thirdparty（第三方库或者脚本文件）
ansistrm子目录
1、ansistrm.py（颜色、format等设置）
#
# Copyright (C) 2010-2012 Vinay Sajip. All rights reserved. Licensed under the new BSD license.
#
import logging
import os
import re
from lib.core.convert import stdoutencode
class ColorizingStreamHandler(logging.StreamHandler):
# color names to indices
color_map = {
'black': 0,
'red': 1,
'green': 2,
'yellow': 3,
'blue': 4,
'magenta': 5,
'cyan': 6,
'white': 7,
}
# levels to (background, foreground, bold/intense)
if os.name == 'nt':
level_map = {
logging.DEBUG: (None, 'blue', False),
logging.INFO: (None, 'green', False),
logging.WARNING: (None, 'yellow', False),
logging.ERROR: (None, 'red', False),
logging.CRITICAL: ('red', 'white', False)
}
else:
level_map = {
logging.DEBUG: (None, 'blue', False),
logging.INFO: (None, 'green', False),
logging.WARNING: (None, 'yellow', False),
logging.ERROR: (None, 'red', False),
logging.CRITICAL: ('red', 'white', False)
}
csi = '\x1b['
reset = '\x1b[0m'
disable_coloring = False
@property
def is_tty(self):
isatty = getattr(self.stream, 'isatty', None)
return isatty and isatty() and not self.disable_coloring
def emit(self, record):
try:
message = stdoutencode(self.format(record))
stream = self.stream
if not self.is_tty:
if message and message[0] == "\r":
message = message[1:]
stream.write(message)
else:
self.output_colorized(message)
stream.write(getattr(self, 'terminator', '\n'))
self.flush()
except (KeyboardInterrupt, SystemExit):
raise
except IOError:
pass
except Exception:
self.handleError(record)
if os.name != 'nt':
def output_colorized(self, message):
self.stream.write(message)
else:
ansi_esc = re.compile(r'\x1b\[((?:\d+)(?:;(?:\d+))*)m')
nt_color_map = {
0: 0x00, # black
1: 0x04, # red
2: 0x02, # green
3: 0x06, # yellow
4: 0x01, # blue
5: 0x05, # magenta
6: 0x03, # cyan
7: 0x07, # white
}
def output_colorized(self, message):
import ctypes
parts = self.ansi_esc.split(message)
write = self.stream.write
h = None
fd = getattr(self.stream, 'fileno', None)
if fd is not None:
fd = fd()
if fd in (1, 2): # stdout or stderr
h = ctypes.windll.kernel32.GetStdHandle(-10 - fd)
while parts:
text = parts.pop(0)
if text:
write(text)
if parts:
params = parts.pop(0)
if h is not None:
params = [int(p) for p in params.split(';')]
color = 0
for p in params:
if 40 <= p <= 47:
color |= self.nt_color_map[p - 40] << 4
elif 30 <= p <= 37:
color |= self.nt_color_map[p - 30]
elif p == 1:
color |= 0x08 # foreground intensity on
elif p == 0: # reset to default color
color = 0x07
else:
pass # error condition ignored
ctypes.windll.kernel32.SetConsoleTextAttribute(h, color)
def colorize(self, message, record):
if record.levelno in self.level_map and self.is_tty:
bg, fg, bold = self.level_map[record.levelno]
params = []
if bg in self.color_map:
params.append(str(self.color_map[bg] + 40))
if fg in self.color_map:
params.append(str(self.color_map[fg] + 30))
if bold:
params.append('1')
if params and message:
if message.lstrip() != message:
prefix = re.search(r"\s+", message).group(0)
message = message[len(prefix):]
else:
prefix = ""
message = "%s%s" % (prefix, ''.join((self.csi, ';'.join(params),
'm', message, self.reset)))
return message
def format(self, record):
message = logging.StreamHandler.format(self, record)
return self.colorize(message, record)
colorama子目录
1.ansi.py（颜色设置）
'''
This module generates ANSI character codes to printing colors to terminals.
See: http://en.wikipedia.org/wiki/ANSI_escape_code
'''
CSI = '\033['
def code_to_chars(code):
return CSI + str(code) + 'm'
class AnsiCodes(object):
def __init__(self, codes):
for name in dir(codes):
if not name.startswith('_'):
value = getattr(codes, name)
setattr(self, name, code_to_chars(value))
class AnsiFore:
BLACK = 30
RED = 31
GREEN = 32
YELLOW = 33
BLUE = 34
MAGENTA = 35
CYAN = 36
WHITE = 37
RESET = 39
class AnsiBack:
BLACK = 40
RED = 41
GREEN = 42
YELLOW = 43
BLUE = 44
MAGENTA = 45
CYAN = 46
WHITE = 47
RESET = 49
class AnsiStyle:
BRIGHT = 1
DIM = 2
NORMAL = 22
RESET_ALL = 0
Fore = AnsiCodes( AnsiFore )
Back = AnsiCodes( AnsiBack )
Style = AnsiCodes( AnsiStyle )
2、ansitowin32.py（window 32位的color参数设置）
import re
import sys
from .ansi import AnsiFore, AnsiBack, AnsiStyle, Style
from .winterm import WinTerm, WinColor, WinStyle
from .win32 import windll
if windll is not None:
winterm = WinTerm()
def is_a_tty(stream):
return hasattr(stream, 'isatty') and stream.isatty()
class StreamWrapper(object):
'''
Wraps a stream (such as stdout), acting as a transparent proxy for all
attribute access apart from method 'write()', which is delegated to our
Converter instance.
'''
def __init__(self, wrapped, converter):
# double-underscore everything to prevent clashes with names of
# attributes on the wrapped stream object.
self.__wrapped = wrapped
self.__convertor = converter
def __getattr__(self, name):
return getattr(self.__wrapped, name)
def write(self, text):
self.__convertor.write(text)
class AnsiToWin32(object):
'''
Implements a 'write()' method which, on Windows, will strip ANSI character
sequences from the text, and if outputting to a tty, will convert them into
win32 function calls.
'''
ANSI_RE = re.compile('\033\[((?:\d|;)*)([a-zA-Z])')
def __init__(self, wrapped, convert=None, strip=None, autoreset=False):
# The wrapped stream (normally sys.stdout or sys.stderr)
self.wrapped = wrapped
# should we reset colors to defaults after every .write()
self.autoreset = autoreset
# create the proxy wrapping our output stream
self.stream = StreamWrapper(wrapped, self)
on_windows = sys.platform.startswith('win')
# should we strip ANSI sequences from our output?
if strip is None:
strip = on_windows
self.strip = strip
# should we should convert ANSI sequences into win32 calls?
if convert is None:
convert = on_windows and is_a_tty(wrapped)
self.convert = convert
# dict of ansi codes to win32 functions and parameters
self.win32_calls = self.get_win32_calls()
# are we wrapping stderr?
self.on_stderr = self.wrapped is sys.stderr
def should_wrap(self):
'''
True if this class is actually needed. If false, then the output
stream will not be affected, nor will win32 calls be issued, so
wrapping stdout is not actually required. This will generally be
False on non-Windows platforms, unless optional functionality like
autoreset has been requested using kwargs to init()
'''
return self.convert or self.strip or self.autoreset
def get_win32_calls(self):
if self.convert and winterm:
return {
AnsiStyle.RESET_ALL: (winterm.reset_all, ),
AnsiStyle.BRIGHT: (winterm.style, WinStyle.BRIGHT),
AnsiStyle.DIM: (winterm.style, WinStyle.NORMAL),
AnsiStyle.NORMAL: (winterm.style, WinStyle.NORMAL),
AnsiFore.BLACK: (winterm.fore, WinColor.BLACK),
AnsiFore.RED: (winterm.fore, WinColor.RED),
AnsiFore.GREEN: (winterm.fore, WinColor.GREEN),
AnsiFore.YELLOW: (winterm.fore, WinColor.YELLOW),
AnsiFore.BLUE: (winterm.fore, WinColor.BLUE),
AnsiFore.MAGENTA: (winterm.fore, WinColor.MAGENTA),
AnsiFore.CYAN: (winterm.fore, WinColor.CYAN),
AnsiFore.WHITE: (winterm.fore, WinColor.GREY),
AnsiFore.RESET: (winterm.fore, ),
AnsiBack.BLACK: (winterm.back, WinColor.BLACK),
AnsiBack.RED: (winterm.back, WinColor.RED),
AnsiBack.GREEN: (winterm.back, WinColor.GREEN),
AnsiBack.YELLOW: (winterm.back, WinColor.YELLOW),
AnsiBack.BLUE: (winterm.back, WinColor.BLUE),
AnsiBack.MAGENTA: (winterm.back, WinColor.MAGENTA),
AnsiBack.CYAN: (winterm.back, WinColor.CYAN),
AnsiBack.WHITE: (winterm.back, WinColor.GREY),
AnsiBack.RESET: (winterm.back, ),
}
def write(self, text):
if self.strip or self.convert:
self.write_and_convert(text)
else:
self.wrapped.write(text)
self.wrapped.flush()
if self.autoreset:
self.reset_all()
def reset_all(self):
if self.convert:
self.call_win32('m', (0,))
elif is_a_tty(self.wrapped):
self.wrapped.write(Style.RESET_ALL)
def write_and_convert(self, text):
'''
Write the given text to our wrapped stream, stripping any ANSI
sequences from the text, and optionally converting them into win32
calls.
'''
cursor = 0
for match in self.ANSI_RE.finditer(text):
start, end = match.span()
self.write_plain_text(text, cursor, start)
self.convert_ansi(*match.groups())
cursor = end
self.write_plain_text(text, cursor, len(text))
def write_plain_text(self, text, start, end):
if start < end:
self.wrapped.write(text[start:end])
self.wrapped.flush()
def convert_ansi(self, paramstring, command):
if self.convert:
params = self.extract_params(paramstring)
self.call_win32(command, params)
def extract_params(self, paramstring):
def split(paramstring):
for p in paramstring.split(';'):
if p != '':
yield int(p)
return tuple(split(paramstring))
def call_win32(self, command, params):
if params == []:
params = [0]
if command == 'm':
for param in params:
if param in self.win32_calls:
func_args = self.win32_calls[param]
func = func_args[0]
args = func_args[1:]
kwargs = dict(on_stderr=self.on_stderr)
func(*args, **kwargs)
elif command in ('H', 'f'): # set cursor position
func = winterm.set_cursor_position
func(params, on_stderr=self.on_stderr)
elif command in ('J'):
func = winterm.erase_data
func(params, on_stderr=self.on_stderr)
elif command == 'A':
if params == () or params == None:
num_rows = 1
else:
num_rows = params[0]
func = winterm.cursor_up
func(num_rows, on_stderr=self.on_stderr)
3、initialise.py(各种初始化)
import atexit
import sys
from .ansitowin32 import AnsiToWin32
orig_stdout = sys.stdout
orig_stderr = sys.stderr
wrapped_stdout = sys.stdout
wrapped_stderr = sys.stderr
atexit_done = False
def reset_all():
AnsiToWin32(orig_stdout).reset_all()
def init(autoreset=False, convert=None, strip=None, wrap=True):
if not wrap and any([autoreset, convert, strip]):
raise ValueError('wrap=False conflicts with any other arg=True')
global wrapped_stdout, wrapped_stderr
sys.stdout = wrapped_stdout = \
wrap_stream(orig_stdout, convert, strip, autoreset, wrap)
sys.stderr = wrapped_stderr = \
wrap_stream(orig_stderr, convert, strip, autoreset, wrap)
global atexit_done
if not atexit_done:
atexit.register(reset_all)
atexit_done = True
def deinit():
sys.stdout = orig_stdout
sys.stderr = orig_stderr
def reinit():
sys.stdout = wrapped_stdout
sys.stderr = wrapped_stdout
def wrap_stream(stream, convert, strip, autoreset, wrap):
if wrap:
wrapper = AnsiToWin32(stream,
convert=convert, strip=strip, autoreset=autoreset)
if wrapper.should_wrap():
stream = wrapper.stream
return stream4、win32.py（windows 32位终端设置）
# from winbase.h
STDOUT = -11
STDERR = -12
try:
from ctypes import windll
except ImportError:
windll = None
SetConsoleTextAttribute = lambda *_: None
else:
from ctypes import (
byref, Structure, c_char, c_short, c_uint32, c_ushort
)
handles = {
STDOUT: windll.kernel32.GetStdHandle(STDOUT),
STDERR: windll.kernel32.GetStdHandle(STDERR),
}
SHORT = c_short
WORD = c_ushort
DWORD = c_uint32
TCHAR = c_char
class COORD(Structure):
"""struct in wincon.h"""
_fields_ = [
('X', SHORT),
('Y', SHORT),
]
class SMALL_RECT(Structure):
"""struct in wincon.h."""
