first commit

This commit is contained in:
Victor
2023-11-16 16:57:13 +01:00
commit a8fb1fd811
1477 changed files with 275005 additions and 0 deletions

View File

@@ -0,0 +1,102 @@
"""
Python HTTP library with thread-safe connection pooling, file post support, user friendly, and more
"""
from __future__ import absolute_import
# Set default logging handler to avoid "No handler found" warnings.
import logging
import warnings
from logging import NullHandler
from . import exceptions
from ._version import __version__
from .connectionpool import HTTPConnectionPool, HTTPSConnectionPool, connection_from_url
from .filepost import encode_multipart_formdata
from .poolmanager import PoolManager, ProxyManager, proxy_from_url
from .response import HTTPResponse
from .util.request import make_headers
from .util.retry import Retry
from .util.timeout import Timeout
from .util.url import get_host
# === NOTE TO REPACKAGERS AND VENDORS ===
# Please delete this block, this logic is only
# for urllib3 being distributed via PyPI.
# See: https://github.com/urllib3/urllib3/issues/2680
try:
import urllib3_secure_extra # type: ignore # noqa: F401
except ImportError:
pass
else:
warnings.warn(
"'urllib3[secure]' extra is deprecated and will be removed "
"in a future release of urllib3 2.x. Read more in this issue: "
"https://github.com/urllib3/urllib3/issues/2680",
category=DeprecationWarning,
stacklevel=2,
)
__author__ = "Andrey Petrov (andrey.petrov@shazow.net)"
__license__ = "MIT"
__version__ = __version__
__all__ = (
"HTTPConnectionPool",
"HTTPSConnectionPool",
"PoolManager",
"ProxyManager",
"HTTPResponse",
"Retry",
"Timeout",
"add_stderr_logger",
"connection_from_url",
"disable_warnings",
"encode_multipart_formdata",
"get_host",
"make_headers",
"proxy_from_url",
)
logging.getLogger(__name__).addHandler(NullHandler())
def add_stderr_logger(level=logging.DEBUG):
"""
Helper for quickly adding a StreamHandler to the logger. Useful for
debugging.
Returns the handler after adding it.
"""
# This method needs to be in this __init__.py to get the __name__ correct
# even if urllib3 is vendored within another package.
logger = logging.getLogger(__name__)
handler = logging.StreamHandler()
handler.setFormatter(logging.Formatter("%(asctime)s %(levelname)s %(message)s"))
logger.addHandler(handler)
logger.setLevel(level)
logger.debug("Added a stderr logging handler to logger: %s", __name__)
return handler
# ... Clean up.
del NullHandler
# All warning filters *must* be appended unless you're really certain that they
# shouldn't be: otherwise, it's very hard for users to use most Python
# mechanisms to silence them.
# SecurityWarning's always go off by default.
warnings.simplefilter("always", exceptions.SecurityWarning, append=True)
# SubjectAltNameWarning's should go off once per host
warnings.simplefilter("default", exceptions.SubjectAltNameWarning, append=True)
# InsecurePlatformWarning's don't vary between requests, so we keep it default.
warnings.simplefilter("default", exceptions.InsecurePlatformWarning, append=True)
# SNIMissingWarnings should go off only once.
warnings.simplefilter("default", exceptions.SNIMissingWarning, append=True)
def disable_warnings(category=exceptions.HTTPWarning):
"""
Helper for quickly disabling all urllib3 warnings.
"""
warnings.simplefilter("ignore", category)

