"""
Counting unique words, and the frequencies of words,
from the text stored in a specific file.

Improvements from word_count_take_3:
- We change print_word_frequencies so that the results are
printed sorted by frequency. To do this, we:
1. define a helper function called inverse_dictionary.
2. we modify print_word_frequenciecs to use that helper function.
"""

import os


"""
function count_words:
    argument: a filename
    return value: a dictionary, mapping words to frequencies.
    description: this function builds a dictionary such that
        dictionary[word] is the number of times the word occurs in
        the filename.
"""
def count_words(filename):
    if (os.path.isfile(filename) == False):
        print("\nError: file " + filename + " does not exist.\n")
        return

    in_file = open(filename, "r")

    # initialize the dictionary to empty
    result = {}
    for line in in_file:
        words = process_line(line)
        for word in words:
            if (word in result):
                result[word] += 1
            else:
                result[word] = 1

    return result

"""
function process_line:
    argument: a line of text
    return value: a list of words in that line, ignoring case and punctuation.
        Words appearing multiple times in the line will be included multiple
        lines in the output.

    Note: the dash character '-' is replaced by space, as dashes separate
    individual words from each other.
"""
def process_line(line):
    line = line.lower()
    new_line = ""

    for letter in line:
        # note: since we want to include double quotes and single quotes
        # in the list of characters to ignore, we must use triple quotes
        # here.
        if letter in """,.!"'()""":
            continue
        elif letter == '-':
            letter = ' '
        
        new_line = new_line + letter

    words = new_line.split()
    return words


"""
function print_word_frequencies:
    argument: a dictionary called in_dictionary
    return value: an inverse dictionary out_dictionary, such that
        out_dictionary[value] is the list of all keys in in_dictionary
        that contain that value.
    description: this function prints the contents of the dictionary in
        a way that is easy to read.
"""
def inverse_dictionary(in_dictionary):
    out_dictionary = {}
    for key in in_dictionary:
        value = in_dictionary[key]
        if (value in out_dictionary):
            list_of_keys = out_dictionary[value]
            list_of_keys.append(key)
        else:
            # note that we store a list containing the key, not just
            # the key itself.
            out_dictionary[value] = [key] 

    return out_dictionary


"""
function print_word_frequencies:
    argument: a dictionary, mapping words to frequencies
    return value: nothing is returned.
    description: this function prints the contents of the dictionary in
        a way that is easy to read.
"""
def print_word_frequencies(dictionary):
    print()
    inverse = inverse_dictionary(dictionary)
    frequencies = inverse.keys()

    # convert frequencies to a list, so that we can sort it.
    frequencies = list(frequencies)
    frequencies.sort()

    # the sort() method sorts in ascending order. We want descending order,
    # so we reverse the list.
    frequencies.reverse()
    
    for frequency in frequencies:
        list_of_words = inverse[frequency]

        # for words with the same frequency, we want them sorted in
        # alphabetical order, so we sort them.
        list_of_words.sort()
        for word in list_of_words:
            print(word + ":", frequency)

    print()
    print(len(dictionary), 'words found\n')

def main():
    filename = "file1.txt"
    dictionary = count_words(filename)
    print_word_frequencies(dictionary)

main()