10.3

# #!/usr/bin/python
#
# By dvdjaco
# Exercise 10.3
#

# most_frequent takes a string and prints the letters in decreasing order of frequency

import string

def most_frequent(s):
    # first remove all punctuation and spaces, and lowercase the string:
    s = s.translate(None,string.punctuation)
    s = s.translate(None,' ')
    s = s.lower()
#    print s

    # build a dictionary where keys are letters and values are counts
    d = dict()
    totalchars = 0
    for char in s:
        d[char] = d.get(char,0) + 1
        totalchars = totalchars + 1

    # build a list of tuples to sort the dictionary by descending value
    l = list()
    for char,count in d.items(): l.append( (count,char) )
    l.sort(reverse=True)

    # print the list including the relative frequency of each letter
    for count,char in l:
        r = 100*float(count)/float(totalchars)
        print char, count, (str("%.2f" % r)) + "%"

f = raw_input("Enter a filename: ")

try:
    fhand = open(f)
except err:
    print "Error opening file ", f, err

s = ''
for line in fhand:
    s = s + line.rstrip()
most_frequent(s)

#
# Below the results on a few books and a mark in the first letter with a different order when compared to http://en.wikipedia.org/wiki/Letter_frequency
#
# Hitchhiker's Guide to the Galaxy (EN)
# $ python exercise-10.3.py
# Enter a filename: hh.txt
# e 29391 10.26%
# t 23494 8.20%
# a 23143 8.08%
# s 18725 6.54% <--
# i 18246 6.37%
# o 16918 5.91%
# l 15218 5.31%
# r 15006 5.24%
# h 14849 5.18%
# n 14720 5.14%
#
#
# The Selfish Gene (EN)
# $ python exercise-10.3.py
# Enter a filename: sg.txt
# e 72639 12.56%
# t 52166 9.02%
# a 46869 8.10%
# i 43425 7.51% <--
# o 41231 7.13%
# s 41077 7.10%
# n 37575 6.49%
# r 32962 5.70%
# h 28264 4.89%
# l 28116 4.86%
#
# Don Quijote de la Mancha (ES)
# $ python exercise-10.3.py
# Enter a filename: pg2000.txt
# e 223827 13.15%
# a 194365 11.42%
# o 154709 9.09%
# s 126489 7.43%
# n 109490 6.43% <--
# r 102124 6.00%
# l 89630 5.27%
# d 87776 5.16%
# i 78739 4.63%
# u 78717 4.63%