View File

@@ -0,0 +1,337 @@
from __future__ import absolute_import
try:
from collections.abc import Mapping, MutableMapping
except ImportError:
from collections import Mapping, MutableMapping
try:
from threading import RLock
except ImportError: # Platform-specific: No threads available
class RLock:
def __enter__(self):
pass
def __exit__(self, exc_type, exc_value, traceback):
pass
from collections import OrderedDict
from .exceptions import InvalidHeader
from .packages import six
from .packages.six import iterkeys, itervalues
__all__ = ["RecentlyUsedContainer", "HTTPHeaderDict"]
_Null = object()
class RecentlyUsedContainer(MutableMapping):
"""
Provides a thread-safe dict-like container which maintains up to
``maxsize`` keys while throwing away the least-recently-used keys beyond
``maxsize``.
:param maxsize:
Maximum number of recent elements to retain.
:param dispose_func:
Every time an item is evicted from the container,
``dispose_func(value)`` is called. Callback which will get called
"""
ContainerCls = OrderedDict
def __init__(self, maxsize=10, dispose_func=None):
self._maxsize = maxsize
self.dispose_func = dispose_func
self._container = self.ContainerCls()
self.lock = RLock()
def __getitem__(self, key):
# Re-insert the item, moving it to the end of the eviction line.
with self.lock:
item = self._container.pop(key)
self._container[key] = item
return item
def __setitem__(self, key, value):
evicted_value = _Null
with self.lock:
# Possibly evict the existing value of 'key'
evicted_value = self._container.get(key, _Null)
self._container[key] = value
# If we didn't evict an existing value, we might have to evict the
# least recently used item from the beginning of the container.
if len(self._container) > self._maxsize:
_key, evicted_value = self._container.popitem(last=False)
if self.dispose_func and evicted_value is not _Null:
self.dispose_func(evicted_value)
def __delitem__(self, key):
with self.lock:
value = self._container.pop(key)
if self.dispose_func:
self.dispose_func(value)
def __len__(self):
with self.lock:
return len(self._container)
def __iter__(self):
raise NotImplementedError(
"Iteration over this class is unlikely to be threadsafe."
)
def clear(self):
with self.lock:
# Copy pointers to all values, then wipe the mapping
values = list(itervalues(self._container))
self._container.clear()
if self.dispose_func:
for value in values:
self.dispose_func(value)
def keys(self):
with self.lock:
return list(iterkeys(self._container))
class HTTPHeaderDict(MutableMapping):
"""
:param headers:
An iterable of field-value pairs. Must not contain multiple field names
when compared case-insensitively.
:param kwargs:
Additional field-value pairs to pass in to ``dict.update``.
A ``dict`` like container for storing HTTP Headers.
Field names are stored and compared case-insensitively in compliance with
RFC 7230. Iteration provides the first case-sensitive key seen for each
case-insensitive pair.
Using ``__setitem__`` syntax overwrites fields that compare equal
case-insensitively in order to maintain ``dict``'s api. For fields that
compare equal, instead create a new ``HTTPHeaderDict`` and use ``.add``
in a loop.
If multiple fields that are equal case-insensitively are passed to the
constructor or ``.update``, the behavior is undefined and some will be
lost.
>>> headers = HTTPHeaderDict()
>>> headers.add('Set-Cookie', 'foo=bar')
>>> headers.add('set-cookie', 'baz=quxx')
>>> headers['content-length'] = '7'
>>> headers['SET-cookie']
'foo=bar, baz=quxx'
>>> headers['Content-Length']
'7'
"""
def __init__(self, headers=None, **kwargs):
super(HTTPHeaderDict, self).__init__()
self._container = OrderedDict()
if headers is not None:
if isinstance(headers, HTTPHeaderDict):
self._copy_from(headers)
else:
self.extend(headers)
if kwargs:
self.extend(kwargs)
def __setitem__(self, key, val):
self._container[key.lower()] = [key, val]
return self._container[key.lower()]
def __getitem__(self, key):
val = self._container[key.lower()]
return ", ".join(val[1:])
def __delitem__(self, key):
del self._container[key.lower()]
def __contains__(self, key):
return key.lower() in self._container
def __eq__(self, other):
if not isinstance(other, Mapping) and not hasattr(other, "keys"):
return False
if not isinstance(other, type(self)):
