X-Git-Url: https://git.njae.me.uk/?a=blobdiff_plain;f=cipherbreak.py;h=669d1330b898e888890f4cdd6e74f41cac98dcc4;hb=2199c12469620a8de4274518e5d0090142468d5b;hp=0ac8ae57f7ed11a443dfedc8366ddc51086eda8c;hpb=a9d938167b636e599586ebbc5aeec75c877e120f;p=cipher-tools.git diff --git a/cipherbreak.py b/cipherbreak.py index 0ac8ae5..669d133 100644 --- a/cipherbreak.py +++ b/cipherbreak.py @@ -13,11 +13,22 @@ from multiprocessing import Pool import matplotlib.pyplot as plt -logger = logging.getLogger(__name__) -logger.addHandler(logging.FileHandler('cipher.log')) +# logging.basicConfig(filename="cipher.log", level=logging.INFO) +# logger = logging.getLogger(__name__) + +logger = logging.getLogger('cipherbreak') logger.setLevel(logging.WARNING) -#logger.setLevel(logging.INFO) -#logger.setLevel(logging.DEBUG) +# logger.setLevel(logging.INFO) +# logger.setLevel(logging.DEBUG) + +# create the logging file handler +fh = logging.FileHandler("cipher.log") +formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s') +fh.setFormatter(formatter) + +# add handler to logger object +logger.addHandler(fh) + from cipher import * from language_models import * @@ -204,7 +215,7 @@ def monoalphabetic_break_hillclimbing(message, max_iterations=10000000, if not alphabet: alphabet = list(string.ascii_lowercase) random.shuffle(alphabet) - alphabet = ''.join(alphabet) + alphabet = cat(alphabet) return monoalphabetic_break_hillclimbing_worker(ciphertext, alphabet, max_iterations, fitness) @@ -218,7 +229,7 @@ def monoalphabetic_break_hillclimbing_mp(message, workers=10, else: this_alphabet = list(string.ascii_lowercase) random.shuffle(this_alphabet) - this_alphabet = ''.join(this_alphabet) + this_alphabet = cat(this_alphabet) worker_args.append((ciphertext, this_alphabet, max_iterations, fitness)) with Pool() as pool: breaks = pool.starmap(monoalphabetic_break_hillclimbing_worker, @@ -255,7 +266,7 @@ def vigenere_keyword_break_mp(message, wordlist=keywords, fitness=Pletters, >>> vigenere_keyword_break_mp(vigenere_encipher(sanitise('this is a test ' \ 'message for the vigenere decipherment'), 'cat'), \ wordlist=['cat', 'elephant', 'kangaroo']) # doctest: +ELLIPSIS - ('cat', -52.947271216...) + ('cat', -52.9472712...) """ with Pool() as pool: helper_args = [(message, word, fitness) @@ -286,11 +297,11 @@ def vigenere_frequency_break(message, max_key_length=20, fitness=Pletters): "certain that the theft has been discovered and that I will " \ "be caught. The SS officer visits less often now that he is " \ "sure"), 'florence')) # doctest: +ELLIPSIS - ('florence', -307.5473096791...) + ('florence', -307.5473096...) """ def worker(message, key_length, fitness): splits = every_nth(sanitised_message, key_length) - key = ''.join([chr(caesar_break(s)[0] + ord('a')) for s in splits]) + key = cat([chr(caesar_break(s)[0] + ord('a')) for s in splits]) plaintext = vigenere_decipher(message, key) fit = fitness(plaintext) return key, fit @@ -314,7 +325,7 @@ def beaufort_frequency_break(message, max_key_length=20, fitness=Pletters): """ def worker(message, key_length, fitness): splits = every_nth(sanitised_message, key_length) - key = ''.join([chr(-caesar_break(s)[0] % 26 + ord('a')) + key = cat([chr(-caesar_break(s)[0] % 26 + ord('a')) for s in splits]) plaintext = beaufort_decipher(message, key) fit = fitness(plaintext) @@ -325,6 +336,78 @@ def beaufort_frequency_break(message, max_key_length=20, fitness=Pletters): return max(results, key=lambda k: k[1]) +def