html: add nofollow/noopener to links

"nofollow" is a deterrent for spammers: they cannot put links and hope
to increase their SEO when all these links have the nofollow
relationship.

"noopener" is a security for links opening a new window. They ensure
the target cannot control us.

Fix #373
This commit is contained in:
Vincent Bernat 2018-04-21 10:25:12 +02:00
parent 07ce742b77
commit 8d8f9c8c59
2 changed files with 7 additions and 2 deletions

View File

@ -65,7 +65,7 @@ class TestHTML(unittest.TestCase):
examples = [
('Look: <img src="..." />', 'Look: '),
('<a href="http://example.org/">Ha</a>',
'<a href="http://example.org/">Ha</a>'),
'<a href="http://example.org/" rel="nofollow noopener">Ha</a>'),
('<a href="sms:+1234567890">Ha</a>', '<a>Ha</a>'),
('<p style="visibility: hidden;">Test</p>', '<p>Test</p>'),
('<script>alert("Onoe")</script>', 'alert("Onoe")')]
@ -93,4 +93,4 @@ class TestHTML(unittest.TestCase):
})
renderer = html.Markup(conf.section("markup")).render
self.assertEqual(renderer("http://example.org/ and sms:+1234567890"),
'<p><a href="http://example.org/">http://example.org/</a> and sms:+1234567890</p>')
'<p><a href="http://example.org/" rel="nofollow noopener">http://example.org/</a> and sms:+1234567890</p>')

View File

@ -50,6 +50,11 @@ def sanitize(tokenizer, document):
if HTML5LIB_VERSION > HTML5LIB_SIMPLETREE:
builder = "etree"
for link in domtree.findall(".//{http://www.w3.org/1999/xhtml}a"):
if link.get('href', None):
link.set("rel", "nofollow noopener")
else:
builder = "simpletree"