How to use bleach - 10 common examples

def clean_html(input):
    # Reaplce html tags from user input, see utils.test for examples

    ok_tags = [u"a", u"img", u"strong", u"b", u"em", u"i", u"u", u"ul", u"li", u"p", u"br",  u"blockquote", u"code"]
    ok_attributes = {u"a": [u"href", u"rel"], u"img": [u"src", u"alt", u"title"]}
    # all other tags: replace with the content of the tag

    # If input contains link in the format:  then convert it to < http:// >
    # This is because otherwise the library recognizes it as a tag and breaks the link.
    input = re.sub("\<(http\S+?)\>", r'< \1 >', input)

    cleaner = bleach.Cleaner(
            filters=[
                EmptyLinkFilter,
                partial(bleach.linkifier.LinkifyFilter, callbacks=[nofollow]),
                ],
            attributes=ok_attributes,
            tags=ok_tags,
            strip=True)
    output = cleaner.clean(input)
    return output

def clean_html(input):
    # Reaplce html tags from user input, see utils.test for examples

    ok_tags = [u"a", u"img", u"strong", u"b", u"em", u"i", u"u", u"ul", u"li", u"p", u"br",  u"blockquote", u"code"]
    ok_attributes = {u"a": [u"href", u"rel"], u"img": [u"src", u"alt", u"title"]}
    # all other tags: replace with the content of the tag

    # If input contains link in the format:  then convert it to < http:// >
    # This is because otherwise the library recognizes it as a tag and breaks the link.
    input = re.sub("\<(http\S+?)\>", r'< \1 >', input)

    cleaner = bleach.Cleaner(
            filters=[
                EmptyLinkFilter,
                partial(bleach.linkifier.LinkifyFilter, callbacks=[nofollow]),
                ],
            attributes=ok_attributes,
            tags=ok_tags,
            strip=True)
    output = cleaner.clean(input)
    return output

def test_no_href_links(self):
        s = '<a name="anchor">x</a>'
        assert linkify(s) == s

def test_ports(data, expected_data):
    """URLs can contain port numbers."""
    out = '<a rel="nofollow" href="{0}">{0}</a>{1}'
    assert linkify(data) == out.format(*expected_data)

def test_mangle_text():
    """We can muck with the inner text of a link."""

    def ft(attrs, new=False):
        attrs['_text'] = 'bar'
        return attrs

    assert (
        linkify('http://ex.mp <a href="http://ex.mp/foo">foo</a>', callbacks=[ft]) ==
        '<a href="http://ex.mp">bar</a> <a href="http://ex.mp/foo">bar</a>'
    )

def test_has_nofollow_already(self):
        attrs = {
            (None, 'href'): 'http://example.com',
            (None, 'rel'): 'nofollow',
        }
        assert nofollow(attrs) == attrs

def test_mailto(self):
        attrs = {(None, 'href'): 'mailto:joe@example.com'}
        assert nofollow(attrs) == attrs

def test_other_rel(self):
        attrs = {
            (None, 'href'): 'http://example.com',
            (None, 'rel'): 'next',
        }
        assert (
            nofollow(attrs) ==
            {(None, 'href'): 'http://example.com', (None, 'rel'): 'next nofollow'}
        )

def test_bleach_html_parser(parser_args, data, expected):
    args = {
        'tags': None,
        'strip': True,
        'consume_entities': True
    }
    args.update(parser_args)

    # Build a parser, walker, and serializer just like we do in clean()
    parser = html5lib_shim.BleachHTMLParser(**args)
    walker = html5lib_shim.getTreeWalker('etree')
    serializer = html5lib_shim.BleachHTMLSerializer(
        quote_attr_values='always',
        omit_optional_tags=False,
        escape_lt_in_attrs=True,
        resolve_entities=False,
        sanitize=False,
        alphabetical_attributes=False,
    )

    # Parse, walk, and then serialize the output
    dom = parser.parseFragment(data)
    serialized = serializer.render(walker(dom))

    assert serialized == expected

def test_convert_entities(data, expected):
    assert html5lib_shim.convert_entities(data) == expected