Merge pull request #1397 from amezhenin/issue_1386

update regex for matching URLs, fixes issue #1386

2 files changed, 47 insertions, 5 deletions

diff --git a/rest_framework/templatetags/rest_framework.py b/rest_framework/templatetags/rest_framework.py
index 83c046f9..beb8c5b0 100644
--- a/rest_framework/templatetags/rest_framework.py
+++ b/rest_framework/templatetags/rest_framework.py
@@ -6,7 +6,7 @@ from django.utils.encoding import iri_to_uri
 from django.utils.html import escape
 from django.utils.safestring import SafeData, mark_safe
 from rest_framework.compat import urlparse, force_text, six, smart_urlquote
-import re, string
+import re
 
 register = template.Library()
 
@@ -189,6 +189,17 @@ simple_url_2_re = re.compile(r'^www\.|^(?!http)\w[^@]+\.(com|edu|gov|int|mil|net
 simple_email_re = re.compile(r'^\S+@\S+\.\S+$')
 
 
+def smart_urlquote_wrapper(matched_url):
+    """
+    Simple wrapper for smart_urlquote. ValueError("Invalid IPv6 URL") can
+    be raised here, see issue #1386
+    """
+    try:
+        return smart_urlquote(matched_url)
+    except ValueError:
+        return None
+
+
 @register.filter
 def urlize_quoted_links(text, trim_url_limit=None, nofollow=True, autoescape=True):
     """
@@ -211,7 +222,6 @@ def urlize_quoted_links(text, trim_url_limit=None, nofollow=True, autoescape=Tru
     safe_input = isinstance(text, SafeData)
     words = word_split_re.split(force_text(text))
     for i, word in enumerate(words):
-        match = None
         if '.' in word or '@' in word or ':' in word:
             # Deal with punctuation.
             lead, middle, trail = '', word, ''
@@ -233,9 +243,9 @@ def urlize_quoted_links(text, trim_url_limit=None, nofollow=True, autoescape=Tru
             url = None
             nofollow_attr = ' rel="nofollow"' if nofollow else ''
             if simple_url_re.match(middle):
-                url = smart_urlquote(middle)
+                url = smart_urlquote_wrapper(middle)
             elif simple_url_2_re.match(middle):
-                url = smart_urlquote('http://%s' % middle)
+                url = smart_urlquote_wrapper('http://%s' % middle)
             elif not ':' in middle and simple_email_re.match(middle):
                 local, domain = middle.rsplit('@', 1)
                 try:
diff --git a/rest_framework/tests/test_templatetags.py b/rest_framework/tests/test_templatetags.py
index 609a9e08..d4da0c23 100644
--- a/rest_framework/tests/test_templatetags.py
+++ b/rest_framework/tests/test_templatetags.py
@@ -2,7 +2,7 @@
 from __future__ import unicode_literals
 from django.test import TestCase
 from rest_framework.test import APIRequestFactory
-from rest_framework.templatetags.rest_framework import add_query_param
+from rest_framework.templatetags.rest_framework import add_query_param, urlize_quoted_links
 
 factory = APIRequestFactory()
 
@@ -17,3 +17,35 @@ class TemplateTagTests(TestCase):
         json_url = add_query_param(request, "format", "json")
         self.assertIn("q=%E6%9F%A5%E8%AF%A2", json_url)
         self.assertIn("format=json", json_url)
+
+
+class Issue1386Tests(TestCase):
+    """
+    Covers #1386
+    """
+
+    def test_issue_1386(self):
+        """
+        Test function urlize_quoted_links with different args
+        """
+        correct_urls = [
+            "asdf.com",
+            "asdf.net",
+            "www.as_df.org",
+            "as.d8f.ghj8.gov",
+        ]
+        for i in correct_urls:
+            res = urlize_quoted_links(i)
+            self.assertNotEqual(res, i)
+            self.assertIn(i, res)
+
+        incorrect_urls = [
+            "mailto://asdf@fdf.com",
+            "asdf.netnet",
+        ]
+        for i in incorrect_urls:
+            res = urlize_quoted_links(i)
+            self.assertEqual(i, res)
+
+        # example from issue #1386, this shouldn't raise an exception
+        _ = urlize_quoted_links("asdf:[/p]zxcv.com")