Home‎ > ‎

Hadoop: How to count number of times a word appeared in a file using Map Reduce framework

The following programs counts the number of times each word appeared in input file/directory and places the results in the output directory
 
1. Create file WordCount.java and paste the following code in it
 
 package org.myorg;
 import java.io.IOException;
 import java.util.*;
 import org.apache.hadoop.fs.Path;
 import org.apache.hadoop.conf.*;
 import org.apache.hadoop.io.*;
 import org.apache.hadoop.mapred.*;
 import org.apache.hadoop.util.*;
 public class WordCount {
    public static class Map extends MapReduceBase implements 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, OutputCollector<Text, IntWritable> output, Reporter reporter) throws IOException {
        String line = value.toString();
        StringTokenizer tokenizer = new StringTokenizer(line);
        while (tokenizer.hasMoreTokens()) {
          word.set(tokenizer.nextToken());
          output.collect(word, one);
        }
      }
    }
    public static class Reduce extends MapReduceBase implements Reducer<Text, IntWritable, Text, IntWritable> {
      public void reduce(Text key, Iterator<IntWritable> values, OutputCollector<Text, IntWritable> output, Reporter reporter) throws IOException {
        int sum = 0;
        while (values.hasNext()) {
          sum += values.next().get();
        }
        output.collect(key, new IntWritable(sum));
      }
    }
    public static void main(String[] args) throws Exception {
      JobConf conf = new JobConf(WordCount.class);
      conf.setJobName("wordcount");
      conf.setOutputKeyClass(Text.class);
      conf.setOutputValueClass(IntWritable.class);
      conf.setMapperClass(Map.class);
      conf.setCombinerClass(Reduce.class);
      conf.setReducerClass(Reduce.class);
      conf.setInputFormat(TextInputFormat.class);
      conf.setOutputFormat(TextOutputFormat.class);
      FileInputFormat.setInputPaths(conf, new Path(args[0]));
      FileOutputFormat.setOutputPath(conf, new Path(args[1]));
      JobClient.runJob(conf);
    }
 }

2. Compile, create jar and run
javac -classpath hadoop-0.20.1-dev-core.jar -d WordCount/ WordCount.java
jar -cvf WordCount.jar -C WordCount/ .
hadoop jar WordCount.jar org.myorg.WordCount in out
Comments