other = type(self)(other)
return dict((k.lower(), v) for k, v in self.itermerged()) == dict(
(k.lower(), v) for k, v in other.itermerged()
)
def __ne__(self, other):
return not self.__eq__(other)
if six.PY2: # Python 2
iterkeys = MutableMapping.iterkeys
itervalues = MutableMapping.itervalues
__marker = object()
def __len__(self):
return len(self._container)
def __iter__(self):
# Only provide the originally cased names
for vals in self._container.values():
yield vals[0]
def pop(self, key, default=__marker):
"""D.pop(k[,d]) -> v, remove specified key and return the corresponding value.
If key is not found, d is returned if given, otherwise KeyError is raised.
"""
# Using the MutableMapping function directly fails due to the private marker.
# Using ordinary dict.pop would expose the internal structures.
# So let's reinvent the wheel.
try:
value = self[key]
except KeyError:
if default is self.__marker:
raise
return default
else:
del self[key]
return value
def discard(self, key):
try:
del self[key]
except KeyError:
pass
def add(self, key, val):
"""Adds a (name, value) pair, doesn't overwrite the value if it already
exists.
>>> headers = HTTPHeaderDict(foo='bar')
>>> headers.add('Foo', 'baz')
>>> headers['foo']
'bar, baz'
"""
key_lower = key.lower()
new_vals = [key, val]
# Keep the common case aka no item present as fast as possible
vals = self._container.setdefault(key_lower, new_vals)
if new_vals is not vals:
vals.append(val)
def extend(self, *args, **kwargs):
"""Generic import function for any type of header-like object.
Adapted version of MutableMapping.update in order to insert items
with self.add instead of self.__setitem__
"""
if len(args) > 1:
raise TypeError(
"extend() takes at most 1 positional "
"arguments ({0} given)".format(len(args))
)
other = args[0] if len(args) >= 1 else ()
if isinstance(other, HTTPHeaderDict):
for key, val in other.iteritems():
self.add(key, val)
elif isinstance(other, Mapping):
for key in other:
self.add(key, other[key])
elif hasattr(other, "keys"):
for key in other.keys():
self.add(key, other[key])
else:
for key, value in other:
self.add(key, value)
for key, value in kwargs.items():
self.add(key, value)
def getlist(self, key, default=__marker):
"""Returns a list of all the values for the named field. Returns an
empty list if the key doesn't exist."""
try:
vals = self._container[key.lower()]
except KeyError:
if default is self.__marker:
return []
return default
else:
return vals[1:]
# Backwards compatibility for httplib
getheaders = getlist
getallmatchingheaders = getlist
iget = getlist
# Backwards compatibility for http.cookiejar
get_all = getlist
def __repr__(self):
return "%s(%s)" % (type(self).__name__, dict(self.itermerged()))
def _copy_from(self, other):
for key in other:
val = other.getlist(key)
if isinstance(val, list):
# Don't need to convert tuples
val = list(val)
self._container[key.lower()] = [key] + val
def copy(self):
clone = type(self)()
clone._copy_from(self)
return clone
def iteritems(self):
"""Iterate over all header lines, including duplicate ones."""
for key in self:
vals = self._container[key.lower()]
for val in vals[1:]:
yield vals[0], val
def itermerged(self):
"""Iterate over all headers, merging duplicate ones together."""
for key in self:
val = self._container[key.lower()]
yield val[0], ", ".join(val[1:])
def items(self):
return list(self.iteritems())
@classmethod
def from_httplib(cls, message): # Python 2
"""Read headers from a Python 2 httplib message object."""
# python2.7 does not expose a proper API for exporting multiheaders
# efficiently. This function re-reads raw lines from the message
# object and extracts the multiheaders properly.
obs_fold_continued_leaders = (" ", "\t")
headers = []
for line in message.headers:
if line.startswith(obs_fold_continued_leaders):
if not headers:
# We received a header line that starts with OWS as described
# in RFC-7230 S3.2.4. This indicates a multiline header, but
# there exists no previous header to which we can attach it.
raise InvalidHeader(
"Header continuation with no previous header: %s" % line
)
else:
key, value = headers[-1]
headers[-1] = (key, value + " " + line.strip())
continue
key, value = line.split(":", 1)
headers.append((key, value.strip()))
return cls(headers)

