SQL join语法案例

Data:

order.txt

order011,u001,300
order012,u002,200
order023,u006,100
order056,u007,300
order066,u003,500
order055,u004,300
order021,u005,300
order014,u001,100
order025,u005,300
order046,u007,30
order067,u003,340
order098,u008,310

user.txt

u001,hls,22,fengjie
u002,wangwu,31,lisi
u003,zhangyanru,22,tananpengyou
u004,laocao,26,fengyi
u005,mengqi,12,nvmengqi
u006,haolei,38,sb
u007,wanghongjing,24,wife
u009,wanghongjing,24,wife



返回一个结果：order011  u001   300  hls  22   fengjie

代码示例：

package com.doit.day03

import scala.io.{BufferedSource, Source}

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

    //u001,hls,22,fengjie
    val bs1= Source.fromFile("D:\develop\ideaWorkSpace\myself\study\scalaDemo\data\user.txt")/*.getLines().toList*/

    //order011,u001,300
    val bs2 = Source.fromFile("D:\develop\ideaWorkSpace\myself\study\scalaDemo\data\order.txt")/*.getLines().toList*/

    //实现left join
/*
    //将用户数据转换成map集合
    val users: Iterator[String] = bs1.getLines()
    val iters: Iterator[(String, (String, String, String, String))] = users.map(_.split(",", -1)).map(x => (x(0), (x(0), x(1), x(2), x(3))))
    val map: Map[String, (String, String, String, String)] = iters.toMap
    // 将订单数据转换成list集合
    val orders: Iterator[String] = bs2.getLines()
    val iters2: Iterator[(String, (String, String))] = orders.map(_.split(",", -1)).map(x => (x(1), (x(0), x(1))))
    val list2: List[(String, (String, String))] = iters2.toList

    //遍历每个订单  拼接用户信息
    var r = list2.map(x => {
      val user = map.getOrElse(x._1, ("null", "null", "null", "null"))
      (user._1, user._2, user._3, user._4, x._2._1)
    })
    // 打印结果
    r.sortBy(_._1).foreach(println)

    */

    //实现join
    /*
    val userTuple: List[(String, String, String, String)] = users.map(line => {
      val arr: Array[String] = line.split(",")
      //user_id，user_name，age,name
      (arr(0), arr(1), arr(2), arr(3))
    })

    val orderTuple: List[(String, String, String)] = orders.map(line => {
      val arr: Array[String] = line.split(",")
      //order_id  user_id   amount
      (arr(0), arr(1), arr(2))
    })

    //join关联条件是user_id = user_id

    for (user <- userTuple) {
      for (order <- orderTuple) {
        if(user._1 == order._2){
          println(user._1,user._2,user._3,user._4,order._1,order._3)
        }
      }
    }
*/
  }
}

pv，uv案例

数据：

site1,user1,2018-03-01 02:12:22
site1,user2,2018-03-05 04:12:22
site1,user2,2018-03-05 04:13:22
site1,user2,2018-03-05 04:14:22
site1,user2,2018-03-05 04:15:22
site4,user7,
site1,user2,2018-03-05 05:15:22
site1,user2,2018-03-05 08:15:22
site1,user3,2018-03-05 04:15:22
site1,user4,2018-03-05 05:15:22
site1,user3,2018-03-07 11:12:22
site1,user3,2018-03-08 11:12:22
site2,user4,2018-03-07 15:12:22
site3,user5,2018-03-07 08:12:22
site3,user6,2018-03-05 08:12:22
site1,user1,2018-03-08 11:12:22
site1,,2018-03-08 11:12:22
site2,user2,2018-03-07 15:12:22
site3,user5,2018-03-07 08:12:22
site3,user5,2018-03-07 18:12:22
site3,user6,2018-03-05 08:12:22
site4,user7,2018-03-03 10:12:22
site2,,2018-03-08 11:12:22
site3,user5,2018-03-07 08:12:22
site3,user6,2018-03-05 08:12:22
site4,user5,2018-03-03 10:12:22
site4,user7,2018-02-20 11:12:22

代码：

package com.doit.day03
import scala.io.{BufferedSource, Source}
/**
 * 需求：计算每天的pv和uv
 * pv：浏览次数
 * uv：访客数
 */
object PVUVDemo {
  def main(args: Array[String]): Unit = {

    val source: BufferedSource = Source.fromFile("D:\develop\ideaWorkSpace\myself\study\scalaDemo\data\pvuv.txt")

    val list: List[String] = source.getLines().toList

    //过滤一些脏数据
    val filtered: List[String] = list.filter(line => {
      val arr: Array[String] = line.split(",",-1)
      arr.length >= 0 && !arr.exists(_.isEmpty)
    })

    val events: List[(String, String, String)] = filtered.map(line => {
      val arr: Array[String] = line.split(",")
      val date: String = arr(2).substring(0, 10)
      //site1,user1,2018-03-01 02:12:22
      (arr(0), arr(1),date)
    })

    //pv：该页面被浏览了多少次
    val tuples: List[((String, String), String)] = events.map(tp => {
      ((tp._3, tp._1), tp._2)
    })

    val pv: Map[(String, String), Int] = tuples.groupBy(_._1).map(tp => (tp._1, tp._2.size))
    val uv: Map[(String, String), Int] = tuples.groupBy(_._1).map(tp => (tp._1, tp._2.distinct.size))
    println("============pv================")
    pv.foreach(println)
    println("============uv================")
    uv.foreach(println)
  }
}

线段重叠案例

data：

1,4
2,5
4,6
2,4
3,6
4,6
1,5

代码：

package com.doit.day03

import scala.io.Source

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

    val list: List[String] = Source.fromFile("D:\develop\ideaWorkSpace\myself\study\scalaDemo\data\line.txt").getLines().toList

    //生成一个个的点
    val points: List[Range.Inclusive] = list.map(line => {
      val arr: Array[String] = line.split(",")
      val start: String = arr(0)
      val end: String = arr(1)
      //按照起始得位置确定这一行中一共有哪些点，后面才能判断哪些点是重合的
      //根据开始和结束得点
      val range: Range.Inclusive = start.toInt to end.toInt
      range
    })

    //将点压平
    val flattened: List[Int] = points.flatten
    //对点分组
    val intToInts: Map[Int, List[Int]] = flattened.groupBy(point => point)
    //获取到最后的个数
    val res: Map[Int, Int] = intToInts.map(tp => (tp._1, tp._2.size))

    res.foreach(println)
  }
}

• 还没有人评论，欢迎说说您的想法！

相关课程

大数据项目用户画像实战

6417 9.8元

大数据企业级实战 13天玩转千亿级数仓项目

4203 9.8元

大数据技术之Hadoop

4152 9.8元 98元 1折