def unquote(s):
s = _unquote(s)
# PY3 always returns unicode. PY2 seems to always return what you give it,
# which differs from quote's behavior. Just to be safe, make sure it is
# unicode before we return.
if isinstance(s, bytes):
s = s.decode('utf-8')
return s
def parse_options_header(value, multiple=False):
"""Parse a ``Content-Type`` like header into a tuple with the content
type and the options:
>>> parse_options_header('text/html; charset=utf8')
('text/html', {'charset': 'utf8'})
This should not be used to parse ``Cache-Control`` like headers that use
a slightly different format. For these headers use the
:func:`parse_dict_header` function.
.. versionadded:: 0.5
:param value: the header to parse.
:param multiple: Whether try to parse and return multiple MIME types
:return: (mimetype, options) or (mimetype, options, mimetype, options, …)
if multiple=True
"""
if not value:
return '', {}
result = []
value = "," + value.replace("\n", ",")
while value:
match = _option_header_start_mime_type.match(value)
if not match:
break
result.append(match.group(1)) # mimetype
options = {}
# Parse options
rest = match.group(2)
while rest:
optmatch = _option_header_piece_re.match(rest)
if not optmatch:
break
option, encoding, _, option_value = optmatch.groups()
option = unquote_header_value(option)
if option_value is not None:
option_value = unquote_header_value(
option_value,
option == 'filename')
if encoding is not None:
option_value = _unquote(option_value).decode(encoding)
options[option] = option_value
rest = rest[optmatch.end():]
result.append(options)
if multiple is False:
return tuple(result)
value = rest
return tuple(result) if result else ('', {})
def parse_options_header(value, multiple=False):
"""Parse a ``Content-Type`` like header into a tuple with the content
type and the options:
>>> parse_options_header('text/html; charset=utf8')
('text/html', {'charset': 'utf8'})
This should not be used to parse ``Cache-Control`` like headers that use
a slightly different format. For these headers use the
:func:`parse_dict_header` function.
.. versionadded:: 0.5
:param value: the header to parse.
:param multiple: Whether try to parse and return multiple MIME types
:return: (mimetype, options) or (mimetype, options, mimetype, options, …)
if multiple=True
"""
if not value:
return '', {}
result = []
value = "," + value.replace("\n", ",")
while value:
match = _option_header_start_mime_type.match(value)
if not match:
break
result.append(match.group(1)) # mimetype
options = {}
# Parse options
rest = match.group(2)
while rest:
optmatch = _option_header_piece_re.match(rest)
if not optmatch:
break
option, encoding, _, option_value = optmatch.groups()
option = unquote_header_value(option)
if option_value is not None:
option_value = unquote_header_value(
option_value,
option == 'filename')
if encoding is not None:
option_value = _unquote(option_value).decode(encoding)
options[option] = option_value
rest = rest[optmatch.end():]
result.append(options)
if multiple is False:
return tuple(result)
value = rest
return tuple(result) if result else ('', {})
def create_uri(self, redirect_uri, state):
description = _unquote(self.description)
# See:
# http://openid.net/specs/openid-connect-core-1_0.html#ImplicitAuthError
hash_or_question = "#" if self.grant_type == "implicit" else "?"
uri = "{0}{1}error={2}&error_description={3}".format(
redirect_uri, hash_or_question, self.error, description)
# Add state if present.
uri = uri + ("&state={0}".format(state) if state else "")
return uri
def _extract_credentials(url):
"""Extract authentication (user name and password) credentials from the
given URL.
>>> extract_credentials('http://*****:*****@localhost:5984/_config/')
('http://*****:*****@localhost:5984/_config/')
('http://*****:*****@example.com', 'secret'))
"""
parts = urlsplit(url)
netloc = parts[1]
if '@' in netloc:
creds, netloc = netloc.split('@')
credentials = tuple(_unquote(i) for i in creds.split(':'))
parts = list(parts)
parts[1] = netloc
else:
credentials = None
return urlunsplit(parts), credentials
def _extract_credentials(url):
"""
Extract authentication (user name and password) credentials from the
given URL.
>>> extract_credentials('http://*****:*****@localhost:5984/_config/')
('http://*****:*****@localhost:5984/_config/')
('http://*****:*****@example.com', 'secret'))
"""
parts = urlsplit(url)
netloc = parts[1]
if '@' in netloc:
creds, netloc = netloc.split('@')
credentials = tuple(_unquote(i) for i in creds.split(':'))
parts = list(parts)
parts[1] = netloc
else:
credentials = None
return urlunsplit(parts), credentials
def parse_options_header(value, multiple=False):
"""Parse a ``Content-Type`` like header into a tuple with the content
type and the options:
>>> parse_options_header('text/html; charset=utf8')
('text/html', {'charset': 'utf8'})
This should not be used to parse ``Cache-Control`` like headers that use
a slightly different format. For these headers use the
:func:`parse_dict_header` function.
