import string
import collections
-import norms
-import logging
import math
-from itertools import zip_longest
-from segment import segment
-from multiprocessing import Pool
-
-# To time a run:
-#
-# import timeit
-# c5a = open('2012/5a.ciphertext', 'r').read()
-# timeit.timeit('keyword_break(c5a)', setup='gc.enable() ; from __main__ import c5a ; from cipher import keyword_break', number=1)
-# timeit.repeat('keyword_break_mp(c5a, chunksize=500)', setup='gc.enable() ; from __main__ import c5a ; from cipher import keyword_break_mp', repeat=5, number=1
-
-logger = logging.getLogger(__name__)
-logger.addHandler(logging.FileHandler('cipher.log'))
-logger.setLevel(logging.WARNING)
-#logger.setLevel(logging.INFO)
-#logger.setLevel(logging.DEBUG)
-
-english_counts = collections.defaultdict(int)
-with open('count_1l.txt', 'r') as f:
- for line in f:
- (letter, count) = line.split("\t")
- english_counts[letter] = int(count)
-normalised_english_counts = norms.normalise(english_counts)
-
-english_bigram_counts = collections.defaultdict(int)
-with open('count_2l.txt', 'r') as f:
- for line in f:
- (bigram, count) = line.split("\t")
- english_bigram_counts[bigram] = int(count)
-normalised_english_bigram_counts = norms.normalise(english_bigram_counts)
-
-with open('words.txt', 'r') as f:
- keywords = [line.rstrip() for line in f]
-
-modular_division_table = [[0]*26 for x in range(26)]
+from enum import Enum
+from itertools import zip_longest, cycle, chain
+from language_models import *
+
+
+modular_division_table = [[0]*26 for _ in range(26)]
for a in range(26):
for b in range(26):
c = (a * b) % 26
modular_division_table[b][c] = a
-def letters(text):
- """Remove all non-alphabetic characters from a text
- >>> letters('The Quick')
- 'TheQuick'
- >>> letters('The Quick BROWN fox jumped! over... the (9lazy) DOG')
- 'TheQuickBROWNfoxjumpedoverthelazyDOG'
- """
- return ''.join([c for c in text if c in string.ascii_letters])
-
-def sanitise(text):
- """Remove all non-alphabetic characters and convert the text to lowercase
-
- >>> sanitise('The Quick')
- 'thequick'
- >>> sanitise('The Quick BROWN fox jumped! over... the (9lazy) DOG')
- 'thequickbrownfoxjumpedoverthelazydog'
- """
- # sanitised = [c.lower() for c in text if c in string.ascii_letters]
- # return ''.join(sanitised)
- return letters(text).lower()
-
-def ngrams(text, n):
- """Returns all n-grams of a text
-
- >>> ngrams(sanitise('the quick brown fox'), 2) # doctest: +NORMALIZE_WHITESPACE
- ['th', 'he', 'eq', 'qu', 'ui', 'ic', 'ck', 'kb', 'br', 'ro', 'ow', 'wn',
- 'nf', 'fo', 'ox']
- >>> ngrams(sanitise('the quick brown fox'), 4) # doctest: +NORMALIZE_WHITESPACE
- ['theq', 'hequ', 'equi', 'quic', 'uick', 'ickb', 'ckbr', 'kbro', 'brow',
- 'rown', 'ownf', 'wnfo', 'nfox']
- """
- return [text[i:i+n] for i in range(len(text)-n+1)]
-def every_nth(text, n):
+def every_nth(text, n, fillvalue=''):
"""Returns n strings, each of which consists of every nth character,
starting with the 0th, 1st, 2nd, ... (n-1)th character
>>> every_nth(string.ascii_lowercase, 5)
['afkpuz', 'bglqv', 'chmrw', 'dinsx', 'ejoty']
- >>> every_nth(string.ascii_lowercase, 1)
+ >>> every_nth(string.ascii_lowercase, 1)
['abcdefghijklmnopqrstuvwxyz']
>>> every_nth(string.ascii_lowercase, 26) # doctest: +NORMALIZE_WHITESPACE
['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n',
'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z']
+ >>> every_nth(string.ascii_lowercase, 5, fillvalue='!')
+ ['afkpuz', 'bglqv!', 'chmrw!', 'dinsx!', 'ejoty!']
"""
- split_text = [text[i:i+n] for i in range(0, len(text), n)]
- return [''.join(l) for l in zip_longest(*split_text, fillvalue='')]
+ split_text = chunks(text, n, fillvalue)
+ return [''.join(l) for l in zip_longest(*split_text, fillvalue=fillvalue)]
def combine_every_nth(split_text):
"""Reforms a text split into every_nth strings
return ''.join([''.join(l)
for l in zip_longest(*split_text, fillvalue='')])
+def chunks(text, n, fillvalue=None):
+ """Split a text into chunks of n characters
-def frequencies(text):
- """Count the number of occurrences of each character in text
+ >>> chunks('abcdefghi', 3)
+ ['abc', 'def', 'ghi']
+ >>> chunks('abcdefghi', 4)
+ ['abcd', 'efgh', 'i']
+ >>> chunks('abcdefghi', 4, fillvalue='!')
+ ['abcd', 'efgh', 'i!!!']