_fields_ = [
("Left", SHORT),
("Top", SHORT),
("Right", SHORT),
("Bottom", SHORT),
]
class CONSOLE_SCREEN_BUFFER_INFO(Structure):
"""struct in wincon.h."""
_fields_ = [
("dwSize", COORD),
("dwCursorPosition", COORD),
("wAttributes", WORD),
("srWindow", SMALL_RECT),
("dwMaximumWindowSize", COORD),
]
def __str__(self):
return '(%d,%d,%d,%d,%d,%d,%d,%d,%d,%d,%d)' % (
self.dwSize.Y, self.dwSize.X
, self.dwCursorPosition.Y, self.dwCursorPosition.X
, self.wAttributes
, self.srWindow.Top, self.srWindow.Left, self.srWindow.Bottom, self.srWindow.Right
, self.dwMaximumWindowSize.Y, self.dwMaximumWindowSize.X
)
def GetConsoleScreenBufferInfo(stream_id=STDOUT):
handle = handles[stream_id]
csbi = CONSOLE_SCREEN_BUFFER_INFO()
success = windll.kernel32.GetConsoleScreenBufferInfo(
handle, byref(csbi))
return csbi
def SetConsoleTextAttribute(stream_id, attrs):
handle = handles[stream_id]
return windll.kernel32.SetConsoleTextAttribute(handle, attrs)
def SetConsoleCursorPosition(stream_id, position):
position = COORD(*position)
# If the position is out of range, do nothing.
if position.Y <= 0 or position.X <= 0:
return
# Adjust for Windows' SetConsoleCursorPosition:
# 1. being 0-based, while ANSI is 1-based.
# 2. expecting (x,y), while ANSI uses (y,x).
adjusted_position = COORD(position.Y - 1, position.X - 1)
# Adjust for viewport's scroll position
sr = GetConsoleScreenBufferInfo(STDOUT).srWindow
adjusted_position.Y += sr.Top
adjusted_position.X += sr.Left
# Resume normal processing
handle = handles[stream_id]
return windll.kernel32.SetConsoleCursorPosition(handle, adjusted_position)
def FillConsoleOutputCharacter(stream_id, char, length, start):
handle = handles[stream_id]
char = TCHAR(char)
length = DWORD(length)
num_written = DWORD(0)
# Note that this is hard-coded for ANSI (vs wide) bytes.
success = windll.kernel32.FillConsoleOutputCharacterA(
handle, char, length, start, byref(num_written))
return num_written.value
def FillConsoleOutputAttribute(stream_id, attr, length, start):
''' FillConsoleOutputAttribute( hConsole, csbi.wAttributes, dwConSize, coordScreen, &cCharsWritten )'''
handle = handles[stream_id]
attribute = WORD(attr)
length = DWORD(length)
num_written = DWORD(0)
# Note that this is hard-coded for ANSI (vs wide) bytes.
return windll.kernel32.FillConsoleOutputAttribute(
handle, attribute, length, start, byref(num_written))
5、winterm.py(color、data等处理)
from . import win32
# from wincon.h
class WinColor(object):
BLACK = 0
BLUE = 1
GREEN = 2
CYAN = 3
RED = 4
MAGENTA = 5
YELLOW = 6
GREY = 7
# from wincon.h
class WinStyle(object):
NORMAL = 0x00 # dim text, dim background
BRIGHT = 0x08 # bright text, dim background
class WinTerm(object):
def __init__(self):
self._default = win32.GetConsoleScreenBufferInfo(win32.STDOUT).wAttributes
self.set_attrs(self._default)
self._default_fore = self._fore
self._default_back = self._back
self._default_style = self._style
def get_attrs(self):
return self._fore + self._back * 16 + self._style
def set_attrs(self, value):
self._fore = value & 7
self._back = (value >> 4) & 7
self._style = value & WinStyle.BRIGHT
def reset_all(self, on_stderr=None):
self.set_attrs(self._default)
self.set_console(attrs=self._default)
def fore(self, fore=None, on_stderr=False):
if fore is None:
fore = self._default_fore
self._fore = fore
self.set_console(on_stderr=on_stderr)
def back(self, back=None, on_stderr=False):
if back is None:
back = self._default_back
self._back = back
self.set_console(on_stderr=on_stderr)
def style(self, style=None, on_stderr=False):
if style is None:
style = self._default_style
self._style = style
self.set_console(on_stderr=on_stderr)
def set_console(self, attrs=None, on_stderr=False):
if attrs is None:
attrs = self.get_attrs()
handle = win32.STDOUT
if on_stderr:
handle = win32.STDERR
win32.SetConsoleTextAttribute(handle, attrs)
def get_position(self, handle):
position = win32.GetConsoleScreenBufferInfo(handle).dwCursorPosition
# Because Windows coordinates are 0-based,
# and win32.SetConsoleCursorPosition expects 1-based.
position.X += 1
position.Y += 1
return position
def set_cursor_position(self, position=None, on_stderr=False):
if position is None:
#I'm not currently tracking the position, so there is no default.
#position = self.get_position()
return
handle = win32.STDOUT
if on_stderr:
handle = win32.STDERR
win32.SetConsoleCursorPosition(handle, position)
def cursor_up(self, num_rows=0, on_stderr=False):
if num_rows == 0:
return
handle = win32.STDOUT
if on_stderr:
handle = win32.STDERR
position = self.get_position(handle)
adjusted_position = (position.Y - num_rows, position.X)
self.set_cursor_position(adjusted_position, on_stderr)
def erase_data(self, mode=0, on_stderr=False):
# 0 (or None) should clear from the cursor to the end of the screen.
# 1 should clear from the cursor to the beginning of the screen.
# 2 should clear the entire screen. (And maybe move cursor to (1,1)?)
#
# At the moment, I only support mode 2. From looking at the API, it
# should be possible to calculate a different number of bytes to clear,
# and to do so relative to the cursor position.
if mode[0] not in (2,):
return
handle = win32.STDOUT
if on_stderr:
handle = win32.STDERR
# here's where we'll home the cursor
coord_screen = win32.COORD(0,0)
csbi = win32.GetConsoleScreenBufferInfo(handle)
# get the number of character cells in the current buffer
dw_con_size = csbi.dwSize.X * csbi.dwSize.Y
# fill the entire screen with blanks
win32.FillConsoleOutputCharacter(handle, ord(' '), dw_con_size, coord_screen)
# now set the buffer's attributes accordingly
win32.FillConsoleOutputAttribute(handle, self.get_attrs(), dw_con_size, coord_screen );
# put the cursor at (0, 0)
win32.SetConsoleCursorPosition(handle, (coord_screen.X, coord_screen.Y))
httplib2子目录
1、__init__.py（ssl、超时、url、cache等初始化处理）
from __future__ import generators
"""
httplib2
A caching http interface that supports ETags and gzip
to conserve bandwidth.
Requires Python 2.3 or later
Changelog:
2007-08-18, Rick: Modified so it's able to use a socks proxy if needed.
"""
__author__ = "Joe Gregorio (joe@bitworking.org)"
__copyright__ = "Copyright 2006, Joe Gregorio"
__contributors__ = ["Thomas Broyer (t.broyer@ltgt.net)",
"James Antill",
"Xavier Verges Farrero",
"Jonathan Feinberg",
"Blair Zajac",
"Sam Ruby",
"Louis Nyffenegger"]
__license__ = "MIT"
__version__ = "0.9.2"
import re
import sys
import email
import email.Utils
import email.Message
import email.FeedParser
import StringIO
import gzip
import zlib
import httplib
import urlparse
import urllib
import base64
import os
import copy
import calendar
import time
import random
import errno
try:
from hashlib import sha1 as _sha, md5 as _md5
except ImportError:
# prior to Python 2.5, these were separate modules
import sha
import md5
_sha = sha.new
_md5 = md5.new
import hmac
from gettext import gettext as _
import socket
try:
from httplib2 import socks
except ImportError:
try:
import socks
except (ImportError, AttributeError):
socks = None
# Build the appropriate socket wrapper for ssl
try:
import ssl # python 2.6
ssl_SSLError = ssl.SSLError
def _ssl_wrap_socket(sock, key_file, cert_file,
disable_validation, ca_certs):
if disable_validation:
cert_reqs = ssl.CERT_NONE
else:
cert_reqs = ssl.CERT_REQUIRED
# We should be specifying SSL version 3 or TLS v1, but the ssl module
# doesn't expose the necessary knobs. So we need to go with the default
# of SSLv23.
return ssl.wrap_socket(sock, keyfile=key_file, certfile=cert_file,
cert_reqs=cert_reqs, ca_certs=ca_certs)
except (AttributeError, ImportError):
ssl_SSLError = None
def _ssl_wrap_socket(sock, key_file, cert_file,
disable_validation, ca_certs):
if not disable_validation:
raise CertificateValidationUnsupported(
"SSL certificate validation is not supported without "
"the ssl module installed. To avoid this error, install "
"the ssl module, or explicity disable validation.")
ssl_sock = socket.ssl(sock, key_file, cert_file)
return httplib.FakeSocket(sock, ssl_sock)
if sys.version_info >= (2,3):
from iri2uri import iri2uri
else:
def iri2uri(uri):
return uri
def has_timeout(timeout): # python 2.6
if hasattr(socket, '_GLOBAL_DEFAULT_TIMEOUT'):
return (timeout is not None and timeout is not socket._GLOBAL_DEFAULT_TIMEOUT)
return (timeout is not None)
__all__ = [
'Http', 'Response', 'ProxyInfo', 'HttpLib2Error', 'RedirectMissingLocation',
'RedirectLimit', 'FailedToDecompressContent',
'UnimplementedDigestAuthOptionError',
'UnimplementedHmacDigestAuthOptionError',
'debuglevel', 'ProxiesUnavailableError']
# The httplib debug level, set to a non-zero value to get debug output
debuglevel = 0
# A request will be tried 'RETRIES' times if it fails at the socket/connection level.
RETRIES = 2
# Python 2.3 support
if sys.version_info < (2,4):
def sorted(seq):
seq.sort()
return seq
# Python 2.3 support
def HTTPResponse__getheaders(self):
"""Return list of (header, value) tuples."""
