package org.myorg;

import java.io.IOException;
import java.util.regex.Pattern; // utilisation de regex.Pattern pour extraire les mots des fichiers d'entrée (input files).
import org.apache.hadoop.conf.Configured; // Cette application etends (hérite de) la classe Configured, 
import org.apache.hadoop.util.Tool;       // et implemente l'interface Tool. 
import org.apache.hadoop.util.ToolRunner; //Hadoop exécutera le programme dans une objet configuration object. 
										  // We use ToolRunner to run the  MapReduce application.
import org.apache.hadoop.mapreduce.Job; //  We  need the Job class to create, configure, and run an instance 
										 // of the   MapReduce application
import org.apache.hadoop.mapreduce.Mapper; // We extend the Mapper class with our own Mapclass 
										   // and add our own processing instructions
import org.apache.hadoop.mapreduce.Reducer; // We  extend Reducer class and customize our  own Reduce class.
import org.apache.hadoop.fs.Path;                               // Use the Path class to access files in HDFS. 
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;   // In our job configuration instructions, 
															    // we pass required paths using the FileInputFormat 
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat; // and FileOutputFormat classes.
import org.apache.hadoop.io.IntWritable;  // Writable objects have convenience methods for writing, reading, and comparing 
import org.apache.hadoop.io.LongWritable; //values during map and reduce processing
import org.apache.hadoop.io.Text;

import org.apache.log4j.Logger;// The Logger class sends debugging messages from inside the mapper and reducer classes
								// Messages  passed to Logger are displayed in the map or reduce logs for the job on the  Hadoop server.

//******  Tp_CC includes main and run methods, and the inner classes Map andReduce. 
//******  The class begins by initializing the logger
public class Tp_CC extends Configured implements Tool {

  //private static final Logger LOG = Logger.getLogger(Tp_CC.class);

  
  //**** The main method invokes ToolRunner, which creates and runs a new instance of Tp_CC, 
  //**** passing the command line arguments.
  public static void main(String[] args) throws Exception {
    int res = ToolRunner.run(new Tp_CC(), args);
    System.exit(res);
  }

  
  //**** The run method configures the job (which includes setting paths passed in at the command line), 
  //**** starts the job, waits for the job to complete, and then returns an integer value as the success flag.
  public int run(String[] args) throws Exception {
	
	  //***** Create a new Job  
    Job job = Job.getInstance(getConf(), this.getClass().toString());  // Job Class:used to configure the job, 
    																   // submit it, control its execution, 
    																   // and query the state.
    job.setJarByClass(this.getClass()); //Hadoop will find out the relevant jar by finding out that the class specified as
                                        // it's parameter to be present as part of that jar
    
    //************   Specify various job-specific parameters 
    //** Set the input and output paths for the  application. 
    //** We store our input files in HDFS, and then pass the input and output paths as command-line arguments at runtime.
    FileInputFormat.addInputPath(job, new Path(args[0])); // we can have more than one input path
    FileOutputFormat.setOutputPath(job, new Path(args[1]));
    
    //** Set the map class and reduce class for the job. 
    job.setMapperClass(Map.class);     // In this case, use the Map and Reduce inner classes defined in this class.
    job.setReducerClass(Reduce.class);
    
    //** Use a Text object to output the key (in this case, the word being counted) 
    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(IntWritable.class); // and the value (in this case, the number of times the word appears).
    
    
   //**** Launch the job and wait for it to finish. The method syntax iswaitForCompletion(boolean verbose). 
   //**** When true, the method reports its progress as the Map and Reduce classes run. 
   //**** When false, the method reports progress up to, but not including, the Map and Reduce processes.
    return job.waitForCompletion(true) ? 0 : 1; // 0=Success, anything other than 0 indicates a failure like 1.
  }

//**** Hadoop invokes the map method once for every key/value pair from your input source. 
//**** the map method receives the offset of the first character in the current line of input as the key, 
//**** and a Text object representing an entire line of text from the input file as the value. 
  public static class Map extends Mapper<LongWritable, Text, Text, IntWritable> {
 
 @Override
  public void map(LongWritable offset, Text lineText, Context context) throws IOException, InterruptedException 
  {
      String []line = lineText.toString().split(";");   
	  .........................
      .........................   
	  context.write(new Text(..........), new IntWritable(........));
    }
  }

  public static class Reduce extends Reducer<Text, IntWritable, Text, IntWritable> {
    @Override
    public void reduce(Text A, Iterable<IntWritable> counts, Context context) throws IOException, InterruptedException 
	{
         ........................
      for (IntWritable count : counts) 
	  {
         ........................
      }
         context.write(A, new IntWritable(........));
    }
  }
}

/*
 Jars can have files inside them that point to other classes and jars located elsewhere and when those other things 
 they point to don't exist, you see warnings like the ones you're getting
 */