Java基础（下）

1. HDFS Java端API

package com.itheima.demo01_hdfs;

import com.sun.jndi.toolkit.url.Uri;
import org.apache.commons.io.IOUtils;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FSDataInputStream;
import org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.LocatedFileStatus;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.RemoteIterator;
import org.junit.Test;

import java.io.File;
import java.net.URI;

/*
    案例: 演示Java操作HDFS文件系统.

    细节:
        你需要在你的windows系统中, 配置Hadoop运行环境, 具体步骤如下:
            1. 把hadoop-2.7.5拷贝到一个没有中文, 空格等特殊符号的合法路径下.
            2. 配置Path环境变量.
            3. 把 hadoop-2.7.5/bin/hadoop.dll 拷贝到  c:/windows/system32 这个路径下.
            4. 重启windows系统.
 */
public class Demo01 {
    /*
        需求: 演示如何获取 FileSystem 文件系统对象.
        Java操作HDFS, 本质上就是 Java连接HDFS文件系统, 然后进行操作, 具体的获取方式有如下三种.
     */
    @Test
    public void show1() throws Exception {
        //方式1: 直接获取FileSystem文件系统对象, 采用默认配置, 即: 获取当前系统(即: windows系统)
        //1. 创建配置文件对象.
        Configuration conf = new Configuration();
        //2. 根据配置文件对象, 获取其对应的 文件系统对象.
        FileSystem fs1 = FileSystem.get(conf);
        //3. 打印结果.
        System.out.println("fs1: " + fs1);
        System.out.println("---------------------------");

        //方式2: 获取FileSystem文件系统对象, 采用指定配置, 例如: windows系统, HDFS文件系统. 采用当前文件系统的默认用户名.
        //1. 创建配置文件对象.
        Configuration conf2 = new Configuration();
        //设置配置信息, 连接: HDFS文件系统.
        conf2.set("fs.defaultFS", "hdfs://node1:8020");
        //2. 根据配置文件对象, 获取其对应的 文件系统对象.
        FileSystem fs2 = FileSystem.get(conf2);
        //3. 打印结果.
        System.out.println("fs2: " + fs2);
        System.out.println("---------------------------");

        //方式3: 获取FileSystem文件系统对象, 采用指定配置, 例如: windows系统, HDFS文件系统, 自定义用户名(术语叫: 伪装用户)
        //HDFS是一个很单纯的孩子, 你告诉它你是谁, 它就认为你是谁, 它是用来防止"好人做错事儿", 而不是解决"坏人做坏事儿"
        //1. 创建配置文件对象.
        Configuration conf3 = new Configuration();
        URI uri = new URI("hdfs://node1:8020");
        //2. 根据配置文件对象, 获取其对应的 文件系统对象.
        FileSystem fs3 = FileSystem.get(uri, conf, "root");
        //3. 打印结果.
        System.out.println("fs3: " + fs3);
        System.out.println("---------------------------");

        //方式4: 实际开发版, 即: 方式三的优化版.
        FileSystem fs4 = FileSystem.get(new URI("hdfs://node1:8020"), new Configuration());
        System.out.println("fs4: " + fs4);
    }

    //Java操作HDFS之遍历HDFS中所有文件
    @Test
    public void show2() throws Exception {
        //1. 获取 HDFS的文件系统对象.
        FileSystem fs = FileSystem.get(new URI("hdfs://node1:8020"), new Configuration(), "root");
        //2. 获取指定目录下所有的文件(目录)的 文件状态对象, 因为是多个, 所以是一个迭代器.
        //参数1: 路径.  参数2: 是否递归
        RemoteIterator<LocatedFileStatus> listFiles = fs.listFiles(new Path("/"), true);
        //3. 遍历迭代器, 获取每一个 文件状态对象.
        while (listFiles.hasNext()) {
            //fileStatus: 文件状态对象, 即: 表示每一个文件(目录)
            LocatedFileStatus fileStatus = listFiles.next();
            //4. 根据文件状态对象, 获取文件(目录)的 路径或者名字, 并打印.
            System.out.println("路径: " + fileStatus.getPath());
            System.out.println("名字: " + fileStatus.getPath().getName());
            System.out.println("------------------------");
        }
        //5. 释放资源.
        fs.close();
    }

