Ticket #13704: 13704-4.patch

django/utils/html.py

@@ -2 +2 @@
 
 import re
 import string
+import urlparse
 
 from django.utils.safestring import SafeData, mark_safe
 from django.utils.encoding import force_unicode
@@ -22 +23 @@
 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+\.\S+$')
 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)
@@ -103 +104 @@
 
 def smart_urlquote(url):
     """Quotes an URL if it isn't already quoted."""
+    # Handle IDN first
+    scheme, netloc, path, query, fragment = urlparse.urlsplit(url)
+    try:
+        netloc = netloc.encode('idna') # IDN -> ACE
+    except UnicodeError: # invalid domain part
+        pass
+    else:
+        url = urlparse.urlunsplit((scheme, netloc, path, query, fragment))
+
     # An URL is considered unquoted if it contains no % character, or if it
     # contains a % not followed by two hexadecimal digits. See #9655.
     if '%' not in url or unquoted_percents_re.search(url):
         # See http://bugs.python.org/issue2637
-        return urlquote(url, safe='!*\'();:@&=+$,/?#[]~')
+        url = urlquote(url, safe='!*\'();:@&=+$,/?#[]~')
+
     return url
 
 def urlize(text, trim_url_limit=None, nofollow=False, autoescape=False):
@@ -145 +156 @@
                     middle and middle[0] in string.ascii_letters + string.digits and \
                     (middle.endswith('.org') or middle.endswith('.net') or middle.endswith('.com'))):
                 url = smart_urlquote('http://%s' % middle)
-            elif '@' in middle and not ':' in middle and simple_email_re.match(middle):
-                url = 'mailto:%s' % middle
+            elif not ':' in middle and simple_email_re.match(middle):
+                local, domain = middle.rsplit('@', 1)
+                domain = domain.encode('idna')
+                url = 'mailto:%s@%s' % (local, domain)
                 nofollow_attr = ''
             # Make link.
             if url:

tests/regressiontests/defaultfilters/tests.py

@@ -238 +238 @@
         # Check urlize with https addresses
         self.assertEqual(urlize('https://google.com'),
             u'<a href="https://google.com" rel="nofollow">https://google.com</a>')
+
         # Check urlize doesn't overquote already quoted urls - see #9655
         self.assertEqual(urlize('http://hi.baidu.com/%D6%D8%D0%C2%BF'),
             u'<a href="http://hi.baidu.com/%D6%D8%D0%C2%BF" rel="nofollow">'
@@ -252 +253 @@
             u'<a href="http://en.wikipedia.org/wiki/Caf%C3%A9" rel="nofollow">'
             u'http://en.wikipedia.org/wiki/Café</a>')
 
+        # Check urlize handles IDN correctly - see #13704
+        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)
         self.assertEqual(wordcount(u'oneword'), 1)