if self.msg is None:
raise httplib.ResponseNotReady()
return self.msg.items()
if not hasattr(httplib.HTTPResponse, 'getheaders'):
httplib.HTTPResponse.getheaders = HTTPResponse__getheaders
# All exceptions raised here derive from HttpLib2Error
class HttpLib2Error(Exception): pass
# Some exceptions can be caught and optionally
# be turned back into responses.
class HttpLib2ErrorWithResponse(HttpLib2Error):
def __init__(self, desc, response, content):
self.response = response
self.content = content
HttpLib2Error.__init__(self, desc)
class RedirectMissingLocation(HttpLib2ErrorWithResponse): pass
class RedirectLimit(HttpLib2ErrorWithResponse): pass
class FailedToDecompressContent(HttpLib2ErrorWithResponse): pass
class UnimplementedDigestAuthOptionError(HttpLib2ErrorWithResponse): pass
class UnimplementedHmacDigestAuthOptionError(HttpLib2ErrorWithResponse): pass
class MalformedHeader(HttpLib2Error): pass
class RelativeURIError(HttpLib2Error): pass
class ServerNotFoundError(HttpLib2Error): pass
class ProxiesUnavailableError(HttpLib2Error): pass
class CertificateValidationUnsupported(HttpLib2Error): pass
class SSLHandshakeError(HttpLib2Error): pass
class NotSupportedOnThisPlatform(HttpLib2Error): pass
class CertificateHostnameMismatch(SSLHandshakeError):
def __init__(self, desc, host, cert):
HttpLib2Error.__init__(self, desc)
self.host = host
self.cert = cert
# Open Items:
# -----------
# Proxy support
# Are we removing the cached content too soon on PUT (only delete on 200 Maybe?)
# Pluggable cache storage (supports storing the cache in
# flat files by default. We need a plug-in architecture
# that can support Berkeley DB and Squid)
# == Known Issues ==
# Does not handle a resource that uses conneg and Last-Modified but no ETag as a cache validator.
# Does not handle Cache-Control: max-stale
# Does not use Age: headers when calculating cache freshness.
# The number of redirections to follow before giving up.
# Note that only GET redirects are automatically followed.
# Will also honor 301 requests by saving that info and never
# requesting that URI again.
DEFAULT_MAX_REDIRECTS = 5
try:
# Users can optionally provide a module that tells us where the CA_CERTS
# are located.
import ca_certs_locater
CA_CERTS = ca_certs_locater.get()
except ImportError:
# Default CA certificates file bundled with httplib2.
CA_CERTS = os.path.join(
os.path.dirname(os.path.abspath(__file__ )), "cacerts.txt")
# Which headers are hop-by-hop headers by default
HOP_BY_HOP = ['connection', 'keep-alive', 'proxy-authenticate', 'proxy-authorization', 'te', 'trailers', 'transfer-encoding', 'upgrade']
def _get_end2end_headers(response):
hopbyhop = list(HOP_BY_HOP)
hopbyhop.extend([x.strip() for x in response.get('connection', '').split(',')])
return [header for header in response.keys() if header not in hopbyhop]
URI = re.compile(r"^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\?([^#]*))?(#(.*))?")
def parse_uri(uri):
"""Parses a URI using the regex given in Appendix B of RFC 3986.
(scheme, authority, path, query, fragment) = parse_uri(uri)
"""
groups = URI.match(uri).groups()
return (groups[1], groups[3], groups[4], groups[6], groups[8])
def urlnorm(uri):
(scheme, authority, path, query, fragment) = parse_uri(uri)
if not scheme or not authority:
raise RelativeURIError("Only absolute URIs are allowed. uri = %s" % uri)
authority = authority.lower()
scheme = scheme.lower()
if not path:
path = "/"
# Could do syntax based normalization of the URI before
# computing the digest. See Section 6.2.2 of Std 66.
request_uri = query and "?".join([path, query]) or path
scheme = scheme.lower()
defrag_uri = scheme + "://" + authority + request_uri
return scheme, authority, request_uri, defrag_uri
# Cache filename construction (original borrowed from Venus http://intertwingly.net/code/venus/)
re_url_scheme = re.compile(r'^\w+://')
re_slash = re.compile(r'[?/:|]+')
def safename(filename):
"""Return a filename suitable for the cache.
Strips dangerous and common characters to create a filename we
can use to store the cache in.
"""
try:
if re_url_scheme.match(filename):
if isinstance(filename,str):
filename = filename.decode('utf-8')
filename = filename.encode('idna')
else:
filename = filename.encode('idna')
except UnicodeError:
pass
if isinstance(filename,unicode):
filename=filename.encode('utf-8')
filemd5 = _md5(filename).hexdigest()
filename = re_url_scheme.sub("", filename)
filename = re_slash.sub(",", filename)
# limit length of filename
if len(filename)>200:
filename=filename[:200]
return ",".join((filename, filemd5))
NORMALIZE_SPACE = re.compile(r'(?:\r\n)?[ \t]+')
def _normalize_headers(headers):
return dict([ (key.lower(), NORMALIZE_SPACE.sub(value, ' ').strip()) for (key, value) in headers.iteritems()])
def _parse_cache_control(headers):
retval = {}
if headers.has_key('cache-control'):
parts = headers.get('cache-control').split(',')
parts_with_args = [tuple([x.strip().lower() for x in part.split("=", 1)]) for part in parts if -1 != part.find("=")]
parts_wo_args = [(name.strip().lower(), 1) for name in parts if -1 == name.find("=")]
retval = dict(parts_with_args + parts_wo_args)
return retval
# Whether to use a strict mode to parse WWW-Authenticate headers
# Might lead to bad results in case of ill-formed header value,
# so disabled by default, falling back to relaxed parsing.
# Set to true to turn on, usefull for testing servers.
USE_WWW_AUTH_STRICT_PARSING = 0
# In regex below:
# [^\0-\x1f\x7f-\xff()<>@,;:\\\"/[\]?={} \t]+ matches a "token" as defined by HTTP
# "(?:[^\0-\x08\x0A-\x1f\x7f-\xff\\\"]|\\[\0-\x7f])*?" matches a "quoted-string" as defined by HTTP, when LWS have already been replaced by a single space
# Actually, as an auth-param value can be either a token or a quoted-string, they are combined in a single pattern which matches both:
# \"?((?<=\")(?:[^\0-\x1f\x7f-\xff\\\"]|\\[\0-\x7f])*?(?=\")|(?<!\")[^\0-\x08\x0A-\x1f\x7f-\xff()<>@,;:\\\"/[\]?={} \t]+(?!\"))\"?
WWW_AUTH_STRICT = re.compile(r"^(?:\s*(?:,\s*)?([^\0-\x1f\x7f-\xff()<>@,;:\\\"/[\]?={} \t]+)\s*=\s*\"?((?<=\")(?:[^\0-\x08\x0A-\x1f\x7f-\xff\\\"]|\\[\0-\x7f])*?(?=\")|(?<!\")[^\0-\x1f\x7f-\xff()<>@,;:\\\"/[\]?={} \t]+(?!\"))\"?)(.*)$")
WWW_AUTH_RELAXED = re.compile(r"^(?:\s*(?:,\s*)?([^ \t\r\n=]+)\s*=\s*\"?((?<=\")(?:[^\\\"]|\\.)*?(?=\")|(?<!\")[^ \t\r\n,]+(?!\"))\"?)(.*)$")
UNQUOTE_PAIRS = re.compile(r'\\(.)')
def _parse_www_authenticate(headers, headername='www-authenticate'):
"""Returns a dictionary of dictionaries, one dict
per auth_scheme."""
retval = {}
if headers.has_key(headername):
try:
authenticate = headers[headername].strip()
www_auth = USE_WWW_AUTH_STRICT_PARSING and WWW_AUTH_STRICT or WWW_AUTH_RELAXED
while authenticate:
# Break off the scheme at the beginning of the line
if headername == 'authentication-info':
(auth_scheme, the_rest) = ('digest', authenticate)
else:
(auth_scheme, the_rest) = authenticate.split(" ", 1)
# Now loop over all the key value pairs that come after the scheme,
# being careful not to roll into the next scheme
match = www_auth.search(the_rest)
auth_params = {}
while match:
if match and len(match.groups()) == 3:
(key, value, the_rest) = match.groups()
auth_params[key.lower()] = UNQUOTE_PAIRS.sub(r'\1', value) # '\\'.join([x.replace('\\', '') for x in value.split('\\\\')])
match = www_auth.search(the_rest)
retval[auth_scheme.lower()] = auth_params
authenticate = the_rest.strip()
except ValueError:
raise MalformedHeader("WWW-Authenticate")
return retval
def _entry_disposition(response_headers, request_headers):
"""Determine freshness from the Date, Expires and Cache-Control headers.
We don't handle the following:
1. Cache-Control: max-stale
2. Age: headers are not used in the calculations.
Not that this algorithm is simpler than you might think
because we are operating as a private (non-shared) cache.
This lets us ignore 's-maxage'. We can also ignore
'proxy-invalidate' since we aren't a proxy.
We will never return a stale document as
fresh as a design decision, and thus the non-implementation
of 'max-stale'. This also lets us safely ignore 'must-revalidate'
since we operate as if every server has sent 'must-revalidate'.
Since we are private we get to ignore both 'public' and
'private' parameters. We also ignore 'no-transform' since
we don't do any transformations.
The 'no-store' parameter is handled at a higher level.
So the only Cache-Control parameters we look at are:
no-cache
only-if-cached
max-age
min-fresh
"""
retval = "STALE"
cc = _parse_cache_control(request_headers)
cc_response = _parse_cache_control(response_headers)
if request_headers.has_key('pragma') and request_headers.get('pragma').lower().find('no-cache') != -1:
retval = "TRANSPARENT"
if 'cache-control' not in request_headers:
request_headers['cache-control'] = 'no-cache'
elif cc.has_key('no-cache'):
retval = "TRANSPARENT"
elif cc_response.has_key('no-cache'):
retval = "STALE"
elif cc.has_key('only-if-cached'):
retval = "FRESH"
elif response_headers.has_key('date'):
date = calendar.timegm(email.Utils.parsedate_tz(response_headers.get('date')))
now = time.time()
current_age = max(0, now - date)
if cc_response.has_key('max-age'):
try:
freshness_lifetime = int(cc_response.get('max-age'))
except ValueError:
freshness_lifetime = 0
elif response_headers.has_key('expires'):
expires = email.Utils.parsedate_tz(response_headers.get('expires'))
if None == expires:
freshness_lifetime = 0
else:
freshness_lifetime = max(0, calendar.timegm(expires) - date)
else:
freshness_lifetime = 0
if cc.has_key('max-age'):
try:
freshness_lifetime = int(cc.get('max-age'))
except ValueError:
freshness_lifetime = 0
if cc.has_key('min-fresh'):
try:
min_fresh = int(cc.get('min-fresh'))
except ValueError:
min_fresh = 0
current_age += min_fresh
if freshness_lifetime > current_age:
retval = "FRESH"
return retval
def _decompressContent(response, new_content):
content = new_content
try:
encoding = response.get('content-encoding', None)
if encoding in ['gzip', 'deflate']:
if encoding == 'gzip':
content = gzip.GzipFile(fileobj=StringIO.StringIO(new_content)).read()
if encoding == 'deflate':
content = zlib.decompress(content)
response['content-length'] = str(len(content))
# Record the historical presence of the encoding in a way the won't interfere.
response['-content-encoding'] = response['content-encoding']
del response['content-encoding']
except IOError:
content = ""
raise FailedToDecompressContent(_("Content purported to be compressed with %s but failed to decompress.") % response.get('content-encoding'), response, content)
return content
def _updateCache(request_headers, response_headers, content, cache, cachekey):
if cachekey:
cc = _parse_cache_control(request_headers)
cc_response = _parse_cache_control(response_headers)
if cc.has_key('no-store') or cc_response.has_key('no-store'):
cache.delete(cachekey)
else:
info = email.Message.Message()
for key, value in response_headers.iteritems():
if key not in ['status','content-encoding','transfer-encoding']:
info[key] = value
# Add annotations to the cache to indicate what headers
# are variant for this request.
vary = response_headers.get('vary', None)
if vary:
vary_headers = vary.lower().replace(' ', '').split(',')
for header in vary_headers:
key = '-varied-%s' % header
try:
info[key] = request_headers[header]
except KeyError:
pass
status = response_headers.status
if status == 304:
status = 200
status_header = 'status: %d\r\n' % status
header_str = info.as_string()
header_str = re.sub("\r(?!\n)|(?<!\r)\n", "\r\n", header_str)
text = "".join([status_header, header_str, content])
cache.set(cachekey, text)
def _cnonce():
dig = _md5("%s:%s" % (time.ctime(), ["0123456789"[random.randrange(0, 9)] for i in range(20)])).hexdigest()
return dig[:16]
def _wsse_username_token(cnonce, iso_now, password):
return base64.b64encode(_sha("%s%s%s" % (cnonce, iso_now, password)).digest()).strip()
# For credentials we need two things, first
# a pool of credential to try (not necesarily tied to BAsic, Digest, etc.)
# Then we also need a list of URIs that have already demanded authentication
# That list is tricky since sub-URIs can take the same auth, or the
# auth scheme may change as you descend the tree.
# So we also need each Auth instance to be able to tell us
# how close to the 'top' it is.
class Authentication(object):
def __init__(self, credentials, host, request_uri, headers, response, content, http):
(scheme, authority, path, query, fragment) = parse_uri(request_uri)
self.path = path
self.host = host
self.credentials = credentials
self.http = http
def depth(self, request_uri):
(scheme, authority, path, query, fragment) = parse_uri(request_uri)
return request_uri[len(self.path):].count("/")
def inscope(self, host, request_uri):
# XXX Should we normalize the request_uri?
(scheme, authority, path, query, fragment) = parse_uri(request_uri)
return (host == self.host) and path.startswith(self.path)
def request(self, method, request_uri, headers, content):
"""Modify the request headers to add the appropriate
Authorization header. Over-ride this in sub-classes."""
pass
def response(self, response, content):
"""Gives us a chance to update with new nonces
or such returned from the last authorized response.
