def __init__(self):
self.fave_ep = None
self.fave_char = None
self.prev_quote = None
self.no_response_count = 0
self.context_count = 0
try:
from seinfeld import Seinfeld
except ImportError:
import pip
print('Crazy Joe: Installing necessary dependency: \'seinfeld\'...')
pip.main(['install', 'seinfeld'])
finally:
from seinfeld import Seinfeld
if not os.path.isfile(self.db_name): #looking in current dir only.
import urllib.request as rq
print('Crazy Joe: Downloading Seinfeld Database...')
rq.urlretrieve(self.db_loc, self.db_name)
self.seinfeld = Seinfeld(self.db_name)
self.eps = {v.title.lower() : k for k, v in self.seinfeld.episode().items()}
self.chars = ['jerry', 'elaine', 'george', 'kramer', 'leo', 'newman', 'maestro', 'puddy', 'peterman',]
self.def_subjects = ['keys', 'tips', 'tipping', 'kavorka', 'tv', 'parents', 'sex', 'comedy', 'nothing',]
self.new_subjects = []
self.jerry = False
self.larry = False
def setUp(self):
self.seinfeld = Seinfeld()
self.seinfeld.open()
class TestDatabase(unittest.TestCase):
def setUp(self):
self.seinfeld = Seinfeld()
self.seinfeld.open()
def tearDown(self):
self.seinfeld.close()
def test_db_counts(self):
self.assertEqual(len(self.seinfeld.episode()), 173)
self.assertEqual(len(self.seinfeld.season()), 9)
self.assertEqual(len(self.seinfeld.speaker()), 726)
c = self.seinfeld.cursor()
c.execute('select id from utterance order by id desc limit 1')
count, = c.fetchone()
self.assertEqual(count, 52206)
self.assertEqual(self.seinfeld.MAX_QUOTE_ID, 52206)
def test_data(self):
season = self.seinfeld.season(2)
self.assertEqual(season.number, 2)
self.assertIsInstance(season.episodes, dict)
self.assertEqual(len(season.episodes), 12)
episode = season.episodes[3]
self.assertEqual(episode.season, 2)
self.assertEqual(episode.number, 3)
self.assertEqual(episode.title, 'The Jacket')
self.assertIsInstance(episode.date, dt.date)
self.assertEqual(episode.date, dt.date(1991, 2, 6))
self.assertIsInstance(episode.writers, set)
self.assertIn('Larry David', episode.writers)
self.assertIn('Jerry Seinfeld', episode.writers)
self.assertEqual(episode.director, 'Tom Cherones')
def test_quote(self):
speaker = self.seinfeld.speaker('GEORGE')
episode = self.seinfeld.season(4).episodes[3]
quote = self.seinfeld.quote(34663)
self.assertEqual(quote.id, 34663)
self.assertEqual(quote.episode, episode)
self.assertEqual(quote.number, 250)
self.assertEqual(quote.speaker, speaker)
self.assertEqual(quote.text, '(Smiling) Nothing.')
def test_search_single_quote(self):
speaker = self.seinfeld.speaker('CUSTOMER')
episode = self.seinfeld.season(2).episodes[3]
quotes = self.seinfeld.search(speaker=speaker, subject='Alton Benes')
self.assertEquals(len(quotes), 1)
quote = quotes[0]
self.assertEqual(quote.id, 1442)
self.assertEqual(quote.episode, episode)
self.assertEqual(quote.number, 6)
self.assertEqual(quote.speaker, speaker)
self.assertEqual(quote.text, 'Alton Benes is your father?')
