yaklang/hack-skills/classical-cipher-analysis

def analyze_charset(ciphertext):
    """Identify encoding/cipher by character set."""
    chars = set(ciphertext.strip())
    if chars <= set('01 \n'):
        return "Binary encoding"
    if chars <= set('.-/ \n'):
        return "Morse code"
    if chars <= set('0123456789abcdef \n'):
        return "Hex encoding"
    if chars <= set('ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/=\n'):
        if '=' in ciphertext or len(ciphertext) % 4 == 0:
            return "Base64 encoding"
    if chars <= set('ABCDEFGHIJKLMNOPQRSTUVWXYZ \n'):
        return "Uppercase only — classical cipher"
    if all(c in '12345' for c in ciphertext.replace(' ', '').replace('\n', '')):
        return "Polybius square (digits 1-5)"
    return "Mixed charset — needs further analysis"

from collections import Counter
def frequency_analysis(text):
    """Compute letter frequency distribution."""
    text = text.upper()
    letters = [c for c in text if c.isalpha()]
    total = len(letters)
    freq = Counter(letters)
    print("Letter frequencies:")
    for letter, count in freq.most_common():
        pct = count / total * 100
        bar = '#' * int(pct)
        print(f"  {letter}: {pct:5.1f}% {bar}")
    return freq
# English letter frequency (for comparison):
# E T A O I N S H R D L C U M W F G Y P B V K J X Q Z
# 12.7 9.1 8.2 7.5 7.0 6.7 6.3 6.1 6.0 4.3 4.0 2.8 ...

def index_of_coincidence(text):
    """
    IC ≈ 0.065 → English / monoalphabetic substitution
    IC ≈ 0.038 → random / polyalphabetic cipher
    """
    text = [c for c in text.upper() if c.isalpha()]
    N = len(text)
    freq = Counter(text)
    ic = sum(f * (f - 1) for f in freq.values()) / (N * (N - 1))
    return ic
# Interpretation:
# IC > 0.060 → monoalphabetic (Caesar, simple substitution, Playfair)
# IC ≈ 0.045-0.055 → polyalphabetic with short key (Vigenere key < 10)
# IC ≈ 0.038-0.042 → polyalphabetic with long key or random

from math import gcd
from functools import reduce
def kasiski(ciphertext, min_len=3):
    """Find repeated sequences and their distances → key length."""
    text = ''.join(c for c in ciphertext.upper() if c.isalpha())
    distances = []
    for length in range(min_len, min(20, len(text) // 3)):
        for i in range(len(text) - length):
            seq = text[i:i+length]
            j = text.find(seq, i + 1)
            while j != -1:
                distances.append(j - i)
                j = text.find(seq, j + 1)
    if not distances:
        return None
    # Key length is likely GCD of common distances
    common_gcds = Counter()
    for d in distances:
        for factor in range(2, min(d + 1, 30)):
            if d % factor == 0:
                common_gcds[factor] += 1
    print("Likely key lengths (by frequency):")
    for length, count in common_gcds.most_common(5):
        print(f"  Key length {length}: {count} occurrences")
    return common_gcds.most_common(1)[0][0]

def solve_substitution(ciphertext, interactive=False):
    """Solve monoalphabetic substitution via frequency analysis."""
    freq = frequency_analysis(ciphertext)
    # English frequency order
    eng_order = "ETAOINSRHLDCUMWFGYPBVKJXQZ"
    cipher_order = ''.join(c for c, _ in freq.most_common())
    # Initial mapping (frequency-based guess)
    mapping = {}
    for i, c in enumerate(cipher_order):
        if i < len(eng_order):
            mapping[c] = eng_order[i]
    # Apply mapping
    result = ""
    for c in ciphertext.upper():
        result += mapping.get(c, c)
    return result, mapping
# Better approach: use automated solvers
# quipqiup.com — online substitution solver
# dcode.fr/monoalphabetic-substitution — with word pattern matching

def crib_drag_substitution(ciphertext, known_plain, known_cipher):
    """Build partial mapping from known plaintext-ciphertext pair."""
    mapping = {}
    for p, c in zip(known_plain.upper(), known_cipher.upper()):
        mapping[c] = p
    # Apply partial mapping
    result = ""
    for c in ciphertext.upper():
        result += mapping.get(c, '?')
    return result, mapping