+ """
+ if fillvalue:
+ padding = fillvalue[0] * (n - len(text) % n)
+ else:
+ padding = ''
+ return [(text+padding)[i:i+n] for i in range(0, len(text), n)]
+
+def transpose(items, transposition):
+ """Moves items around according to the given transposition
- >>> sorted(frequencies('abcdefabc').items())
- [('a', 2), ('b', 2), ('c', 2), ('d', 1), ('e', 1), ('f', 1)]
- >>> sorted(frequencies('the quick brown fox jumped over the lazy ' \
- 'dog').items()) # doctest: +NORMALIZE_WHITESPACE
- [(' ', 8), ('a', 1), ('b', 1), ('c', 1), ('d', 2), ('e', 4), ('f', 1),
- ('g', 1), ('h', 2), ('i', 1), ('j', 1), ('k', 1), ('l', 1), ('m', 1),
- ('n', 1), ('o', 4), ('p', 1), ('q', 1), ('r', 2), ('t', 2), ('u', 2),
- ('v', 1), ('w', 1), ('x', 1), ('y', 1), ('z', 1)]
- >>> sorted(frequencies('The Quick BROWN fox jumped! over... the ' \
- '(9lazy) DOG').items()) # doctest: +NORMALIZE_WHITESPACE
- [(' ', 8), ('!', 1), ('(', 1), (')', 1), ('.', 3), ('9', 1), ('B', 1),
- ('D', 1), ('G', 1), ('N', 1), ('O', 2), ('Q', 1), ('R', 1), ('T', 1),
- ('W', 1), ('a', 1), ('c', 1), ('d', 1), ('e', 4), ('f', 1), ('h', 2),
- ('i', 1), ('j', 1), ('k', 1), ('l', 1), ('m', 1), ('o', 2), ('p', 1),
- ('r', 1), ('t', 1), ('u', 2), ('v', 1), ('x', 1), ('y', 1), ('z', 1)]
- >>> sorted(frequencies(sanitise('The Quick BROWN fox jumped! over... ' \
- 'the (9lazy) DOG')).items()) # doctest: +NORMALIZE_WHITESPACE
- [('a', 1), ('b', 1), ('c', 1), ('d', 2), ('e', 4), ('f', 1), ('g', 1),
- ('h', 2), ('i', 1), ('j', 1), ('k', 1), ('l', 1), ('m', 1), ('n', 1),
- ('o', 4), ('p', 1), ('q', 1), ('r', 2), ('t', 2), ('u', 2), ('v', 1),
- ('w', 1), ('x', 1), ('y', 1), ('z', 1)]
- >>> frequencies('abcdefabcdef')['x']
- 0
+ >>> transpose(['a', 'b', 'c', 'd'], (0,1,2,3))
+ ['a', 'b', 'c', 'd']
+ >>> transpose(['a', 'b', 'c', 'd'], (3,1,2,0))
+ ['d', 'b', 'c', 'a']
+ >>> transpose([10,11,12,13,14,15], (3,2,4,1,5,0))
+ [13, 12, 14, 11, 15, 10]
"""
- #counts = collections.defaultdict(int)
- #for c in text:
- # counts[c] += 1
- #return counts
- return collections.Counter(c for c in text)
-letter_frequencies = frequencies
+ transposed = [''] * len(transposition)
+ for p, t in enumerate(transposition):
+ transposed[p] = items[t]
+ return transposed
+
+def untranspose(items, transposition):
+ """Undoes a transpose
+
+ >>> untranspose(['a', 'b', 'c', 'd'], [0,1,2,3])
+ ['a', 'b', 'c', 'd']
+ >>> untranspose(['d', 'b', 'c', 'a'], [3,1,2,0])
+ ['a', 'b', 'c', 'd']
+ >>> untranspose([13, 12, 14, 11, 15, 10], [3,2,4,1,5,0])
+ [10, 11, 12, 13, 14, 15]
+ """
+ transposed = [''] * len(transposition)
+ for p, t in enumerate(transposition):
+ transposed[t] = items[p]
+ return transposed
def deduplicate(text):
return list(collections.OrderedDict.fromkeys(text))
-
-def caesar_encipher_letter(letter, shift):
+def caesar_encipher_letter(accented_letter, shift):
"""Encipher a letter, given a shift amount
>>> caesar_encipher_letter('a', 1)
'y'
>>> caesar_encipher_letter('a', -1)
'z'
+ >>> caesar_encipher_letter('A', 1)
+ 'B'
+ >>> caesar_encipher_letter('é', 1)
+ 'f'
"""
+ letter = unaccent(accented_letter)
if letter in string.ascii_letters:
if letter in string.ascii_uppercase:
alphabet_start = ord('A')
'cdezab'
>>> caesar_encipher('ab cx yz', 2)
'cd ez ab'
+ >>> caesar_encipher('Héllo World!', 2)
+ 'Jgnnq Yqtnf!'
"""
enciphered = [caesar_encipher_letter(l, shift) for l in message]
return ''.join(enciphered)
def caesar_decipher(message, shift):
- """Encipher a message with the Caesar cipher of given shift
+ """Decipher a message with the Caesar cipher of given shift
>>> caesar_decipher('bcd', 1)
'abc'
'abc'
>>> caesar_decipher('cd ez ab', 2)
'ab cx yz'
+ >>> caesar_decipher('Jgnnq Yqtnf!', 2)
+ 'Hello World!'
