-with open('words.txt', 'r') as f:
- keywords = [line.rstrip() for line in f]
-
-transpositions = collections.defaultdict(list)
-for word in keywords:
- transpositions[transpositions_of(word)] += [word]
-
-def frequencies(text):
- """Count the number of occurrences of each character in text
-
- >>> 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
- """
- #counts = collections.defaultdict(int)
- #for c in text:
- # counts[c] += 1
- #return counts
- return collections.Counter(c for c in text)
-letter_frequencies = frequencies
-
-
-def bigram_likelihood(bigram, bf, lf):
- return bf[bigram] / (lf[bigram[0]] * lf[bigram[1]])
-
-
-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, 0.080345432737...)
- >>> caesar_break('wxwmaxdgheetgwuxztgptedbgznitgwwhpguxyhkxbmhvvtlbhgtee' \
- 'raxlmhiixweblmxgxwmhmaxybkbgztgwztsxwbgmxgmert') # doctest: +ELLIPSIS
- (19, 0.11189290326...)
- >>> caesar_break('yltbbqnqnzvguvaxurorgenafsbezqvagbnornfgsbevpnaabjurer' \
- 'svaquvzyvxrnznazlybequrvfohgriraabjtbaruraprur') # doctest: +ELLIPSIS
- (13, 0.08293968842...)
- """
- 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), 0.0598745365924...)
- """
- 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), 0.1066453448861...)
- """
- 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), 0.106645344886...)
- """
- 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_worker, helper_args, chunksize)
- return min(breaks, key=lambda k: k[1])
-
-def keyword_break_worker(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