Determine ROT encoding

Question

I want to determine which type of ROT encoding is used and based off that, do the correct decode.

Also, I have found the following code which will indeed decode rot13 "sbbone" to "foobart" correctly:

import codecs
codecs.decode('sbbone', 'rot_13')

The thing is I'd like to run this python file against an existing file which has rot13 encoding. (for example rot13.py encoded.txt).

Thank you!

Answer 1

To answer the second part of your first question, decode something in ROT-x, you can use the following code:

def encode(s, ROT_number=13):
    """Encodes a string (s) using ROT (ROT_number) encoding."""
    ROT_number %= 26  # To avoid IndexErrors
    alpha = "abcdefghijklmnopqrstuvwxyz" * 2
    alpha += alpha.upper()
    def get_i():
        for i in range(26):
            yield i  # indexes of the lowercase letters
        for i in range(53, 78):
            yield i  # indexes of the uppercase letters
    ROT = {alpha[i]: alpha[i + ROT_number] for i in get_i()}
    return "".join(ROT.get(i, i) for i in s)


def decode(s, ROT_number=13):
    """Decodes a string (s) using ROT (ROT_number) encoding."""
    return encrypt(s, abs(ROT_number % 26 - 26))

To answer the first part of your first question, find the rot encoding of an arbitrarily encoded string, you probably want to brute-force. Uses all rot-encodings, and check which one makes the most sense. A quick(-ish) way to do this is to get a space-delimited (e.g. cat\ndog\nmouse\nsheep\nsay\nsaid\nquick\n... where \n is a newline) file containing most common words in the English language, and then check which encoding has the most words in it.

with open("words.txt") as f:
    words = frozenset(f.read().lower().split("\n"))
    # frozenset for speed
def get_most_likely_encoding(s, delimiter=" "):
    alpha = "abcdefghijklmnopqrstuvwxyz" + delimiter
    for punctuation in "\n\t,:; .()":
        s.replace(punctuation, delimiter)
    s = "".join(c for c in s if c.lower() in alpha)
    word_count = [sum(w.lower() in words for w in encode(
            s, enc).split(delimiter)) for enc in range(26)]
    return word_count.index(max(word_count))

A file on Unix machines that you could use is /usr/dict/words, which can also be found here