"""
return caesar_encipher(message, -shift)
-def affine_encipher_letter(letter, multiplier=1, adder=0, one_based=True):
+def affine_encipher_letter(accented_letter, multiplier=1, adder=0, one_based=True):
"""Encipher a letter, given a multiplier and adder
>>> ''.join([affine_encipher_letter(l, 3, 5, True) \
for l in string.ascii_uppercase])
'FILORUXADGJMPSVYBEHKNQTWZC'
"""
+ letter = unaccent(accented_letter)
if letter in string.ascii_letters:
if letter in string.ascii_uppercase:
alphabet_start = ord('A')
alphabet_start = ord('a')
cipher_number = ord(letter) - alphabet_start
if one_based: cipher_number += 1
- plaintext_number = ( modular_division_table[multiplier]
- [(cipher_number - adder) % 26] )
+ plaintext_number = (
+ modular_division_table[multiplier]
+ [(cipher_number - adder) % 26])
if one_based: plaintext_number -= 1
return chr(plaintext_number % 26 + alphabet_start)
else:
return ''.join(enciphered)
-def keyword_cipher_alphabet_of(keyword, wrap_alphabet=0):
+class KeywordWrapAlphabet(Enum):
+ from_a = 1
+ from_last = 2
+ from_largest = 3
+
+
+def keyword_cipher_alphabet_of(keyword, wrap_alphabet=KeywordWrapAlphabet.from_a):
"""Find the cipher alphabet given a keyword.
wrap_alphabet controls how the rest of the alphabet is added
after the keyword.
- 0 : from 'a'
- 1 : from the last letter in the sanitised keyword
- 2 : from the largest letter in the sanitised keyword
>>> keyword_cipher_alphabet_of('bayes')
'bayescdfghijklmnopqrtuvwxz'
- >>> keyword_cipher_alphabet_of('bayes', 0)
+ >>> keyword_cipher_alphabet_of('bayes', KeywordWrapAlphabet.from_a)
'bayescdfghijklmnopqrtuvwxz'
- >>> keyword_cipher_alphabet_of('bayes', 1)
+ >>> keyword_cipher_alphabet_of('bayes', KeywordWrapAlphabet.from_last)
'bayestuvwxzcdfghijklmnopqr'
- >>> keyword_cipher_alphabet_of('bayes', 2)
+ >>> keyword_cipher_alphabet_of('bayes', KeywordWrapAlphabet.from_largest)
'bayeszcdfghijklmnopqrtuvwx'
"""
- if wrap_alphabet == 0:
+ if wrap_alphabet == KeywordWrapAlphabet.from_a:
cipher_alphabet = ''.join(deduplicate(sanitise(keyword) +
string.ascii_lowercase))
else:
- if wrap_alphabet == 1:
+ if wrap_alphabet == KeywordWrapAlphabet.from_last:
last_keyword_letter = deduplicate(sanitise(keyword))[-1]
else:
last_keyword_letter = sorted(sanitise(keyword))[-1]
return cipher_alphabet
-def keyword_encipher(message, keyword, wrap_alphabet=0):
+def keyword_encipher(message, keyword, wrap_alphabet=KeywordWrapAlphabet.from_a):
"""Enciphers a message with a keyword substitution cipher.
wrap_alphabet controls how the rest of the alphabet is added
after the keyword.
>>> keyword_encipher('test message', 'bayes')
'rsqr ksqqbds'
- >>> keyword_encipher('test message', 'bayes', 0)
+ >>> keyword_encipher('test message', 'bayes', KeywordWrapAlphabet.from_a)
'rsqr ksqqbds'
- >>> keyword_encipher('test message', 'bayes', 1)
+ >>> keyword_encipher('test message', 'bayes', KeywordWrapAlphabet.from_last)
'lskl dskkbus'
- >>> keyword_encipher('test message', 'bayes', 2)
+ >>> keyword_encipher('test message', 'bayes', KeywordWrapAlphabet.from_largest)
'qspq jsppbcs'
"""
cipher_alphabet = keyword_cipher_alphabet_of(keyword, wrap_alphabet)
cipher_translation = ''.maketrans(string.ascii_lowercase, cipher_alphabet)
- return message.lower().translate(cipher_translation)
+ return unaccent(message).lower().translate(cipher_translation)
-def keyword_decipher(message, keyword, wrap_alphabet=0):
+def keyword_decipher(message, keyword, wrap_alphabet=KeywordWrapAlphabet.from_a):
"""Deciphers a message with a keyword substitution cipher.
wrap_alphabet controls how the rest of the alphabet is added
after the keyword.