    //Java操作HDFS之HDFS上创建文件夹,  HDFS指令: mkdir
    @Test
    public void show3() throws Exception {
        //1. 获取HDFS文件系统对象.
        FileSystem fs = FileSystem.newInstance(new URI("hdfs://node1:8020"), new Configuration(), "root");
        //2. 创建文件.
        boolean flag = fs.mkdirs(new Path("/hangge"));
        System.out.println(flag ? "创建成功" : "创建失败");
        //3. 释放资源
        fs.close();
    }

    //Java操作HDFS之下载文件, HDFS指令: get, getmerge, copyToLocal, moveToLocal
    @Test
    public void show4() throws Exception {
        //1. 获取HDFS文件系统对象.
        FileSystem fs = FileSystem.newInstance(new URI("hdfs://node1:8020"), new Configuration(), "root");
        //2. 下载文件.
        //如果是操作本地文件, 建议加上:  file:///
        //复制, 下载.
        //fs.copyToLocalFile(new Path("/1.txt"), new Path("D:\\abc\\ceshi"));

        //剪切, 下载.
        fs.moveToLocalFile(new Path("/1.txt"), new Path("D:\\abc\\ceshi"));

        //3. 释放资源
        fs.close();
    }

    //Java操作HDFS之上传文件.  HDFS指令: put, copyFromLocal, moveFromLocal
    @Test
    public void show5() throws Exception {
        //1. 获取HDFS文件系统对象.
        FileSystem fs = FileSystem.newInstance(new URI("hdfs://node1:8020"), new Configuration(), "root");
        //2. 上传文件.
        //如果是操作本地文件, 建议加上:  file:///
        //复制, 上传.
        //fs.copyFromLocalFile(new Path("file:///D:\\abc\\ceshi\\绕口令.txt"), new Path("/hangge"));

        //剪切, 上传.
        fs.moveFromLocalFile(new Path("file:///D:\\abc\\ceshi\\绕口令.txt"), new Path("/hangge"));

        //3. 释放资源
        fs.close();
    }

    //Java操作HDFS之 合并上传
    @Test
    public void show6() throws Exception {
        //1. 获取本地文件系统对象, 即: windows
        FileSystem localFS = FileSystem.get(new Configuration());
        //2. 获取HDFS文件系统对象.
        FileSystem hdfsFS = FileSystem.get(new URI("hdfs://node1:8020"), new Configuration(), "root");

        //3. 获取HDFS文件的输出流对象, 将来可以往这个文件中写数据.
        FSDataOutputStream fos = hdfsFS.create(new Path("/hangge/abc.txt"));

        //4. 根据本地(windows)文件系统对象, 获取指定目录下所有的文件对象(文件状态对象的迭代器形式)
        RemoteIterator<LocatedFileStatus> listFiles = localFS.listFiles(new Path("D:\\abc\\a123"), false);

        //5. 遍历, 获取到每一个具体的 文件状态对象.
        while (listFiles.hasNext()) {
            //fileStatus: 就是具体的每一个文件状态对象, 即: core-site.xml, hdfs-site.xml, yarn-site.xml
            LocatedFileStatus fileStatus = listFiles.next();

            //6. 通过本地(windows)文件系统对象, 结合 该文件的路径, 获取该文件的输入流, 可以读取文件信息.
            FSDataInputStream fis = localFS.open(fileStatus.getPath());
            //7. 具体的拷贝文件的动作.
            IOUtils.copy(fis, fos);
            //8. 关闭此输入流
            IOUtils.closeQuietly(fis);
        }

        //9. 释放资源.
        hdfsFS.close();
        localFS.close();
    }

    //Java操作HDFS之权限问题
    @Test
    public void show7() throws Exception {
        //1. 获取HDFS文件系统对象.
        //FileSystem fs = FileSystem.newInstance(new URI("hdfs://node1:8020"), new Configuration(), "zhangsan");
        FileSystem fs = FileSystem.newInstance(new URI("hdfs://node1:8020"), new Configuration(), "root");

        //2. 下载文件.
        //如果是操作本地文件, 建议加上:  file:///
        //复制, 下载.
        fs.copyToLocalFile(new Path("/hangge/abc.txt"), new Path("D:\\abc\\ceshi"));

        //剪切, 下载.
        //fs.moveToLocalFile(new Path("/1.txt"), new Path("D:\\abc\\ceshi"));

