def test_add(self):
gen = MarkovGenerator([], random=self.random)
gen.add(SAMPLE_TEXT)
result = gen.generate(3)
self.assertEqual('One two three', result)
class MailGenerator(object):
NAMES = ['alice', 'bob', 'eve']
def __init__(self, receiver, domain_name, sample_mail_list, random=None):
self._random = random if random else Random()
self._receiver = receiver
self._domain_name = domain_name
self._subjects, self._bodies = load_all_mails(sample_mail_list)
self._potential_tags = search_for_tags(' '.join(self._bodies))
self._subject_markov = MarkovGenerator(self._subjects,
random=self._random)
self._body_markov = MarkovGenerator(self._bodies,
random=self._random,
add_paragraph_on_empty_chain=True)
def generate_mail(self):
body = self._body_markov.generate(150)
mail = MIMEText(body)
mail['Subject'] = self._subject_markov.generate(8)
mail['To'] = '%s@%s' % (self._receiver, self._domain_name)
mail['From'] = self._random_from()
mail['Date'] = self._random_date()
mail['X-Tags'] = self._random_tags()
mail['X-Leap-Encryption'] = self._random_encryption_state()
mail['X-Leap-Signature'] = self._random_signature_state()
return mail
def _random_date(self):
now = int(time.time())
ten_days = 60 * 60 * 24 * 10
mail_time = self._random.randint(now - ten_days, now)
return formatdate(mail_time)
def _random_encryption_state(self):
return self._random.choice(['true', 'decrypted'])
def _random_signature_state(self):
return self._random.choice(['could not verify', 'valid'])
def _random_from(self):
name = self._random.choice(
filter(lambda name: name != self._receiver, MailGenerator.NAMES))
return '%s@%s' % (name, self._domain_name)
def _random_tags(self):
barrier = 0.5
tags = set()
while self._random.random() > barrier:
tags.add(self._random.choice(self._potential_tags))
barrier += 0.15
return ' '.join(tags)
def __init__(self, receiver, domain_name, sample_mail_list, random=None):
self._random = random if random else Random()
self._receiver = receiver
self._domain_name = domain_name
self._subjects, self._bodies = load_all_mails(sample_mail_list)
self._potential_tags = search_for_tags(' '.join(self._bodies))
self._subject_markov = MarkovGenerator(self._subjects, random=self._random)
self._body_markov = MarkovGenerator(self._bodies, random=self._random, add_paragraph_on_empty_chain=True)
class MailGenerator(object):
NAMES = ['alice', 'bob', 'eve']
def __init__(self, receiver, domain_name, sample_mail_list, random=None):
self._random = random if random else Random()
self._receiver = receiver
self._domain_name = domain_name
self._subjects, self._bodies = load_all_mails(sample_mail_list)
self._potential_tags = search_for_tags(' '.join(self._bodies))
self._subject_markov = MarkovGenerator(self._subjects, random=self._random)
self._body_markov = MarkovGenerator(self._bodies, random=self._random, add_paragraph_on_empty_chain=True)
def generate_mail(self):
body = self._body_markov.generate(150)
mail = MIMEText(body)
mail['Subject'] = self._subject_markov.generate(8)
mail['To'] = '%s@%s' % (self._receiver, self._domain_name)
mail['From'] = self._random_from()
mail['Date'] = self._random_date()
mail['X-Tags'] = self._random_tags()
mail['X-Leap-Encryption'] = self._random_encryption_state()
mail['X-Leap-Signature'] = self._random_signature_state()
return mail
def _random_date(self):
now = int(time.time())
ten_days = 60 * 60 * 24 * 10
mail_time = self._random.randint(now - ten_days, now)
return formatdate(mail_time)
def _random_encryption_state(self):
return self._random.choice(['true', 'decrypted'])
def _random_signature_state(self):
return self._random.choice(['could not verify', 'valid'])
def _random_from(self):
name = self._random.choice(filter(lambda name: name != self._receiver, MailGenerator.NAMES))
return '%s@%s' % (name, self._domain_name)
def _random_tags(self):
barrier = 0.5
tags = set()
while self._random.random() > barrier:
tags.add(self._random.choice(self._potential_tags))
barrier += 0.15
return ' '.join(tags)
def __init__(self, receiver, domain_name, sample_mail_list, random=None):
self._random = random if random else Random()
self._receiver = receiver
self._domain_name = domain_name
self._subjects, self._bodies = load_all_mails(sample_mail_list)
self._potential_tags = search_for_tags(' '.join(self._bodies))
self._subject_markov = MarkovGenerator(self._subjects,
random=self._random)
self._body_markov = MarkovGenerator(self._bodies,
random=self._random,
add_paragraph_on_empty_chain=True)
def test_multiple_word_occurences(self):
gen = MarkovGenerator(['One Two One Three One Two One Four'], random=self.random)
result = gen.generate(2)
self.assertEqual('Two One', result)
def test_multiple_inputs(self):
gen = MarkovGenerator([SAMPLE_TEXT, 'Five Six seven eight'], random=self.random)
result = gen.generate(3)
self.assertEqual('Five Six seven', result)
def test_add_paragraph_on_empty_chain(self):
gen = MarkovGenerator([SAMPLE_TEXT], random=self.random, add_paragraph_on_empty_chain=True)
result = gen.generate(5)
self.assertEqual('One two three four \n\n One', result)
def test_minimum_three_words(self):
self.assertRaises(ValueError, MarkovGenerator([]).generate, 1)
self.assertRaises(ValueError, MarkovGenerator, ['1'])
self.assertRaises(ValueError, MarkovGenerator, ['1', '2'])
self.assertRaises(ValueError, MarkovGenerator, ['1', '2', '3'])
def test_aborts_if_no_upper_letter_word_found(self):
gen = MarkovGenerator(['all lower case'], random=self.random)
self.assertRaises(ValueError, gen.generate, 1)
def test_starts_with_capital_case_workd(self):
gen = MarkovGenerator(['lower Upper smaller Capital'], random=self.random)
result = gen.generate(1)
self.assertTrue(result.startswith('Upper'))