Over-rise this in sub-classes if necessary.
Return TRUE is the request is to be retried, for
example Digest may return stale=true.
"""
return False
class BasicAuthentication(Authentication):
def __init__(self, credentials, host, request_uri, headers, response, content, http):
Authentication.__init__(self, credentials, host, request_uri, headers, response, content, http)
def request(self, method, request_uri, headers, content):
"""Modify the request headers to add the appropriate
Authorization header."""
headers['authorization'] = 'Basic ' + base64.b64encode("%s:%s" % self.credentials).strip()
class DigestAuthentication(Authentication):
"""Only do qop='auth' and MD5, since that
is all Apache currently implements"""
def __init__(self, credentials, host, request_uri, headers, response, content, http):
Authentication.__init__(self, credentials, host, request_uri, headers, response, content, http)
challenge = _parse_www_authenticate(response, 'www-authenticate')
self.challenge = challenge.get('digest')
qop = self.challenge.get('qop', 'auth')
self.challenge['qop'] = ('auth' in [x.strip() for x in qop.split()]) and 'auth' or None
if self.challenge.get('qop') is None:
raise UnimplementedDigestAuthOptionError( _("Unsupported value for qop: %s." % qop))
self.challenge['algorithm'] = self.challenge.get('algorithm', 'MD5').upper()
if self.challenge.get('algorithm') != 'MD5':
raise UnimplementedDigestAuthOptionError( _("Unsupported value for algorithm: %s." % self.challenge('algorithm')))
self.A1 = "".join([self.credentials[0], ":", self.challenge.get('realm'), ":", self.credentials[1]])
self.challenge['nc'] = 1
def request(self, method, request_uri, headers, content, cnonce = None):
"""Modify the request headers"""
H = lambda x: _md5(x).hexdigest()
KD = lambda s, d: H("%s:%s" % (s, d))
A2 = "".join([method, ":", request_uri])
self.challenge['cnonce'] = cnonce or _cnonce()
request_digest = '"%s"' % KD(H(self.A1), "%s:%s:%s:%s:%s" % (
self.challenge['nonce'],
'%08x' % self.challenge['nc'],
self.challenge['cnonce'],
self.challenge['qop'], H(A2)))
headers['authorization'] = 'Digest username="%s", realm="%s", nonce="%s", uri="%s", algorithm=%s, response=%s, qop=%s, nc=%08x, cnonce="%s"' % (
self.credentials[0],
self.challenge['realm'],
self.challenge['nonce'],
request_uri,
self.challenge['algorithm'],
request_digest,
self.challenge['qop'],
self.challenge['nc'],
self.challenge['cnonce'])
if self.challenge.get('opaque'):
headers['authorization'] += ', opaque="%s"' % self.challenge.get('opaque')
self.challenge['nc'] += 1
def response(self, response, content):
if not response.has_key('authentication-info'):
challenge = _parse_www_authenticate(response, 'www-authenticate').get('digest', {})
if 'true' == challenge.get('stale'):
self.challenge['nonce'] = challenge.get('nonce')
self.challenge['nc'] = 1
return True
else:
updated_challenge = _parse_www_authenticate(response, 'authentication-info').get('digest', {})
if updated_challenge.has_key('nextnonce'):
self.challenge['nonce'] = updated_challenge.get('nextnonce')
self.challenge['nc'] = 1
return False
class HmacDigestAuthentication(Authentication):
"""Adapted from Robert Sayre's code and DigestAuthentication above."""
__author__ = "Thomas Broyer (t.broyer@ltgt.net)"
def __init__(self, credentials, host, request_uri, headers, response, content, http):
Authentication.__init__(self, credentials, host, request_uri, headers, response, content, http)
challenge = _parse_www_authenticate(response, 'www-authenticate')
self.challenge = challenge.get('hmacdigest')
# TODO: self.challenge['domain']
self.challenge['reason'] = self.challenge.get('reason', 'unauthorized')
if self.challenge.get('reason') not in ['unauthorized', 'integrity']:
self.challenge['reason'] = 'unauthorized'
self.challenge['salt'] = self.challenge.get('salt', '')
if not self.challenge.get('snonce'):
raise UnimplementedHmacDigestAuthOptionError( _("The challenge doesn't contain a server nonce, or this one is empty."))
self.challenge['algorithm'] = self.challenge.get('algorithm', 'HMAC-SHA-1')
if self.challenge.get('algorithm') not in ['HMAC-SHA-1', 'HMAC-MD5']:
raise UnimplementedHmacDigestAuthOptionError( _("Unsupported value for algorithm: %s." % self.challenge.get('algorithm')))
self.challenge['pw-algorithm'] = self.challenge.get('pw-algorithm', 'SHA-1')
if self.challenge.get('pw-algorithm') not in ['SHA-1', 'MD5']:
raise UnimplementedHmacDigestAuthOptionError( _("Unsupported value for pw-algorithm: %s." % self.challenge.get('pw-algorithm')))
if self.challenge.get('algorithm') == 'HMAC-MD5':
self.hashmod = _md5
else:
self.hashmod = _sha
if self.challenge.get('pw-algorithm') == 'MD5':
self.pwhashmod = _md5
else:
self.pwhashmod = _sha
self.key = "".join([self.credentials[0], ":",
self.pwhashmod.new("".join([self.credentials[1], self.challenge['salt']])).hexdigest().lower(),
":", self.challenge['realm']])
self.key = self.pwhashmod.new(self.key).hexdigest().lower()
def request(self, method, request_uri, headers, content):
"""Modify the request headers"""
keys = _get_end2end_headers(headers)
keylist = "".join(["%s " % k for k in keys])
headers_val = "".join([headers[k] for k in keys])
created = time.strftime('%Y-%m-%dT%H:%M:%SZ',time.gmtime())
cnonce = _cnonce()
request_digest = "%s:%s:%s:%s:%s" % (method, request_uri, cnonce, self.challenge.get('snonce'), headers_val)
request_digest = hmac.new(self.key, request_digest, self.hashmod).hexdigest().lower()
headers['authorization'] = 'HMACDigest username="%s", realm="%s", snonce="%s", cnonce="%s", uri="%s", created="%s", response="%s", headers="%s"' % (
self.credentials[0],
self.challenge['realm'],
self.challenge['snonce'],
cnonce,
request_uri,
created,
request_digest,
keylist)
def response(self, response, content):
challenge = _parse_www_authenticate(response, 'www-authenticate').get('hmacdigest', {})
if challenge.get('reason') in ['integrity', 'stale']:
return True
return False
class WsseAuthentication(Authentication):
"""This is thinly tested and should not be relied upon.
At this time there isn't any third party server to test against.
Blogger and TypePad implemented this algorithm at one point
but Blogger has since switched to Basic over HTTPS and
TypePad has implemented it wrong, by never issuing a 401
challenge but instead requiring your client to telepathically know that
their endpoint is expecting WSSE profile="UsernameToken"."""
def __init__(self, credentials, host, request_uri, headers, response, content, http):
Authentication.__init__(self, credentials, host, request_uri, headers, response, content, http)
def request(self, method, request_uri, headers, content):
"""Modify the request headers to add the appropriate
Authorization header."""
headers['authorization'] = 'WSSE profile="UsernameToken"'
iso_now = time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime())
cnonce = _cnonce()
password_digest = _wsse_username_token(cnonce, iso_now, self.credentials[1])
headers['X-WSSE'] = 'UsernameToken Username="%s", PasswordDigest="%s", Nonce="%s", Created="%s"' % (
self.credentials[0],
password_digest,
cnonce,
iso_now)
class GoogleLoginAuthentication(Authentication):
def __init__(self, credentials, host, request_uri, headers, response, content, http):
from urllib import urlencode
Authentication.__init__(self, credentials, host, request_uri, headers, response, content, http)
challenge = _parse_www_authenticate(response, 'www-authenticate')
service = challenge.get('googlelogin', {}).get('service', 'xapi')
# Bloggger actually returns the service in the challenge
# For the rest we guess based on the URI
if service == 'xapi' and request_uri.find("calendar") > 0:
service = "cl"
# No point in guessing Base or Spreadsheet
#elif request_uri.find("spreadsheets") > 0:
# service = "wise"
auth = dict(Email=credentials[0], Passwd=credentials[1], service=service, source=headers.get('user-agent'))
resp, content = self.http.request("https://www.google.com/accounts/ClientLogin", method="POST", body=urlencode(auth), headers={'Content-Type': 'application/x-www-form-urlencoded'})
lines = content.split('\n')
d = dict([tuple(line.split("=", 1)) for line in lines if line])
if resp.status == 403:
self.Auth = ""
else:
self.Auth = d['Auth']
def request(self, method, request_uri, headers, content):
"""Modify the request headers to add the appropriate
Authorization header."""
headers['authorization'] = 'GoogleLogin Auth=' + self.Auth
AUTH_SCHEME_CLASSES = {
"basic": BasicAuthentication,
"wsse": WsseAuthentication,
"digest": DigestAuthentication,
"hmacdigest": HmacDigestAuthentication,
"googlelogin": GoogleLoginAuthentication
}
AUTH_SCHEME_ORDER = ["hmacdigest", "googlelogin", "digest", "wsse", "basic"]
class FileCache(object):
"""Uses a local directory as a store for cached files.
Not really safe to use if multiple threads or processes are going to
be running on the same cache.
"""
def __init__(self, cache, safe=safename): # use safe=lambda x: md5.new(x).hexdigest() for the old behavior
self.cache = cache
self.safe = safe
if not os.path.exists(cache):
os.makedirs(self.cache)
def get(self, key):
retval = None
cacheFullPath = os.path.join(self.cache, self.safe(key))
try:
f = file(cacheFullPath, "rb")
retval = f.read()
f.close()
except IOError:
pass
return retval
def set(self, key, value):
cacheFullPath = os.path.join(self.cache, self.safe(key))
f = file(cacheFullPath, "wb")
f.write(value)
f.close()
def delete(self, key):
cacheFullPath = os.path.join(self.cache, self.safe(key))
if os.path.exists(cacheFullPath):
os.remove(cacheFullPath)
class Credentials(object):
def __init__(self):
self.credentials = []
def add(self, name, password, domain=""):
self.credentials.append((domain.lower(), name, password))
def clear(self):
self.credentials = []
def iter(self, domain):
for (cdomain, name, password) in self.credentials:
if cdomain == "" or domain == cdomain:
yield (name, password)
class KeyCerts(Credentials):
"""Identical to Credentials except that
name/password are mapped to key/cert."""
pass
class AllHosts(object):
pass
class ProxyInfo(object):
"""Collect information required to use a proxy."""
bypass_hosts = ()
def __init__(self, proxy_type, proxy_host, proxy_port,
proxy_rdns=True, proxy_user=None, proxy_pass=None):
"""
Args:
proxy_type: The type of proxy server. This must be set to one of
socks.PROXY_TYPE_XXX constants. For example:
p = ProxyInfo(proxy_type=socks.PROXY_TYPE_HTTP,
proxy_host='localhost', proxy_port=8000)
proxy_host: The hostname or IP address of the proxy server.
proxy_port: The port that the proxy server is running on.
proxy_rdns: If True (default), DNS queries will not be performed
locally, and instead, handed to the proxy to resolve. This is useful
if the network does not allow resolution of non-local names. In
httplib2 0.9 and earlier, this defaulted to False.
proxy_user: The username used to authenticate with the proxy server.
proxy_pass: The password used to authenticate with the proxy server.
