Commit e3a7bfcc authored by Aymeric Augustin's avatar Aymeric Augustin
Browse files

Fixed #9655 -- Prevented the urlize template filter from double-quoting URLs.... 

Fixed #9655 -- Prevented the urlize template filter from double-quoting URLs. Thanks Claude Paroz for writing the tests.


git-svn-id: http://code.djangoproject.com/svn/django/trunk@17347 bcc190cf-cafb-0310-a4f2-bffc1f526a37
parent c3697a09

django/utils/html.py

+12 −2
Original line number Diff line number Diff line
@@ -17,6 +17,7 @@ TRAILING_PUNCTUATION = ['.', ',', ')', '>', '\n', '>'] 
DOTS = [u'·', u'*', u'\u2022', u'•', u'•', u'•']



unencoded_ampersands_re = re.compile(r'&(?!(\w+|#\d+);)')

unquoted_percents_re = re.compile(r'%(?![0-9A-Fa-f]{2})')

word_split_re = re.compile(r'(\s+)')

punctuation_re = re.compile('^(?P<lead>(?:%s)*)(?P<middle>.*?)(?P<trail>(?:%s)*)$' % \

    ('|'.join([re.escape(x) for x in LEADING_PUNCTUATION]),
@@ -100,6 +101,15 @@ def fix_ampersands(value): 
    return unencoded_ampersands_re.sub('&amp;', force_unicode(value))

fix_ampersands = allow_lazy(fix_ampersands, unicode)



def smart_urlquote(url):

    """Quotes an URL if it isn't already quoted."""

    # 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='!*\'();:@&=+$,/?#[]~')

    return url



def urlize(text, trim_url_limit=None, nofollow=False, autoescape=False):

    """

    Converts any URLs in text into clickable links.
@@ -130,11 +140,11 @@ def urlize(text, trim_url_limit=None, nofollow=False, autoescape=False): 
            # Make URL we want to point to.

            url = None

            if middle.startswith('http://') or middle.startswith('https://'):

                url = urlquote(middle, safe='/&=:;#?+*')

                url = smart_urlquote(middle)

            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 = smart_urlquote('http://%s' % middle)

            elif '@' in middle and not ':' in middle and simple_email_re.match(middle):

                url = 'mailto:%s' % middle

                nofollow_attr = ''

docs/releases/1.4.txt

+9 −0
Original line number Diff line number Diff line
@@ -1044,6 +1044,15 @@ Now, the flags are keyword arguments of :meth:`@register.filter 


See :ref:`filters and auto-escaping <filters-auto-escaping>` for more information.



The :tfilter:`urlize` filter no longer escapes every URL

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



When an URL contains a ``%xx`` sequence, where ``xx`` are two hexadecimal

digits, :tfilter:`urlize` assumes that the URL is already escaped, and doesn't

apply URL escaping again. This is wrong for URLs whose unquoted form contains

a ``%xx`` sequence, but such URLs are very unlikely to happen in the wild,

since they would confuse browsers too.



Session cookies now have the ``httponly`` flag by default

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

tests/regressiontests/defaultfilters/tests.py

+13 −0
Original line number Diff line number Diff line
@@ -238,6 +238,19 @@ class DefaultFiltersTests(TestCase): 
        # 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">'

            u'http://hi.baidu.com/%D6%D8%D0%C2%BF</a>')

        self.assertEqual(urlize('www.mystore.com/30%OffCoupons!'),

            u'<a href="http://www.mystore.com/30%25OffCoupons!" rel="nofollow">'

            u'www.mystore.com/30%OffCoupons!</a>')

        self.assertEqual(urlize('http://en.wikipedia.org/wiki/Caf%C3%A9'),

            u'<a href="http://en.wikipedia.org/wiki/Caf%C3%A9" rel="nofollow">'

            u'http://en.wikipedia.org/wiki/Caf%C3%A9</a>')

        self.assertEqual(urlize('http://en.wikipedia.org/wiki/Café'),

            u'<a href="http://en.wikipedia.org/wiki/Caf%C3%A9" rel="nofollow">'

            u'http://en.wikipedia.org/wiki/Café</a>')



    def test_wordcount(self):

        self.assertEqual(wordcount(''), 0)