>>> keyword_decipher('rsqr ksqqbds', 'bayes')
'test message'
- >>> keyword_decipher('rsqr ksqqbds', 'bayes', 0)
+ >>> keyword_decipher('rsqr ksqqbds', 'bayes', KeywordWrapAlphabet.from_a)
'test message'
- >>> keyword_decipher('lskl dskkbus', 'bayes', 1)
+ >>> keyword_decipher('lskl dskkbus', 'bayes', KeywordWrapAlphabet.from_last)
'test message'
- >>> keyword_decipher('qspq jsppbcs', 'bayes', 2)
+ >>> keyword_decipher('qspq jsppbcs', 'bayes', KeywordWrapAlphabet.from_largest)
'test message'
"""
cipher_alphabet = keyword_cipher_alphabet_of(keyword, wrap_alphabet)
cipher_translation = ''.maketrans(cipher_alphabet, string.ascii_lowercase)
return message.lower().translate(cipher_translation)
-def scytale_encipher(message, rows):
+
+def vigenere_encipher(message, keyword):
+ """Vigenere encipher
+
+ >>> vigenere_encipher('hello', 'abc')
+ 'hfnlp'
+ """
+ shifts = [ord(l) - ord('a') for l in sanitise(keyword)]
+ pairs = zip(message, cycle(shifts))
+ return ''.join([caesar_encipher_letter(l, k) for l, k in pairs])
+
+def vigenere_decipher(message, keyword):
+ """Vigenere decipher
+
+ >>> vigenere_decipher('hfnlp', 'abc')
+ 'hello'
+ """
+ shifts = [ord(l) - ord('a') for l in sanitise(keyword)]
+ pairs = zip(message, cycle(shifts))
+ return ''.join([caesar_decipher_letter(l, k) for l, k in pairs])
+
+beaufort_encipher=vigenere_decipher
+beaufort_decipher=vigenere_encipher
+
+
+def transpositions_of(keyword):
+ """Finds the transpostions given by a keyword. For instance, the keyword
+ 'clever' rearranges to 'celrv', so the first column (0) stays first, the
+ second column (1) moves to third, the third column (2) moves to second,
+ and so on.
+
+ If passed a tuple, assume it's already a transposition and just return it.
+
+ >>> transpositions_of('clever')
+ (0, 2, 1, 4, 3)
+ >>> transpositions_of('fred')
+ (3, 2, 0, 1)
+ >>> transpositions_of((3, 2, 0, 1))
+ (3, 2, 0, 1)
+ """
+ if isinstance(keyword, tuple):
+ return keyword
+ else:
+ key = deduplicate(keyword)
+ transpositions = tuple(key.index(l) for l in sorted(key))
+ return transpositions
+
+def pad(message_len, group_len, fillvalue):
+ padding_length = group_len - message_len % group_len
+ if padding_length == group_len: padding_length = 0
+ padding = ''
+ for i in range(padding_length):
+ if callable(fillvalue):
+ padding += fillvalue()
+ else:
+ padding += fillvalue
+ return padding
+
+def column_transposition_encipher(message, keyword, fillvalue=' ',
+ fillcolumnwise=False,
+ emptycolumnwise=False):
+ """Enciphers using the column transposition cipher.
+ Message is padded to allow all rows to be the same length.
+
+ >>> column_transposition_encipher('hellothere', 'abcdef', fillcolumnwise=True)
+ 'hlohr eltee '
+ >>> column_transposition_encipher('hellothere', 'abcdef', fillcolumnwise=True, emptycolumnwise=True)
+ 'hellothere '
+ >>> column_transposition_encipher('hellothere', 'abcdef')
+ 'hellothere '
+ >>> column_transposition_encipher('hellothere', 'abcde')
+ 'hellothere'
+ >>> column_transposition_encipher('hellothere', 'abcde', fillcolumnwise=True, emptycolumnwise=True)
+ 'hellothere'
+ >>> column_transposition_encipher('hellothere', 'abcde', fillcolumnwise=True, emptycolumnwise=False)
+ 'hlohreltee'
+ >>> column_transposition_encipher('hellothere', 'abcde', fillcolumnwise=False, emptycolumnwise=True)
+ 'htehlelroe'
+ >>> column_transposition_encipher('hellothere', 'abcde', fillcolumnwise=False, emptycolumnwise=False)
+ 'hellothere'
+ >>> column_transposition_encipher('hellothere', 'clever', fillcolumnwise=True, emptycolumnwise=True)
+ 'heotllrehe'
+ >>> column_transposition_encipher('hellothere', 'clever', fillcolumnwise=True, emptycolumnwise=False)
+ 'holrhetlee'
+ >>> column_transposition_encipher('hellothere', 'clever', fillcolumnwise=False, emptycolumnwise=True)
+ 'htleehoelr'
+ >>> column_transposition_encipher('hellothere', 'clever', fillcolumnwise=False, emptycolumnwise=False)
+ 'hleolteher'
+ >>> column_transposition_encipher('hellothere', 'cleverly')
+ 'hleolthre e '
+ >>> column_transposition_encipher('hellothere', 'cleverly', fillvalue='!')
+ 'hleolthre!e!'
+ >>> column_transposition_encipher('hellothere', 'cleverly', fillvalue=lambda: '*')
+ 'hleolthre*e*'
+ """
+ transpositions = transpositions_of(keyword)
+ message += pad(len(message), len(transpositions), fillvalue)
+ if fillcolumnwise:
+ rows = every_nth(message, len(message) // len(transpositions))
+ else:
+ rows = chunks(message, len(transpositions))
+ transposed = [transpose(r, transpositions) for r in rows]
+ if emptycolumnwise:
+ return combine_every_nth(transposed)
+ else:
+ return ''.join(chain(*transposed))
+
+def column_transposition_decipher(message, keyword, fillvalue=' ',
+ fillcolumnwise=False,
+ emptycolumnwise=False):
+ """Deciphers using the column transposition cipher.
+ Message is padded to allow all rows to be the same length.