.. versionchanged:: 0.15
:rfc:`2231` parameter continuations are handled.
.. versionadded:: 0.5
:param value: the header to parse.
:param multiple: Whether try to parse and return multiple MIME types
:return: (mimetype, options) or (mimetype, options, mimetype, options, …)
if multiple=True
"""
if not value:
return "", {}
result = []
value = "," + value.replace("\n", ",")
while value:
match = _option_header_start_mime_type.match(value)
if not match:
break
result.append(match.group(1)) # mimetype
options = {}
# Parse options
rest = match.group(2)
continued_encoding = None
while rest:
optmatch = _option_header_piece_re.match(rest)
if not optmatch:
break
option, count, encoding, language, option_value = optmatch.groups()
# Continuations don't have to supply the encoding after the
# first line. If we're in a continuation, track the current
# encoding to use for subsequent lines. Reset it when the
# continuation ends.
if not count:
continued_encoding = None
else:
if not encoding:
encoding = continued_encoding
continued_encoding = encoding
option = unquote_header_value(option)
if option_value is not None:
option_value = unquote_header_value(option_value, option == "filename")
if encoding is not None:
option_value = _unquote(option_value).decode(encoding)
if count:
# Continuations append to the existing value. For
# simplicity, this ignores the possibility of
# out-of-order indices, which shouldn't happen anyway.
options[option] = options.get(option, "") + option_value
else:
options[option] = option_value
rest = rest[optmatch.end() :]
result.append(options)
if multiple is False:
return tuple(result)
value = rest
return tuple(result) if result else ("", {})
def run_viewvc(self):
"""Run ViewVC to field a single request."""
### Much of this is adapter from Python's standard library
### module CGIHTTPServer.
# Is this request even aimed at ViewVC? If not, complain.
if not self.is_viewvc():
raise NotViewVCLocationException()
# If htpasswd authentication is enabled, try to authenticate the user.
self.username = None
if options.htpasswd_file:
authn = self.headers.get('authorization')
if not authn:
raise AuthenticationException()
try:
kind, data = authn.split(' ', 1)
if kind == 'Basic':
data = base64.b64decode(data)
username, password = data.split(':', 1)
except:
raise AuthenticationException()
if not self.validate_password(options.htpasswd_file, username,
password):
raise AuthenticationException()
self.username = username
# Setup the environment in preparation of executing ViewVC's core code.
env = os.environ
scriptname = options.script_alias and '/' + options.script_alias or ''
viewvc_url = self.server.url[:-1] + scriptname
rest = self.path[len(scriptname):]
i = rest.rfind('?')
if i >= 0:
rest, query = rest[:i], rest[i + 1:]
else:
query = ''
# Since we're going to modify the env in the parent, provide empty
# values to override previously set values
for k in env.keys():
if k[:5] == 'HTTP_':
del env[k]
for k in ('QUERY_STRING', 'REMOTE_HOST', 'CONTENT_LENGTH',
'HTTP_USER_AGENT', 'HTTP_COOKIE'):
if k in env:
env[k] = ""
# XXX Much of the following could be prepared ahead of time!
env['SERVER_SOFTWARE'] = self.version_string()
env['SERVER_NAME'] = self.server.server_name
env['GATEWAY_INTERFACE'] = 'CGI/1.1'
env['SERVER_PROTOCOL'] = self.protocol_version
env['SERVER_PORT'] = str(self.server.server_port)
env['REQUEST_METHOD'] = self.command
uqrest = _unquote(rest)
env['PATH_INFO'] = uqrest
env['SCRIPT_NAME'] = scriptname
if query:
env['QUERY_STRING'] = query
env['HTTP_HOST'] = self.server.address[0]
host = self.address_string()
if host != self.client_address[0]:
env['REMOTE_HOST'] = host
env['REMOTE_ADDR'] = self.client_address[0]
if self.username:
env['REMOTE_USER'] = self.username
env['CONTENT_TYPE'] = self.headers.get_content_type()
length = self.headers.get('content-length', None)
if length:
env['CONTENT_LENGTH'] = length
accept = []
for line in self.headers.getallmatchingheaders('accept'):
if line[:1] in string.whitespace:
accept.append(line.strip())
else:
accept = accept + line[7:].split(',')
env['HTTP_ACCEPT'] = ','.join(accept)
ua = self.headers.get('user-agent', None)
if ua:
env['HTTP_USER_AGENT'] = ua
modified = self.headers.get('if-modified-since', None)
if modified:
env['HTTP_IF_MODIFIED_SINCE'] = modified
etag = self.headers.get('if-none-match', None)
if etag:
env['HTTP_IF_NONE_MATCH'] = etag
# AUTH_TYPE
# REMOTE_IDENT
# XXX Other HTTP_* headers
try:
try:
viewvc.main(StandaloneServer(self), cfg)
finally:
if not self.wfile.closed:
self.wfile.flush()
except SystemExit as status:
self.log_error("ViewVC exit status %s", str(status))
else:
self.log_error("ViewVC exited ok")
def unquote(s):
if isinstance(s, bytes):
s = s.decode("ascii")
quoted = _unquote(s)
return quoted.encode("ascii")
def unquote(s):
return to_unicode(_unquote(s))
def parse_options_header(value, multiple=False):
"""Parse a ``Content-Type`` like header into a tuple with the content
type and the options:
>>> parse_options_header('text/html; charset=utf8')
('text/html', {'charset': 'utf8'})
This should not be used to parse ``Cache-Control`` like headers that use
a slightly different format. For these headers use the
:func:`parse_dict_header` function.