        //3. 释放资源
        fs.close();
    }

    //Java操作HDFS之删除功能.
    @Test
    public void show8() throws Exception {
        //1. 获取HDFS文件系统对象.
        //FileSystem fs = FileSystem.newInstance(new URI("hdfs://node1:8020"), new Configuration(), "zhangsan");
        FileSystem fs = FileSystem.newInstance(new URI("hdfs://node1:8020"), new Configuration(), "root");

        //2. 删除文件.
        fs.delete(new Path("/hangge/绕口令.txt"), false);

        //3. 释放资源
        fs.close();
    }
}

2. Java编写MapReduce程序

2.1 wordcount程序

一个最简单的MR程序由Map，Reduce和主程序入口3个部分组成。下面是一个最简单的wordcount程序。

首先是MapTask

package cn.itcast.word_count;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.io.Text;

import java.io.IOException;

public class WordCountMapTask extends Mapper<LongWritable, Text, Text, IntWritable> {
    @Override
    protected void map(LongWritable key, Text value, Mapper<LongWritable, Text, Text, IntWritable>.Context context) throws IOException, InterruptedException {
        System.out.println("行偏移量: " + key.get());
        String line = value.toString();
        System.out.println(line);

        if (line!=null && !"".equals(line)){
            String[] arr = line.split(" ");
            for (String word : arr){
                context.write(new Text(word), new IntWritable(1));
            }
        }

    }
}

然后是ReduceTask

package cn.itcast.word_count;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.io.Text;

import java.io.IOException;

public class WordCountMapTask extends Mapper<LongWritable, Text, Text, IntWritable> {
    @Override
    protected void map(LongWritable key, Text value, Mapper<LongWritable, Text, Text, IntWritable>.Context context) throws IOException, InterruptedException {
        System.out.println("行偏移量: " + key.get());
        String line = value.toString();
        System.out.println(line);

        if (line!=null && !"".equals(line)){
            String[] arr = line.split(" ");
            for (String word : arr){
                context.write(new Text(word), new IntWritable(1));
            }
        }

    }
}

最后是主程序的入口

package cn.itcast.word_count;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.io.Text;

import java.io.IOException;

public class WordCountMapTask extends Mapper<LongWritable, Text, Text, IntWritable> {
    @Override
    protected void map(LongWritable key, Text value, Mapper<LongWritable, Text, Text, IntWritable>.Context context) throws IOException, InterruptedException {
        System.out.println("行偏移量: " + key.get());
        String line = value.toString();
        System.out.println(line);

        if (line!=null && !"".equals(line)){
            String[] arr = line.split(" ");
            for (String word : arr){
                context.write(new Text(word), new IntWritable(1));
            }
        }

    }
}

用Maven打包程序

Mvn package

提交任务

hadoop jar hdfs_maven-1.0-SNAPSHOT.jar cn.itcast.word_count.WordCountMain03 /mapreduce/wordcount/word.txt /mapreduce/wordcount/output

hadoop后台监控

2.2 分区

错误提示：如果找不到class，注意需要在Main函数添加下面的这句话。

job.setJarByClass(LotteryMain.class);

mapper

package org.example.demo01_lottery;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;

import java.io.IOException;

/*
     自定义的Map阶段, 负责 拆 的动作.
        k1: 行偏移量, LongWritable
        v1: 整行数据, Text
        k2: 中奖号, IntWritable
        v2: 整行数据, Text
 */
public class LotteryMapper extends Mapper<LongWritable, Text, IntWritable, Text> {
    @Override
    protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
        System.out.println("行偏移量: " + key.get());

        //1. 获取v1, 即: 整行数据.
        String line = value.toString();
        //2. 非空校验
        if (line != null && !"".equals(line)) {
            //3. 获取中奖号.
            int num = Integer.parseInt(line.split("\t")[5]);
            System.out.println(num);
            //4. 输出数据, 即: k2, v2
            context.write(new IntWritable(num), value);
        }
    }
}

reducer

package org.example.demo01_lottery;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;

import java.io.IOException;

