[utils] Support attributes with no values in get_elements_by_attribute()

This commit is contained in:
Yen Chi Hsuan 2017-07-05 23:23:35 +08:00
parent b6c9fe4162
commit 609ff8ca19
2 changed files with 8 additions and 2 deletions

View File

@ -1228,6 +1228,12 @@ def test_get_element_by_attribute(self):
self.assertEqual(get_element_by_attribute('class', 'foo', html), None)
self.assertEqual(get_element_by_attribute('class', 'no-such-foo', html), None)
html = '''
<div itemprop="author" itemscope>foo</div>
'''
self.assertEqual(get_element_by_attribute('itemprop', 'author', html), 'foo')
def test_get_elements_by_class(self):
html = '''
<span class="foo bar">nice</span><span class="foo bar">also nice</span>

View File

@ -365,9 +365,9 @@ def get_elements_by_attribute(attribute, value, html, escape_value=True):
retlist = []
for m in re.finditer(r'''(?xs)
<([a-zA-Z0-9:._-]+)
(?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
(?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
\s+%s=['"]?%s['"]?
(?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
(?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
\s*>
(?P<content>.*?)
</\1>