wordgen

wordgen.py

import random
import collections
import re

Token = collections.namedtuple('Token', ['type', 'value'])

consonants = "bcdfghjklmnpqrstvwxzy"
vowels = "aeiou"

wordblock = "(C)(C)(C)V(C)(C)(C)(C)(C)"


def tokenize(pattern):
    token_specification = [
        ('CONSONANT', r'C'),
        ('VOWEL', r'V'),
        ('OPEN_PARENS', r'\('),
        ('CLOSE_PARENS', r'\)'),
        ('LITERAL', r'.')
    ]
    token_regex = '|'.join('(?P<%s>%s)' % pair for pair in token_specification)
    for item in re.finditer(token_regex, pattern):
        kind = item.lastgroup
        value = item.group(kind)
        yield Token(kind, value)


def parse(pattern):
    output = ''
    parens = False
    discard = False

    for token in tokenize(pattern):
        if (discard == True) and (token.type != 'CLOSE_PARENS'):
            continue

        if token.type == 'CONSONANT':
            output += random.choice(consonants)
        elif token.type == 'VOWEL':
            output += random.choice(vowels)
        elif token.type == 'OPEN_PARENS':
            if parens:
                raise ValueError('You have mismatched parentheses in your phonotactic constraints.')  # throw exception
            parens = True
            if random.getrandbits(1) == 1:
                discard = True
        elif token.type == 'CLOSE_PARENS':
            if not parens:
                raise ValueError('You have mismatched parentheses in your phonotactic constraints.')  # throw exception
            parens = False
            discard = False
        elif token.type == 'LITERAL':
            output += token.value

    return output

for i in range(0, 10):
    print(parse(wordblock))