/*
     自定义的Reduce阶段, 负责 合 的动作.
        k2: 中奖号, IntWritable
        v2: 整行数据, Text
        k3: 整行数据, Text
        v3: 空
 */
public class LotteryReducer extends Reducer<IntWritable, Text, Text, NullWritable> {
    /**
     * 自定义的Reduce阶段, 负责 合 的动作.
     *
     * @param key     k2, 即: 中奖号
     * @param values  v2的集合, 即: 整行数据
     * @param context 用于输出k3, v3的
     * @throws IOException
     * @throws InterruptedException
     */
    @Override
    protected void reduce(IntWritable key, Iterable<Text> values, Context context) throws IOException, InterruptedException {
        //因为values记录的是所有的 整行数据(已经分好区, 排好序, 规好约, 分好组)的数据, 直接写出即可.
        for (Text value : values) {
            context.write(value, NullWritable.get());
        }

    }
}

partitioner

package org.example.demo01_lottery;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Partitioner;

//自定义的分区类, 按照 中奖号(即:k2)分区, 15以下一组, 15及其以上1组.
public class MyPartitioner extends Partitioner<IntWritable, Text> {

    /**
     * 具体的自定义分区规则
     *
     * @param intWritable   k2的数据类型, 即: 中奖号
     * @param text          v2的数据类型, 即: 整行数据
     * @param numReduceTask ReduceTask任务的个数.
     * @return 分区编号, 即: 打标记的结果.
     */
    @Override
    public int getPartition(IntWritable intWritable, Text text, int numReduceTask) {
        //1. 获取中奖号.
        int num = intWritable.get();

        //2. 自定义分区规则.
        if (num < 15)
            return 0;
        else
            return 1;
    }
}

main

package org.example.demo01_lottery;


import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat;

//驱动类.
public class LotteryMain {
    public static void main(String[] args) throws Exception {
        //1. 创建Job任务.
        Job job = Job.getInstance(new Configuration(), "lotteryMR");

        job.setJarByClass(LotteryMain.class);

        //2. 封装MR程序的核心8步.
        //2 .1 封装输入组件, 读取: 源数据.
        job.setInputFormatClass(TextInputFormat.class);
        TextInputFormat.addInputPath(job, new Path(args[0]));
        //2.2 封装Map任务, 指定: k2, v2的类型.
        job.setMapperClass(LotteryMapper.class);
        job.setMapOutputKeyClass(IntWritable.class);
        job.setMapOutputValueClass(Text.class);
        //2.3 分区, 用自定义分组
        job.setPartitionerClass(MyPartitioner.class);
        //2.4 排序, 用默认的
        //2.5 规约, 用默认的
        //2.6 分组, 用默认的
        //2.7 封装Reduce任务, 指定k3,v3的类型
        job.setReducerClass(LotteryReducer.class);
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(NullWritable.class);
        //2.8 封装输出组件, 写出数据到 目的地文件中.
        job.setOutputFormatClass(TextOutputFormat.class);
        TextOutputFormat.setOutputPath(job, new Path(args[1]));

        //3. 设置reduce的个数.
        job.setNumReduceTasks(2);

        //4. 提交任务, 等待执行结果.
        boolean flag = job.waitForCompletion(true);

        //5. 关闭程序.
        System.exit(flag ? 0 : 1);
    }
}

运行命令

hadoop jar MapReduce-1.0-SNAPSHOT.jar org.example.demo01_lottery.LotteryMain /mapreduce/lottery/partition.csv /mapreduce/lottery/output

2.2 一个排序的程序

数据

a	1
a	9
b	3
a	7
b	8
a	19
b	10
a	5
a	9
b	10

定义SortBean类

package org.example.demo02_sort;

import org.apache.hadoop.io.WritableComparable;

import java.io.DataInput;
import java.io.DataOutput;
import java.io.IOException;

public class SortBean implements WritableComparable<SortBean> {

    private String first;
    private int second;

    public SortBean(){}

    public SortBean(String first, int second){
        this.first = first;
        this.second = second;
    }

    public String getFirst() {
        return first;
    }

    public void setFirst(String first){
        this.first = first;
    }


    public int getSecond() {
        return second;
    }

    public void setSecond(int second) {
        this.second = second;
    }

    @Override
    public String toString() {
        return first + "\t" + second;
    }

    @Override
    public int compareTo(SortBean sb) {
        int num = this.first.compareTo(sb.first);
        int num2 = num == 0 ? sb.second-this.second:num;
        return num2;
    }

    @Override
    public void write(DataOutput dataOutput) throws IOException {
        dataOutput.writeUTF(first);
        dataOutput.writeInt(second);
    }

    @Override
    public void readFields(DataInput dataInput) throws IOException {
        first = dataInput.readUTF();
        second = dataInput.readInt();
    }
}

