creepy shakespeare

README

Requires shakespeare_sonnets.json from https://github.com/dariusk/corpora

python transform_data.py > lines.txt
python creep.py

creep.py

import random
import sys

PATTERNS = (
    '{} ;)',
    '{} ;)',
    '{} ;)',
    '{} ;)',
    '{} ;)',
    '{} ;)',
    '{} ;)',
    '{} ;) x',
    '{} :/',
    '{} lol',
    '{} lol ;)',
    '{} ;) lol',
    'haha {} ;)',
    '{} haha ;)',
    '{} ;) haha',
)

def source_lines():
    lines = []
    with open('lines.txt') as fp:
        for line in fp:
            lines.append(line.strip())
    return lines

def creep():
    line = random.choice(source_lines())
    pattern = random.choice(PATTERNS)
    return pattern.format(line)

if __name__ == '__main__':
    for _ in range(10):
        print creep()

transform_data.py

import json

SEP_CHARS = '.,;?!:'
REMOVE_CHARS = '\'"'

def parse_lines(s):
    lines = []

    s = s.lower().replace('--', ' ')
    for sep in SEP_CHARS:
        s = s.replace(sep, '\n')
    for rem in REMOVE_CHARS:
        s = s.replace(rem, '')

    def replace_word(word):
        if word in ('you', 'thou'):
            return 'u'
        elif word in ('your', 'thine'):
            return 'ur'
        elif word in ('why', ):
            return 'y'
        else:
            return word

    for line in s.split('\n'):
        words = line.split()
        if 1 < len(words) < 5:
            lines.append(' '.join(replace_word(word) for word in words))

    return lines

if __name__ == '__main__':
    lines = []
    with open('shakespeare_sonnets.json') as fp:
        for sonnet in json.load(fp)['sonnets']:
            for line in sonnet['lines']:
                lines.extend(parse_lines(line))
    print '\n'.join(lines)