"""
self.proxy_type = proxy_type
self.proxy_host = proxy_host
self.proxy_port = proxy_port
self.proxy_rdns = proxy_rdns
self.proxy_user = proxy_user
self.proxy_pass = proxy_pass
def astuple(self):
return (self.proxy_type, self.proxy_host, self.proxy_port,
self.proxy_rdns, self.proxy_user, self.proxy_pass)
def isgood(self):
return (self.proxy_host != None) and (self.proxy_port != None)
def applies_to(self, hostname):
return not self.bypass_host(hostname)
def bypass_host(self, hostname):
"""Has this host been excluded from the proxy config"""
if self.bypass_hosts is AllHosts:
return True
bypass = False
for domain in self.bypass_hosts:
if hostname.endswith(domain):
bypass = True
return bypass
def proxy_info_from_environment(method='http'):
"""
Read proxy info from the environment variables.
"""
if method not in ['http', 'https']:
return
env_var = method + '_proxy'
url = os.environ.get(env_var, os.environ.get(env_var.upper()))
if not url:
return
pi = proxy_info_from_url(url, method)
no_proxy = os.environ.get('no_proxy', os.environ.get('NO_PROXY', ''))
bypass_hosts = []
if no_proxy:
bypass_hosts = no_proxy.split(',')
# special case, no_proxy=* means all hosts bypassed
if no_proxy == '*':
bypass_hosts = AllHosts
pi.bypass_hosts = bypass_hosts
return pi
def proxy_info_from_url(url, method='http'):
"""
Construct a ProxyInfo from a URL (such as http_proxy env var)
"""
url = urlparse.urlparse(url)
username = None
password = None
port = None
if '@' in url[1]:
ident, host_port = url[1].split('@', 1)
if ':' in ident:
username, password = ident.split(':', 1)
else:
password = ident
else:
host_port = url[1]
if ':' in host_port:
host, port = host_port.split(':', 1)
else:
host = host_port
if port:
port = int(port)
else:
port = dict(https=443, http=80)[method]
proxy_type = 3 # socks.PROXY_TYPE_HTTP
return ProxyInfo(
proxy_type = proxy_type,
proxy_host = host,
proxy_port = port,
proxy_user = username or None,
proxy_pass = password or None,
)
class HTTPConnectionWithTimeout(httplib.HTTPConnection):
"""
HTTPConnection subclass that supports timeouts
All timeouts are in seconds. If None is passed for timeout then
Python's default timeout for sockets will be used. See for example
the docs of socket.setdefaulttimeout():
http://docs.python.org/library/socket.html#socket.setdefaulttimeout
"""
def __init__(self, host, port=None, strict=None, timeout=None, proxy_info=None):
httplib.HTTPConnection.__init__(self, host, port, strict)
self.timeout = timeout
self.proxy_info = proxy_info
def connect(self):
"""Connect to the host and port specified in __init__."""
# Mostly verbatim from httplib.py.
if self.proxy_info and socks is None:
raise ProxiesUnavailableError(
'Proxy support missing but proxy use was requested!')
msg = "getaddrinfo returns an empty list"
if self.proxy_info and self.proxy_info.isgood():
use_proxy = True
proxy_type, proxy_host, proxy_port, proxy_rdns, proxy_user, proxy_pass = self.proxy_info.astuple()
host = proxy_host
port = proxy_port
else:
use_proxy = False
host = self.host
port = self.port
for res in socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM):
af, socktype, proto, canonname, sa = res
try:
if use_proxy:
self.sock = socks.socksocket(af, socktype, proto)
self.sock.setproxy(proxy_type, proxy_host, proxy_port, proxy_rdns, proxy_user, proxy_pass)
else:
self.sock = socket.socket(af, socktype, proto)
self.sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)
# Different from httplib: support timeouts.
if has_timeout(self.timeout):
self.sock.settimeout(self.timeout)
# End of difference from httplib.
if self.debuglevel > 0:
print "connect: (%s, %s) ************" % (self.host, self.port)
if use_proxy:
print "proxy: %s ************" % str((proxy_host, proxy_port, proxy_rdns, proxy_user, proxy_pass))
self.sock.connect((self.host, self.port) + sa[2:])
except socket.error, msg:
if self.debuglevel > 0:
print "connect fail: (%s, %s)" % (self.host, self.port)
if use_proxy:
print "proxy: %s" % str((proxy_host, proxy_port, proxy_rdns, proxy_user, proxy_pass))
if self.sock:
self.sock.close()
self.sock = None
continue
break
if not self.sock:
raise socket.error, msg
class HTTPSConnectionWithTimeout(httplib.HTTPSConnection):
"""
This class allows communication via SSL.
All timeouts are in seconds. If None is passed for timeout then
Python's default timeout for sockets will be used. See for example
the docs of socket.setdefaulttimeout():
http://docs.python.org/library/socket.html#socket.setdefaulttimeout
"""
def __init__(self, host, port=None, key_file=None, cert_file=None,
strict=None, timeout=None, proxy_info=None,
ca_certs=None, disable_ssl_certificate_validation=False):
httplib.HTTPSConnection.__init__(self, host, port=port,
key_file=key_file,
cert_file=cert_file, strict=strict)
self.timeout = timeout
self.proxy_info = proxy_info
if ca_certs is None:
ca_certs = CA_CERTS
self.ca_certs = ca_certs
self.disable_ssl_certificate_validation = \
disable_ssl_certificate_validation
# The following two methods were adapted from https_wrapper.py, released
# with the Google Appengine SDK at
# http://googleappengine.googlecode.com/svn-history/r136/trunk/python/google/appengine/tools/https_wrapper.py
# under the following license:
#
# Copyright 2007 Google Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
def _GetValidHostsForCert(self, cert):
"""Returns a list of valid host globs for an SSL certificate.
Args:
cert: A dictionary representing an SSL certificate.
Returns:
list: A list of valid host globs.
"""
if 'subjectAltName' in cert:
return [x[1] for x in cert.get('subjectAltName')
if x[0].lower() == 'dns']
else:
return [x[0][1] for x in cert.get('subject')
if x[0][0].lower() == 'commonname']
def _ValidateCertificateHostname(self, cert, hostname):
"""Validates that a given hostname is valid for an SSL certificate.
Args:
cert: A dictionary representing an SSL certificate.
hostname: The hostname to test.
Returns:
bool: Whether or not the hostname is valid for this certificate.
"""
hosts = self._GetValidHostsForCert(cert)
for host in hosts:
host_re = host.replace('.', '\.').replace('*', '[^.]*')
if re.search('^%s$' % (host_re,), hostname, re.I):
return True
return False
def connect(self):
"Connect to a host on a given (SSL) port."
msg = "getaddrinfo returns an empty list"
if self.proxy_info and self.proxy_info.isgood():
use_proxy = True
proxy_type, proxy_host, proxy_port, proxy_rdns, proxy_user, proxy_pass = self.proxy_info.astuple()
host = proxy_host
port = proxy_port
else:
use_proxy = False
host = self.host
port = self.port
address_info = socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM)
for family, socktype, proto, canonname, sockaddr in address_info:
try:
if use_proxy:
sock = socks.socksocket(family, socktype, proto)
sock.setproxy(proxy_type, proxy_host, proxy_port, proxy_rdns, proxy_user, proxy_pass)
else:
sock = socket.socket(family, socktype, proto)
sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)
if has_timeout(self.timeout):
sock.settimeout(self.timeout)
sock.connect((self.host, self.port))
self.sock =_ssl_wrap_socket(
sock, self.key_file, self.cert_file,
self.disable_ssl_certificate_validation, self.ca_certs)
if self.debuglevel > 0:
print "connect: (%s, %s)" % (self.host, self.port)
if use_proxy:
print "proxy: %s" % str((proxy_host, proxy_port, proxy_rdns, proxy_user, proxy_pass))
if not self.disable_ssl_certificate_validation:
cert = self.sock.getpeercert()
hostname = self.host.split(':', 0)[0]
if not self._ValidateCertificateHostname(cert, hostname):
raise CertificateHostnameMismatch(
'Server presented certificate that does not match '
'host %s: %s' % (hostname, cert), hostname, cert)
except ssl_SSLError, e:
if sock:
sock.close()
if self.sock:
self.sock.close()
self.sock = None
# Unfortunately the ssl module doesn't seem to provide any way
# to get at more detailed error information, in particular
# whether the error is due to certificate validation or
# something else (such as SSL protocol mismatch).
if e.errno == ssl.SSL_ERROR_SSL:
raise SSLHandshakeError(e)
else:
raise
except (socket.timeout, socket.gaierror):
raise
except socket.error, msg:
if self.debuglevel > 0:
print "connect fail: (%s, %s)" % (self.host, self.port)
if use_proxy:
print "proxy: %s" % str((proxy_host, proxy_port, proxy_rdns, proxy_user, proxy_pass))
if self.sock:
self.sock.close()
self.sock = None
continue
break
if not self.sock:
raise socket.error, msg
SCHEME_TO_CONNECTION = {
'http': HTTPConnectionWithTimeout,
'https': HTTPSConnectionWithTimeout
}
# Use a different connection object for Google App Engine
try:
try:
from google.appengine.api import apiproxy_stub_map
if apiproxy_stub_map.apiproxy.GetStub('urlfetch') is None:
raise ImportError # Bail out; we're not actually running on App Engine.
from google.appengine.api.urlfetch import fetch
from google.appengine.api.urlfetch import InvalidURLError
except (ImportError, AttributeError):
from google3.apphosting.api import apiproxy_stub_map
if apiproxy_stub_map.apiproxy.GetStub('urlfetch') is None:
raise ImportError # Bail out; we're not actually running on App Engine.
from google3.apphosting.api.urlfetch import fetch
from google3.apphosting.api.urlfetch import InvalidURLError
def _new_fixed_fetch(validate_certificate):
def fixed_fetch(url, payload=None, method="GET", headers={},
allow_truncated=False, follow_redirects=True,
deadline=None):
if deadline is None:
deadline = socket.getdefaulttimeout() or 5
return fetch(url, payload=payload, method=method, headers=headers,
allow_truncated=allow_truncated,
follow_redirects=follow_redirects, deadline=deadline,
validate_certificate=validate_certificate)
return fixed_fetch
class AppEngineHttpConnection(httplib.HTTPConnection):
"""Use httplib on App Engine, but compensate for its weirdness.
The parameters key_file, cert_file, proxy_info, ca_certs, and
disable_ssl_certificate_validation are all dropped on the ground.
"""
def __init__(self, host, port=None, key_file=None, cert_file=None,
strict=None, timeout=None, proxy_info=None, ca_certs=None,
disable_ssl_certificate_validation=False):
httplib.HTTPConnection.__init__(self, host, port=port,
strict=strict, timeout=timeout)
class AppEngineHttpsConnection(httplib.HTTPSConnection):
"""Same as AppEngineHttpConnection, but for HTTPS URIs."""
def __init__(self, host, port=None, key_file=None, cert_file=None,
strict=None, timeout=None, proxy_info=None, ca_certs=None,
disable_ssl_certificate_validation=False):
httplib.HTTPSConnection.__init__(self, host, port=port,
key_file=key_file,
cert_file=cert_file, strict=strict,
timeout=timeout)
self._fetch = _new_fixed_fetch(
not disable_ssl_certificate_validation)
# Update the connection classes to use the Googel App Engine specific ones.
SCHEME_TO_CONNECTION = {
'http': AppEngineHttpConnection,
'https': AppEngineHttpsConnection
}
except (ImportError, AttributeError):
pass
class Http(object):
"""An HTTP client that handles:
- all methods
- caching
- ETags
- compression,
- HTTPS
- Basic
- Digest
- WSSE
and more.
"""
def __init__(self, cache=None, timeout=None,
proxy_info=proxy_info_from_environment,
ca_certs=None, disable_ssl_certificate_validation=False):
"""If 'cache' is a string then it is used as a directory name for
a disk cache. Otherwise it must be an object that supports the
same interface as FileCache.
