DStream实例——实现网站热词排序

1、进入数据库：mysql -uroot -pPassword123$

 

创建数据库名字为spark：create database spark

创建数据表sww ：create table sww(insert_time date,keyword varchar(30),search_count integer)

2、在pom.xml文件添加mysql数据库的依赖，添加的内容如下：

            mysql
            mysql-connector-java
            5.1.38
        

3、在spark03项目的/src/main/scala/itcast目录下创建一个名为HotWordBySort的scala类，编写以下内容：

import java.sql.{DriverManager, Statement}

import org.apache.spark.streaming.dstream.{DStream, ReceiverInputDStream}
import org.apache.spark.streaming.{Seconds, StreamingContext}
import org.apache.spark.{SparkConf, SparkContext}
//import org.scalatest.time.Second

object HotWordBySort {

  def main(args: Array[String]): Unit = {
    //创建SparkConf对象
    val sparkConf: SparkConf = new SparkConf()
      .setAppName("HotWordBySort").setMaster("local[2]")
    //创建SparkContext对象
    val sc : SparkContext = new SparkContext(sparkConf)
    //设置日志打印级别
    sc.setLogLevel("WARN")
    //创建StreamingContetxt，需要两个参数
    val ssc:StreamingContext = new StreamingContext(sc,Seconds(5))
    //连接socket服务
    val dstream:ReceiverInputDStream[String] = ssc.socketTextStream("192.168.196.101",9999)
    //逗号分隔
    val itemPairs:DStream[(String,Int)]  = dstream.map(line => (line.split(",")(0),1))

    //调用操作
    val itemCount:DStream[(String,Int)] = itemPairs.reduceByKeyAndWindow((v1:Int,v2:Int)=>v1+v2,Seconds(60),Seconds(10))
    //Dstream没有sortByKey操作
    val hotWord = itemCount.transform(itemRDD=>{
      val top3:Array[(String,Int)]=itemRDD.map(pair =>(pair._2,pair._1)).sortByKey(false).map(pair=>(pair._2,pair._1)).take(3)
      //本地集合转成RDD
      ssc.sparkContext.makeRDD(top3)
    })

    //调用foreachRDD操作
    hotWord.foreachRDD(rdd=>{
      val url = "jdbc:mysql://192.168.196.101:3306/spark"
      val user = "root"
      val password = "Password123$"
      Class.forName("com.mysql.jdbc.Driver")
      val conn1 = DriverManager.getConnection(url,user,password)
      conn1.prepareStatement("delete from searchKeyWord where 1=1").executeUpdate()

      conn1.close()
      rdd.foreachPartition(partitionOfRecords=>{
        val url = "jdbc:mysql://192.168.196.101:3306/spark"
        val user = "root"
        val password = "Password123$"

        Class.forName("com.mysql.jdbc.Driver")
        val conn2 = DriverManager.getConnection(url,user,password)
        conn2.setAutoCommit(false)
        val stat:Statement = conn2.createStatement()
        partitionOfRecords.foreach(record=>{
          stat.addBatch("insert into searchKeyWord(insert_time,keyword,search_count) values (now(),'"+record._1+"','"+record._2+"')")
        })
        stat.executeBatch()
        conn2.commit()
        conn2.close()
      })
    })
    ssc.start()
    ssc.awaitTermination()
    ssc.stop()
  }
}

4、运行HotWordBySort代码，并在master 9999端口输入以下数据：

nc -lk 9999

Hadoop,111

Spark,222

Hadoop,222

Hadoop,222

Hive,333

Hive,222

在mysql中查看数据表是否插入数据：

由上面结果可以看出刚刚创建的sww数据表是已经添加数据成功的。