def linkify(self, text):
"""Linkify specified text
:arg str text: the text to add links to
:returns: linkified text as unicode
:raises TypeError: if ``text`` is not a text type
"""
if not isinstance(text, six.string_types):
raise TypeError('argument must be of text type')
text = force_unicode(text)
if not text:
return ''
dom = self.parser.parseFragment(text)
filtered = LinkifyFilter(
source=self.walker(dom),
callbacks=self.callbacks,
skip_tags=self.skip_tags,
parse_email=self.parse_email,
url_re=self.url_re,
email_re=self.email_re,
)
return self.serializer.render(filtered)
def linkify(self, text):
"""Linkify specified text
:arg str text: the text to add links to
:returns: linkified text as unicode
:raises TypeError: if ``text`` is not a text type
"""
if not isinstance(text, six.string_types):
raise TypeError('argument must be of text type')
text = force_unicode(text)
if not text:
return u''
dom = self.parser.parseFragment(text)
filtered = LinkifyFilter(
source=self.walker(dom),
callbacks=self.callbacks,
skip_tags=self.skip_tags,
parse_email=self.parse_email,
url_re=self.url_re,
email_re=self.email_re,
)
return self.serializer.render(filtered)
def clean(self, text, allowed_domains=False):
if not allowed_domains:
allowed_domains = []
if not isinstance(text, str):
message = "argument cannot be of '{name}' type, must be of text type".format(
name=text.__class__.__name__)
raise TypeError(message)
if not text:
return ""
text = force_unicode(text)
dom = self.parser.parseFragment(text)
filtered = BleachSanitizerFilter(
source=self.walker(dom),
# Bleach-sanitizer-specific things
attributes=self.attributes,
strip_disallowed_elements=self.strip,
strip_html_comments=self.strip_comments,
# html5lib-sanitizer things
allowed_elements=self.tags,
allowed_css_properties=self.styles,
allowed_protocols=self.protocols,
allowed_svg_properties=[],
)
# Apply any filters after the BleachSanitizerFilter
for filter_class in self.filters:
fc = filter_class(source=filtered)
filtered = fc.__iter__(allowed_domains=allowed_domains)
return self.serializer.render(filtered)
def handle_email_addresses(self, src_iter):
"""Handle email addresses in character tokens"""
for token in src_iter:
if token['type'] == 'Characters':
text = token['data']
new_tokens = []
end = 0
# For each email address we find in the text
for match in self.email_re.finditer(text):
if match.start() > end:
new_tokens.append(
{'type': 'Characters', 'data': text[end:match.start()]}
)
# Run attributes through the callbacks to see what we
# should do with this match
attrs = {
(None, 'href'): 'mailto:%s' % match.group(0),
'_text': match.group(0)
}
attrs = self.apply_callbacks(attrs, True)
if attrs is None:
# Just add the text--but not as a link
new_tokens.append(
{'type': 'Characters', 'data': match.group(0)}
)
else:
# Add an "a" tag for the new link
_text = attrs.pop('_text', '')
attrs = alphabetize_attributes(attrs)
new_tokens.extend([
{'type': 'StartTag', 'name': 'a', 'data': attrs},
{'type': 'Characters', 'data': force_unicode(_text)},
{'type': 'EndTag', 'name': 'a'}
])
end = match.end()
if new_tokens:
# Yield the adjusted set of tokens and then continue
# through the loop
if end < len(text):
new_tokens.append({'type': 'Characters', 'data': text[end:]})
for new_token in new_tokens:
yield new_token
continue
yield token
def handle_email_addresses(self, src_iter):
"""Handle email addresses in character tokens"""
for token in src_iter:
if token['type'] == 'Characters':
text = token['data']
new_tokens = []
end = 0
# For each email address we find in the text
for match in self.email_re.finditer(text):
if match.start() > end:
new_tokens.append(
{u'type': u'Characters', u'data': text[end:match.start()]}
)
# Run attributes through the callbacks to see what we
# should do with this match
attrs = {
(None, u'href'): u'mailto:%s' % match.group(0),
u'_text': match.group(0)
}
attrs = self.apply_callbacks(attrs, True)
if attrs is None:
# Just add the text--but not as a link
new_tokens.append(
{u'type': u'Characters', u'data': match.group(0)}
)
else:
# Add an "a" tag for the new link
_text = attrs.pop(u'_text', '')
attrs = alphabetize_attributes(attrs)
new_tokens.extend([
{u'type': u'StartTag', u'name': u'a', u'data': attrs},
{u'type': u'Characters', u'data': force_unicode(_text)},
{u'type': u'EndTag', u'name': 'a'}
])
end = match.end()
if new_tokens:
# Yield the adjusted set of tokens and then continue
# through the loop
if end < len(text):
new_tokens.append({u'type': u'Characters', u'data': text[end:]})
for new_token in new_tokens:
yield new_token
continue
yield token
def handle_a_tag(self, token_buffer):
"""Handle the "a" tag
This could adjust the link or drop it altogether depending on what the
callbacks return.
