Generate a regular expression matching all numbers of the following form: - Any number of digits as written out English words (case insensitive) followed by at least one whitespace. Any number witch typos (e.g. eihgt instead of eight) are allowed, but no character must move more than one position. - followed by any number of digits.

xmasregex.py

#!/usr/bin/nev python


def gen_switch_typo_variants(string):
    if len(string) < 2:
        return [string]

    var1 = [string[0] + v for v in gen_switch_typo_variants(string[1:])]
    var2 = [string[1::-1] + v for v in gen_switch_typo_variants(string[2:])]
    return var1 + var2


def make_case_insensitive(string):
    return reduce(
        lambda a, b: a + b, ('[{}{}]'.format(c, c.upper()) for c in string))


number_words = ['one', 'two', 'three', 'four', 'five', 'six', 'seven', 'eight',
                'nine', 'zero']


if __name__ == '__main__':
    all_lower_number_writings = reduce(
        lambda a, b: a + b,
        (gen_switch_typo_variants(word) for word in number_words))
    all_lower_number_writings_case_insensitive = [
        make_case_insensitive(w) for w in all_lower_number_writings]
    regex = r'^(({writings})\s+)*\d*$'.format(
        writings='|'.join(all_lower_number_writings_case_insensitive))
    print regex