Ticket #13704: 13704-3.diff

django/utils/html.py

@@ -2 +2 @@
 
 import re
 import string
+from urlparse import urlparse, urlunparse
 
 from django.utils.safestring import SafeData, mark_safe
 from django.utils.encoding import force_unicode
@@ -21 +22 @@
 punctuation_re = re.compile('^(?P<lead>(?:%s)*)(?P<middle>.*?)(?P<trail>(?:%s)*)$' % \
     ('|'.join([re.escape(x) for x in LEADING_PUNCTUATION]),
     '|'.join([re.escape(x) for x in TRAILING_PUNCTUATION])))
-simple_email_re = re.compile(r'^\S+@[a-zA-Z0-9._-]+\.[a-zA-Z0-9._-]+$')
+simple_email_re = re.compile(r'^\S+@\S+\.[a-zA-Z0-9._-]+$')
 link_target_attribute_re = re.compile(r'(<a [^>]*?)target=[^\s>]+')
 html_gunk_re = re.compile(r'(?:<br clear="all">|<i><\/i>|<b><\/b>|<em><\/em>|<strong><\/strong>|<\/?smallcaps>|<\/?uppercase>)', re.IGNORECASE)
 hard_coded_bullets_re = re.compile(r'((?:<p>(?:%s).*?[a-zA-Z].*?</p>\s*)+)' % '|'.join([re.escape(x) for x in DOTS]), re.DOTALL)
@@ -118 +119 @@
     If autoescape is True, the link text and URLs will get autoescaped.
     """
     trim_url = lambda x, limit=trim_url_limit: limit is not None and (len(x) > limit and ('%s...' % x[:max(0, limit - 3)])) or x
+
+    def clean_netloc(url):
+        """ Handle idna encoding """
+        try:
+            parsed = urlparse(url)
+            url = urlunparse(
+                (parsed.scheme, parsed.netloc.encode('idna'), parsed.path,
+                 parsed.params, parsed.query, parsed.fragment))
+        except ValueError:
+            pass
+        return url
+
     safe_input = isinstance(text, SafeData)
     words = word_split_re.split(force_unicode(text))
     nofollow_attr = nofollow and ' rel="nofollow"' or ''
@@ -130 +143 @@
             # Make URL we want to point to.
             url = None
             if middle.startswith('http://') or middle.startswith('https://'):
-                url = urlquote(middle, safe='/&=:;#?+*')
+                url = urlquote(clean_netloc(middle), safe='/&=:;#?+*')
             elif middle.startswith('www.') or ('@' not in middle and \
                     middle and middle[0] in string.ascii_letters + string.digits and \
                     (middle.endswith('.org') or middle.endswith('.net') or middle.endswith('.com'))):
-                url = urlquote('http://%s' % middle, safe='/&=:;#?+*')
+                url = urlquote(clean_netloc('http://%s' % middle), safe='/&=:;#?+*')
             elif '@' in middle and not ':' in middle and simple_email_re.match(middle):
-                url = 'mailto:%s' % middle
+                parts = middle.split('@', 1)
+                url = 'mailto:%s@%s' % (parts[0], clean_netloc('http://%s' % parts[1])[7:])
                 nofollow_attr = ''
             # Make link.
             if url:

tests/regressiontests/defaultfilters/tests.py

@@ -236 +236 @@
         # Check urlize with https addresses
         self.assertEqual(urlize('https://google.com'),
             u'<a href="https://google.com" rel="nofollow">https://google.com</a>')
+        # IDN domain names
+        self.assertEqual(urlize('http://c✶.ws'),
+            u'<a href="http://xn--c-lgq.ws" rel="nofollow">http://c✶.ws</a>')
+        self.assertEqual(urlize('www.c✶.ws'),
+            u'<a href="http://www.xn--c-lgq.ws" rel="nofollow">www.c✶.ws</a>')
+        self.assertEqual(urlize('c✶.org'),
+            u'<a href="http://xn--c-lgq.org" rel="nofollow">c✶.org</a>')
+        self.assertEqual(urlize('info@c✶.org'),
+            u'<a href="mailto:info@xn--c-lgq.org">info@c✶.org</a>')
 
     def test_wordcount(self):
         self.assertEqual(wordcount(''), 0)