/***********************************************************************
* Program:
*    Assignment 16, Assignment 16
*    Brother Ercanbrack, CS235 
* Author:
*    Derek Hildreth
* Summary: 
*    Searches through the full Book of Mormon and returns the top 100
*    words used and how many times that word occured.  Uses map.
*
* ********************************************************************
* ****NOTE for Bro. Ercanbrack:***************************************
*       There is some unused commented code down below, and it is
*       there so that you know of the conditions I tested for, and
*       still never matched the output word for word.  I was one off
*       in the overall words processed and a for some reason "below"
*       was on the list of top 100.
**********************************************************************/
#include <fstream>
#include <iostream>
#include <iomanip>
#include <string>
#include <map>
using namespace std;

/***********************************************************************
* This is Main!  
***********************************************************************/
int main(int argc, char* argv[])
{
   char* inFileName;
   string word;
   string seperate;
   int howMany;
   int wordCount = 0;
   int counter = 0;
   
   //****If there is argument after a.out, then...****//
   if (argc > 1)
   {
      inFileName = argv[1];
   }

   //****If there is not argument after a.out, then...****//
   else
   {
      cout << "Enter input filename: "; 
      inFileName = new char[80]; 
      cin >> inFileName;
   }
   
   ifstream file;
   file.open(inFileName);
      
   //****Checks to see that file is valid.****//
   if (file.fail())
   {
      cout << "Invalid" << endl;
      exit(1);
   }

   file >> word;
   
   map < string, int > theMap;
   map < int, string > theMap2;
   
   //****Read file into the list****//
   while (!file.eof())
   {
      for (int i = 0; i < word.length(); i++)
      {
         //*****Converting all words to lowercase *****//
         word[i] = tolower(word[i]);
         
//          //*****Checking for case "Lamanites--Written" *****//
//          if (word[i] == '-' && word[i + 1] == '-' && isalpha(word[i + 2]))
//          {
//             seperate = word.substr(i + 2);
//             word.erase(i);

//             //****After seperating, the word needs to be lowercased*****//
//             for (int i = 0; i < seperate.length(); i++)
//             {
//                seperate[i] = tolower(seperate[i]);
//             }
//             theMap[seperate]++;
//          }
         
         //****Checking for cases of punctuation including "they're" ****//
         if (!isalpha(word[i]) && word[i] != '-')
         {
            word.replace(i, 1, "", 0);
            
            //****After deleting, the word needs to be lowercased...
            //****Particularly helpful in cases of "(Alalikiah)" *****//
            for (int i = 0; i < word.length(); i++)
            {
               word[i] = tolower(word[i]);
            }
         }

         //****Checking for cases of "Mormon--" ******//
         if (word[i] == '-' && word[i + 1] == '-')
         {
            word.erase(i);
         }

//          //****Checking for digits between 0 and 9 ******//
//          if (isdigit(word[i]))
//          {
//             word.erase(i);
//          }
      }
      
      //*****Checking for return or blank characters*****//
//      if (word.length() == 0)
//      {
//         file >> word;
//      }
//      else
//      {      
      theMap[word]++;
      file >> word;
//      }
   }
   
   map < string, int > :: iterator it;
   map < int, string > :: reverse_iterator rit2;
   
   for (it = theMap.begin(); it != theMap.end(); it++)
   {
      howMany = it->second;
      theMap2[howMany] = it->first;
      wordCount += it->second;
   }
   
   cout << "\nNumber of words processed: " << wordCount << endl;
   cout << "100 most common words found and their frequencies: " << endl;
   
   for (rit2 = theMap2.rbegin(); rit2 != theMap2.rend(); ++rit2)
   {
      if (counter < 100)
         
         cout << setw(23) << setfill(' ') << rit2->second << " - "
              << rit2->first << endl;
      else
         break;
      
      counter++;
   }
   
}