定义mapper类

package org.example.demo02_sort;

import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;

import java.io.IOException;

public class SortMapper extends Mapper<LongWritable, Text, SortBean, NullWritable> {
    @Override
    protected void map(LongWritable key, Text value, Mapper<LongWritable, Text, SortBean, NullWritable>.Context context) throws IOException, InterruptedException {
        String line = value.toString();
        if (line!=null && !"".equals(line)){
            String[] arr = line.split("\t");
            SortBean sb = new SortBean(arr[0], Integer.parseInt(arr[1]));
            context.write(sb, NullWritable.get());
        }

    }
}

定义reducer类

package org.example.demo02_sort;

import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;

import java.io.IOException;

public class SortReducer extends Reducer<SortBean, NullWritable, Text, NullWritable> {
    @Override
    protected void reduce(SortBean key, Iterable<NullWritable> values, Reducer<SortBean, NullWritable, Text, NullWritable>.Context context) throws IOException, InterruptedException {
        for (NullWritable value : values) {
            context.write(new Text(key.toString()), NullWritable.get());
        }
    }
}

main类

package org.example.demo02_sort;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat;

public class SortMain {
    public static void main(String[] args) throws Exception{
        Job job = Job.getInstance(new Configuration(), "sortMR");
        job.setJarByClass(SortMain.class);
        job.setInputFormatClass(TextInputFormat.class);
        TextInputFormat.addInputPath(job, new Path("file:///D:\\sort\\sort.txt"));

        job.setMapperClass(SortMapper.class);
        job.setMapOutputKeyClass(SortBean.class);
        job.setMapOutputValueClass(NullWritable.class);

        job.setReducerClass(SortReducer.class);
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(NullWritable.class);

        job.setOutputFormatClass(TextOutputFormat.class);
        TextOutputFormat.setOutputPath(job, new Path("file:///D:\\sort\\output"));
        //3. 提交任务, 等待执行结果.
        boolean flag = job.waitForCompletion(true);
        System.exit(flag? 0:1);

    }

}

2.3 combiner实现

数据集

<<java入门宝典>>
<<Python入门宝典>>
<<spark入门宝典>>
<<UI入门宝典>>
<<乾坤大挪移>>
<<hive入门宝典>>
<<天龙八部>>
<<凌波微步>>
<<葵花点穴手>>
<<史记>>
<<PHP入门宝典>>
<<铁砂掌>>
<<葵花宝典>>
<<hadoop入门宝典>>
<<论清王朝的腐败>>

mapper

package org.example.demo03_combiner;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;

import java.io.IOException;

public class BookMapper extends Mapper<LongWritable, Text, Text, IntWritable> {
    @Override
    protected void map(LongWritable key, Text value, Mapper<LongWritable, Text, Text, IntWritable>.Context context) throws IOException, InterruptedException {
        String line = value.toString();
        if (line!=null && !"".equals(line)){
            if (line.contains("入门"))
                context.write(new Text("计算机类图书"), new IntWritable(1));
            else if (line.contains("史记") || line.contains("论清王朝的腐败")) {
                context.write(new Text("史书"), new IntWritable(1));
            }
            else
                context.write(new Text("武林秘籍"), new IntWritable(1));
        }

    }
}

reducer

package org.example.demo03_combiner;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;

import java.io.IOException;

public class BookMapper extends Mapper<LongWritable, Text, Text, IntWritable> {
    @Override
    protected void map(LongWritable key, Text value, Mapper<LongWritable, Text, Text, IntWritable>.Context context) throws IOException, InterruptedException {
        String line = value.toString();
        if (line!=null && !"".equals(line)){
            if (line.contains("入门"))
                context.write(new Text("计算机类图书"), new IntWritable(1));
            else if (line.contains("史记") || line.contains("论清王朝的腐败")) {
                context.write(new Text("史书"), new IntWritable(1));
            }
            else
                context.write(new Text("武林秘籍"), new IntWritable(1));
        }

