// Allison Obourn
// CSE 143 - lecture 2
// Counts the occurrences of words in a large file and prints the most 
// common ones in order of popularity.
// It demonstrates the use of a Map collection.

import java.io.*;
import java.util.*;

public class WordCount2 {
   public static void main(String[] args) throws FileNotFoundException {
      System.out.println("Reading file...");
      Scanner input = new Scanner(new File("mobydick.txt"));

      Map<String, Integer> words = new TreeMap<String, Integer>();
            
      while(input.hasNext()) {
         String word = input.next().toLowerCase();
         
         // when we get the value of a key that's not there,
         // we get null! Adding 1 to null causes an exception.
         // we therefore need to separate the cases where the key is already
         // in the map from the one where it's not.
         if(words.containsKey(word)) {
            words.put(word, words.get(word) + 1);
         } else {
            words.put(word, 1);
         }
         
      }
      
      // create a reversed version of the map so we can order by count
      Map<Integer, Set<String>> counts = new TreeMap<Integer, Set<String>>();
      for(String word : words.keySet()) {
         int count = words.get(word);
         if(count > 40) {
            if(!counts.containsKey(count)) {
               counts.put(count, new HashSet<String>());
            }
            counts.get(count).add(word);
         }
      }
      
      // iterate over the map using the set of keys
      for(int count : counts.keySet()) {
         System.out.println(count + " " + counts.get(count));
      }
   }
}