All timeouts are in seconds. If None is passed for timeout
then Python's default timeout for sockets will be used. See
for example the docs of socket.setdefaulttimeout():
http://docs.python.org/library/socket.html#socket.setdefaulttimeout
`proxy_info` may be:
- a callable that takes the http scheme ('http' or 'https') and
returns a ProxyInfo instance per request. By default, uses
proxy_nfo_from_environment.
- a ProxyInfo instance (static proxy config).
- None (proxy disabled).
ca_certs is the path of a file containing root CA certificates for SSL
server certificate validation. By default, a CA cert file bundled with
httplib2 is used.
If disable_ssl_certificate_validation is true, SSL cert validation will
not be performed.
"""
self.proxy_info = proxy_info
self.ca_certs = ca_certs
self.disable_ssl_certificate_validation = \
disable_ssl_certificate_validation
# Map domain name to an httplib connection
self.connections = {}
# The location of the cache, for now a directory
# where cached responses are held.
if cache and isinstance(cache, basestring):
self.cache = FileCache(cache)
else:
self.cache = cache
# Name/password
self.credentials = Credentials()
# Key/cert
self.certificates = KeyCerts()
# authorization objects
self.authorizations = []
# If set to False then no redirects are followed, even safe ones.
self.follow_redirects = True
# Which HTTP methods do we apply optimistic concurrency to, i.e.
# which methods get an "if-match:" etag header added to them.
self.optimistic_concurrency_methods = ["PUT", "PATCH"]
# If 'follow_redirects' is True, and this is set to True then
# all redirecs are followed, including unsafe ones.
self.follow_all_redirects = False
self.ignore_etag = False
self.force_exception_to_status_code = False
self.timeout = timeout
# Keep Authorization: headers on a redirect.
self.forward_authorization_headers = False
def __getstate__(self):
state_dict = copy.copy(self.__dict__)
# In case request is augmented by some foreign object such as
# credentials which handle auth
if 'request' in state_dict:
del state_dict['request']
if 'connections' in state_dict:
del state_dict['connections']
return state_dict
def __setstate__(self, state):
self.__dict__.update(state)
self.connections = {}
def _auth_from_challenge(self, host, request_uri, headers, response, content):
"""A generator that creates Authorization objects
that can be applied to requests.
"""
challenges = _parse_www_authenticate(response, 'www-authenticate')
for cred in self.credentials.iter(host):
for scheme in AUTH_SCHEME_ORDER:
if challenges.has_key(scheme):
yield AUTH_SCHEME_CLASSES[scheme](cred, host, request_uri, headers, response, content, self)
def add_credentials(self, name, password, domain=""):
"""Add a name and password that will be used
any time a request requires authentication."""
self.credentials.add(name, password, domain)
def add_certificate(self, key, cert, domain):
"""Add a key and cert that will be used
any time a request requires authentication."""
self.certificates.add(key, cert, domain)
def clear_credentials(self):
"""Remove all the names and passwords
that are used for authentication"""
self.credentials.clear()
self.authorizations = []
def _conn_request(self, conn, request_uri, method, body, headers):
i = 0
seen_bad_status_line = False
while i < RETRIES:
i += 1
try:
if hasattr(conn, 'sock') and conn.sock is None:
conn.connect()
conn.request(method, request_uri, body, headers)
except socket.timeout:
raise
except socket.gaierror:
conn.close()
raise ServerNotFoundError("Unable to find the server at %s" % conn.host)
except ssl_SSLError:
conn.close()
raise
except socket.error, e:
err = 0
if hasattr(e, 'args'):
err = getattr(e, 'args')[0]
else:
err = e.errno
if err in (errno.ENETUNREACH, errno.EADDRNOTAVAIL) and i < RETRIES:
continue # retry on potentially transient socket errors
raise
except httplib.HTTPException:
# Just because the server closed the connection doesn't apparently mean
# that the server didn't send a response.
if hasattr(conn, 'sock') and conn.sock is None:
if i < RETRIES-1:
conn.close()
conn.connect()
continue
else:
conn.close()
raise
if i < RETRIES-1:
conn.close()
conn.connect()
continue
try:
response = conn.getresponse()
except httplib.BadStatusLine:
# If we get a BadStatusLine on the first try then that means
# the connection just went stale, so retry regardless of the
# number of RETRIES set.
if not seen_bad_status_line and i == 1:
i = 0
seen_bad_status_line = True
conn.close()
conn.connect()
continue
else:
conn.close()
raise
except (socket.error, httplib.HTTPException):
if i < RETRIES-1:
conn.close()
conn.connect()
continue
else:
conn.close()
raise
else:
content = ""
if method == "HEAD":
conn.close()
else:
content = response.read()
response = Response(response)
if method != "HEAD":
content = _decompressContent(response, content)
break
return (response, content)
def _request(self, conn, host, absolute_uri, request_uri, method, body, headers, redirections, cachekey):
"""Do the actual request using the connection object
and also follow one level of redirects if necessary"""
auths = [(auth.depth(request_uri), auth) for auth in self.authorizations if auth.inscope(host, request_uri)]
auth = auths and sorted(auths)[0][1] or None
if auth:
auth.request(method, request_uri, headers, body)
(response, content) = self._conn_request(conn, request_uri, method, body, headers)
if auth:
if auth.response(response, body):
auth.request(method, request_uri, headers, body)
(response, content) = self._conn_request(conn, request_uri, method, body, headers )
response._stale_digest = 1
if response.status == 401:
for authorization in self._auth_from_challenge(host, request_uri, headers, response, content):
authorization.request(method, request_uri, headers, body)
(response, content) = self._conn_request(conn, request_uri, method, body, headers, )
if response.status != 401:
self.authorizations.append(authorization)
authorization.response(response, body)
break
if (self.follow_all_redirects or (method in ["GET", "HEAD"]) or response.status == 303):
if self.follow_redirects and response.status in [300, 301, 302, 303, 307]:
# Pick out the location header and basically start from the beginning
# remembering first to strip the ETag header and decrement our 'depth'
if redirections:
if not response.has_key('location') and response.status != 300:
raise RedirectMissingLocation( _("Redirected but the response is missing a Location: header."), response, content)
# Fix-up relative redirects (which violate an RFC 2616 MUST)
if response.has_key('location'):
location = response['location']
(scheme, authority, path, query, fragment) = parse_uri(location)
if authority == None:
response['location'] = urlparse.urljoin(absolute_uri, location)
if response.status == 301 and method in ["GET", "HEAD"]:
response['-x-permanent-redirect-url'] = response.get('location')
if not response.has_key('content-location'):
response['content-location'] = absolute_uri
_updateCache(headers, response, content, self.cache, cachekey)
if headers.has_key('if-none-match'):
del headers['if-none-match']
if headers.has_key('if-modified-since'):
del headers['if-modified-since']
if 'authorization' in headers and not self.forward_authorization_headers:
del headers['authorization']
if response.has_key('location'):
location = response.get('location')
old_response = copy.deepcopy(response)
if not old_response.has_key('content-location'):
old_response['content-location'] = absolute_uri
redirect_method = method
if response.status in [302, 303]:
redirect_method = "GET"
body = None
(response, content) = self.request(
location, method=redirect_method,
body=body, headers=headers,
redirections=redirections - 1)
response.previous = old_response
else:
raise RedirectLimit("Redirected more times than rediection_limit allows.", response, content)
elif response.status in [200, 203] and method in ["GET", "HEAD"]:
# Don't cache 206's since we aren't going to handle byte range requests
if not response.has_key('content-location'):
response['content-location'] = absolute_uri
_updateCache(headers, response, content, self.cache, cachekey)
return (response, content)
def _normalize_headers(self, headers):
return _normalize_headers(headers)
# Need to catch and rebrand some exceptions
# Then need to optionally turn all exceptions into status codes
# including all socket.* and httplib.* exceptions.
def request(self, uri, method="GET", body=None, headers=None, redirections=DEFAULT_MAX_REDIRECTS, connection_type=None):
""" Performs a single HTTP request.
The 'uri' is the URI of the HTTP resource and can begin with either
'http' or 'https'. The value of 'uri' must be an absolute URI.
The 'method' is the HTTP method to perform, such as GET, POST, DELETE,
etc. There is no restriction on the methods allowed.
The 'body' is the entity body to be sent with the request. It is a
string object.
Any extra headers that are to be sent with the request should be
provided in the 'headers' dictionary.
The maximum number of redirect to follow before raising an
exception is 'redirections. The default is 5.
The return value is a tuple of (response, content), the first
being and instance of the 'Response' class, the second being
a string that contains the response entity body.
"""
try:
if headers is None:
headers = {}
else:
headers = self._normalize_headers(headers)
if not headers.has_key('user-agent'):
headers['user-agent'] = "Python-httplib2/%s (gzip)" % __version__
uri = iri2uri(uri)
(scheme, authority, request_uri, defrag_uri) = urlnorm(uri)
domain_port = authority.split(":")[0:2]
if len(domain_port) == 2 and domain_port[1] == '443' and scheme == 'http':
scheme = 'https'
authority = domain_port[0]
proxy_info = self._get_proxy_info(scheme, authority)
conn_key = scheme+":"+authority
if conn_key in self.connections:
conn = self.connections[conn_key]
else:
if not connection_type:
connection_type = SCHEME_TO_CONNECTION[scheme]
certs = list(self.certificates.iter(authority))
if scheme == 'https':
if certs:
conn = self.connections[conn_key] = connection_type(
authority, key_file=certs[0][0],
cert_file=certs[0][1], timeout=self.timeout,
proxy_info=proxy_info,
ca_certs=self.ca_certs,
disable_ssl_certificate_validation=
self.disable_ssl_certificate_validation)
else:
conn = self.connections[conn_key] = connection_type(
authority, timeout=self.timeout,
proxy_info=proxy_info,
ca_certs=self.ca_certs,
disable_ssl_certificate_validation=
self.disable_ssl_certificate_validation)
else:
conn = self.connections[conn_key] = connection_type(
authority, timeout=self.timeout,
proxy_info=proxy_info)
conn.set_debuglevel(debuglevel)
if 'range' not in headers and 'accept-encoding' not in headers:
headers['accept-encoding'] = 'gzip, deflate'
info = email.Message.Message()
cached_value = None
if self.cache:
cachekey = defrag_uri.encode('utf-8')
cached_value = self.cache.get(cachekey)
if cached_value:
# info = email.message_from_string(cached_value)
#
# Need to replace the line above with the kludge below
# to fix the non-existent bug not fixed in this
# bug report: http://mail.python.org/pipermail/python-bugs-list/2005-September/030289.html
try:
info, content = cached_value.split('\r\n\r\n', 1)
feedparser = email.FeedParser.FeedParser()
feedparser.feed(info)
info = feedparser.close()
feedparser._parse = None
except (IndexError, ValueError):
self.cache.delete(cachekey)
cachekey = None
cached_value = None
else:
cachekey = None
if method in self.optimistic_concurrency_methods and self.cache and info.has_key('etag') and not self.ignore_etag and 'if-match' not in headers:
# http://www.w3.org/1999/04/Editing/
headers['if-match'] = info.get('etag')
if method not in ["GET", "HEAD"] and self.cache and cachekey:
# RFC 2616 Section 13.10
self.cache.delete(cachekey)
# Check the vary header in the cache to see if this request
# matches what varies in the cache.
if method in ['GET', 'HEAD'] and 'vary' in info:
vary = info.get('vary')
vary_headers = vary.lower().replace(' ', '').split(',')
for header in vary_headers:
key = '-varied-%s' % header
value = info[key]
if headers.get(header, None) != value:
cached_value = None
break
if cached_value and method in ["GET", "HEAD"] and self.cache and 'range' not in headers:
if info.has_key('-x-permanent-redirect-url'):
# Should cached permanent redirects be counted in our redirection count? For now, yes.
if redirections <= 0:
raise RedirectLimit("Redirected more times than rediection_limit allows.", {}, "")
(response, new_content) = self.request(
info['-x-permanent-redirect-url'], method='GET',
headers=headers, redirections=redirections - 1)
response.previous = Response(info)
response.previous.fromcache = True
else:
