[1.7.x] Improved strip_tags and clarified documentation

The fact that strip_tags cannot guarantee to really strip all
non-safe HTML content was not clear enough. Also see:
https://www.djangoproject.com/weblog/2014/mar/22/strip-tags-advisory/
Backport of 6ca6c36 from master.

django/utils/html.py

@@ -120,7 +120,10 @@ def linebreaks(value, autoescape=False):
 
 class MLStripper(HTMLParser):
     def __init__(self):
-        HTMLParser.__init__(self)
+        if six.PY2:
+            HTMLParser.__init__(self)
+        else:
+            HTMLParser.__init__(self, strict=False)
         self.reset()
         self.fed = []
 
@@ -137,16 +140,36 @@ def get_data(self):
         return ''.join(self.fed)
 
 
-def strip_tags(value):
-    """Returns the given HTML with all tags stripped."""
+def _strip_once(value):
+    """
+    Internal tag stripping utility used by strip_tags.
+    """
     s = MLStripper()
     try:
         s.feed(value)
-        s.close()
     except HTMLParseError:
         return value
+    try:
+        s.close()
+    except (HTMLParseError, UnboundLocalError) as err:
+        # UnboundLocalError because of http://bugs.python.org/issue17802
+        # on Python 3.2, triggered by strict=False mode of HTMLParser
+        return s.get_data() + s.rawdata
     else:
         return s.get_data()
+
+
+def strip_tags(value):
+    """Returns the given HTML with all tags stripped."""
+    while True:
+        if not ('<' in value or '>' in value):
+            return value
+        new_value = _strip_once(value)
+        if new_value == value:
+            # _strip_once was not able to detect more tags
+            return value
+        else:
+            value = new_value
 strip_tags = allow_lazy(strip_tags)
 
 

docs/ref/templates/builtins.txt

@@ -2050,7 +2050,7 @@ If ``value`` is ``10``, the output will be ``1.000000E+01``.
 striptags
 ^^^^^^^^^
 
-Strips all [X]HTML tags.
+Makes all possible efforts to strip all [X]HTML tags.
 
 For example::
 
@@ -2059,6 +2059,16 @@ For example::
 If ``value`` is ``"<b>Joel</b> <button>is</button> a <span>slug</span>"``, the
 output will be ``"Joel is a slug"``.
 
+.. admonition:: No safety guarantee
+
+    Note that ``striptags`` doesn't give any guarantee about its output being
+    entirely HTML safe, particularly with non valid HTML input. So **NEVER**
+    apply the ``safe`` filter to a ``striptags`` output.
+    If you are looking for something more robust, you can use the ``bleach``
+    Python library, notably its `clean`_ method.
+
+.. _clean: http://bleach.readthedocs.org/en/latest/clean.html
+
 .. templatefilter:: time
 
 time

docs/ref/utils.txt

@@ -595,17 +595,23 @@ escaping HTML.
 
 .. function:: strip_tags(value)
 
-    Removes anything that looks like an html tag from the string, that is
-    anything contained within ``<>``.
+    Tries to remove anything that looks like an HTML tag from the string, that
+    is anything contained within ``<>``.
+    Absolutely NO guaranty is provided about the resulting string being entirely
+    HTML safe. So NEVER mark safe the result of a ``strip_tag`` call without
+    escaping it first, for example with :func:`~django.utils.html.escape`.
 
     For example::
 
         strip_tags(value)
 
     If ``value`` is ``"<b>Joel</b> <button>is</button> a <span>slug</span>"``
-    the return value will be ``"Joel is a slug"``. Note that ``strip_tags``
-    result may still contain unsafe HTML content, so you might use
-    :func:`~django.utils.html.escape` to make it a safe string.
+    the return value will be ``"Joel is a slug"``.
+
+    If you are looking for a more robust solution, take a look at the `bleach`_
+    Python library.
+
+    .. _bleach: https://pypi.python.org/pypi/bleach
 
     .. versionchanged:: 1.6
 

tests/utils_tests/test_html.py

@@ -82,6 +82,8 @@ def test_strip_tags(self):
             ('a<p a >b</p>c', 'abc'),
             ('d<a:b c:d>e</p>f', 'def'),
             ('<strong>foo</strong><a href="http://example.com">bar</a>', 'foobar'),
+            ('<sc<!-- -->ript>test<<!-- -->/script>', 'test'),
+            ('<script>alert()</script>&h', 'alert()&h'),
         )
         for value, output in items:
             self.check_output(f, value, output)