Add markov-generate to load-mails

- Allows to generat mails based
  on a sample mails

- use it with:
   pixelated-maintenance markov-generate --seed 21 --limit 10

2 files changed, 125 insertions, 0 deletions

diff --git a/service/test/unit/support/mail_generator_test.py b/service/test/unit/support/mail_generator_test.py
new file mode 100644
index 00000000..9d604378
--- /dev/null
+++ b/service/test/unit/support/mail_generator_test.py
@@ -0,0 +1,42 @@
+#
+# Copyright (c) 2015 ThoughtWorks, Inc.
+#
+# Pixelated is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# Pixelated is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with Pixelated. If not, see <http://www.gnu.org/licenses/>.
+from mailbox import mbox
+import unittest
+import pkg_resources
+import random
+from mock import patch
+
+from pixelated.support.mail_generator import MailGenerator
+
+
+class MailGeneratorTest(unittest.TestCase):
+
+    def test_generator(self):
+        mail = """Content-Type: text/plain; charset="us-ascii"\nMIME-Version: 1.0\nContent-Transfer-Encoding: 7bit\nSubject: Scott\'s Laws with a longer subject Scott\'s Laws\nTo: alice@domain.test\nFrom: bob@domain.test\nDate: Sun, 18 Oct 2015 21:45:13 -0000\nX-Tags: \nX-Leap-Encryption: true\nX-Leap-Signature: valid\n\nFirst Law: No matter what goes wrong, it will probably look right. Scott\'s Second Law: When an error has been detected and corrected, it will probably look right. Scott\'s Second Law: When an error has been found in error, it will probably look right. Scott\'s Second Law: When an error has been found in error, it will probably look right. Scott\'s Second Law: When an error has been found in error, it will be impossible to fit the original quantity back into the \n\n First Law: No matter what goes wrong, it will be impossible to fit the original quantity back into the \n\n Scott\'s Second Law: When an error has been found in error, it will be found to have been wrong in the first place. After the correction has been found in error, it will be impossible to fit the original quantity back into the \n\n Second Law: When an error"""
+        receiver = 'alice'
+        domain_name = 'domain.test'
+        mbox_file = pkg_resources.resource_filename('test.unit.fixtures', 'mbox')
+        mails = mbox(mbox_file)
+        rnd = random.Random(0)
+
+        with patch('pixelated.support.mail_generator.time.time') as time_mock:
+            time_mock.return_value = 1446029232.636018
+
+            gen = MailGenerator(receiver, domain_name, mails, rnd)
+
+            result = gen.generate_mail()
+
+            self.assertEqual(mail, result.as_string())
diff --git a/service/test/unit/support/test_markov.py b/service/test/unit/support/test_markov.py
new file mode 100644
index 00000000..f0b0277d
--- /dev/null
+++ b/service/test/unit/support/test_markov.py
@@ -0,0 +1,83 @@
+#
+# Copyright (c) 2015 ThoughtWorks, Inc.
+#
+# Pixelated is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# Pixelated is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with Pixelated. If not, see <http://www.gnu.org/licenses/>.
+
+
+import unittest
+from pixelated.support.markov import MarkovGenerator
+import random
+
+
+SAMPLE_TEXT = 'One two three four'
+
+
+class MarkovGeneratorTest(unittest.TestCase):
+
+    def setUp(self):
+        self.random = random.Random(0)
+
+    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'))
+
+    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_generate(self):
+        gen = MarkovGenerator([SAMPLE_TEXT], random=self.random)
+
+        result = gen.generate(3)
+
+        self.assertEqual('One two three', 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_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_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(self):
+        gen = MarkovGenerator([], random=self.random)
+
+        gen.add(SAMPLE_TEXT)
+        result = gen.generate(3)
+
+        self.assertEqual('One two three', result)
+
+    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)