def test_search_multi_quote(self):
speaker = self.seinfeld.speaker('ELAINE')
episode = self.seinfeld.season(3).episodes[14]
# test default limit
quotes = self.seinfeld.search(episode=episode, speaker=speaker)
self.assertEquals(len(quotes), 10)
# test custom limit
quotes = self.seinfeld.search(episode=episode, speaker=speaker,
limit=5)
self.assertEquals(len(quotes), 5)
# test no limit
quotes = self.seinfeld.search(episode=episode, speaker=speaker,
limit=None)
self.assertEquals(len(quotes), 37)
last_number = 0
for quote in quotes:
self.assertEqual(quote.episode, episode)
self.assertGreater(quote.number, last_number)
self.assertEqual(quote.speaker, speaker)
self.assertTrue(quote.text)
last_number = quote.number
def test_passage(self):
speaker = self.seinfeld.speaker('GEORGE')
episode = self.seinfeld.season(4).episodes[3]
quote = self.seinfeld.quote(34663)
# custom length
passage = self.seinfeld.passage(quote, length=10)
self.assertEqual(len(passage.quotes), 10)
# default length
passage = self.seinfeld.passage(quote)
self.assertEqual(len(passage.quotes), 5)
self.assertTrue(passage)
self.assertEqual(passage.id, 34663)
self.assertEqual(passage.episode, episode)
# passage beginning
quote = passage.quotes[0]
self.assertEqual(quote.id, 34661)
self.assertEqual(quote.episode, episode)
self.assertEqual(quote.number, 248)
self.assertEqual(quote.speaker, speaker)
self.assertTrue(quote.text.startswith('I think I can sum up the show'))
# passage ending
quote = passage.quotes[-1]
self.assertEqual(quote.id, 34665)
self.assertEqual(quote.episode, episode)
self.assertEqual(quote.number, 252)
self.assertEqual(quote.speaker, speaker)
self.assertEqual(quote.text, 'The show is about nothing.')
class CrazyJoe(object):
db_name = 'seinfeld.db'
db_loc = 'https://noswap.com/pub/seinfeld.db'
agent_name = 'Crazy Joe'
def __init__(self):
self.fave_ep = None
self.fave_char = None
self.prev_quote = None
self.no_response_count = 0
self.context_count = 0
try:
from seinfeld import Seinfeld
except ImportError:
import pip
print('Crazy Joe: Installing necessary dependency: \'seinfeld\'...')
pip.main(['install', 'seinfeld'])
finally:
from seinfeld import Seinfeld
if not os.path.isfile(self.db_name): #looking in current dir only.
import urllib.request as rq
print('Crazy Joe: Downloading Seinfeld Database...')
rq.urlretrieve(self.db_loc, self.db_name)
self.seinfeld = Seinfeld(self.db_name)
self.eps = {v.title.lower() : k for k, v in self.seinfeld.episode().items()}
self.chars = ['jerry', 'elaine', 'george', 'kramer', 'leo', 'newman', 'maestro', 'puddy', 'peterman',]
self.def_subjects = ['keys', 'tips', 'tipping', 'kavorka', 'tv', 'parents', 'sex', 'comedy', 'nothing',]
self.new_subjects = []
self.jerry = False
self.larry = False
@classmethod
def kmp(self, text, pattern):
'''returns index of first subsequence in text matching pattern. -1 if DNE'''
shifts = [1] * (len(pattern) + 1)
shift = 1
for pos in range(len(pattern)):
while shift <= pos and pattern[pos] != pattern[pos-shift]:
shift += shifts[pos-shift]
shifts[pos + 1] = shift
start = 0
match_len = 0
for c in text:
while match_len == len(pattern) or match_len >= 0 and not match(pattern[match_len], c):
start += shifts[match_len]
match_len -= shifts[match_len]
match_len += 1
if match_len == len(pattern):
return start
return -1
def introduce(self):
return '''
Good evening ladies and gentlemen, how are you doing tonight?
Good? Good. I'm Crazy Joe Devola, I used to write for acclaimed sitcom 'Seinfeld.'
I have a hard time differentiating between myself and my characters so WATCH OUT.
I can only respond in 'Seinfeld' quotes, so bear with me.
If anything I say or do offends you, too bad!...Although you could try contacting
my creator Graham Kelly at [email protected].
What do you want to hear, quotes? Do you have a favorite episode? A favorite character?
You can read my manual too if you need direction. (type 'help')'''
def respond(self, the_input):
'''Returns a response based on the passed input.'''
if not the_input.strip(): # rule 0, respond to no input
self.prev_quote = None
if self.no_response_count >= len(no_response):
self.no_response_count = 0
return self._get_good_quote() # default to just saying quote if no_responses exhausted.
else:
self.no_response_count += 1
return no_response[self.no_response_count - 1]
clean_input = remove_punctuation(the_input)
# filtered_input = sw_pattern.sub('', clean_input) # filter stop words
wordlist = split(' ', clean_input)
wordlist[0] = wordlist[0].lower()
sub_wordlist = [w for w in wordlist if w not in stop_words]
return self.make_response(wordlist, sub_wordlist)
def make_response(self, wordlist, sub_wordlist):
'''generates a response to the passed wordlist based on several rules'''
if wordlist[0] == 'help': #rule 1: HELP ME!
self.prev_quote = None
return '\n\t'.join(commands)
if wordlist[0] == 'random': #rule 2: random quote.