    }
}

combiner

package org.example.demo03_combiner;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.mapreduce.Reducer;
import org.apache.hadoop.io.Text;

import java.io.IOException;

public class MyCombiner extends Reducer<Text, IntWritable, Text, IntWritable> {
    @Override
    protected void reduce(Text key, Iterable<IntWritable> values, Context context) throws IOException, InterruptedException {
        int count = 0;
        for (IntWritable value:values){
            count += value.get();
        }
        context.write(key, new IntWritable(count));
    }
}

main

package org.example.demo03_combiner;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat;
import org.apache.hadoop.yarn.webapp.hamlet.Hamlet;

public class BookMain {
    public static void main(String[] args) throws Exception{
        Job job = Job.getInstance(new Configuration(), "bookMR");
        job.setJarByClass(BookMain.class);

        job.setInputFormatClass(TextInputFormat.class);
        TextInputFormat.addInputPath(job, new Path("file:///D:\\book\\combiner.txt"));

        job.setMapperClass(BookMapper.class);
        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(IntWritable.class);

        job.setCombinerClass(MyCombiner.class);

        job.setReducerClass(BookReducer.class);
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(IntWritable.class);

        job.setOutputFormatClass(TextOutputFormat.class);
        TextOutputFormat.setOutputPath(job, new Path("file:///D:\\book\\output"));

        boolean flag = job.waitForCompletion(true);

        //4. 退出程序.
        System.exit(flag ? 0 : 1);




    }
}

2.4 好友案例

找出两两之间的共同好友

数据

A:B,C,D,F,E,O
B:A,C,E,K
C:F,A,D,I
D:A,E,F,L
E:B,C,D,M,L
F:A,B,C,D,E,O,M
G:A,C,D,E,F
H:A,C,D,E,O
I:A,O
J:B,O
K:A,C,D
L:D,E,F
M:E,F,G
O:A,H,I,J

思路如下图

map1

package com.itheima.demo05_friend;

import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;

import java.io.IOException;

//第一个MR的Map阶段,  k2: 好友    v2: 用户
public class FriendMapper1 extends Mapper<LongWritable, Text, Text, Text> {
    @Override
    protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
        //1. 获取整行数据
        String line = value.toString();         //"A:B,C,D"
        //2. 非空校验.
        if (line != null && !"".equals(line)) {
            //3. 切割数据, 获取: 用户, 该用户的好友列表
            String[] arr = line.split(":");  //["A", "B,C,D"]
            //4. 遍历好友列表, 获取该用户所有的好友
            for (String friend : arr[1].split(",")) {  //"B", "C", "D"
                //5. 写出结果, 即: k2, v2
                context.write(new Text(friend), new Text(arr[0]));  //B->A, C->A, D m->A
            }
        }
    }
}

reduce1

package com.itheima.demo05_friend;

import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;

import java.io.IOException;

/*
    自定义的Reduce阶段, 负责: 合
        k2: 好友
        v2: 用户
        k3: 好友
        v3: 有该好友的所有的用户

    目前数据格式:
          好友          有该好友的所有的用户
        {{C,C,C}	: 	{B,A,D}}    ->   C:B-A-D-
 */
public class FriendReduce1 extends Reducer<Text, Text, Text, Text> {
    @Override
    protected void reduce(Text key, Iterable<Text> values, Context context) throws IOException, InterruptedException {
        //1. 拼接用户列表.
        String users = "";
        //2. 遍历, 获取到每一个用户.
        for (Text value : values) {
            users += (value.toString() + "-");   //B-A-D-
        }
        //3. 写出结果, 即: k3, v3
        context.write(key, new Text(users));
    }
}

main1

package com.itheima.demo05_friend;

import com.itheima.demo01_lottery.LotteryMapper;
import com.itheima.demo01_lottery.LotteryReducer;
import com.itheima.demo01_lottery.MyPartitioner;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat;

//第一个MR的驱动类: 用来实现  谁有A, 而不是 A有谁.
public class FriendMain1 {
    public static void main(String[] args) throws Exception {
        //1. 创建Job任务.
        Job job = Job.getInstance(new Configuration(), "friendMR1");

        //2. 封装MR程序的核心8步.
        //2.1 封装输入组件, 读取: 源数据.
        job.setInputFormatClass(TextInputFormat.class);
        TextInputFormat.addInputPath(job, new Path("file:///D:\\abc\\共同好友\\input\\friends.txt"));
        //2.2 封装Map任务, 指定: k2, v2的类型.
        job.setMapperClass(FriendMapper1.class);
        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(Text.class);
        //2.3 分区, 用默认的.
        //2.4 排序, 用默认的
        //2.5 规约, 用默认的
        //2.6 分组, 用默认的
        //2.7 封装Reduce任务, 指定k3,v3的类型
        job.setReducerClass(FriendReduce1.class);
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(Text.class);
        //2.8 封装输出组件, 写出数据到 目的地文件中.
        job.setOutputFormatClass(TextOutputFormat.class);
        TextOutputFormat.setOutputPath(job, new Path("file:///D:\\abc\\共同好友\\output"));

