MapReduce案例之ReduceJoin

MapReduce案例之ReduceJoin

需求

​ 如下兩個文件

• order.txt是用戶的訂單信息

  order001,u001
  order002,u001
  order003,u005
  order004,u002
  order005,u003
  order006,u004
  

• user.txt是基本用戶信息

  u001,senge,18,angelababy
  u002,laozhao,48,ruhua
  u003,xiaoxu,16,chunge
  u004,laoyang,28,zengge
  u005,nana,14,huangbo
  

• 輸出用戶與訂單的關聯關係

  order001,u001,senge,18,angelababy
  order002,u001,senge,18,angelababy
  order003,u005,nana,14,huangbo
  ……
  

分析

• 確定map輸入類型<LongWritable, Text>

  根據輸入文件order.txt和user.txt分析：map輸入<key,value>爲<行索引，每行的內容>。

  <0,"order001,u001">
  

• 確定reduce輸出類型<Text,Text>

  根據輸出結果分析，reduce輸出的<key,value>爲

  <"order001,u001,senge,18,angelababy","">
  

• 確定map輸出類型<Text,Text>

  • 因爲需要根據userId字段進行關聯，所以userId字段爲關聯鍵。mr作業就是通過map輸出的key進行關聯和分組的，所以map輸出的key爲userId。

  • 兩個輸入切片，需要區分才能進行關聯。可爲order.txt文件map輸出value添加1前綴；爲user.txt文件map輸出value添加2前綴。

    <"u001","1:order001">
    <"u001","2:senge,18,angelababy">
    

• 確定map邏輯

• 確定reduce邏輯

  • 分別使用orderList和userList存放指定userId的訂單信息和用戶信息。
  • 嵌套遍歷orderList和userList，輸出最終結果。

實現

導入依賴包

​

代碼

package mapreduce;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.Reducer;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.input.FileSplit;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;

import java.io.IOException;
import java.util.ArrayList;
import java.util.List;

public class ReduceJoin2 {
    //本地運行時，指定靜態代碼塊
    static {
        System.setProperty("hadoop.home.dir", "D:\\software\\hadoop-2.9.2");
    }

    public static class MyMapper extends Mapper<LongWritable, Text, Text, Text> {
        @Override
        protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
           //如果第一行是orderid,userid列字段說明，需要過濾掉
            //if (key.get() == 0) return;
            String fileName = ((FileSplit) context.getInputSplit()).getPath().getName();
            String line = value.toString();
            String[] lineArr = line.split(",");
            //order.txt 訂單表 格式：order,userid
            if("order.txt".equals(fileName)) {
                context.write(new Text(lineArr[1]), new Text("1:" + lineArr[0]));
            } else {
                //user.txt 用戶表  格式：userid,nickname,age,username
                context.write(new Text(lineArr[0]), new Text("2:" + lineArr[1] + "," + lineArr[2] + "," + lineArr[3]));
            }
        }
    }

    public static class MyReducer extends Reducer<Text, Text, Text, Text> {
        @Override
        protected void reduce(Text key, Iterable<Text> values, Context context) throws IOException, InterruptedException {
            List<String> orderList = new ArrayList<>();
            List<String> userList = new ArrayList<>();
            //1.根據user_id，遍歷對應的訂單信息和用戶信息，拆分到相應的集合中
            for (Text value : values) {
                String valueStr = value.toString();
                String[] valueArr = valueStr.split(":");
                if("1".equals(valueArr[0])) {
                    orderList.add(valueArr[1]);
                } else {
                    userList.add(valueStr.substring(2));
                }
            }
            //2.將訂單列表和用戶列表進行嵌套遍歷
            for (String order : orderList) {
                for (String user : userList) {
                    context.write(new Text(order + "," + key.toString() + "," + user), new Text(""));
                }
            }
        }
    }
    /**
     * 驅動方法
     * @param args
     */
    public static void main(String[] args) throws IOException, ClassNotFoundException, InterruptedException {
        //0.初始化一個job
        Configuration conf = new Configuration();

        Job job = Job.getInstance(conf, "reduce_join2");
        job.setJarByClass(ReduceJoin2.class);
        //1.輸入文件
        FileInputFormat.addInputPaths(job, args[0]);
        //2.map並行計算
        job.setMapperClass(MyMapper.class);
        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(Text.class);
        //3.shuffle流程(內部實現)
        //4.reduce計算
        job.setReducerClass(MyReducer.class);
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(Text.class);
        //5.輸出文件
        FileOutputFormat.setOutputPath(job, new Path(args[1]));

        //6.提交作業(總入口)
        boolean result = job.waitForCompletion(true);
        System.exit(result ? 1 : 0);
    }
}

運行

本地運行

​ 指定參數：比如,D://data/order.txt,D://data/user.txt D://output

集羣運行

• 打jar包

• 將jar包放到linux上

• 輸入文件order.txt和user.txt放到hdfs 的/test目錄中

• 運行jar

  bin/yarn jar hadoop_demo-1.0-SNAPSHOT.jar mapreduce.ReduceJoin2 /test/order.txt,/test/user.txt /test/ouput