+
+ >>> column_transposition_decipher('hellothere', 'abcde', fillcolumnwise=True, emptycolumnwise=True)
+ 'hellothere'
+ >>> column_transposition_decipher('hlohreltee', 'abcde', fillcolumnwise=True, emptycolumnwise=False)
+ 'hellothere'
+ >>> column_transposition_decipher('htehlelroe', 'abcde', fillcolumnwise=False, emptycolumnwise=True)
+ 'hellothere'
+ >>> column_transposition_decipher('hellothere', 'abcde', fillcolumnwise=False, emptycolumnwise=False)
+ 'hellothere'
+ >>> column_transposition_decipher('heotllrehe', 'clever', fillcolumnwise=True, emptycolumnwise=True)
+ 'hellothere'
+ >>> column_transposition_decipher('holrhetlee', 'clever', fillcolumnwise=True, emptycolumnwise=False)
+ 'hellothere'
+ >>> column_transposition_decipher('htleehoelr', 'clever', fillcolumnwise=False, emptycolumnwise=True)
+ 'hellothere'
+ >>> column_transposition_decipher('hleolteher', 'clever', fillcolumnwise=False, emptycolumnwise=False)
+ 'hellothere'
+ """
+ transpositions = transpositions_of(keyword)
+ message += pad(len(message), len(transpositions), '*')
+ if emptycolumnwise:
+ rows = every_nth(message, len(message) // len(transpositions))
+ else:
+ rows = chunks(message, len(transpositions))
+ untransposed = [untranspose(r, transpositions) for r in rows]
+ if fillcolumnwise:
+ return combine_every_nth(untransposed)
+ else:
+ return ''.join(chain(*untransposed))
+
+def scytale_encipher(message, rows, fillvalue=' '):
"""Enciphers using the scytale transposition cipher.
Message is padded with spaces to allow all rows to be the same length.
>>> scytale_encipher('thequickbrownfox', 7)
'tqcrnx hukof eibwo '
"""
- if len(message) % rows != 0:
- message += ' '*(rows - len(message) % rows)
- row_length = round(len(message) / rows)
- slices = [message[i:i+row_length]
- for i in range(0, len(message), row_length)]
- return ''.join([''.join(r) for r in zip_longest(*slices, fillvalue='')])
+ # transpositions = [i for i in range(math.ceil(len(message) / rows))]
+ # return column_transposition_encipher(message, transpositions,
+ # fillvalue=fillvalue, fillcolumnwise=False, emptycolumnwise=True)
+ transpositions = [i for i in range(rows)]
+ return column_transposition_encipher(message, transpositions,
+ fillvalue=fillvalue, fillcolumnwise=True, emptycolumnwise=False)
def scytale_decipher(message, rows):
"""Deciphers using the scytale transposition cipher.
>>> scytale_decipher('tqcrnx hukof eibwo ', 7)
'thequickbrownfox '
"""
- cols = round(len(message) / rows)
- columns = [message[i:i+rows] for i in range(0, cols * rows, rows)]
- return ''.join([''.join(c) for c in zip_longest(*columns, fillvalue='')])
-
-
-def caesar_break(message,
- metric=norms.euclidean_distance,
- target_counts=normalised_english_counts,
- message_frequency_scaling=norms.normalise):
- """Breaks a Caesar cipher using frequency analysis
-
- >>> caesar_break('ibxcsyorsaqcheyklxivoexlevmrimwxsfiqevvmihrsasrxliwyrh' \
- 'ecjsppsamrkwleppfmergefifvmhixscsymjcsyqeoixlm') # doctest: +ELLIPSIS
- (4, 0.31863952890183...)
- >>> caesar_break('wxwmaxdgheetgwuxztgptedbgznitgwwhpguxyhkxbmhvvtlbhgtee' \
- 'raxlmhiixweblmxgxwmhmaxybkbgztgwztsxwbgmxgmert') # doctest: +ELLIPSIS
- (19, 0.42152901235832...)
- >>> caesar_break('yltbbqnqnzvguvaxurorgenafsbezqvagbnornfgsbevpnaabjurer' \
- 'svaquvzyvxrnznazlybequrvfohgriraabjtbaruraprur') # doctest: +ELLIPSIS
- (13, 0.316029208075451...)
+ # transpositions = [i for i in range(math.ceil(len(message) / rows))]
+ # return column_transposition_decipher(message, transpositions,
+ # fillcolumnwise=False, emptycolumnwise=True)
+ transpositions = [i for i in range(rows)]
+ return column_transposition_decipher(message, transpositions,
+ fillcolumnwise=True, emptycolumnwise=False)
+
+
+def railfence_encipher(message, height, fillvalue=' '):
+ """Railfence cipher
+
+ >>> railfence_encipher('hellothereavastmeheartiesthisisalongpieceoftextfortestingrailfenceciphers', 2, fillvalue='!')
+ 'hlohraateerishsslnpeefetotsigaleccpeselteevsmhatetiiaogicotxfretnrifneihr!'
+ >>> railfence_encipher('hellothereavastmeheartiesthisisalongpieceoftextfortestingrailfenceciphers', 3, fillvalue='!')
+ 'horaersslpeeosglcpselteevsmhatetiiaogicotxfretnrifneihr!!lhateihsnefttiaece!'
+ >>> railfence_encipher('hellothereavastmeheartiesthisisalongpieceoftextfortestingrailfenceciphers', 5, fillvalue='!')
+ 'hresleogcseeemhetaocofrnrner!!lhateihsnefttiaece!!ltvsatiigitxetifih!!oarspeslp!'
+ >>> railfence_encipher('hellothereavastmeheartiesthisisalongpieceoftextfortestingrailfenceciphers', 10, fillvalue='!')
