diff options
| author | Luke Plant <L.Plant.98@cantab.net> | 2011-04-28 14:08:53 +0000 |
|---|---|---|
| committer | Luke Plant <L.Plant.98@cantab.net> | 2011-04-28 14:08:53 +0000 |
| commit | cf11e3789b6643cf451d79d675a97c4de94542b0 (patch) | |
| tree | a1b7c10559f18a1daef2a6d63b75d98c5821d274 | |
| parent | 2ac4f175ec7ec5f6122d079e7f1dbac4800e7657 (diff) | |
Fixed #7267 - UnicodeDecodeError in clean_html
Thanks to Nikolay for the report, and gav and aaugustin for the patch.
git-svn-id: http://code.djangoproject.com/svn/django/trunk@16118 bcc190cf-cafb-0310-a4f2-bffc1f526a37
| -rw-r--r-- | django/utils/html.py | 8 | ||||
| -rw-r--r-- | tests/regressiontests/utils/html.py | 12 |
2 files changed, 16 insertions, 4 deletions
diff --git a/django/utils/html.py b/django/utils/html.py index 094bc6660d..7fda015840 100644 --- a/django/utils/html.py +++ b/django/utils/html.py @@ -13,7 +13,7 @@ LEADING_PUNCTUATION = ['(', '<', '<'] TRAILING_PUNCTUATION = ['.', ',', ')', '>', '\n', '>'] # List of possible strings used for bullets in bulleted lists. -DOTS = ['·', '*', '\xe2\x80\xa2', '•', '•', '•'] +DOTS = [u'·', u'*', u'\u2022', u'•', u'•', u'•'] unencoded_ampersands_re = re.compile(r'&(?!(\w+|#\d+);)') word_split_re = re.compile(r'(\s+)') @@ -180,13 +180,13 @@ def clean_html(text): text = html_gunk_re.sub('', text) # Convert hard-coded bullets into HTML unordered lists. def replace_p_tags(match): - s = match.group().replace('</p>', '</li>') + s = match.group().replace(u'</p>', u'</li>') for d in DOTS: - s = s.replace('<p>%s' % d, '<li>') + s = s.replace(u'<p>%s' % d, u'<li>') return u'<ul>\n%s\n</ul>' % s text = hard_coded_bullets_re.sub(replace_p_tags, text) # Remove stuff like "<p> </p>", but only if it's at the bottom # of the text. - text = trailing_empty_content_re.sub('', text) + text = trailing_empty_content_re.sub(u'', text) return text clean_html = allow_lazy(clean_html, unicode) diff --git a/tests/regressiontests/utils/html.py b/tests/regressiontests/utils/html.py index 3acb218cd1..d8b9bde8bf 100644 --- a/tests/regressiontests/utils/html.py +++ b/tests/regressiontests/utils/html.py @@ -121,3 +121,15 @@ class TestUtilsHtml(unittest.TestCase): ) for value, output in items: self.check_output(f, value, output) + + def test_clean_html(self): + f = html.clean_html + items = ( + (u'<p>I <i>believe</i> in <b>semantic markup</b>!</p>', u'<p>I <em>believe</em> in <strong>semantic markup</strong>!</p>'), + (u'I escape & I don\'t <a href="#" target="_blank">target</a>', u'I escape & I don\'t <a href="#" >target</a>'), + (u'<p>I kill whitespace</p><br clear="all"><p> </p>', u'<p>I kill whitespace</p>'), + # also a regression test for #7267: this used to raise an UnicodeDecodeError + (u'<p>* foo</p><p>* bar</p>', u'<ul>\n<li> foo</li><li> bar</li>\n</ul>'), + ) + for value, output in items: + self.check_output(f, value, output) |