return self._get_random_quote()
if wordlist == ['i', 'am', 'the', 'master', 'of', 'my', 'domain']: # rule 11: master of my domain
self.fave_ep = self.seinfeld.episode(id=self.eps['the contest'])
return self._get_good_quote(episode=self.fave_ep)
if wordlist[0:2] == ['who', 'is']:
self.prev_quote = None
if wordlist[2:4] == ['jerry', 'seinfeld']: #rule 13: learn about jerry seinfeld
if not self.jerry:
return 'Jerome Allen "Jerry" Seinfeld (born April 29, 1954) is an American comedian, actor, writer, producer, and director. He is known for playing a semifictional version of himself in the sitcom Seinfeld, which he created and wrote with Larry David. Seinfeld was heavily involved in the Bee Movie, in which he voiced its protagonist. In 2010, he premiered a reality series called The Marriage Ref. He directed Colin Quinn in the Broadway show Long Story Short at the Helen Hayes Theater, which ran until January 2011. He is the creator and host of the web series Comedians in Cars Getting Coffee. In his stand-up comedy career, Seinfeld is known for specializing in observational comedy, often ranting about relationships and embarrassing social situations.'
else:
return 'I already told you about him!'
if wordlist[2:4] == ['larry', 'david']: #rule 14: learn about larry david
if not self.larry:
return 'Lawrence Gene "Larry" David (born July 2, 1947) is an American comedian, writer, actor, playwright, and television producer. He and Jerry Seinfeld created the television series Seinfeld, where he served as its head writer and executive producer from 1989 to 1996. David has subsequently gained further recognition for the HBO series Curb Your Enthusiasm, which he also created, in which he stars as a semi-fictionalized version of himself. He is worth an estimated $900 million US dollars. David\'s work won him a Primetime Emmy Award for Outstanding Comedy Series in 1993. Formerly a stand-up comedian, David went into television comedy, writing and starring in ABC\'s Fridays, as well as writing briefly for Saturday Night Live. He has won two Primetime Emmy Awards, and was voted by fellow comedians and comedy insiders as the 23rd greatest comedy star ever in a 2004 British poll to select "The Comedian\'s Comedian."'
else:
return 'I already told you about him!'
if 'opera' in sub_wordlist: #rule 6: opera house
self.fave_char = self.seinfeld.speaker(name='Devola')
self.fave_ep = self.seinfeld.episode(id=self.eps['the opera'])
lines = ['You have entered the opera house and unleashed my true persona: Crazy Joe Devola.',
'Muahahaha, I have changed your favorite episode and character settings.']
lines.append(self._get_good_quote())
return '\n'.join(lines)
if self.kmp(wordlist, ['from', 'my', 'favorite', 'episode']) >= 0: #rule 12: quote from favorite episode
if self.fave_ep:
return self._get_good_quote(episode=self.fave_ep)
if self.kmp(wordlist, ['my', 'favorite']) >= 0:
if 'episode' in sub_wordlist: #rule 3: set fave ep
try:
try:
e_loc = wordlist.index('episode')
i_loc = wordlist.index('is')
if i_loc < e_loc:
ep_name = ' '.join(wordlist[0: i_loc]).lower() #pattern: XYZ is my favorite episode...
else:
ep_name = ' '.join(wordlist[i_loc + 1:]).lower() #pattern: ...my favorite episode is XYZ
self.fave_ep = self.seinfeld.episode(id=self.eps[ep_name])
except KeyError:
for k in self.eps.keys():
if SequenceMatcher(None, k, ep_name).ratio() >= .75:
self.fave_ep = self.seinfeld.episode(id=self.eps[k])
if not self.fave_ep:
raise NoEpisodeException
except (IndexError, NoEpisodeException):
return '\n'.join(['I can\'t seem to figure out what episode you meant.',
# 'This is what I got: "' + ep_name + '."',
self._get_random_quote(subject='stupid'),])
else:
return ' '.join(['Oh yeah,', self.fave_ep.title, 'is a great episode.', 'Here\'s a quote:',
self._get_good_quote()
])
if 'character' in sub_wordlist: #rule 4: set fave character
try:
try:
c_loc = wordlist.index('character')
i_loc = wordlist.index('is')
if i_loc < c_loc:
c_name = ' '.join(wordlist[0: i_loc]).lower() #pattern: XYZ is my favorite char...