View File

@@ -0,0 +1,2 @@
# This file is protected via CODEOWNERS
__version__ = "1.26.16"

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,36 @@
"""
This module provides means to detect the App Engine environment.
"""
import os
def is_appengine():
return is_local_appengine() or is_prod_appengine()
def is_appengine_sandbox():
"""Reports if the app is running in the first generation sandbox.
The second generation runtimes are technically still in a sandbox, but it
is much less restrictive, so generally you shouldn't need to check for it.
see https://cloud.google.com/appengine/docs/standard/runtimes
"""
return is_appengine() and os.environ["APPENGINE_RUNTIME"] == "python27"
def is_local_appengine():
return "APPENGINE_RUNTIME" in os.environ and os.environ.get(
"SERVER_SOFTWARE", ""
).startswith("Development/")
def is_prod_appengine():
return "APPENGINE_RUNTIME" in os.environ and os.environ.get(
"SERVER_SOFTWARE", ""
).startswith("Google App Engine/")
def is_prod_appengine_mvms():
"""Deprecated."""
return False

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,397 @@
"""
Low-level helpers for the SecureTransport bindings.
These are Python functions that are not directly related to the high-level APIs
but are necessary to get them to work. They include a whole bunch of low-level
CoreFoundation messing about and memory management. The concerns in this module
are almost entirely about trying to avoid memory leaks and providing
appropriate and useful assistance to the higher-level code.
"""
import base64
import ctypes
import itertools
import os
import re
import ssl
import struct
import tempfile
from .bindings import CFConst, CoreFoundation, Security
# This regular expression is used to grab PEM data out of a PEM bundle.
_PEM_CERTS_RE = re.compile(
b"-----BEGIN CERTIFICATE-----\n(.*?)\n-----END CERTIFICATE-----", re.DOTALL
)
def _cf_data_from_bytes(bytestring):
"""
Given a bytestring, create a CFData object from it. This CFData object must
be CFReleased by the caller.
"""
return CoreFoundation.CFDataCreate(
CoreFoundation.kCFAllocatorDefault, bytestring, len(bytestring)
)
def _cf_dictionary_from_tuples(tuples):
"""
Given a list of Python tuples, create an associated CFDictionary.
"""
dictionary_size = len(tuples)
# We need to get the dictionary keys and values out in the same order.
keys = (t[0] for t in tuples)
values = (t[1] for t in tuples)
cf_keys = (CoreFoundation.CFTypeRef * dictionary_size)(*keys)
cf_values = (CoreFoundation.CFTypeRef * dictionary_size)(*values)
return CoreFoundation.CFDictionaryCreate(
CoreFoundation.kCFAllocatorDefault,
cf_keys,
cf_values,
dictionary_size,
CoreFoundation.kCFTypeDictionaryKeyCallBacks,
CoreFoundation.kCFTypeDictionaryValueCallBacks,
)
def _cfstr(py_bstr):
"""
Given a Python binary data, create a CFString.
The string must be CFReleased by the caller.
"""
c_str = ctypes.c_char_p(py_bstr)
cf_str = CoreFoundation.CFStringCreateWithCString(
CoreFoundation.kCFAllocatorDefault,
c_str,
CFConst.kCFStringEncodingUTF8,
)
return cf_str
def _create_cfstring_array(lst):
"""
Given a list of Python binary data, create an associated CFMutableArray.
The array must be CFReleased by the caller.
Raises an ssl.SSLError on failure.
"""
cf_arr = None
try:
cf_arr = CoreFoundation.CFArrayCreateMutable(
CoreFoundation.kCFAllocatorDefault,
0,
ctypes.byref(CoreFoundation.kCFTypeArrayCallBacks),
)
if not cf_arr:
raise MemoryError("Unable to allocate memory!")
for item in lst:
cf_str = _cfstr(item)
if not cf_str:
raise MemoryError("Unable to allocate memory!")
try:
CoreFoundation.CFArrayAppendValue(cf_arr, cf_str)
finally:
CoreFoundation.CFRelease(cf_str)
except BaseException as e:
if cf_arr:
CoreFoundation.CFRelease(cf_arr)
raise ssl.SSLError("Unable to allocate array: %s" % (e,))
return cf_arr
def _cf_string_to_unicode(value):
"""
Creates a Unicode string from a CFString object. Used entirely for error
reporting.
Yes, it annoys me quite a lot that this function is this complex.
"""
value_as_void_p = ctypes.cast(value, ctypes.POINTER(ctypes.c_void_p))
string = CoreFoundation.CFStringGetCStringPtr(
value_as_void_p, CFConst.kCFStringEncodingUTF8
)
if string is None:
buffer = ctypes.create_string_buffer(1024)
result = CoreFoundation.CFStringGetCString(
value_as_void_p, buffer, 1024, CFConst.kCFStringEncodingUTF8
)
if not result:
raise OSError("Error copying C string from CFStringRef")
string = buffer.value
if string is not None:
string = string.decode("utf-8")