+ 'hepisehagitnr!!lernesge!!lmtocerh!!otiletap!!tseaorii!!hassfolc!!evtitffe!!rahsetec!!eixn!'
"""
- sanitised_message = sanitise(message)
- best_shift = 0
- best_fit = float("inf")
- for shift in range(26):
- plaintext = caesar_decipher(sanitised_message, shift)
- counts = message_frequency_scaling(letter_frequencies(plaintext))
- fit = metric(target_counts, counts)
- logger.debug('Caesar break attempt using key {0} gives fit of {1} '
- 'and decrypt starting: {2}'.format(shift, fit, plaintext[:50]))
- if fit < best_fit:
- best_fit = fit
- best_shift = shift
- logger.info('Caesar break best fit: key {0} gives fit of {1} and '
- 'decrypt starting: {2}'.format(best_shift, best_fit,
- caesar_decipher(sanitised_message, best_shift)[:50]))
- return best_shift, best_fit
-
-def affine_break(message,
- metric=norms.euclidean_distance,
- target_counts=normalised_english_counts,
- message_frequency_scaling=norms.normalise):
- """Breaks an affine cipher using frequency analysis
-
- >>> affine_break('lmyfu bkuusd dyfaxw claol psfaom jfasd snsfg jfaoe ls ' \
- 'omytd jlaxe mh jm bfmibj umis hfsul axubafkjamx. ls kffkxwsd jls ' \
- 'ofgbjmwfkiu olfmxmtmwaokttg jlsx ls kffkxwsd jlsi zg tsxwjl. jlsx ' \
- 'ls umfjsd jlsi zg hfsqysxog. ls dmmdtsd mx jls bats mh bkbsf. ls ' \
- 'bfmctsd kfmyxd jls lyj, mztanamyu xmc jm clm cku tmmeaxw kj lai kxd ' \
- 'clm ckuxj.') # doctest: +ELLIPSIS
- ((15, 22, True), 0.23570361818655...)
+ sections = chunks(message, (height - 1) * 2, fillvalue=fillvalue)
+ # Add the top row
+ rows = [s[0] for s in sections]
+ # process the middle rows of the grid
+ for r in range(1, height - 1):
+ rows += [s[r] + s[-r] for s in sections]
+ # process the bottom row
+ rows += [s[height - 1] for s in sections]
+ return ''.join(rows)
+
+def railfence_decipher(message, height):
+ """Railfence decipher. Assumes the message is already the correct length.
+
+ >>> railfence_decipher('hlohraateerishsslnpeefetotsigaleccpeselteevsmhatetiiaogicotxfretnrifneihr!', 2).strip('!')
+ 'hellothereavastmeheartiesthisisalongpieceoftextfortestingrailfenceciphers'
+ >>> railfence_decipher('horaersslpeeosglcpselteevsmhatetiiaogicotxfretnrifneihr!!lhateihsnefttiaece!', 3).strip('!')
+ 'hellothereavastmeheartiesthisisalongpieceoftextfortestingrailfenceciphers'
+ >>> railfence_decipher('hresleogcseeemhetaocofrnrner!!lhateihsnefttiaece!!ltvsatiigitxetifih!!oarspeslp!', 5).strip('!')
+ 'hellothereavastmeheartiesthisisalongpieceoftextfortestingrailfenceciphers'
+ >>> railfence_decipher('hepisehagitnr!!lernesge!!lmtocerh!!otiletap!!tseaorii!!hassfolc!!evtitffe!!rahsetec!!eixn!', 10).strip('!')
+ 'hellothereavastmeheartiesthisisalongpieceoftextfortestingrailfenceciphers'
"""
- sanitised_message = sanitise(message)
- best_multiplier = 0
- best_adder = 0
- best_one_based = True
- best_fit = float("inf")
- for one_based in [True, False]:
- for multiplier in range(1, 26, 2):
- for adder in range(26):
- plaintext = affine_decipher(sanitised_message,
- multiplier, adder, one_based)
- counts = message_frequency_scaling(letter_frequencies(plaintext))
- fit = metric(target_counts, counts)
- logger.debug('Affine break attempt using key {0}x+{1} ({2}) '
- 'gives fit of {3} and decrypt starting: {4}'.
- format(multiplier, adder, one_based, fit,
- plaintext[:50]))
- if fit < best_fit:
- best_fit = fit
- best_multiplier = multiplier
- best_adder = adder
- best_one_based = one_based
- logger.info('Affine break best fit with key {0}x+{1} ({2}) gives fit of {3} '
- 'and decrypt starting: {4}'.format(
- best_multiplier, best_adder, best_one_based, best_fit,
- affine_decipher(sanitised_message, best_multiplier,
- best_adder, best_one_based)[:50]))
- return (best_multiplier, best_adder, best_one_based), best_fit
-
-def keyword_break(message,
- wordlist=keywords,
- metric=norms.euclidean_distance,
- target_counts=normalised_english_counts,
- message_frequency_scaling=norms.normalise):
- """Breaks a keyword substitution cipher using a dictionary and
- frequency analysis
-
- >>> keyword_break(keyword_encipher('this is a test message for the ' \
- 'keyword decipherment', 'elephant', 1), \
- wordlist=['cat', 'elephant', 'kangaroo']) # doctest: +ELLIPSIS
- (('elephant', 1), 0.41643991598441...)
