Extract unicode words from string

extract_unicode_words.py

# -*- coding: utf-8 -*-

import unicodedata as mod_unicodedata

all_unicode = ''.join(unichr(i) for i in xrange(65536))
UNICODE_LETTERS = ''.join(c for c in all_unicode if mod_unicodedata.category(c)=='Lu' or mod_unicodedata.category(c)=='Ll')
UNICODE_LETTERS_AND_NUMBERS = '1234567890' + UNICODE_LETTERS

def extract_unicode_words(text, allowed_chars=None):
    if text.__class__ != unicode:
        text = text.decode('utf-8')
    if not allowed_chars:
        allowed_chars = UNICODE_LETTERS_AND_NUMBERS
    words = []
    current_word = ''
    for c in text:
        if c in allowed_chars:
            current_word += c
        elif current_word:
            words.append(current_word)
            current_word = ''
    if current_word:
        words.append(current_word)
    return words

if __name__ == '__main__':
    print extract_unicode_words('dsuifouio eršešđšĐŠĐŠ zuizu 374637 ČĆeuzu')