.. versionchanged:: 0.15
:rfc:`2231` parameter continuations are handled.
.. versionadded:: 0.5
:param value: the header to parse.
:param multiple: Whether try to parse and return multiple MIME types
:return: (mimetype, options) or (mimetype, options, mimetype, options, …)
if multiple=True
"""
if not value:
return "", {}
result = []
value = "," + value.replace("\n", ",")
while value:
match = _option_header_start_mime_type.match(value)
if not match:
break
result.append(match.group(1)) # mimetype
options = {}
# Parse options
rest = match.group(2)
continued_encoding = None
while rest:
optmatch = _option_header_piece_re.match(rest)
if not optmatch:
break
option, count, encoding, language, option_value = optmatch.groups()
# Continuations don't have to supply the encoding after the
# first line. If we're in a continuation, track the current
# encoding to use for subsequent lines. Reset it when the
# continuation ends.
if not count:
continued_encoding = None
else:
if not encoding:
encoding = continued_encoding
continued_encoding = encoding
option = unquote_header_value(option)
if option_value is not None:
option_value = unquote_header_value(option_value, option == "filename")
if encoding is not None:
option_value = _unquote(option_value).decode(encoding)
if count:
# Continuations append to the existing value. For
# simplicity, this ignores the possibility of
# out-of-order indices, which shouldn't happen anyway.
options[option] = options.get(option, "") + option_value
else:
options[option] = option_value
rest = rest[optmatch.end() :]
result.append(options)
if multiple is False:
return tuple(result)
value = rest
return tuple(result) if result else ("", {})
def uri_path(uri: str, unquote: bool = False) -> PurePosixPath:
raw = urlsplit(uri).path
path = _unquote(raw) if unquote else raw
return PurePosixPath(path)
# coding=utf-8
# Copyright 2013 Elsie Powell - embolalia.com
# Licensed under the Eiffel Forum License 2.
from sopel import web, tools
from sopel.config.types import StaticSection, ValidatedAttribute
from sopel.module import NOLIMIT, commands, example, rule
import json
import re
import sys
if sys.version_info.major < 3:
from urllib.parse import unquote as _unquote
unquote = lambda s: _unquote(s.encode('utf-8')).decode('utf-8')
else:
from urllib.parse import unquote
REDIRECT = re.compile(r'^REDIRECT (.*)')
class WikipediaSection(StaticSection):
default_lang = ValidatedAttribute('default_lang', default='en')
"""The default language to find articles from."""