# Determine our course of action:
# Is the cached entry fresh or stale?
# Has the client requested a non-cached response?
#
# There seems to be three possible answers:
# 1. [FRESH] Return the cache entry w/o doing a GET
# 2. [STALE] Do the GET (but add in cache validators if available)
# 3. [TRANSPARENT] Do a GET w/o any cache validators (Cache-Control: no-cache) on the request
entry_disposition = _entry_disposition(info, headers)
if entry_disposition == "FRESH":
if not cached_value:
info['status'] = '504'
content = ""
response = Response(info)
if cached_value:
response.fromcache = True
return (response, content)
if entry_disposition == "STALE":
if info.has_key('etag') and not self.ignore_etag and not 'if-none-match' in headers:
headers['if-none-match'] = info.get('etag')
if info.has_key('last-modified') and not 'last-modified' in headers:
headers['if-modified-since'] = info.get('last-modified')
elif entry_disposition == "TRANSPARENT":
pass
(response, new_content) = self._request(conn, authority, uri, request_uri, method, body, headers, redirections, cachekey)
if response.status == 304 and method == "GET":
# Rewrite the cache entry with the new end-to-end headers
# Take all headers that are in response
# and overwrite their values in info.
# unless they are hop-by-hop, or are listed in the connection header.
for key in _get_end2end_headers(response):
info[key] = response[key]
merged_response = Response(info)
if hasattr(response, "_stale_digest"):
merged_response._stale_digest = response._stale_digest
_updateCache(headers, merged_response, content, self.cache, cachekey)
response = merged_response
response.status = 200
response.fromcache = True
elif response.status == 200:
content = new_content
else:
self.cache.delete(cachekey)
content = new_content
else:
cc = _parse_cache_control(headers)
if cc.has_key('only-if-cached'):
info['status'] = '504'
response = Response(info)
content = ""
else:
(response, content) = self._request(conn, authority, uri, request_uri, method, body, headers, redirections, cachekey)
except Exception, e:
if self.force_exception_to_status_code:
if isinstance(e, HttpLib2ErrorWithResponse):
response = e.response
content = e.content
response.status = 500
response.reason = str(e)
elif isinstance(e, socket.timeout):
content = "Request Timeout"
response = Response({
"content-type": "text/plain",
"status": "408",
"content-length": len(content)
})
response.reason = "Request Timeout"
else:
content = str(e)
response = Response({
"content-type": "text/plain",
"status": "400",
"content-length": len(content)
})
response.reason = "Bad Request"
else:
raise
return (response, content)
def _get_proxy_info(self, scheme, authority):
"""Return a ProxyInfo instance (or None) based on the scheme
and authority.
"""
hostname, port = urllib.splitport(authority)
proxy_info = self.proxy_info
if callable(proxy_info):
proxy_info = proxy_info(scheme)
if (hasattr(proxy_info, 'applies_to')
and not proxy_info.applies_to(hostname)):
proxy_info = None
return proxy_info
class Response(dict):
"""An object more like email.Message than httplib.HTTPResponse."""
"""Is this response from our local cache"""
fromcache = False
"""HTTP protocol version used by server. 10 for HTTP/1.0, 11 for HTTP/1.1. """
version = 11
"Status code returned by server. "
status = 200
"""Reason phrase returned by server."""
reason = "Ok"
previous = None
def __init__(self, info):
# info is either an email.Message or
# an httplib.HTTPResponse object.
if isinstance(info, httplib.HTTPResponse):
for key, value in info.getheaders():
self[key.lower()] = value
self.status = info.status
self['status'] = str(self.status)
self.reason = info.reason
self.version = info.version
elif isinstance(info, email.Message.Message):
for key, value in info.items():
self[key.lower()] = value
self.status = int(self.get('status'))
else:
for key, value in info.iteritems():
self[key.lower()] = value
self.status = int(self.get('status', self.status))
self.reason = self.get('reason', self.reason)
def __getattr__(self, name):
if name == 'dict':
return self
else:
raise AttributeError, name
2、iri2uri.py
"""
iri2uri
Converts an IRI to a URI.
"""
__author__ = "Joe Gregorio (joe@bitworking.org)"
__copyright__ = "Copyright 2006, Joe Gregorio"
__contributors__ = []
__version__ = "1.0.0"
__license__ = "MIT"
__history__ = """
"""
import urlparse
# Convert an IRI to a URI following the rules in RFC 3987
#
# The characters we need to enocde and escape are defined in the spec:
#
# iprivate = %xE000-F8FF / %xF0000-FFFFD / %x100000-10FFFD
# ucschar = %xA0-D7FF / %xF900-FDCF / %xFDF0-FFEF
# / %x10000-1FFFD / %x20000-2FFFD / %x30000-3FFFD
# / %x40000-4FFFD / %x50000-5FFFD / %x60000-6FFFD
# / %x70000-7FFFD / %x80000-8FFFD / %x90000-9FFFD
# / %xA0000-AFFFD / %xB0000-BFFFD / %xC0000-CFFFD
# / %xD0000-DFFFD / %xE1000-EFFFD
escape_range = [
(0xA0, 0xD7FF),
(0xE000, 0xF8FF),
(0xF900, 0xFDCF),
(0xFDF0, 0xFFEF),
(0x10000, 0x1FFFD),
(0x20000, 0x2FFFD),
(0x30000, 0x3FFFD),
(0x40000, 0x4FFFD),
(0x50000, 0x5FFFD),
(0x60000, 0x6FFFD),
(0x70000, 0x7FFFD),
(0x80000, 0x8FFFD),
(0x90000, 0x9FFFD),
(0xA0000, 0xAFFFD),
(0xB0000, 0xBFFFD),
(0xC0000, 0xCFFFD),
(0xD0000, 0xDFFFD),
(0xE1000, 0xEFFFD),
(0xF0000, 0xFFFFD),
(0x100000, 0x10FFFD),
]
def encode(c):
retval = c
i = ord(c)
for low, high in escape_range:
if i < low:
break
if i >= low and i <= high:
retval = "".join(["%%%2X" % ord(o) for o in c.encode('utf-8')])
break
return retval
def iri2uri(uri):
"""Convert an IRI to a URI. Note that IRIs must be
passed in a unicode strings. That is, do not utf-8 encode
the IRI before passing it into the function."""
if isinstance(uri ,unicode):
(scheme, authority, path, query, fragment) = urlparse.urlsplit(uri)
authority = authority.encode('idna')
# For each character in 'ucschar' or 'iprivate'
# 1. encode as utf-8
# 2. then %-encode each octet of that utf-8
uri = urlparse.urlunsplit((scheme, authority, path, query, fragment))
uri = "".join([encode(c) for c in uri])
return uri
if __name__ == "__main__":
import unittest
class Test(unittest.TestCase):
def test_uris(self):
"""Test that URIs are invariant under the transformation."""
invariant = [
u"ftp://ftp.is.co.za/rfc/rfc1808.txt",
u"http://www.ietf.org/rfc/rfc2396.txt",
u"ldap://[2001:db8::7]/c=GB?objectClass?one",
u"mailto:John.Doe@example.com",
u"news:comp.infosystems.www.servers.unix",
u"tel:+1-816-555-1212",
u"telnet://192.0.2.16:80/",
u"urn:oasis:names:specification:docbook:dtd:xml:4.1.2" ]
for uri in invariant:
self.assertEqual(uri, iri2uri(uri))
def test_iri(self):
""" Test that the right type of escaping is done for each part of the URI."""
self.assertEqual("http://xn--o3h.com/%E2%98%84", iri2uri(u"http://\N{COMET}.com/\N{COMET}"))
self.assertEqual("http://bitworking.org/?fred=%E2%98%84", iri2uri(u"http://bitworking.org/?fred=\N{COMET}"))
self.assertEqual("http://bitworking.org/#%E2%98%84", iri2uri(u"http://bitworking.org/#\N{COMET}"))
self.assertEqual("#%E2%98%84", iri2uri(u"#\N{COMET}"))
self.assertEqual("/fred?bar=%E2%98%9A#%E2%98%84", iri2uri(u"/fred?bar=\N{BLACK LEFT POINTING INDEX}#\N{COMET}"))
self.assertEqual("/fred?bar=%E2%98%9A#%E2%98%84", iri2uri(iri2uri(u"/fred?bar=\N{BLACK LEFT POINTING INDEX}#\N{COMET}")))
self.assertNotEqual("/fred?bar=%E2%98%9A#%E2%98%84", iri2uri(u"/fred?bar=\N{BLACK LEFT POINTING INDEX}#\N{COMET}".encode('utf-8')))
unittest.main()
3、socks.py
"""SocksiPy - Python SOCKS module.
Version 1.00
Copyright 2006 Dan-Haim. All rights reserved.
Redistribution and use in source and binary forms, with or without modification,
are permitted provided that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice, this
list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
3. Neither the name of Dan Haim nor the names of his contributors may be used
to endorse or promote products derived from this software without specific
prior written permission.
THIS SOFTWARE IS PROVIDED BY DAN HAIM "AS IS" AND ANY EXPRESS OR IMPLIED
WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
EVENT SHALL DAN HAIM OR HIS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA
OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMANGE.
This module provides a standard socket-like interface for Python
for tunneling connections through SOCKS proxies.
"""
"""
Minor modifications made by Christopher Gilbert (http://motomastyle.com/)
for use in PyLoris (http://pyloris.sourceforge.net/)
Minor modifications made by Mario Vilas (http://breakingcode.wordpress.com/)
mainly to merge bug fixes found in Sourceforge
"""
import base64
import socket
import struct
import sys
if getattr(socket, 'socket', None) is None:
raise ImportError('socket.socket missing, proxy support unusable')
PROXY_TYPE_SOCKS4 = 1
PROXY_TYPE_SOCKS5 = 2
PROXY_TYPE_HTTP = 3
PROXY_TYPE_HTTP_NO_TUNNEL = 4
_defaultproxy = None
_orgsocket = socket.socket
class ProxyError(Exception): pass
class GeneralProxyError(ProxyError): pass
class Socks5AuthError(ProxyError): pass
class Socks5Error(ProxyError): pass
class Socks4Error(ProxyError): pass
class HTTPError(ProxyError): pass
_generalerrors = ("success",
"invalid data",
"not connected",
"not available",
"bad proxy type",
"bad input")
_socks5errors = ("succeeded",
"general SOCKS server failure",
"connection not allowed by ruleset",
"Network unreachable",
"Host unreachable",
"Connection refused",
"TTL expired",
"Command not supported",
"Address type not supported",
"Unknown error")
_socks5autherrors = ("succeeded",
"authentication is required",
"all offered authentication methods were rejected",
"unknown username or invalid password",
"unknown error")
_socks4errors = ("request granted",
"request rejected or failed",
"request rejected because SOCKS server cannot connect to identd on the client",
"request rejected because the client program and identd report different user-ids",
"unknown error")
def setdefaultproxy(proxytype=None, addr=None, port=None, rdns=True, username=None, password=None):
"""setdefaultproxy(proxytype, addr[, port[, rdns[, username[, password]]]])
Sets a default proxy which all further socksocket objects will use,
unless explicitly changed.
"""
global _defaultproxy
_defaultproxy = (proxytype, addr, port, rdns, username, password)
def wrapmodule(module):
"""wrapmodule(module)
Attempts to replace a module's socket library with a SOCKS socket. Must set
a default proxy using setdefaultproxy(...) first.
This will only work on modules that import socket directly into the namespace;
most of the Python Standard Library falls into this category.
"""
if _defaultproxy != None:
module.socket.socket = socksocket
else:
raise GeneralProxyError((4, "no proxy specified"))
class socksocket(socket.socket):
"""socksocket([family[, type[, proto]]]) -> socket object
Open a SOCKS enabled socket. The parameters are the same as
those of the standard socket init. In order for SOCKS to work,
you must specify family=AF_INET, type=SOCK_STREAM and proto=0.