polybius_break_mp(message, column_labels, row_labels, + letters_to_merge=None, + wordlist=keywords, fitness=Pletters, + number_of_solutions=1, chunksize=500): + """Breaks a Polybius substitution cipher using a dictionary and + frequency analysis + + >>> polybius_break_mp(polybius_encipher('this is a test message for the ' \ + 'polybius decipherment', 'elephant', 'abcde', 'abcde'), \ + 'abcde', 'abcde', \ + wordlist=['cat', 'elephant', 'kangaroo']) # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE + (('elephant', , 'abcde', 'abcde', False), \ + -54.53880...) + >>> polybius_break_mp(polybius_encipher('this is a test message for the ' \ + 'polybius decipherment', 'elephant', 'abcde', 'abcde', column_first=True), \ + 'abcde', 'abcde', \ + wordlist=['cat', 'elephant', 'kangaroo']) # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE + (('elephant', , 'abcde', 'abcde', True), \ + -54.53880...) + >>> polybius_break_mp(polybius_encipher('this is a test message for the ' \ + 'polybius decipherment', 'elephant', 'abcde', 'abcde', column_first=False), \ + 'abcde', 'abcde', \ + wordlist=['cat', 'elephant', 'kangaroo']) # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE + (('elephant', , 'abcde', 'abcde', False), \ + -54.53880...) + >>> polybius_break_mp(polybius_encipher('this is a test message for the ' \ + 'polybius decipherment', 'elephant', 'abcde', 'pqrst', column_first=True), \ + 'abcde', 'pqrst', \ + wordlist=['cat', 'elephant', 'kangaroo']) # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE + (('elephant', , 'abcde', 'pqrst', True), \ + -54.53880...) + """ + if letters_to_merge is None: + letters_to_merge = {'j': 'i'} + with Pool() as pool: + helper_args = [(message, word, wrap, + column_labels, row_labels, column_first, + letters_to_merge, + fitness) + for word in wordlist + for wrap in KeywordWrapAlphabet + for column_first in [False, True]] + # Gotcha: the helper function here needs to be defined at the top level + # (limitation of Pool.starmap) + breaks = pool.starmap(polybius_break_worker, helper_args, chunksize) + if number_of_solutions == 1: + return max(breaks, key=lambda k: k[1]) + else: + return sorted(breaks, key=lambda k: k[1], reverse=True)[:number_of_solutions] + +def polybius_break_worker(message, keyword, wrap_alphabet, + column_order, row_order, column_first, + letters_to_merge, + fitness): + plaintext = polybius_decipher(message, keyword, + column_order, row_order, + column_first=column_first, + letters_to_merge=letters_to_merge, + wrap_alphabet=wrap_alphabet) + if plaintext: + fit = fitness(plaintext) + else: + fit = float('-inf') + logger.debug('Polybius break attempt using key {0} (wrap={1}, merging {2}), ' + 'columns as {3}, rows as {4} (column_first={5}) ' + 'gives fit of {6} and decrypt starting: ' + '{7}'.format(keyword, wrap_alphabet, letters_to_merge, + column_order, row_order, column_first, + fit, sanitise(plaintext)[:50])) + return (keyword, wrap_alphabet, column_order, row_order, column_first), fit + + def column_transposition_break_mp(message, translist=transpositions, fitness=Pbigrams, chunksize=500): """Breaks a column transposition cipher using a dictionary and @@ -359,7 +442,7 @@ def column_transposition_break_mp(message, translist=transpositions, with Pool() as pool: helper_args = [(message, trans, fillcolumnwise, emptycolumnwise, fitness) - for trans in translist.keys() + for trans in translist for fillcolumnwise in [True, False] for emptycolumnwise in [True, False]] # Gotcha: the helper function here needs to be defined at the top level @@ -489,7 +572,7 @@ def