return string
def _assert_no_error(error, exception_class=None):
"""
Checks the return code and throws an exception if there is an error to
report
"""
if error == 0:
return
cf_error_string = Security.SecCopyErrorMessageString(error, None)
output = _cf_string_to_unicode(cf_error_string)
CoreFoundation.CFRelease(cf_error_string)
if output is None or output == u"":
output = u"OSStatus %s" % error
if exception_class is None:
exception_class = ssl.SSLError
raise exception_class(output)
def _cert_array_from_pem(pem_bundle):
"""
Given a bundle of certs in PEM format, turns them into a CFArray of certs
that can be used to validate a cert chain.
"""
# Normalize the PEM bundle's line endings.
pem_bundle = pem_bundle.replace(b"\r\n", b"\n")
der_certs = [
base64.b64decode(match.group(1)) for match in _PEM_CERTS_RE.finditer(pem_bundle)
]
if not der_certs:
raise ssl.SSLError("No root certificates specified")
cert_array = CoreFoundation.CFArrayCreateMutable(
CoreFoundation.kCFAllocatorDefault,
0,
ctypes.byref(CoreFoundation.kCFTypeArrayCallBacks),
)
if not cert_array:
raise ssl.SSLError("Unable to allocate memory!")
try:
for der_bytes in der_certs:
certdata = _cf_data_from_bytes(der_bytes)
if not certdata:
raise ssl.SSLError("Unable to allocate memory!")
cert = Security.SecCertificateCreateWithData(
CoreFoundation.kCFAllocatorDefault, certdata
)
CoreFoundation.CFRelease(certdata)
if not cert:
raise ssl.SSLError("Unable to build cert object!")
CoreFoundation.CFArrayAppendValue(cert_array, cert)
CoreFoundation.CFRelease(cert)
except Exception:
# We need to free the array before the exception bubbles further.
# We only want to do that if an error occurs: otherwise, the caller
# should free.
CoreFoundation.CFRelease(cert_array)
raise
return cert_array
def _is_cert(item):
"""
Returns True if a given CFTypeRef is a certificate.
"""
expected = Security.SecCertificateGetTypeID()
return CoreFoundation.CFGetTypeID(item) == expected
def _is_identity(item):
"""
Returns True if a given CFTypeRef is an identity.
"""
expected = Security.SecIdentityGetTypeID()
return CoreFoundation.CFGetTypeID(item) == expected
def _temporary_keychain():
"""
This function creates a temporary Mac keychain that we can use to work with
credentials. This keychain uses a one-time password and a temporary file to
store the data. We expect to have one keychain per socket. The returned
SecKeychainRef must be freed by the caller, including calling
SecKeychainDelete.
Returns a tuple of the SecKeychainRef and the path to the temporary
directory that contains it.
"""
# Unfortunately, SecKeychainCreate requires a path to a keychain. This
# means we cannot use mkstemp to use a generic temporary file. Instead,
# we're going to create a temporary directory and a filename to use there.
# This filename will be 8 random bytes expanded into base64. We also need
# some random bytes to password-protect the keychain we're creating, so we
# ask for 40 random bytes.
random_bytes = os.urandom(40)
filename = base64.b16encode(random_bytes[:8]).decode("utf-8")
password = base64.b16encode(random_bytes[8:]) # Must be valid UTF-8
tempdirectory = tempfile.mkdtemp()
keychain_path = os.path.join(tempdirectory, filename).encode("utf-8")
# We now want to create the keychain itself.
keychain = Security.SecKeychainRef()
status = Security.SecKeychainCreate(
keychain_path, len(password), password, False, None, ctypes.byref(keychain)
)
_assert_no_error(status)
# Having created the keychain, we want to pass it off to the caller.
return keychain, tempdirectory
def _load_items_from_file(keychain, path):
"""
Given a single file, loads all the trust objects from it into arrays and
the keychain.
Returns a tuple of lists: the first list is a list of identities, the
second a list of certs.
"""
certificates = []
identities = []
result_array = None
with open(path, "rb") as f:
raw_filedata = f.read()
try:
filedata = CoreFoundation.CFDataCreate(
CoreFoundation.kCFAllocatorDefault, raw_filedata, len(raw_filedata)
)
result_array = CoreFoundation.CFArrayRef()
result = Security.SecItemImport(
filedata, # cert data
None, # Filename, leaving it out for now
None, # What the type of the file is, we don't care
None, # what's in the file, we don't care
0, # import flags
None, # key params, can include passphrase in the future
keychain, # The keychain to insert into
ctypes.byref(result_array), # Results
)
_assert_no_error(result)
