HTMLParser is now able to handle slashes in the start tag.

This commit is contained in:
Ezio Melotti 2012-02-21 09:25:00 +02:00
parent 178e5ea305
commit 29877e8e04
3 changed files with 34 additions and 7 deletions

View file

@ -389,6 +389,27 @@ class HTMLParserTolerantTestCase(HTMLParserStrictTestCase):
self._run_check("<a foo='>'", [('data', "<a foo='>'")])
self._run_check("<a foo='>", [('data', "<a foo='>")])
def test_slashes_in_starttag(self):
self._run_check('<a foo="var"/>', [('startendtag', 'a', [('foo', 'var')])])
html = ('<img width=902 height=250px '
'src="/sites/default/files/images/homepage/foo.jpg" '
'/*what am I doing here*/ />')
expected = [(
'startendtag', 'img',
[('width', '902'), ('height', '250px'),
('src', '/sites/default/files/images/homepage/foo.jpg'),
('*what', None), ('am', None), ('i', None),
('doing', None), ('here*', None)]
)]
self._run_check(html, expected)
html = ('<a / /foo/ / /=/ / /bar/ / />'
'<a / /foo/ / /=/ / /bar/ / >')
expected = [
('startendtag', 'a', [('foo', None), ('=', None), ('bar', None)]),
('starttag', 'a', [('foo', None), ('=', None), ('bar', None)])
]
self._run_check(html, expected)
def test_declaration_junk_chars(self):
self._run_check("<!DOCTYPE foo $ >", [('decl', 'DOCTYPE foo $ ')])