Mapreduce代码疑点（1）

一、Hadoop MultipleInputs.addInputPath 读取多个路径

https://blog.csdn.net/t1dmzks/article/details/76473905

MultipleInputs.addInputPath

作用
可以指定多个输入路径，每个路径都可以指定相应的map方法
使用方法
MultipleInputs.addInputPath
(Job job, Path path, Class<? extends InputFormat> inputFormatClass, Class<? extends Mapper> mapperClass)

举例

使用wordcount来举例
F:\hadooptest\wordcount\input1下有个word.txt，单词用空格分割

aa bb cc

dd ee ff

aa  bb  ff

F:\hadooptest\wordcount\input2下有个word.txt。单词用 ## 分割

aa##bb##cc
ee##gg##kk

代码

package com.myhadoop.multiple;

import com.myhadoop.mapreduce.test.WordCount;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
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.MultipleInputs;
import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat;

import java.io.IOException;
import java.util.StringTokenizer;

/**
 * Created by kaishun on 2017/7/31.
 */
public class TestMultipleInputs {
    public static class MapA extends Mapper<LongWritable, Text, Text, IntWritable>
    {
        private final static IntWritable one = new IntWritable(1);
        private Text word = new Text();
        public void map(LongWritable key,Text value,Context context) throws IOException,InterruptedException
        {
            String lines = value.toString();
            String strs[] = lines.split("\\s+");
            for (int i = 0; i <strs.length ; i++) {
                word.set(strs[i]);
                context.write(word, one);
            }
        }
    }

    public static class MapB extends Mapper<LongWritable, Text, Text, IntWritable>
    {
        private final static IntWritable one = new IntWritable(1);
        private Text word = new Text();
        public void map(LongWritable key,Text value,Context context) throws IOException,InterruptedException
        {
            String lines = value.toString();
            String strs[] = lines.split("##");
            for (int i = 0; i <strs.length ; i++) {
                word.set(strs[i]);
                context.write(word, one);
            }
        }
    }

    public static class Reduce extends Reducer<Text, IntWritable, Text, IntWritable>
    {
        public void reduce(Text key,Iterable<IntWritable> values,Context context) throws IOException,InterruptedException
        {
            int sum = 0;
            for (IntWritable val : values) {
                sum += val.get();
            }
            context.write(key, new IntWritable(sum));
        }
    }

    public static void main(String[] args) throws Exception {
        Configuration conf = new Configuration();
        Job job = Job.getInstance(conf);
        job.setJobName("MultipleWordCount");
        job.setJarByClass(WordCount.class);
        //多个输入，分别对应不同的map
        MultipleInputs.addInputPath(job,new Path("F:\\hadooptest\\wordcount\\input1"),TextInputFormat.class,WordCount.MapA.class);
        MultipleInputs.addInputPath(job,new Path("F:\\hadooptest\\wordcount\\input2"),TextInputFormat.class,WordCount.MapB.class);

        job.setNumReduceTasks(1);
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(IntWritable.class);
        //分到一个reduce
        job.setReducerClass(WordCount.Reduce.class);

        FileOutputFormat.setOutputPath(job, new Path(args[0]));
        System.exit(job.waitForCompletion(true) ? 0 : 1);

    }
}

输出

aa  3
bb  3
cc  2
dd  1
ee  2
ff  2
gg  1
kk  1

二、hadoop中的job.setOutputKeyClass与job.setMapOutputKeyClass

mr程序中一般都会有hadoop中的job.setOutputKeyClass(theClass)与job.setOutputValueClass(theClass)，

但是有的程序处理以上两个外还有job.setMapOutputKeyClass(theClass）与job.setMapOu

tputValueClass(Text.class)，一直没弄懂是怎么回事，网上查了下，原来当mapper与reducer

的输出类型一致时可以用　job.setOutputKeyClass(theClass)与job.setOutputValueClass

(theClass)这两个进行配置就行，但是当mapper用于reducer两个的输出类型不一致的时候就需

要分别进行配置了。