Fixed urlize after smart_urlquote rewrite

Refs #22267.

django/utils/html.py

@@ -251,6 +251,7 @@ def smart_urlquote(url):
 
     return urlunsplit((scheme, netloc, path, query, fragment))
 
+
 def urlize(text, trim_url_limit=None, nofollow=False, autoescape=False):
     """
     Converts any URLs in text into clickable links.
@@ -268,11 +269,31 @@ def urlize(text, trim_url_limit=None, nofollow=False, autoescape=False):
 
     If autoescape is True, the link text and URLs will be autoescaped.
     """
+    safe_input = isinstance(text, SafeData)
+
     def trim_url(x, limit=trim_url_limit):
         if limit is None or len(x) <= limit:
             return x
         return '%s...' % x[:max(0, limit - 3)]
-    safe_input = isinstance(text, SafeData)
+
+    def unescape(text, trail):
+        """
+        If input URL is HTML-escaped, unescape it so as we can safely feed it to
+        smart_urlquote. For example:
+        http://example.com?x=1&amp;y=&lt;2&gt; => http://example.com?x=1&y=<2>
+        """
+        if not safe_input:
+            return text, text, trail
+        unescaped = (text + trail).replace('&amp;', '&').replace('&lt;', '<'
+                                 ).replace('&gt;', '>').replace('&quot;', '"'
+                                 ).replace('&#39;', "'")
+        # ';' in trail can be either trailing punctuation or end-of-entity marker
+        if unescaped.endswith(';'):
+            return text, unescaped[:-1], trail
+        else:
+            text += trail
+            return text, unescaped, ''
+
     words = word_split_re.split(force_text(text))
     for i, word in enumerate(words):
         if '.' in word or '@' in word or ':' in word:
@@ -296,9 +317,11 @@ def urlize(text, trim_url_limit=None, nofollow=False, autoescape=False):
             url = None
             nofollow_attr = ' rel="nofollow"' if nofollow else ''
             if simple_url_re.match(middle):
-                url = smart_urlquote(middle)
+                middle, middle_unescaped, trail = unescape(middle, trail)
+                url = smart_urlquote(middle_unescaped)
             elif simple_url_2_re.match(middle):
-                url = smart_urlquote('http://%s' % middle)
+                middle, middle_unescaped, trail = unescape(middle, trail)
+                url = smart_urlquote('http://%s' % middle_unescaped)
             elif ':' not in middle and simple_email_re.match(middle):
                 local, domain = middle.rsplit('@', 1)
                 try:
@@ -313,7 +336,7 @@ def urlize(text, trim_url_limit=None, nofollow=False, autoescape=False):
                 trimmed = trim_url(middle)
                 if autoescape and not safe_input:
                     lead, trail = escape(lead), escape(trail)
-                    url, trimmed = escape(url), escape(trimmed)
+                    trimmed = escape(trimmed)
                 middle = '<a href="%s"%s>%s</a>' % (url, nofollow_attr, trimmed)
                 words[i] = mark_safe('%s%s%s' % (lead, middle, trail))
             else:

tests/template_tests/filters.py

@@ -151,8 +151,16 @@ def get_filter_tests():
         'filter-upper01': ('{% autoescape off %}{{ a|upper }} {{ b|upper }}{% endautoescape %}', {"a": "a & b", "b": mark_safe("a & b")}, "A & B A & B"),
         'filter-upper02': ('{{ a|upper }} {{ b|upper }}', {"a": "a & b", "b": mark_safe("a & b")}, "A & B A & B"),
 
-        'filter-urlize01': ('{% autoescape off %}{{ a|urlize }} {{ b|urlize }}{% endautoescape %}', {"a": "http://example.com/?x=&y=", "b": mark_safe("http://example.com?x=&amp;y=")}, '<a href="http://example.com/?x=&y=" rel="nofollow">http://example.com/?x=&y=</a> <a href="http://example.com?x=&amp;y=" rel="nofollow">http://example.com?x=&amp;y=</a>'),
-        'filter-urlize02': ('{{ a|urlize }} {{ b|urlize }}', {"a": "http://example.com/?x=&y=", "b": mark_safe("http://example.com?x=&amp;y=")}, '<a href="http://example.com/?x=&amp;y=" rel="nofollow">http://example.com/?x=&amp;y=</a> <a href="http://example.com?x=&amp;y=" rel="nofollow">http://example.com?x=&amp;y=</a>'),
+        'filter-urlize01': (
+            '{% autoescape off %}{{ a|urlize }} {{ b|urlize }}{% endautoescape %}',
+            {"a": "http://example.com/?x=&y=", "b": mark_safe("http://example.com?x=&amp;y=&lt;2&gt;")},
+            '<a href="http://example.com/?x=&y=" rel="nofollow">http://example.com/?x=&y=</a> '
+            '<a href="http://example.com?x=&y=%3C2%3E" rel="nofollow">http://example.com?x=&amp;y=&lt;2&gt;</a>'),
+        'filter-urlize02': (
+            '{{ a|urlize }} {{ b|urlize }}',
+            {"a": "http://example.com/?x=&y=", "b": mark_safe("http://example.com?x=&amp;y=")},
+            '<a href="http://example.com/?x=&y=" rel="nofollow">http://example.com/?x=&amp;y=</a> '
+            '<a href="http://example.com?x=&y=" rel="nofollow">http://example.com?x=&amp;y=</a>'),
         'filter-urlize03': ('{% autoescape off %}{{ a|urlize }}{% endautoescape %}', {"a": mark_safe("a &amp; b")}, 'a &amp; b'),
         'filter-urlize04': ('{{ a|urlize }}', {"a": mark_safe("a &amp; b")}, 'a &amp; b'),
 
