from random import shuffle
from string import ascii_lowercase as letters
import sys
from collections import defaultdict

def encrypt(text, mapping):
  mapping[' '] = ' '
  mapping[','] = ','
  encrypted = ''
  for letter in text:
    if letter in mapping:
      encrypted += encode(letter,mapping) 
  return encrypted

def encode(letter, mapping):
  if letter.isupper(): return mapping[letter.lower()].upper()
  elif letter.islower(): return mapping[letter]
  else: return letter

def freq(seq):
  d = defaultdict(int)
  for e in seq: d[e] += 1
  return d

def ngrams(text,n):
  return [text[i:i+n] for i in range(0,len(text)-n+1)]

def unigrams(text): return ngrams(text,1)
def bigrams(text): return ngrams(text,2)

class Model:
  def __init__(self, text):
    self.text = text
    self.unigrams = unigrams(self.text)
    self.bigrams = bigrams(self.text)
    self.unigram_model = [(v,k) for k,v in sorted(freq(self.unigrams).items())]
    self.bigram_model = freq(bigrams(text))
    self.mapping = self.genmap(letters)
    self.encrypted = encrypt(self.text, self.mapping)
  def genmap(self, original):
    alpha = list(original)
    randomized = alpha[:]
    shuffle(randomized)
    return dict(zip(alpha,randomized))

if __name__ == "__main__":
  model = Model(open('neuromancer.txt').read())
  print model.encrypted[:1000]
  for letter, count in  model.unigram_model: 
    print letter, count