use trigrams, works much better

inspired by soloution from previous year :P

src/mono/break_mono.py

@@ -1,4 +1,36 @@
 from random import randint as rand
+from math import log10
+
+class Scoring:
+    floor = 0
+
+    def __init__(self):
+        self.trigrams = {}
+        # read trigrams from file
+        with open(os.path.dirname(os.path.realpath(__file__)) + '/english_trigrams.txt', 'r') as trigram_file:
+            for line in trigram_file:
+                trigram, score = line.split(' ')
+                self.trigrams[trigram.lower()] = int(score)
+
+        score_sum = sum(self.trigrams.values())
+        self.floor = log10(0.1 / score_sum)
+
+        for i in list(self.trigrams.keys()):
+            self.trigrams[i] = log10(float(self.trigrams[i]) / score_sum)
+        pass
+
+    def get_score(self, text):
+        score = 0
+
+        for k in range(len(text) - 3):
+            trigram = text[k:k + 3]
+            if trigram in self.trigrams:
+                score += self.trigrams[trigram]
+            else:
+                score += self.floor
+        return score
+
+
 def randomize_key(key):
     a_index = rand(0, 25)
     b_index = rand(0, 25)
@@ -34,17 +66,21 @@ if __name__ == "__main__":
     import re
     import mono
 
-    best_score = 0
+    best_score = -10000
     best_key = key
 
+    scoring = Scoring()
+
     trys = 0
-    while trys < 1000:
+    while trys < 20000:
         score = 0
         tmp = randomize_key(key)
         plain = mono.mono_decrypt(t, tmp)
-        for word in words:
-            if re.search(word, plain):
-                score += 1
+        #for word in words:
+        #    search_res = re.search(word, plain, re.IGNORECASE)
+        #    if search_res:
+        #        score += len(word)
+        score = scoring.get_score(plain)
         if score > best_score:
             trys = 0
             best_score = score