+ n_secs = len(message) // ((height - 1) * 2)
+ downrows = [message[:n_secs]]
+ uprows = []
+ for r in range(height-2):
+ midrow = message[(2 * r + 1) * n_secs:(2 * r + 1) * n_secs + n_secs * 2]
+ downrows += [''.join([midrow[i] for i in range(0, len(midrow), 2)])]
+ uprows = [''.join([midrow[i] for i in range(1, len(midrow), 2)])] + uprows
+ downrows += [message[-n_secs:]]
+ rows = downrows + uprows
+ return ''.join(letter for section in zip(*rows) for letter in section)
+
+
+class PocketEnigma(object):
+ """A pocket enigma machine
+ The wheel is internally represented as a 26-element list self.wheel_map,
+ where wheel_map[i] == j shows that the position i places on from the arrow
+ maps to the position j places on.
"""
- best_keyword = ''
- best_wrap_alphabet = True
- best_fit = float("inf")
- for wrap_alphabet in range(3):
- for keyword in wordlist:
- plaintext = keyword_decipher(message, keyword, wrap_alphabet)
- counts = message_frequency_scaling(letter_frequencies(plaintext))
- fit = metric(target_counts, counts)
- logger.debug('Keyword break attempt using key {0} (wrap={1}) '
- 'gives fit of {2} and decrypt starting: {3}'.format(
- keyword, wrap_alphabet, fit,
- sanitise(plaintext)[:50]))
- if fit < best_fit:
- best_fit = fit
- best_keyword = keyword
- best_wrap_alphabet = wrap_alphabet
- logger.info('Keyword break best fit with key {0} (wrap={1}) gives fit of '
- '{2} and decrypt starting: {3}'.format(best_keyword,
- best_wrap_alphabet, best_fit, sanitise(
- keyword_decipher(message, best_keyword,
- best_wrap_alphabet))[:50]))
- return (best_keyword, best_wrap_alphabet), best_fit
-
-def keyword_break_mp(message,
- wordlist=keywords,
- metric=norms.euclidean_distance,
- target_counts=normalised_english_counts,
- message_frequency_scaling=norms.normalise,
- chunksize=500):
- """Breaks a keyword substitution cipher using a dictionary and
- frequency analysis
-
- >>> keyword_break_mp(keyword_encipher('this is a test message for the ' \
- 'keyword decipherment', 'elephant', 1), \
- wordlist=['cat', 'elephant', 'kangaroo']) # doctest: +ELLIPSIS
- (('elephant', 1), 0.41643991598441...)
- """
- with Pool() as pool:
- helper_args = [(message, word, wrap, metric, target_counts,
- message_frequency_scaling)
- for word in wordlist for wrap in range(3)]
- # Gotcha: the helper function here needs to be defined at the top level
- # (limitation of Pool.starmap)
- breaks = pool.starmap(keyword_break_one, helper_args, chunksize)
- return min(breaks, key=lambda k: k[1])
-
-def keyword_break_one(message, keyword, wrap_alphabet, metric, target_counts,
- message_frequency_scaling):
- plaintext = keyword_decipher(message, keyword, wrap_alphabet)
- counts = message_frequency_scaling(letter_frequencies(plaintext))
- fit = metric(target_counts, counts)
- logger.debug('Keyword break attempt using key {0} (wrap={1}) gives fit of '
- '{2} and decrypt starting: {3}'.format(keyword,
- wrap_alphabet, fit, sanitise(plaintext)[:50]))
- return (keyword, wrap_alphabet), fit
-
-def scytale_break(message,
- metric=norms.euclidean_distance,
- target_counts=normalised_english_bigram_counts,
- message_frequency_scaling=norms.normalise):
- """Breaks a Scytale cipher
-
- >>> scytale_break('tfeulchtrtteehwahsdehneoifeayfsondmwpltmaoalhikotoere' \
- 'dcweatehiplwxsnhooacgorrcrcraotohsgullasenylrendaianeplscdriioto' \
- 'aek') # doctest: +ELLIPSIS
- (6, 0.83453041115025...)
- """
- best_key = 0
- best_fit = float("inf")
- for key in range(1, 20):
- if len(message) % key == 0:
- plaintext = scytale_decipher(message, key)
- counts = message_frequency_scaling(frequencies(
- ngrams(sanitise(plaintext), 2)))
- fit = metric(target_counts, counts)
- logger.debug('Scytale break attempt using key {0} gives fit of '
- '{1} and decrypt starting: {2}'.format(key,
- fit, sanitise(plaintext)[:50]))
- if fit < best_fit:
- best_fit = fit
- best_key = key
- logger.info('Scytale break best fit with key {0} gives fit of {1} and '
- 'decrypt starting: {2}'.format(best_key, best_fit,
- sanitise(scytale_decipher(message, best_key))[:50]))
- return best_key, best_fit
+ def __init__(self, wheel=1, position='a'):
+ """initialise the pocket enigma, including which wheel to use and the
+ starting position of the wheel.
+
+ The wheel is either 1 or 2 (the predefined wheels) or a list of letter
+ pairs.
+
+ The position is the letter pointed to by the arrow on the wheel.