else:
c_name = ' '.join(wordlist[i_loc + 1:]).lower() #pattern: ...my favorite char is XYZ
self.fave_char = self.seinfeld.speaker(name=c_name.capitalize())
except KeyError:
for c in self.chars:
if SequenceMatcher(None, c, c_name).ratio() >= .75:
self.fave_char = self.seinfeld.character(name=c)
if not self.fave_char:
raise NoCharacterException
except (IndexError, NoCharacterException):
return '\n'.join(['I can\'t seem to figure out what character you meant.',
# 'This is what I got: "' + ep_name + '."',
self._get_random_quote(subject='character'),])
else:
return ' '.join(['Oh yeah,', self.fave_char.name, 'is funny.', 'Here\'s a quote: \n',
self._get_good_quote()
])
if self.kmp(wordlist, ['add', 'subject']) >= 0: # rule 8: add subject
self.prev_quote = None
try:
subject = wordlist[wordlist.index('subject') + 1]
self.new_subjects.append(subject)
return 'I will remember that one for later.'
except (IndexError, ValueError):
return 'Unable to remember that one...'
if self.kmp(wordlist, ['show', r'\w*', 'subjects']) >= 0 or self.kmp(wordlist, ['show', 'subjects']) >= 0: # rule 9: show subjects
self.prev_quote = None
index = max(self.kmp(wordlist, ['show', r'\w*', 'subjects']), self.kmp(wordlist, ['show', 'subjects']))
try:
if 'default' in wordlist[index : index + 2]:
return 'Here are the default subjects: ' + ', '.join(self.def_subjects) + '.'
elif 'my' in wordlist[index : index + 2]:
return 'Here are the subjects I remember: ' + ', '.join(self.new_subjects) + '.'
elif len(wordlist[index:index+2]) == 2:
return 'These are all the subjects I know: ' + ', '.join(self.def_subjects + self.new_subjects) + '.'
else:
return 'I\'m not quite sure what subjects you want.'
except (IndexError, ValueError):
return 'I\'m not quite sure what subjects you want.'
if self.kmp(wordlist, ['show', 'preferences']) >= 0: # rule 10: show current user prefs.
self.prev_quote = None
lines = ['Here are the preferences I know:',
'Favorite character: ' + (self.fave_char.name if self.fave_char else 'NONE'),
'Favorite episode: ' + (self.fave_ep.title if self.fave_ep else 'NONE')]
return '\n\t'.join(lines)
if self.prev_quote and 'context' in sub_wordlist or self.kmp(wordlist, ['i', 'don\'t', 'understand']) >= 0: #rule 5 give context with cycle
if self.context_count < 5: #CYCLE
p = self.seinfeld.passage(self.prev_quote, length=self.context_count * 3 + 8)
lines = ['Perhaps this will help:']
for q in p.quotes:
lines.append(q.speaker.name + ': ' + q.text)
self.context_count += 1
return '\n\t'.join(lines)
else:
self.context_count = 0
return 'Seems like you\'re just not going to understand this one...\n' + self._get_random_quote(subject='understand')
if self.kmp(wordlist, ['remove', 'preferences']) >= 0: #rule 7: remove preferences
self.fave_char = None
self.fave_ep = None
self.prev_quote = None
return '\n'.join(['I\'ve forgotten everything you told me.', self._get_random_quote(subject='forgot')])
try: #default response
return self._get_good_quote(subject=choice(sub_wordlist) if sub_wordlist else None)
except IndexError:
return self._get_good_quote()
def _get_good_quote(self, subject=None, episode=None):
q = None
qtext = ''
while len(split(' ', qtext)) < 8:
if episode:
q = choice(self.seinfeld.search(episode=episode, random=True, limit=10))
elif any([self.fave_char, subject]):
q = choice(self.seinfeld.search(speaker=self.fave_char, random=True, limit=10, subject=subject))
else:
q = choice(self.seinfeld.search(subject=choice(self.def_subjects + self.new_subjects), random=True, limit=10)) #essentially random quote
qtext = q.text
self.prev_quote = q
self.context_count = 0
return q.speaker.name.strip() + ': ' + qtext
def _get_random_quote(self, subject=None):
q = None
qtext = ''
while len(split(' ', qtext)) < 8:
q = self.seinfeld.random(subject=subject, speaker=choice(self.chars) if not subject else None)
qtext = q.text
self.prev_quote = q
self.context_count = 0
return q.speaker.name.strip() + ': ' + qtext
def agentName(self):
return self.agent_name
"""
Mock app to test building
"""
from seinfeld import Seinfeld # pylint: disable=import-error
SEINFELD = Seinfeld('mock.db')
if __name__ == "__main__":
print(SEINFELD.season(1).episodes[1].title)