@@ -165,8 +173,16 @@ def get_filter_tests():
         'filter-urlize07': ('{{ a|urlize }}', {"a": "Email me at me@example.com"}, 'Email me at <a href="mailto:me@example.com">me@example.com</a>'),
         'filter-urlize08': ('{{ a|urlize }}', {"a": "Email me at <me@example.com>"}, 'Email me at &lt;<a href="mailto:me@example.com">me@example.com</a>&gt;'),
 
-        'filter-urlizetrunc01': ('{% autoescape off %}{{ a|urlizetrunc:"8" }} {{ b|urlizetrunc:"8" }}{% endautoescape %}', {"a": '"Unsafe" http://example.com/x=&y=', "b": mark_safe('&quot;Safe&quot; http://example.com?x=&amp;y=')}, '"Unsafe" <a href="http://example.com/x=&y=" rel="nofollow">http:...</a> &quot;Safe&quot; <a href="http://example.com?x=&amp;y=" rel="nofollow">http:...</a>'),
-        'filter-urlizetrunc02': ('{{ a|urlizetrunc:"8" }} {{ b|urlizetrunc:"8" }}', {"a": '"Unsafe" http://example.com/x=&y=', "b": mark_safe('&quot;Safe&quot; http://example.com?x=&amp;y=')}, '&quot;Unsafe&quot; <a href="http://example.com/x=&amp;y=" rel="nofollow">http:...</a> &quot;Safe&quot; <a href="http://example.com?x=&amp;y=" rel="nofollow">http:...</a>'),
+        'filter-urlizetrunc01': (
+            '{% autoescape off %}{{ a|urlizetrunc:"8" }} {{ b|urlizetrunc:"8" }}{% endautoescape %}',
+            {"a": '"Unsafe" http://example.com/x=&y=', "b": mark_safe('&quot;Safe&quot; http://example.com?x=&amp;y=')},
+            '"Unsafe" <a href="http://example.com/x=&y=" rel="nofollow">http:...</a> '
+            '&quot;Safe&quot; <a href="http://example.com?x=&y=" rel="nofollow">http:...</a>'),
+        'filter-urlizetrunc02': (
+            '{{ a|urlizetrunc:"8" }} {{ b|urlizetrunc:"8" }}',
+            {"a": '"Unsafe" http://example.com/x=&y=', "b": mark_safe('&quot;Safe&quot; http://example.com?x=&amp;y=')},
+            '&quot;Unsafe&quot; <a href="http://example.com/x=&y=" rel="nofollow">http:...</a> '
+            '&quot;Safe&quot; <a href="http://example.com?x=&y=" rel="nofollow">http:...</a>'),
 
         'filter-wordcount01': ('{% autoescape off %}{{ a|wordcount }} {{ b|wordcount }}{% endautoescape %}', {"a": "a & b", "b": mark_safe("a &amp; b")}, "3 3"),
         'filter-wordcount02': ('{{ a|wordcount }} {{ b|wordcount }}', {"a": "a & b", "b": mark_safe("a &amp; b")}, "3 3"),

tests/utils_tests/test_html.py

@@ -174,6 +174,7 @@ class TestUtilsHtml(TestCase):
         self.assertEqual(quote('http://example.com/path/öäü/'), 'http://example.com/path/%C3%B6%C3%A4%C3%BC/')
         self.assertEqual(quote('http://example.com/%C3%B6/ä/'), 'http://example.com/%C3%B6/%C3%A4/')
         self.assertEqual(quote('http://example.com/?x=1&y=2+3&z='), 'http://example.com/?x=1&y=2+3&z=')
+        self.assertEqual(quote('http://example.com/?x=<>"\''), 'http://example.com/?x=%3C%3E%22%27')
         self.assertEqual(quote('http://example.com/?q=http://example.com/?x=1%26q=django'),
                          'http://example.com/?q=http%3A%2F%2Fexample.com%2F%3Fx%3D1%26q%3Ddjango')
         self.assertEqual(quote('http://example.com/?q=http%3A%2F%2Fexample.com%2F%3Fx%3D1%26q%3Ddjango'),