def caesar_bruteforce(ciphertext):
    """Try all 25 shifts, score by English frequency."""
    results = []
    for shift in range(26):
        decrypted = ""
        for c in ciphertext:
            if c.isalpha():
                base = ord('A') if c.isupper() else ord('a')
                decrypted += chr((ord(c) - base - shift) % 26 + base)
            else:
                decrypted += c
        # Chi-squared scoring against English frequency
        score = chi_squared_score(decrypted)
        results.append((shift, score, decrypted))
    results.sort(key=lambda x: x[1])
    return results[0]  # best match
def chi_squared_score(text):
    """Lower score = closer to English."""
    expected = {
        'E': 12.7, 'T': 9.1, 'A': 8.2, 'O': 7.5, 'I': 7.0,
        'N': 6.7, 'S': 6.3, 'H': 6.1, 'R': 6.0, 'D': 4.3,
        'L': 4.0, 'C': 2.8, 'U': 2.8, 'M': 2.4, 'W': 2.4,
        'F': 2.2, 'G': 2.0, 'Y': 2.0, 'P': 1.9, 'B': 1.5,
        'V': 1.0, 'K': 0.8, 'J': 0.2, 'X': 0.2, 'Q': 0.1, 'Z': 0.1,
    }
    text = text.upper()
    letters = [c for c in text if c.isalpha()]
    total = len(letters)
    if total == 0:
        return float('inf')
    freq = Counter(letters)
    score = sum(
        (freq.get(c, 0) / total * 100 - expected.get(c, 0)) ** 2 / max(expected.get(c, 0.1), 0.1)
        for c in 'ABCDEFGHIJKLMNOPQRSTUVWXYZ'
    )
    return score

import codecs
# ROT13 (letters only)
rot13 = codecs.decode(ciphertext, 'rot_13')
# ROT47 (ASCII 33-126)
def rot47(text):
    return ''.join(
        chr(33 + (ord(c) - 33 + 47) % 94) if 33 <= ord(c) <= 126 else c
        for c in text
    )

Step 1: Confirm polyalphabetic (IC ≈ 0.04-0.05)
Step 2: Find key length (Kasiski + IC per period)
Step 3: For each key position, solve as single Caesar cipher
Step 4: Assemble key → decrypt

def find_vigenere_key_length(ciphertext, max_key=20):
    """Use IC to find Vigenere key length."""
    text = [c for c in ciphertext.upper() if c.isalpha()]
    results = []
    for kl in range(1, max_key + 1):
        # Split text into kl columns
        columns = [[] for _ in range(kl)]
        for i, c in enumerate(text):
            columns[i % kl].append(c)
        # Average IC across columns
        avg_ic = sum(
            index_of_coincidence(''.join(col)) for col in columns
        ) / kl
        results.append((kl, avg_ic))
        print(f"  Key length {kl:2d}: IC = {avg_ic:.4f}")
    # Key length with IC closest to 0.065
    best = max(results, key=lambda x: x[1])
    return best[0]

def crack_vigenere(ciphertext, key_length):
    """Crack Vigenere given known key length."""
    text = [c for c in ciphertext.upper() if c.isalpha()]
    key = ""
    for pos in range(key_length):
        column = ''.join(text[i] for i in range(pos, len(text), key_length))
        # Solve as Caesar cipher
        shift, score, _ = caesar_bruteforce(column)
        key += chr(shift + ord('A'))
    # Decrypt
    plaintext = ""
    ki = 0
    for c in ciphertext:
        if c.isalpha():
            shift = ord(key[ki % key_length]) - ord('A')
            base = ord('A') if c.isupper() else ord('a')
            plaintext += chr((ord(c) - base - shift) % 26 + base)
            ki += 1
        else:
            plaintext += c
    return key, plaintext

def crack_affine(ciphertext):
    """Brute force affine cipher: 12 × 26 = 312 combinations."""
    valid_a = [a for a in range(1, 26) if gcd(a, 26) == 1]
    for a in valid_a:
        a_inv = pow(a, -1, 26)
        for b in range(26):
            plaintext = ""
            for c in ciphertext.upper():
                if c.isalpha():
                    y = ord(c) - ord('A')
                    x = (a_inv * (y - b)) % 26
                    plaintext += chr(x + ord('A'))
                else:
                    plaintext += c
            score = chi_squared_score(plaintext)
            if score < 50:  # reasonable English
                print(f"a={a}, b={b}: {plaintext[:50]}...")

def affine_from_known(plain1, cipher1, plain2, cipher2):
    """Recover (a, b) from two known plaintext-ciphertext pairs."""
    p1, c1 = ord(plain1) - ord('A'), ord(cipher1) - ord('A')
    p2, c2 = ord(plain2) - ord('A'), ord(cipher2) - ord('A')
    # c1 = a*p1 + b, c2 = a*p2 + b
    # c1 - c2 = a*(p1 - p2) mod 26
    diff_p = (p1 - p2) % 26
    diff_c = (c1 - c2) % 26
    if gcd(diff_p, 26) != 1:
        return None
    a = (diff_c * pow(diff_p, -1, 26)) % 26
    b = (c1 - a * p1) % 26
    return a, b

import numpy as np
def crack_hill(known_plain, known_cipher, n=2):
    """Recover Hill cipher key from known plaintext-ciphertext (mod 26)."""
    # Convert to numbers
    P = [ord(c) - ord('A') for c in known_plain.upper()]
    C = [ord(c) - ord('A') for c in known_cipher.upper()]