lang_per_channel = ValidatedAttribute('lang_per_channel')
def setup(bot):
bot.config.define_section('wikipedia', WikipediaSection)
regex = re.compile('([a-z]+).(wikipedia.org/wiki/)([^ ]+)')
if not bot.memory.contains('url_callbacks'):
def parse_options_header(value, multiple=False):
"""将像``Content-Type``这样的HTTP头部解析成包含内容类型和选项的元组：
>>> parse_options_header('text/html; charset=utf8')
('text/html', {'charset': 'utf8'})
不应该使用这个函数来解析像``Cache-control``这样子稍微使用了不同格式的HTTP头部。对于这
种头部，使用:func:``parse_dict_header``函数。
.. versionchanged:: 0.15
:rfc:`2231` 处理参数连续
.. versionadded:: 0.5
:param value: 要解析的头部。
:param multiple: 尝试解析并且返回多个MIME types
:return: (mimetype, options) or (mimetype, options, mimetype, options, …)
if multiple=True
"""
if not value:
# value == None or value == ""
return "", {}
result = []
# value = 'text/html; charset=utf-8'
value = "," + value.replace("\n", ",") # ',text/html; charset=utf-8'
while value:
match = _option_header_start_mime_type.match(value)
if not match:
break
result.append(match.group(1)) # mimetype, text/html(Content-type)
options = {}
# 解析选项
rest = match.group(2) # ; charset=utf-8
continued_encoding = None
while rest:
optmatch = _option_header_piece_re.match(rest)
if not optmatch:
break
option, count, encoding, language, option_value = optmatch.groups(
) # ('charset, None, None, None, 'utf-8')
# 首行后的连续行不需要提供编码。如果处于连续行，使用当前的编码当作后续行的编码。
# 当连续行结束的时候进行重置。
if not count:
continued_encoding = None
else:
if not encoding:
encoding = continued_encoding
continued_encoding = encoding
option = unquote_header_value(option)
if option_value is not None:
option_value = unquote_header_value(option_value,
option == "filename")
if encoding is not None:
option_value = _unquote(option_value).decode(encoding)
if count:
# 连续添加到正存在的的值，为了简洁性，忽略了不该产生的乱许索引的可能性。
options[option] = options.get(option, "") + option_value
else:
options[option] = option_value
rest = rest[optmatch.end():]
result.append(options)
if multiple is False:
return tuple(result)
value = rest
return tuple(result) if result else ("", {})
def parse_options_header(
value: t.Optional[str], multiple: "te.Literal[None]" = None
) -> t.Tuple[str, t.Dict[str, str]]:
"""Parse a ``Content-Type``-like header into a tuple with the
value and any options:
>>> parse_options_header('text/html; charset=utf8')
('text/html', {'charset': 'utf8'})
This should is not for ``Cache-Control``-like headers, which use a
different format. For those, use :func:`parse_dict_header`.
:param value: The header value to parse.
.. versionchanged:: 2.1
The ``multiple`` parameter is deprecated and will be removed in
Werkzeug 2.2.
.. versionchanged:: 0.15
:rfc:`2231` parameter continuations are handled.
.. versionadded:: 0.5
"""
if multiple is not None:
import warnings
warnings.warn(
"The 'multiple' parameter of 'parse_options_header' is"
" deprecated and will be removed in Werkzeug 2.2.",
DeprecationWarning,
stacklevel=2,
)
if not value:
return "", {}
result: t.List[t.Any] = []
value = "," + value.replace("\n", ",")
while value:
match = _option_header_start_mime_type.match(value)
if not match:
break
result.append(match.group(1)) # mimetype
options: t.Dict[str, str] = {}
# Parse options
rest = match.group(2)
encoding: t.Optional[str]
continued_encoding: t.Optional[str] = None
while rest:
optmatch = _option_header_piece_re.match(rest)
if not optmatch:
break
option, count, encoding, language, option_value = optmatch.groups()
# Continuations don't have to supply the encoding after the
# first line. If we're in a continuation, track the current
# encoding to use for subsequent lines. Reset it when the
# continuation ends.
if not count:
continued_encoding = None
else:
if not encoding:
encoding = continued_encoding
continued_encoding = encoding
option = unquote_header_value(option)
if option_value is not None:
option_value = unquote_header_value(option_value, option == "filename")
if encoding is not None:
option_value = _unquote(option_value).decode(encoding)
if count:
# Continuations append to the existing value. For
# simplicity, this ignores the possibility of
# out-of-order indices, which shouldn't happen anyway.
if option_value is not None:
options[option] = options.get(option, "") + option_value
else:
options[option] = option_value # type: ignore[assignment]
rest = rest[optmatch.end() :]
result.append(options)
if not multiple:
return tuple(result) # type: ignore[return-value]
value = rest
return tuple(result) if result else ("", {}) # type: ignore[return-value]
# coding=utf-8
# Copyright 2013 Elsie Powell - embolalia.com
# Licensed under the Eiffel Forum License 2.
from sopel import web, tools
from sopel.config.types import StaticSection, ValidatedAttribute
from sopel.module import NOLIMIT, commands, example, rule
import json
import re
import sys
if sys.version_info.major < 3:
from urllib.parse import unquote as _unquote
unquote = lambda s: _unquote(s.encode('utf-8')).decode('utf-8')
else:
from urllib.parse import unquote
REDIRECT = re.compile(r'^REDIRECT (.*)')
class WikipediaSection(StaticSection):
default_lang = ValidatedAttribute('default_lang', default='en')
"""The default language to find articles from."""
lang_per_channel = ValidatedAttribute('lang_per_channel')
def setup(bot):
bot.config.define_section('wikipedia', WikipediaSection)
regex = re.compile('([a-z]+).(wikipedia.org/wiki/)([^ ]+)')
if not bot.memory.contains('url_callbacks'):
