@@ -102,12 +102,14 @@ def sanitize_html(html='<p>No comment</p>', encoding=None, if isinstance(html, unicode) and not encoding: logging.debug("Sanitizing unicode input.") - soup = BeautifulSoup(html) + soup = BeautifulSoup(html, + convertEntities=BeautifulSoup.XHTML_ENTITIES) else: if not encoding: encoding = 'latin-1' logging.debug("Sanitizing string input, assuming %s", encoding) - soup = BeautifulSoup(html.decode(encoding, 'ignore')) + soup = BeautifulSoup(html.decode(encoding, 'ignore'), + convertEntities=BeautifulSoup.XHTML_ENTITIES) for comment in soup.findAll( text = lambda text: isinstance(text, Comment)): comment.extract() utils/sanitizer.py 4bfe1ddd1aaeb8d6abb0a02463cbe8a4af4c2fac Bill Katz billkatz@gmail.com http://github.com/DocSavage/bloog/commit/b5f9622ce02428cd6dac80a8239d6f26460ec16c b5f9622ce02428cd6dac80a8239d6f26460ec16c 2008-08-23T20:29:57-07:00 2008-08-23T12:57:54-07:00 Convert HTML entities on receiving data. This should prevent errors in Atom feed as well. df5b8a6804a1563adf2327d570f307eb8fedd28a Bill Katz billkatz@gmail.com