'''This file was modified 10-08-09 at approximately 6:30pm'''

def vocabulary(word_and_punctuation_instances, punctuation):
    '''word_and_punctuation_instances is a list of strings; punctuation is
    a list of punctuation characters. Return a new list that is the
    vocabulary of word_list, that is, all the distinct elements of
    word_and_punctuation_instances that are not in the punctuation list.
    By all the distinct elements, we mean the list returned does not include
    duplicates.'''
        
    vocabulary = []
    for word in word_and_punctuation_instances:
        if not word in vocabulary and not word in punctuation:
            vocabulary.append(word)
    return vocabulary

def analyze_vocabulary(word_and_punctuation_instances, punctuation):
    '''Determine and print the number of distinct words in word_and_punctuation_instances,
    as well as a sorted list of the distinct words.  "punctuation" is a list of 
    punctuation characters.  They may occur in word_and_punctuation_instances,
    but are not treated as words.'''
    
    v = vocabulary(word_and_punctuation_instances, punctuation)
    print "The number of distinct words is: ", len(v)
    v.sort()
    print "And the word list is: ", v

def print_stats(word_and_punctuation_instances, within_sentence_punctuation, \
		end_sentence_punctuation):    
    '''"word_and_punctuation_instances" is a list of strings;
    "within_sentence_punctuation" and "end_sentence_punctionation" are
    lists of punctuation characters that can occur in the middle
    of a sentence, or at the end of a sentence, respectively.  Print
    statistics about the words, sentences, and readability of
    word_and_punctuation_instances'''
    
    # Add code here to make your calculations.    

    # Calculate the number of distinct words
    numDistinctWords = len(vocabulary(word_and_punctuation_instances, \
                         within_sentence_punctuation + end_sentence_punctuation))

    # following would also be a good function
    # It calculates the number of words, that is, all the elements of
    # word_and_punctuation_instances that are not punctuation

    words = []
    for wp in word_and_punctuation_instances:
        if not wp in within_sentence_punctuation + end_sentence_punctuation:
            words.append(wp)

    totalWordLength = 0
    for w in words:
        totalWordLength = totalWordLength + len(w)
    aveWordLength = totalWordLength / float(len(words))    

    # Now, find the sentences, counting how many there are.
    # Note that we already have the numerator for the average sentence length -
    # it is len(words).  So, we don't need to count the sizes of the sentences
    # as we go along.

    number_sentences = 0
    for wp in word_and_punctuation_instances:
        if wp in end_sentence_punctuation:
            number_sentences += 1
            

    # In each print statment below, replace the value 99 with an appropriate
    # expression.
    print "Number of distinct words: ", numDistinctWords
    print "Number of words: ", len(words)
    print "Average word length: ", aveWordLength
    print "Number of sentences: ", number_sentences
    print "Average sentence length: ", float(len(words))/number_sentences
    print "ARI: ", 4.71 * (totalWordLength/float(len(words))) \
          + 0.5 * (len(words) /float(number_sentences)) - 21.43

    
def separate_punctuation(strings, punctuation_list):
    '''strings is a list of strings.  "punctuation" is a list of 
    punctuation characters.  Return a new list that is the same as strings, 
    but for every string that ends in a punctuation character, the string 
    (without the punctuation) and the punctuation are separated and
    made two elements of the list.'''

    new_strings = []
    for s in strings:
        if s[-1] in punctuation_list:
            word = s[:-1]
            punct = s[-1]
            new_strings.append(word)
            new_strings.append(punct)
        else:
            new_strings.append(s)
    return new_strings            


def print_one_line(list):
    '''Print the strings in 'list' on a single line, with commas separating
    them.'''
    for item in list:
	print item + ",",
	

def get_valid_command(valid):
    '''Prompt for and return a string that is a valid command, i.e., a string
    in the list "valid".'''
    
    # Uncomment out the following two lines and use them, as many times as 
    # needed, as your ONLY input and output in this function:
    
    command = raw_input("Please enter a command: ")
    while not command in valid:
        print "Invalid command"
        command = raw_input("Please enter a command: ")
    
    return command

	
if __name__ == "__main__":
    # Get the name of a file to analyze, and put its contents into a
    # list of strings.
    filename = raw_input("File to analyze: ")
    f = open(filename, 'r')
    contents = f.read().split()

    print "Raw contents, for debugging - don't print this out in your final solution"
    print contents
    # inner_punct is a list of within-sentence punctuation.
    # end_punct is a list of end-of-sentence punctuation.
    # punct is a list of all punctuation characters.
    inner_punct = [",", ";", "-"]
    end_punct = ["!", "?", "."]
    punct = inner_punct + end_punct
    
    # Separate any punctuation that is attached to the end of words.
    contents = separate_punctuation(contents, punct)
    print "Contents after separating out punctuation, for debugging  - don't print this out in your final solution"
    print contents
    
    # Tell the user what the valid commands are, then
    # read and process the user's command.
    valid_commands = ["stats", "vocab"]
    print "The valid commands are:", 
    print_one_line(valid_commands)
    command = get_valid_command(valid_commands)
    if command == "stats":
       print_stats(contents, inner_punct, end_punct)
    elif command == "vocab":
       analyze_vocabulary(contents, punct)