only_ascii boolean with the support of unidecode

README.md

@@ -11,3 +11,13 @@ required more than simple transliteration.
 
     >>> slugify.slugify(u'Bän...g (bang)')
     u'bäng-bang'
+
+    >>> slugify.slugify(u'Bäuma means a tree', only_ascii=True)
+    u'bauma-means-a-tree'
+
+    >>> slugify(u'Bakıcı geldi', only_ascii=True)
+    u'bakici-geldi'
+
+## Thanks
+
+Tomaz Solc, unidecode, https://pypi.python.org/pypi/Unidecode

setup.py

@@ -2,7 +2,7 @@
 
 setup(
     name='unicode-slugify',
-    version='0.1.2',
+    version='0.1.3',
     description='A slug generator that turns strings into unicode slugs.',
     long_description=open('README.md').read(),
     author='Jeff Balogh, Dave Dash',
@@ -13,7 +13,7 @@
     include_package_data=True,
     package_data={'': ['README.md']},
     zip_safe=False,
-    install_requires=['six'],
+    install_requires=['six', 'unidecode'],
     classifiers=[
         'Development Status :: 4 - Beta',
         'Environment :: Web Environment',

slugify/__init__.py

@@ -1,6 +1,7 @@
 import re
 import six
 import unicodedata
+from unidecode import unidecode
 
 
 def smart_text(s, encoding='utf-8', errors='strict'):
@@ -26,9 +27,27 @@ def smart_text(s, encoding='utf-8', errors='strict'):
 SLUG_OK = '-_~'
 
 
-def slugify(s, ok=SLUG_OK, lower=True, spaces=False):
-    # L and N signify letter/number.
-    # http://www.unicode.org/reports/tr44/tr44-4.html#GC_Values_Table
+def slugify(s, ok=SLUG_OK, lower=True, spaces=False, only_ascii=False):
+    """
+    Creates a unicode slug for given string with several options.
+
+    L and N signify letter/number.
+    http://www.unicode.org/reports/tr44/tr44-4.html#GC_Values_Table
+
+    :param s: Your unicode string.
+    :param ok: Extra characters outside of alphanumerics to be allowed.
+    :param lower: Lower the output string.
+    :param spaces: True allows spaces, False replaces a space with a dash (-).
+    :param only_ascii: True to replace non-ASCII unicode characters with their ASCII representations.
+    :type s: String
+    :type ok: String
+    :type lower: Bool
+    :type spaces: Bool
+    :type only_ascii: Bool
+    :return: Slugified unicode string
+
+    """
+
     rv = []
     for c in unicodedata.normalize('NFKC', smart_text(s)):
         cat = unicodedata.category(c)[0]
@@ -39,4 +58,10 @@ def slugify(s, ok=SLUG_OK, lower=True, spaces=False):
     new = ''.join(rv).strip()
     if not spaces:
         new = re.sub('[-\s]+', '-', new)
-    return new.lower() if lower else new
+
+    new = new.lower() if lower else new
+
+    if only_ascii == True:
+        new = unidecode(new)
+
+    return new

slugify/tests.py

@@ -16,6 +16,12 @@ def test_slugify():
     def check(x, y):
         eq_(slugify(x), y)
 
+    def check_only_ascii(x, y):
+        eq_(slugify(x, only_ascii=True), y)
+
+    def check_only_ascii_capital(x, y):
+        eq_(slugify(x, lower=False, only_ascii=True), y)
+
     s = [('xx x  - "#$@ x', 'xx-x-x'),
          (u'Bän...g (bang)', u'bäng-bang'),
          (u, u.lower()),
@@ -33,11 +39,24 @@ def check(x, y):
          # I don't really care what slugify returns.  Just don't crash.
          (u'x𘍿', u'x'),
          (u'ϧ΃𘒬𘓣',  u'\u03e7'),
-         (u'¿x', u'x')]
+         (u'¿x', u'x'),
+         (u'Bakıcı geldi', u'bak\u0131c\u0131-geldi'),
+         (u'Bäuma means tree', u'b\xe4uma-means-tree')]
+
+    only_ascii = [(u'Bakıcı geldi', u'bakici-geldi'), (u'Bäuma means tree', u'bauma-means-tree')]
+
+    only_ascii_capital = [(u'BÄUMA MEANS TREE', u'BAUMA-MEANS-TREE'),
+                             (u'EMİN WAS HERE', u'EMIN-WAS-HERE')]
 
     for val, expected in s:
         yield check, val, expected
 
+    for val, expected in only_ascii:
+        yield check_only_ascii, val, expected
+
+    for val, expected in only_ascii_capital:
+        yield check_only_ascii_capital, val, expected
+
 
 class SmartTextTestCase(unittest.TestCase):