"""
def __init__(self, family=socket.AF_INET, type=socket.SOCK_STREAM, proto=0, _sock=None):
_orgsocket.__init__(self, family, type, proto, _sock)
if _defaultproxy != None:
self.__proxy = _defaultproxy
else:
self.__proxy = (None, None, None, None, None, None)
self.__proxysockname = None
self.__proxypeername = None
self.__httptunnel = True
def __recvall(self, count):
"""__recvall(count) -> data
Receive EXACTLY the number of bytes requested from the socket.
Blocks until the required number of bytes have been received.
"""
data = self.recv(count)
while len(data) < count:
d = self.recv(count-len(data))
if not d: raise GeneralProxyError((0, "connection closed unexpectedly"))
data = data + d
return data
def sendall(self, content, *args):
""" override socket.socket.sendall method to rewrite the header
for non-tunneling proxies if needed
"""
if not self.__httptunnel:
content = self.__rewriteproxy(content)
return super(socksocket, self).sendall(content, *args)
def __rewriteproxy(self, header):
""" rewrite HTTP request headers to support non-tunneling proxies
(i.e. those which do not support the CONNECT method).
This only works for HTTP (not HTTPS) since HTTPS requires tunneling.
"""
host, endpt = None, None
hdrs = header.split("\r\n")
for hdr in hdrs:
if hdr.lower().startswith("host:"):
host = hdr
elif hdr.lower().startswith("get") or hdr.lower().startswith("post"):
endpt = hdr
if host and endpt:
hdrs.remove(host)
hdrs.remove(endpt)
host = host.split(" ")[1]
endpt = endpt.split(" ")
if (self.__proxy[4] != None and self.__proxy[5] != None):
hdrs.insert(0, self.__getauthheader())
hdrs.insert(0, "Host: %s" % host)
hdrs.insert(0, "%s http://%s%s %s" % (endpt[0], host, endpt[1], endpt[2]))
return "\r\n".join(hdrs)
def __getauthheader(self):
auth = self.__proxy[4] + ":" + self.__proxy[5]
return "Proxy-Authorization: Basic " + base64.b64encode(auth)
def setproxy(self, proxytype=None, addr=None, port=None, rdns=True, username=None, password=None):
"""setproxy(proxytype, addr[, port[, rdns[, username[, password]]]])
Sets the proxy to be used.
proxytype - The type of the proxy to be used. Three types
are supported: PROXY_TYPE_SOCKS4 (including socks4a),
PROXY_TYPE_SOCKS5 and PROXY_TYPE_HTTP
addr - The address of the server (IP or DNS).
port - The port of the server. Defaults to 1080 for SOCKS
servers and 8080 for HTTP proxy servers.
rdns - Should DNS queries be preformed on the remote side
(rather than the local side). The default is True.
Note: This has no effect with SOCKS4 servers.
username - Username to authenticate with to the server.
The default is no authentication.
password - Password to authenticate with to the server.
Only relevant when username is also provided.
"""
self.__proxy = (proxytype, addr, port, rdns, username, password)
def __negotiatesocks5(self, destaddr, destport):
"""__negotiatesocks5(self,destaddr,destport)
Negotiates a connection through a SOCKS5 server.
"""
# First we'll send the authentication packages we support.
if (self.__proxy[4]!=None) and (self.__proxy[5]!=None):
# The username/password details were supplied to the
# setproxy method so we support the USERNAME/PASSWORD
# authentication (in addition to the standard none).
self.sendall(struct.pack('BBBB', 0x05, 0x02, 0x00, 0x02))
else:
# No username/password were entered, therefore we
# only support connections with no authentication.
self.sendall(struct.pack('BBB', 0x05, 0x01, 0x00))
# We'll receive the server's response to determine which
# method was selected
chosenauth = self.__recvall(2)
if chosenauth[0:1] != chr(0x05).encode():
self.close()
raise GeneralProxyError((1, _generalerrors[1]))
# Check the chosen authentication method
if chosenauth[1:2] == chr(0x00).encode():
# No authentication is required
pass
elif chosenauth[1:2] == chr(0x02).encode():
# Okay, we need to perform a basic username/password
# authentication.
self.sendall(chr(0x01).encode() + chr(len(self.__proxy[4])) + self.__proxy[4] + chr(len(self.__proxy[5])) + self.__proxy[5])
authstat = self.__recvall(2)
if authstat[0:1] != chr(0x01).encode():
# Bad response
self.close()
raise GeneralProxyError((1, _generalerrors[1]))
if authstat[1:2] != chr(0x00).encode():
# Authentication failed
self.close()
raise Socks5AuthError((3, _socks5autherrors[3]))
# Authentication succeeded
else:
# Reaching here is always bad
self.close()
if chosenauth[1] == chr(0xFF).encode():
raise Socks5AuthError((2, _socks5autherrors[2]))
else:
raise GeneralProxyError((1, _generalerrors[1]))
# Now we can request the actual connection
req = struct.pack('BBB', 0x05, 0x01, 0x00)
# If the given destination address is an IP address, we'll
# use the IPv4 address request even if remote resolving was specified.
try:
ipaddr = socket.inet_aton(destaddr)
req = req + chr(0x01).encode() + ipaddr
except socket.error:
# Well it's not an IP number, so it's probably a DNS name.
if self.__proxy[3]:
# Resolve remotely
ipaddr = None
req = req + chr(0x03).encode() + chr(len(destaddr)).encode() + destaddr
else:
# Resolve locally
ipaddr = socket.inet_aton(socket.gethostbyname(destaddr))
req = req + chr(0x01).encode() + ipaddr
req = req + struct.pack(">H", destport)
self.sendall(req)
# Get the response
resp = self.__recvall(4)
if resp[0:1] != chr(0x05).encode():
self.close()
raise GeneralProxyError((1, _generalerrors[1]))
elif resp[1:2] != chr(0x00).encode():
# Connection failed
self.close()
if ord(resp[1:2])<=8:
raise Socks5Error((ord(resp[1:2]), _socks5errors[ord(resp[1:2])]))
else:
raise Socks5Error((9, _socks5errors[9]))
# Get the bound address/port
elif resp[3:4] == chr(0x01).encode():
boundaddr = self.__recvall(4)
elif resp[3:4] == chr(0x03).encode():
resp = resp + self.recv(1)
boundaddr = self.__recvall(ord(resp[4:5]))
else:
self.close()
raise GeneralProxyError((1,_generalerrors[1]))
boundport = struct.unpack(">H", self.__recvall(2))[0]
self.__proxysockname = (boundaddr, boundport)
if ipaddr != None:
self.__proxypeername = (socket.inet_ntoa(ipaddr), destport)
else:
self.__proxypeername = (destaddr, destport)
def getproxysockname(self):
"""getsockname() -> address info
Returns the bound IP address and port number at the proxy.
"""
return self.__proxysockname
def getproxypeername(self):
"""getproxypeername() -> address info
Returns the IP and port number of the proxy.
"""
return _orgsocket.getpeername(self)
def getpeername(self):
"""getpeername() -> address info
Returns the IP address and port number of the destination
machine (note: getproxypeername returns the proxy)
"""
return self.__proxypeername
def __negotiatesocks4(self,destaddr,destport):
"""__negotiatesocks4(self,destaddr,destport)
Negotiates a connection through a SOCKS4 server.
"""
# Check if the destination address provided is an IP address
rmtrslv = False
try:
ipaddr = socket.inet_aton(destaddr)
except socket.error:
# It's a DNS name. Check where it should be resolved.
if self.__proxy[3]:
ipaddr = struct.pack("BBBB", 0x00, 0x00, 0x00, 0x01)
rmtrslv = True
else:
ipaddr = socket.inet_aton(socket.gethostbyname(destaddr))
# Construct the request packet
req = struct.pack(">BBH", 0x04, 0x01, destport) + ipaddr
# The username parameter is considered userid for SOCKS4
if self.__proxy[4] != None:
req = req + self.__proxy[4]
req = req + chr(0x00).encode()
# DNS name if remote resolving is required
# NOTE: This is actually an extension to the SOCKS4 protocol
# called SOCKS4A and may not be supported in all cases.
if rmtrslv:
req = req + destaddr + chr(0x00).encode()
self.sendall(req)
# Get the response from the server
resp = self.__recvall(8)
if resp[0:1] != chr(0x00).encode():
# Bad data
self.close()
raise GeneralProxyError((1,_generalerrors[1]))
if resp[1:2] != chr(0x5A).encode():
# Server returned an error
self.close()
if ord(resp[1:2]) in (91, 92, 93):
self.close()
raise Socks4Error((ord(resp[1:2]), _socks4errors[ord(resp[1:2]) - 90]))
else:
raise Socks4Error((94, _socks4errors[4]))
# Get the bound address/port
self.__proxysockname = (socket.inet_ntoa(resp[4:]), struct.unpack(">H", resp[2:4])[0])
if rmtrslv != None:
self.__proxypeername = (socket.inet_ntoa(ipaddr), destport)
else:
self.__proxypeername = (destaddr, destport)
def __negotiatehttp(self, destaddr, destport):
"""__negotiatehttp(self,destaddr,destport)
Negotiates a connection through an HTTP server.
"""
# If we need to resolve locally, we do this now
if not self.__proxy[3]:
addr = socket.gethostbyname(destaddr)
else:
addr = destaddr
headers = ["CONNECT ", addr, ":", str(destport), " HTTP/1.1\r\n"]
headers += ["Host: ", destaddr, "\r\n"]
if (self.__proxy[4] != None and self.__proxy[5] != None):
headers += [self.__getauthheader(), "\r\n"]
headers.append("\r\n")
self.sendall("".join(headers).encode())
# We read the response until we get the string "\r\n\r\n"
resp = self.recv(1)
while resp.find("\r\n\r\n".encode()) == -1:
resp = resp + self.recv(1)
# We just need the first line to check if the connection
# was successful
statusline = resp.splitlines()[0].split(" ".encode(), 2)
if statusline[0] not in ("HTTP/1.0".encode(), "HTTP/1.1".encode()):
self.close()
raise GeneralProxyError((1, _generalerrors[1]))
try:
statuscode = int(statusline[1])
except ValueError:
self.close()
raise GeneralProxyError((1, _generalerrors[1]))
if statuscode != 200:
self.close()
raise HTTPError((statuscode, statusline[2]))
self.__proxysockname = ("0.0.0.0", 0)
self.__proxypeername = (addr, destport)
def connect(self, destpair):
"""connect(self, despair)
Connects to the specified destination through a proxy.
destpar - A tuple of the IP/DNS address and the port number.
(identical to socket's connect).
To select the proxy server use setproxy().
"""
# Do a minimal input check first
if (not type(destpair) in (list,tuple)) or (len(destpair) < 2) or (not isinstance(destpair[0], basestring)) or (type(destpair[1]) != int):
raise GeneralProxyError((5, _generalerrors[5]))
if self.__proxy[0] == PROXY_TYPE_SOCKS5:
if self.__proxy[2] != None:
portnum = self.__proxy[2]
else:
portnum = 1080
_orgsocket.connect(self, (self.__proxy[1], portnum))
self.__negotiatesocks5(destpair[0], destpair[1])
elif self.__proxy[0] == PROXY_TYPE_SOCKS4:
if self.__proxy[2] != None:
portnum = self.__proxy[2]
else:
portnum = 1080
_orgsocket.connect(self,(self.__proxy[1], portnum))
self.__negotiatesocks4(destpair[0], destpair[1])
elif self.__proxy[0] == PROXY_TYPE_HTTP:
if self.__proxy[2] != None:
portnum = self.__proxy[2]
else:
portnum = 8080
_orgsocket.connect(self,(self.__proxy[1], portnum))
self.__negotiatehttp(destpair[0], destpair[1])
elif self.__proxy[0] == PROXY_TYPE_HTTP_NO_TUNNEL:
if self.__proxy[2] != None:
portnum = self.__proxy[2]
else:
portnum = 8080
_orgsocket.connect(self,(self.__proxy[1],portnum))
if destpair[1] == 443:
self.__negotiatehttp(destpair[0],destpair[1])
else:
self.__httptunnel = False
elif self.__proxy[0] == None:
_orgsocket.connect(self, (destpair[0], destpair[1]))
else:
raise GeneralProxyError((4, _generalerrors[4]))
IPy子目录
odict子目录
termcolor子目录
未完，待续！
3354
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
