#!/usr/bin/env python

#

#  wordCount.py -- Written by Coty R Miller 13/Feb/18

#   ussage ./wordCount.py bookYouWishToProccess.TXT

#

#  This code is desised to count and list words within a book

#   or any genric text file. For example the book 1984 the word

#   'the' apears 5716 times.

#==================================================================

import sys

#==================================================================

# removePuncuation(ArrayOfStrings)

# return ArrayOfStrings

#==================================================================

def removePuncuation(book):

    # List of symbols to omit, you could also add numbers.

    symbolsToOmit = '`~!@#$%^&*()-_=+[{]}\|;:",<.>/?'

    newBook = {}                        # Create an empty array

    for i in range(0, len(book)):       # Loop until we've check all words.

        newWord = ""                    # Create empty String.

        for j in book[i]:               # Check all letters chars in the word.

            if(j not in symbolsToOmit): # If the symbol is not in our omit list.

                newWord+=j              # Then add it to our new word.

        newBook[i] = newWord            # Place the new word in the new array.

    return newBook                      # return our new array.

#==================================================================

# countWords(ArrayOfStrings)

# return dictionay

#==================================================================

def countWords(listOfWords):

    words = {"the":1}                    # Create a dictionary.

    for i in range(0, len(listOfWords)): # Check the whole array with a loop.

        if listOfWords[i] in words:      # Check if word is in the dictionary.

            a = words[listOfWords[i]]    # If so.. Grab the count.

            a = a + 1                    # Increment it.

            words[listOfWords[i]] = a    # and put it back in.

        else:

            a = listOfWords[i]           # If not insert it into

            words[a] = 1                 # the dictionary with a value of one.

    return words                         # return our new dictionary.

#==================================================================

# printWords(Dictionary)

# return Nothing

#==================================================================

def printWords(words):

    # sort words in dictionary into assending order

    for key, value in sorted(words.iteritems(), key=lambda (k,v): (v,k)):

        print key, "=" , value   # Print them.

#==================================================================

# Main Function

#==================================================================

print("Word counter...")

if len(sys.argv) > 1:                # If user provided a file name.

    file2load = sys.argv[1]          # Find desired file to load

    try:                             # TRY to...

        myfile = open(file2load)     # Load the file.

    except IOError as e:             # If we can't find it, error.

        print "Could not load file! '%s' Check the file name." % (file2load)

    else:

        book = myfile.read().split(" ")  # Split words into an array

        book = removePuncuation(book)    # Remove puctuation.

        words = countWords(book)         # Count the words.

        printWords(words)                # Print the result.

        print "Total word count =", len(book)

else:

    print "No system argument used."

#end