This yields the new set of tokens.
"""
a_token = token_buffer[0]
if a_token['data']:
attrs = a_token['data']
else:
attrs = {}
text = self.extract_character_data(token_buffer)
attrs['_text'] = text
attrs = self.apply_callbacks(attrs, False)
if attrs is None:
# We're dropping the "a" tag and everything else and replacing
# it with character data. So emit that token.
yield {'type': 'Characters', 'data': text}
else:
new_text = attrs.pop('_text', '')
a_token['data'] = alphabetize_attributes(attrs)
if text == new_text:
# The callbacks didn't change the text, so we yield the new "a"
# token, then whatever else was there, then the end "a" token
yield a_token
for mem in token_buffer[1:]:
yield mem
else:
# If the callbacks changed the text, then we're going to drop
# all the tokens between the start and end "a" tags and replace
# it with the new text
yield a_token
yield {'type': 'Characters', 'data': force_unicode(new_text)}
yield token_buffer[-1]
def handle_a_tag(self, token_buffer):
"""Handle the "a" tag
This could adjust the link or drop it altogether depending on what the
callbacks return.
This yields the new set of tokens.
"""
a_token = token_buffer[0]
if a_token['data']:
attrs = a_token['data']
else:
attrs = {}
text = self.extract_character_data(token_buffer)
attrs['_text'] = text
attrs = self.apply_callbacks(attrs, False)
if attrs is None:
# We're dropping the "a" tag and everything else and replacing
# it with character data. So emit that token.
yield {'type': 'Characters', 'data': text}
else:
new_text = attrs.pop('_text', '')
a_token['data'] = alphabetize_attributes(attrs)
if text == new_text:
# The callbacks didn't change the text, so we yield the new "a"
# token, then whatever else was there, then the end "a" token
yield a_token
for mem in token_buffer[1:]:
yield mem
else:
# If the callbacks changed the text, then we're going to drop
# all the tokens between the start and end "a" tags and replace
# it with the new text
yield a_token
yield {'type': 'Characters', 'data': force_unicode(new_text)}
yield token_buffer[-1]
def clean(self, text):
"""Cleans text and returns sanitized result as unicode
:arg str text: text to be cleaned
:returns: sanitized text as unicode
:raises TypeError: if ``text`` is not a text type
"""
if not isinstance(text, six.string_types):
message = "argument cannot be of '{name}' type, must be of text type".format(
name=text.__class__.__name__)
raise TypeError(message)
if not text:
return ''
text = force_unicode(text)
dom = self.parser.parseFragment(text)
filtered = BleachSanitizerFilter(
source=self.walker(dom),
# Bleach-sanitizer-specific things
attributes=self.attributes,
strip_disallowed_elements=self.strip,
strip_html_comments=self.strip_comments,
# html5lib-sanitizer things
allowed_elements=self.tags,
allowed_css_properties=self.styles,
allowed_protocols=self.protocols,
allowed_svg_properties=[],
)
# Apply any filters after the BleachSanitizerFilter
for filter_class in self.filters:
filtered = filter_class(source=filtered)
return self.serializer.render(filtered)
def clean(self, text):
"""Cleans text and returns sanitized result as unicode
:arg str text: text to be cleaned
:returns: sanitized text as unicode
:raises TypeError: if ``text`` is not a text type
"""
if not isinstance(text, six.string_types):
message = "argument cannot be of '{name}' type, must be of text type".format(
name=text.__class__.__name__)
raise TypeError(message)
if not text:
return u''
text = force_unicode(text)
dom = self.parser.parseFragment(text)
filtered = BleachSanitizerFilter(
source=self.walker(dom),
# Bleach-sanitizer-specific things
attributes=self.attributes,
strip_disallowed_elements=self.strip,
strip_html_comments=self.strip_comments,
# html5lib-sanitizer things
allowed_elements=self.tags,
allowed_css_properties=self.styles,
allowed_protocols=self.protocols,
allowed_svg_properties=[],
)
# Apply any filters after the BleachSanitizerFilter
for filter_class in self.filters:
filtered = filter_class(source=filtered)
return self.serializer.render(filtered)
def handle_links(self, src_iter):
"""Handle links in character tokens"""
in_a = False # happens, if parse_email=True and if a mail was found
for token in src_iter:
if in_a:
if token['type'] == 'EndTag' and token['name'] == 'a':
in_a = False
yield token
continue
elif token['type'] == 'StartTag' and token['name'] == 'a':
in_a = True
yield token
continue
if token['type'] == 'Characters':
text = token['data']
new_tokens = []
end = 0
for match in self.url_re.finditer(text):
if match.start() > end:
new_tokens.append(
{'type': 'Characters', 'data': text[end:match.start()]}
)
url = match.group(0)
prefix = suffix = ''
# Sometimes we pick up too much in the url match, so look for
# bits we should drop and remove them from the match
url, prefix, suffix = self.strip_non_url_bits(url)
# If there's no protocol, add one
if PROTO_RE.search(url):
href = url
else:
href = 'http://%s' % url
attrs = {
(None, 'href'): href,
'_text': url
}
attrs = self.apply_callbacks(attrs, True)
if attrs is None:
# Just add the text
new_tokens.append(
{'type': 'Characters', 'data': prefix + url + suffix}
)
else:
# Add the "a" tag!
if prefix:
new_tokens.append(
{'type': 'Characters', 'data': prefix}
)
_text = attrs.pop('_text', '')
attrs = alphabetize_attributes(attrs)
new_tokens.extend([
{'type': 'StartTag', 'name': 'a', 'data': attrs},
{'type': 'Characters', 'data': force_unicode(_text)},
{'type': 'EndTag', 'name': 'a'},
])
if suffix:
new_tokens.append(
{'type': 'Characters', 'data': suffix}
)
end = match.end()
if new_tokens:
# Yield the adjusted set of tokens and then continue
# through the loop
if end < len(text):
new_tokens.append({'type': 'Characters', 'data': text[end:]})
for new_token in new_tokens:
yield new_token
continue
yield token
def handle_links(self, src_iter):
"""Handle links in character tokens"""
in_a = False # happens, if parse_email=True and if a mail was found
for token in src_iter:
if in_a:
if token["type"] == "EndTag" and token["name"] == "a":
in_a = False
yield token
continue
elif token["type"] == "StartTag" and token["name"] == "a":
in_a = True
yield token
continue
if token["type"] == "Characters":
text = token["data"]
new_tokens = []
end = 0
for match in self.url_re.finditer(text):
if match.start() > end:
new_tokens.append({
"type": "Characters",
"data": text[end:match.start()]
})
url = match.group(0)
prefix = suffix = ""
# Sometimes we pick up too much in the url match, so look for
# bits we should drop and remove them from the match
url, prefix, suffix = self.strip_non_url_bits(url)
# If there's no protocol, add one
if PROTO_RE.search(url):
href = url
else:
href = "http://%s" % url
attrs = {(None, "href"): href, "_text": url}
attrs = self.apply_callbacks(attrs, True)
if attrs is None:
# Just add the text
new_tokens.append({
"type": "Characters",
"data": prefix + url + suffix
})
else:
# Add the "a" tag!
if prefix:
new_tokens.append({
"type": "Characters",
"data": prefix
})
_text = attrs.pop("_text", "")
attrs = alphabetize_attributes(attrs)
new_tokens.extend([
{
"type": "StartTag",
"name": "a",
"data": attrs
},
{
"type": "Characters",
"data": force_unicode(_text)
},
{
"type": "EndTag",
"name": "a"
},
])
if suffix:
new_tokens.append({
"type": "Characters",
"data": suffix
})
end = match.end()
if new_tokens:
# Yield the adjusted set of tokens and then continue
# through the loop
if end < len(text):
new_tokens.append({
"type": "Characters",
"data": text[end:]
})
for new_token in new_tokens:
yield new_token
continue
yield token
def handle_email_addresses(self, src_iter):
"""Handle email addresses in character tokens"""
for token in src_iter:
if token["type"] == "Characters":
text = token["data"]
new_tokens = []
end = 0
# For each email address we find in the text
for match in self.email_re.finditer(text):
if match.start() > end:
new_tokens.append({
"type": "Characters",
"data": text[end:match.start()]
})
# Run attributes through the callbacks to see what we
# should do with this match
attrs = {
(None, "href"): "mailto:%s" % match.group(0),
"_text": match.group(0),
}
attrs = self.apply_callbacks(attrs, True)
if attrs is None:
# Just add the text--but not as a link
new_tokens.append({
"type": "Characters",
"data": match.group(0)
})
else:
# Add an "a" tag for the new link
_text = attrs.pop("_text", "")
attrs = alphabetize_attributes(attrs)
new_tokens.extend([
{
"type": "StartTag",
"name": "a",
"data": attrs
},
{
"type": "Characters",
"data": force_unicode(_text)
},
{
"type": "EndTag",
"name": "a"
},
])
end = match.end()
if new_tokens:
# Yield the adjusted set of tokens and then continue
# through the loop
if end < len(text):
new_tokens.append({
"type": "Characters",
"data": text[end:]
})
for new_token in new_tokens:
yield new_token
continue
yield token
def handle_links(self, src_iter):
"""Handle links in character tokens"""
in_a = False # happens, if parse_email=True and if a mail was found
for token in src_iter:
if in_a:
if token['type'] == 'EndTag' and token['name'] == 'a':
in_a = False
yield token
continue
elif token['type'] == 'StartTag' and token['name'] == 'a':
in_a = True
yield token
continue
if token['type'] == 'Characters':
text = token['data']
new_tokens = []
end = 0
for match in self.url_re.finditer(text):
if match.start() > end:
new_tokens.append(
{u'type': u'Characters', u'data': text[end:match.start()]}
)
url = match.group(0)
prefix = suffix = ''
# Sometimes we pick up too much in the url match, so look for
# bits we should drop and remove them from the match
url, prefix, suffix = self.strip_non_url_bits(url)
# If there's no protocol, add one
if PROTO_RE.search(url):
href = url
else:
href = u'http://%s' % url
attrs = {
(None, u'href'): href,
u'_text': url
}
attrs = self.apply_callbacks(attrs, True)
if attrs is None:
# Just add the text
new_tokens.append(
{u'type': u'Characters', u'data': prefix + url + suffix}
)
else:
# Add the "a" tag!
if prefix:
new_tokens.append(
{u'type': u'Characters', u'data': prefix}
)
_text = attrs.pop(u'_text', '')
attrs = alphabetize_attributes(attrs)
new_tokens.extend([
{u'type': u'StartTag', u'name': u'a', u'data': attrs},
{u'type': u'Characters', u'data': force_unicode(_text)},
{u'type': u'EndTag', u'name': 'a'},
])
if suffix:
new_tokens.append(
{u'type': u'Characters', u'data': suffix}
)
end = match.end()
if new_tokens:
# Yield the adjusted set of tokens and then continue
# through the loop
if end < len(text):
new_tokens.append({u'type': u'Characters', u'data': text[end:]})
for new_token in new_tokens:
yield new_token
continue
yield token