMysqlSink)
    // 10.执行任务
    env.execute()
  }
}

3、参考ShopStatMysqlSink.scala 代码，进⾏FlinkSink 编写，存⼊数据到Mysql

package com.gaotu.flink

import org.apache.flink.configuration.Configuration
import org.apache.flink.streaming.api.functions.sink.RichSinkFunction
import java.sql.{Connection, DriverManager, PreparedStatement}

/**
 * Mysql Sink
 *
 * CREATE TABLE `g_orders_stat` (
  `stock_name` varchar(20)DEFAULT NULL COMMENT '商品名称',
  `order_cnt` int(11) DEFAULT NULL COMMENT '商品数量',
  `stock_sales_cnt` int(11) DEFAULT NULL COMMENT '商品数量',
  `stock_sales_amt` double DEFAULT NULL COMMENT '商品数量'
  ) ENGINE=InnoDB AUTO_INCREMENT=120 DEFAULT CHARSET=utf8;
 *
 */
class ShopStatMysqlSink extends RichSinkFunction[(String, Long, Int, Double)] {

  val driver = "com.mysql.cj.jdbc.Driver"
  val url = "jdbc:mysql://localhost/plato"
  val username = "root"
  val password = "root"
  var connection: Connection = null
  var ps: PreparedStatement = null

  override def open(parameters: Configuration): Unit = {
    Class.forName(driver) // 1. 加载驱动
    connection = DriverManager.getConnection(url, username, password)  // 2. 创建连接
    val sql = "insert into g_orders_stat (stock_name,order_cnt,stock_sales_cnt,stock_sales_amt) values (?,?,?,?);" // 3. 获得执行语句
    ps = connection.prepareStatement(sql)
  }
  override def invoke(value: (String, Long, Int, Double)): Unit = {// 4. 插入数据
    try {
      ps.setString(1, value._1)
      ps.setLong(2, value._2)
      ps.setInt(3, value._3)
      ps.setDouble(4, value._4)
      ps.execute()
    } catch {
      case e: Exception => println(e.getMessage)
    }
  }
  // 5. 关闭连接操作
  override def close(): Unit = {
    if(connection != null) {
      connection.close()
    }
    if(ps != null) {
      ps.close()
    }
  }
}