X-Git-Url: https://git.njae.me.uk/?a=blobdiff_plain;f=cipher.py;h=b07cd250fb8263e338a1b35686f28a583f9bf2a2;hb=0a8d248c149a8499fc9b92e8ea7a5f4780be4d76;hp=c4b28ba321ea4a55953af22f914c5b94a7843b6f;hpb=e6aea9194297e503e6b2887f1756ce5dc4d6f078;p=cipher-tools.git

diff --git a/cipher.py b/cipher.py
index c4b28ba..b07cd25 100644
--- a/cipher.py
+++ b/cipher.py
@@ -1,6 +1,36 @@
 import string
 import collections
 
+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)
+
+def sanitise(text):
+    sanitised = [c.lower() for c in text if c in string.ascii_letters]
+    return ''.join(sanitised)
+
+def letter_frequencies(message):
+    frequencies = collections.defaultdict(int)
+    for letter in sanitise(message): 
+        frequencies[letter]+=1
+    return frequencies
+
+def scale_freq(frequencies):
+    total= sum(frequencies.values())
+    scaled_frequencies = collections.defaultdict(int)
+    for letter in frequencies.keys():
+        scaled_frequencies[letter] = frequencies[letter] / total
+    return scaled_frequencies
+
+def value_diff(frequencies1, frequencies2):
+    total= 0
+    for letter in frequencies1.keys():
+        total += abs(frequencies1[letter]-frequencies2[letter])
+    return total
+        
+    
 
 def caesar_cipher_letter(letter, shift):
     if letter in string.ascii_letters:
@@ -23,10 +53,16 @@ def caesar_cipher_message(message, shift):
 def caesar_decipher_message(message, shift):
     return caesar_cipher_message(message, -shift)
 
-def letter_frequencies(message):
-    frequencies = collections.defaultdict(int)
-    for letter in message: 
-        if letter in  string.ascii_letters:
-            frequencies[letter.lower()]+=1
-    return frequencies
-
+def caesar_break(message):
+    best_key = 0
+    best_fit = float("inf")
+    for shift in range(26):
+        plaintxt = caesar_decipher_message(message, shift)
+        lettertxt = letter_frequencies(plaintxt)
+        total1 = scale_freq(lettertxt)
+        total2 = scale_freq(english_counts)
+        fit = value_diff(total2, total1)
+        if fit < best_fit:
+            best_key = shift
+            best_fit = fit
+    return best_key