amsco_break(message, translist=transpositions, patterns = [(1, 2), (2, 1)], (5, 0, 6, 1, 3, 4, 2): ['fourteen'], \ (6, 1, 0, 4, 5, 3, 2): ['keyword']}, \ patterns=[(1, 2)]) # doctest: +ELLIPSIS - (((2, 0, 5, 3, 1, 4, 6), (1, 2)), -709.4646722...) + (((2, 0, 5, 3, 1, 4, 6), (1, 2), ), -709.4646722...) >>> amsco_break(amsco_transposition_encipher(sanitise( \ "It is a truth universally acknowledged, that a single man in \ possession of a good fortune, must be in want of a wife. However \ @@ -502,11 +585,11 @@ def amsco_break(message, translist=transpositions, patterns = [(1, 2), (2, 1)], (5, 0, 6, 1, 3, 4, 2): ['fourteen'], \ (6, 1, 0, 4, 5, 3, 2): ['keyword']}, \ patterns=[(1, 2), (2, 1)], fitness=Ptrigrams) # doctest: +ELLIPSIS - (((2, 0, 5, 3, 1, 4, 6), (2, 1)), -997.0129085...) + (((2, 0, 5, 3, 1, 4, 6), (2, 1), ), -997.0129085...) """ with Pool() as pool: helper_args = [(message, trans, pattern, fillstyle, fitness) - for trans in translist.keys() + for trans in translist for pattern in patterns for fillstyle in fillstyles] # Gotcha: the helper function here needs to be defined at the top level @@ -558,6 +641,43 @@ def hill_break_worker(message, matrix, fitness): fit, sanitise(plaintext)[:50])) return matrix, fit +def bifid_break_mp(message, wordlist=keywords, fitness=Pletters, max_period=10, + number_of_solutions=1, chunksize=500): + """Breaks a keyword substitution cipher using a dictionary and + frequency analysis + + >>> bifid_break_mp(bifid_encipher('this is a test message for the ' \ + 'keyword decipherment', 'elephant', wrap_alphabet=KeywordWrapAlphabet.from_last), \ + wordlist=['cat', 'elephant', 'kangaroo']) # doctest: +ELLIPSIS + (('elephant', , 0), -52.834575011...) + >>> bifid_break_mp(bifid_encipher('this is a test message for the ' \ + 'keyword decipherment', 'elephant', wrap_alphabet=KeywordWrapAlphabet.from_last), \ + wordlist=['cat', 'elephant', 'kangaroo'], \ + number_of_solutions=2) # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE + [(('elephant', , 0), -52.834575011...), + (('elephant', , 0), -52.834575011...)] + """ + with Pool() as pool: + helper_args = [(message, word, wrap, period, fitness) + for word in wordlist + for wrap in KeywordWrapAlphabet + for period in range(max_period+1)] + # Gotcha: the helper function here needs to be defined at the top level + # (limitation of Pool.starmap) + breaks = pool.starmap(bifid_break_worker, helper_args, chunksize) + if number_of_solutions == 1: + return max(breaks, key=lambda k: k[1]) + else: + return sorted(breaks, key=lambda k: k[1], reverse=True)[:number_of_solutions] + +def bifid_break_worker(message, keyword, wrap_alphabet, period, fitness): + plaintext = bifid_decipher(message, keyword, wrap_alphabet, period=period) + fit = fitness(plaintext) + 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, period), fit + def pocket_enigma_break_by_crib(message, wheel_spec, crib, crib_position): """Break a pocket enigma using a crib (some plaintext that's expected to @@ -588,13 +708,13 @@ def pocket_enigma_break_by_crib(message, wheel_spec, crib, crib_position): def plot_frequency_histogram(freqs, sort_key=None): - x = range(len(freqs.keys())) - y = [freqs[l] for l in sorted(freqs.keys(), key=sort_key)] + x = range(len(freqs)) + y = [freqs[l] for l in sorted(freqs, key=sort_key)] f = plt.figure() ax = f.add_axes([0.1, 0.1, 0.9, 0.9]) ax.bar(x, y, align='center') ax.set_xticks(x) - ax.set_xticklabels(sorted(freqs.keys(), key=sort_key)) + ax.set_xticklabels(sorted(freqs, key=sort_key)) f.show()