text_prettify.py

   1 from segment import segment
   2 from cipher import cat
   3 from language_models import sanitise
   4 import string
   5
   6
   7 def tpack(text, width=100):
   8     """Pack a list of words into lines, so long as each line (including
   9     intervening spaces) is no longer than _width_"""
  10     lines = [text[0]]
  11     for word in text[1:]:
  12         if len(lines[-1]) + 1 + len(word) <= width:
  13             lines[-1] += (' ' + word)
  14         else:
  15             lines += [word]
  16     return lines
  17
  18
  19 def depunctuate_character(c):
  20     """Record the punctuation of a character"""
  21     if c in string.ascii_uppercase:
  22         return 'UPPER'
  23     elif c in string.ascii_lowercase:
  24         return 'LOWER'
  25     else:
  26         return c
  27
  28
  29 def depunctuate(text):
  30     """Record the punctuation of a string, so it can be applied to a converted
  31     version of the string.
  32
  33     For example,
  34     punct = depunctuate(ciphertext)
  35     plaintext = decipher(sanitise(ciphertext))
  36     readable_plaintext = repunctuate(plaintext, punct)
  37     """
  38     return [depunctuate_character(c) for c in text]
  39
  40
  41 def repunctuate_character(letters, punctuation):
  42     """Apply the recorded punctuation to a character. The letters must be
  43     an iterator of base characters."""
  44     if punctuation == 'UPPER':
  45         return next(letters).upper()
  46     elif punctuation == 'LOWER':
  47         return next(letters).lower()
  48     else:
  49         return punctuation
  50
  51
  52 def repunctuate(text, punctuation):
  53     """Apply the recored punctuation to a sanitised string.
  54
  55     For example,
  56     punct = depunctuate(ciphertext)
  57     plaintext = decipher(sanitise(ciphertext))
  58     readable_plaintext = repunctuate(plaintext, punct)
  59     """
  60     letters = iter(sanitise(text))
  61     return cat(repunctuate_character(letters, p) for p in punctuation)