Create slugs using Python

Makefile

test:
	clear
	nosetests --with-coverage --cover-package slugfy test_slugfy.py

clean:
	find -regex '.*\.pyc' -exec rm {} \;
	find -regex '.*~' -exec rm {} \;

.PHONY: test clean

slugfy.py

#!/usr/bin/env python
#coding: utf-8

from unicodedata import normalize


def slug(text, encoding=None):
    if isinstance(text, str):
        text = text.decode(encoding or 'ascii')
    clean_text = text.strip().replace(' ', '-')
    while '--' in clean_text:
        clean_text = clean_text.replace('--', '-')
    ascii_text = normalize('NFKD', clean_text).encode('ascii', 'ignore')
    return ascii_text.lower()

test_slugfy.py

#!/usr/bin/env python
# coding: utf-8

import unittest
from slugfy import slug


class TestSlug(unittest.TestCase):
    def test_should_always_return_lowercase_words(self):
        self.assertEquals(slug('ALVAROJUSTEN'), 'alvarojusten')

    def test_should_replace_space_with_dash(self):
        self.assertEquals(slug('Alvaro Justen'), 'alvaro-justen')

    def test_should_ignore_unecessary_spaces(self):
        self.assertEquals(slug('  alvaro   justen  '), 'alvaro-justen')

    def test_should_replace_nonascii_chars_with_corresponding_ascii_chars(self):
        self.assertEquals(slug('áÁàÀãÃâÂäÄ'.decode('utf8')), 'aaaaaaaaaa')
        self.assertEquals(slug('éÉèÈẽẼêÊëË'.decode('utf8')), 'eeeeeeeeee')
        self.assertEquals(slug('íÍìÌĩĨîÎïÏ'.decode('utf8')), 'iiiiiiiiii')
        self.assertEquals(slug('óÓòÒõÕôÔöÖ'.decode('utf8')), 'oooooooooo')
        self.assertEquals(slug('úÚùÙũŨûÛüÜ'.decode('utf8')), 'uuuuuuuuuu')
        self.assertEquals(slug('ćĆĉĈçÇ'.decode('utf8')), 'cccccc')

    def test_should_accept_unicode_text(self):
        self.assertEquals(slug(u'Álvaro Justen'), 'alvaro-justen')

    def test_should_accept_other_input_encodings(self):
        slugged_text = slug(u'Álvaro Justen'.encode('utf16'), 'utf16')
        self.assertEquals(slugged_text, 'alvaro-justen')

Not sure if this was intended to support all languages, but this won't work e.g. with German: 'ß' is dropped, Polish: 'ł' is dropped, Croatian: đ is dropped, and so on.