freqanalysis.py

#!/usr/bin/env python3
# Calculate occurrence of letters in two text files,
# and map out the closest relation
# (ie, perform frequency analysis and resolution)
from sys import argv, exit
from os.path import basename
import string

if len(argv) <= 2:
	print("Usage: {} <file 1> <file b>".format(basename(argv[0])))
	exit(0)

files = argv[1:3]
freqs = {}
freqs_sorted = {}
value_totals = {}

for filename in files:

	# Initialize buffers
	freqs[filename] = dict.fromkeys(string.ascii_lowercase, 0)
	freqs_sorted[filename] = []
	value_totals[filename] = 0

	# Open and read the file
	with open(filename, 'r', encoding='utf-8') as fo:
		buf = fo.read().lower()

	# Count the characters (a-z only)
	for i, char in enumerate(buf):
		val = ord(char)
		if val >= ord(u'a') and val <= ord(u'z'):
			freqs[filename][char] += 1
	
	# Generate tuple lists fit for column printout
	for k, v in sorted(freqs[filename].items(), 
			   key=lambda x:x[1], 
			   reverse=True):
		tup = (k, v)
		value_totals[filename] += v
		freqs_sorted[filename].append(tup)

# Generate columns; fugly
buf = []

fa, fb = files[0], files[1]

left_offset = len(fa) + 10
letter_map = {}

print("Analysis:\n")
print(u'{}{}{}\n'.format(fa, ' ' * 10, fb), u'')

for a, b in zip(freqs_sorted[fa], 
		freqs_sorted[fb]): 

	str_a = u'{}: {:4d} ({:2.2f}%)'.format(
		a[0], a[1], (a[1] / value_totals[fa]) * 100)
	str_b = u'{}: {:4d} ({:2.2f}%)'.format(
		b[0], b[1], (b[1] / value_totals[fb]) * 100)

	print(u'{}{}{}'.format( 
		     str_a, 
		     ' ' * (left_offset - len(str_a)),
		     str_b))

	letter_map[a[0]] = b[0]

print()
with open(fa, 'r', encoding='utf-8') as fo:
	buf = fo.read().lower()

print(buf)
print("translated using the above table:\n")

for i in range(0, len(buf)):
	char = buf[i]
	val = ord(char)
	try:
		print(letter_map[char], end='')
	except KeyError:
		print(char, end='')