+
+ >>> pe.wheel_map
+ [25, 4, 23, 10, 1, 7, 9, 5, 12, 6, 3, 17, 8, 14, 13, 21, 19, 11, 20, 16, 18, 15, 24, 2, 22, 0]
+ >>> pe.position
+ 0
+ """
+ self.wheel1 = [('a', 'z'), ('b', 'e'), ('c', 'x'), ('d', 'k'),
+ ('f', 'h'), ('g', 'j'), ('i', 'm'), ('l', 'r'), ('n', 'o'),
+ ('p', 'v'), ('q', 't'), ('s', 'u'), ('w', 'y')]
+ self.wheel2 = [('a', 'c'), ('b', 'd'), ('e', 'w'), ('f', 'i'),
+ ('g', 'p'), ('h', 'm'), ('j', 'k'), ('l', 'n'), ('o', 'q'),
+ ('r', 'z'), ('s', 'u'), ('t', 'v'), ('x', 'y')]
+ if wheel == 1:
+ self.make_wheel_map(self.wheel1)
+ elif wheel == 2:
+ self.make_wheel_map(self.wheel2)
+ else:
+ self.validate_wheel_spec(wheel)
+ self.make_wheel_map(wheel)
+ if position in string.ascii_lowercase:
+ self.position = ord(position) - ord('a')
+ else:
+ self.position = position
+
+ def make_wheel_map(self, wheel_spec):
+ """Expands a wheel specification from a list of letter-letter pairs
+ into a full wheel_map.
+
+ >>> pe.make_wheel_map(pe.wheel2)
+ [2, 3, 0, 1, 22, 8, 15, 12, 5, 10, 9, 13, 7, 11, 16, 6, 14, 25, 20, 21, 18, 19, 4, 24, 23, 17]
+ """
+ self.validate_wheel_spec(wheel_spec)
+ self.wheel_map = [0] * 26
+ for p in wheel_spec:
+ self.wheel_map[ord(p[0]) - ord('a')] = ord(p[1]) - ord('a')
+ self.wheel_map[ord(p[1]) - ord('a')] = ord(p[0]) - ord('a')
+ return self.wheel_map
+
+ def validate_wheel_spec(self, wheel_spec):
+ """Validates that a wheel specificaiton will turn into a valid wheel
+ map.
+
+ >>> pe.validate_wheel_spec([])
+ Traceback (most recent call last):
+ ...
+ ValueError: Wheel specification has 0 pairs, requires 13
+ >>> pe.validate_wheel_spec([('a', 'b', 'c')]*13)
+ Traceback (most recent call last):
+ ...
+ ValueError: Not all mappings in wheel specificationhave two elements
+ >>> pe.validate_wheel_spec([('a', 'b')]*13)
+ Traceback (most recent call last):
+ ...
+ ValueError: Wheel specification does not contain 26 letters
+ """
+ if len(wheel_spec) != 13:
+ raise ValueError("Wheel specification has {} pairs, requires 13".
+ format(len(wheel_spec)))
+ for p in wheel_spec:
+ if len(p) != 2:
+ raise ValueError("Not all mappings in wheel specification"
+ "have two elements")
+ if len(set([p[0] for p in wheel_spec] +
+ [p[1] for p in wheel_spec])) != 26:
+ raise ValueError("Wheel specification does not contain 26 letters")
+
+ def encipher_letter(self, letter):
+ """Enciphers a single letter, by advancing the wheel before looking up
+ the letter on the wheel.
+
+ >>> pe.set_position('f')
+ 5
+ >>> pe.encipher_letter('k')
+ 'h'
+ """
+ self.advance()
+ return self.lookup(letter)
+ decipher_letter = encipher_letter
+
+ def lookup(self, letter):
+ """Look up what a letter enciphers to, without turning the wheel.
+
+ >>> pe.set_position('f')
+ 5
+ >>> ''.join([pe.lookup(l) for l in string.ascii_lowercase])
+ 'udhbfejcpgmokrliwntsayqzvx'
+ >>> pe.lookup('A')
+ ''
+ """
+ if letter in string.ascii_lowercase:
+ return chr(
+ (self.wheel_map[(ord(letter) - ord('a') - self.position) % 26] +
+ self.position) % 26 +
+ ord('a'))
+ else:
+ return ''
+
+ def advance(self):
+ """Advances the wheel one position.
+
+ >>> pe.set_position('f')
+ 5
+ >>> pe.advance()
+ 6
+ """
+ self.position = (self.position + 1) % 26
+ return self.position
+
+ def encipher(self, message, starting_position=None):
+ """Enciphers a whole message.
+
+ >>> pe.set_position('f')
+ 5
+ >>> pe.encipher('helloworld')
+ 'kjsglcjoqc'
+ >>> pe.set_position('f')
+ 5
+ >>> pe.encipher('kjsglcjoqc')
+ 'helloworld'
+ >>> pe.encipher('helloworld', starting_position = 'x')
+ 'egrekthnnf'
+ """
+ if starting_position:
+ self.set_position(starting_position)
+ transformed = ''
+ for l in message:
+ transformed += self.encipher_letter(l)
+ return transformed
+ decipher = encipher
+
+ def set_position(self, position):
+ """Sets the position of the wheel, by specifying the letter the arrow
+ points to.
+
+ >>> pe.set_position('a')
+ 0
+ >>> pe.set_position('m')
+ 12
+ >>> pe.set_position('z')
+ 25
+ """
+ self.position = ord(position) - ord('a')
+ return self.position
if __name__ == "__main__":
import doctest
- doctest.testmod()
+ doctest.testmod(extraglobs={'pe': PocketEnigma(1, 'a')})