        //3. 提交任务, 等待执行结果.
        boolean flag = job.waitForCompletion(true);

        //4. 关闭程序.
        System.exit(flag ? 0 : 1);
    }
}

mapper2

package com.itheima.demo05_friend;

import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;

import java.io.IOException;
import java.util.Arrays;

//自定义的Map阶段, 求共同好友.
public class FriendMapper2 extends Mapper<LongWritable, Text, Text, Text> {
    @Override
    protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
        //             好友  有该好友的用户
        //目前的数据格式:  C	B-A-D-
        //1. 获取数据.
        String line = value.toString();
        //2. 非空校验.
        if (line != null && !"".equals(line)) {
            String[] arr = line.split("\t");
            //3. 获取所有的 有该好友的用户
            String[] allFriends = arr[1].split("-");     //["B", "A", "D"]
            //4. 对所有的用户 升序排序.
            Arrays.sort(allFriends);         //["A", "B", "D"]
            //5. 写出数据, 即: 写出k2(有该共同好友的两个用户), v2(共同好友)
            for (int i = 0; i < allFriends.length - 1; i++) {       //"A", "B"
                String firstUser = allFriends[i];
                //获取第二个用户
                for (int j = i + 1; j < allFriends.length; j++) {
                    String secondUser = allFriends[j];
                    //写出结果:写出k2(有该共同好友的两个用户), v2(共同好友)
                    //A-B: C,  A-D: C,  B-D: C
                    context.write(new Text(firstUser +"-"+ secondUser), new Text(arr[0]));
                }
            }
        }
    }
}

reduce2

package com.itheima.demo05_friend;

import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;

import java.io.IOException;

//           两个有共同好友的用户                     他们的共同好友
//目前数据格式: {{A-B,A-B} 				: 			{C, D}}         =>   A-B: {C, D}
public class FriendReducer2 extends Reducer<Text, Text, Text, NullWritable> {
    @Override
    protected void reduce(Text key, Iterable<Text> values, Context context) throws IOException, InterruptedException {
        //1. 拼接用户列表.
        String friends = "";
        //2. 遍历, 获取到每一个用户.
        for (Text value : values) {
            friends += (value.toString() + ", ");   //"C, D, "
        }
        //3. 获取最终k3的格式: A-B: {C, D}
        String k3 = key.toString() + ": {" +  friends.substring(0, friends.length() - 2) + "}";
        //4. 写出结果, 即: k3, v3
        context.write(new Text(k3), NullWritable.get());
    }
}

main2

package com.itheima.demo05_friend;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat;

//第一个MR的驱动类: 用来实现  谁有A, 而不是 A有谁.
public class FriendMain2 {
    public static void main(String[] args) throws Exception {
        //1. 创建Job任务.
        Job job = Job.getInstance(new Configuration(), "friendMR2");

        //2. 封装MR程序的核心8步.
        //2.1 封装输入组件, 读取: 源数据.
        job.setInputFormatClass(TextInputFormat.class);
        TextInputFormat.addInputPath(job, new Path("file:///D:\\abc\\共同好友\\output\\part-r-00000"));
        //2.2 封装Map任务, 指定: k2, v2的类型.
        job.setMapperClass(FriendMapper2.class);
        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(Text.class);
        //2.3 分区, 用默认的.
        //2.4 排序, 用默认的
        //2.5 规约, 用默认的
        //2.6 分组, 用默认的
        //2.7 封装Reduce任务, 指定k3,v3的类型
        job.setReducerClass(FriendReducer2.class);
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(NullWritable.class);
        //2.8 封装输出组件, 写出数据到 目的地文件中.
        job.setOutputFormatClass(TextOutputFormat.class);
        TextOutputFormat.setOutputPath(job, new Path("file:///D:\\abc\\共同好友\\output2"));

        //3. 提交任务, 等待执行结果.
        boolean flag = job.waitForCompletion(true);

        //4. 关闭程序.
        System.exit(flag ? 0 : 1);
    }
}