# A CFArray is not very useful to us as an intermediary
# representation, so we are going to extract the objects we want
# and then free the array. We don't need to keep hold of keys: the
# keychain already has them!
result_count = CoreFoundation.CFArrayGetCount(result_array)
for index in range(result_count):
item = CoreFoundation.CFArrayGetValueAtIndex(result_array, index)
item = ctypes.cast(item, CoreFoundation.CFTypeRef)
if _is_cert(item):
CoreFoundation.CFRetain(item)
certificates.append(item)
elif _is_identity(item):
CoreFoundation.CFRetain(item)
identities.append(item)
finally:
if result_array:
CoreFoundation.CFRelease(result_array)
CoreFoundation.CFRelease(filedata)
return (identities, certificates)
def _load_client_cert_chain(keychain, *paths):
"""
Load certificates and maybe keys from a number of files. Has the end goal
of returning a CFArray containing one SecIdentityRef, and then zero or more
SecCertificateRef objects, suitable for use as a client certificate trust
chain.
"""
# Ok, the strategy.
#
# This relies on knowing that macOS will not give you a SecIdentityRef
# unless you have imported a key into a keychain. This is a somewhat
# artificial limitation of macOS (for example, it doesn't necessarily
# affect iOS), but there is nothing inside Security.framework that lets you
# get a SecIdentityRef without having a key in a keychain.
#
# So the policy here is we take all the files and iterate them in order.
# Each one will use SecItemImport to have one or more objects loaded from
# it. We will also point at a keychain that macOS can use to work with the
# private key.
#
# Once we have all the objects, we'll check what we actually have. If we
# already have a SecIdentityRef in hand, fab: we'll use that. Otherwise,
# we'll take the first certificate (which we assume to be our leaf) and
# ask the keychain to give us a SecIdentityRef with that cert's associated
# key.
#
# We'll then return a CFArray containing the trust chain: one
# SecIdentityRef and then zero-or-more SecCertificateRef objects. The
# responsibility for freeing this CFArray will be with the caller. This
# CFArray must remain alive for the entire connection, so in practice it
# will be stored with a single SSLSocket, along with the reference to the
# keychain.
certificates = []
identities = []
# Filter out bad paths.
paths = (path for path in paths if path)
try:
for file_path in paths:
new_identities, new_certs = _load_items_from_file(keychain, file_path)
identities.extend(new_identities)
certificates.extend(new_certs)
# Ok, we have everything. The question is: do we have an identity? If
# not, we want to grab one from the first cert we have.
if not identities:
new_identity = Security.SecIdentityRef()
status = Security.SecIdentityCreateWithCertificate(
keychain, certificates[0], ctypes.byref(new_identity)
)
_assert_no_error(status)
identities.append(new_identity)
# We now want to release the original certificate, as we no longer
# need it.
CoreFoundation.CFRelease(certificates.pop(0))
# We now need to build a new CFArray that holds the trust chain.
trust_chain = CoreFoundation.CFArrayCreateMutable(
CoreFoundation.kCFAllocatorDefault,
0,
ctypes.byref(CoreFoundation.kCFTypeArrayCallBacks),
)
for item in itertools.chain(identities, certificates):
# ArrayAppendValue does a CFRetain on the item. That's fine,
# because the finally block will release our other refs to them.
CoreFoundation.CFArrayAppendValue(trust_chain, item)
return trust_chain
finally:
for obj in itertools.chain(identities, certificates):
CoreFoundation.CFRelease(obj)
TLS_PROTOCOL_VERSIONS = {
"SSLv2": (0, 2),
"SSLv3": (3, 0),
"TLSv1": (3, 1),
"TLSv1.1": (3, 2),
"TLSv1.2": (3, 3),
}
def _build_tls_unknown_ca_alert(version):
"""
Builds a TLS alert record for an unknown CA.
"""
ver_maj, ver_min = TLS_PROTOCOL_VERSIONS[version]
severity_fatal = 0x02
description_unknown_ca = 0x30
msg = struct.pack(">BB", severity_fatal, description_unknown_ca)
msg_len = len(msg)
record_type_alert = 0x15
record = struct.pack(">BBBH", record_type_alert, ver_maj, ver_min, msg_len) + msg
return record

View File

@@ -0,0 +1,314 @@
"""
This module provides a pool manager that uses Google App Engine's
`URLFetch Service <https://cloud.google.com/appengine/docs/python/urlfetch>`_.
Example usage::
from pip._vendor.urllib3 import PoolManager
from pip._vendor.urllib3.contrib.appengine import AppEngineManager, is_appengine_sandbox
if is_appengine_sandbox():
# AppEngineManager uses AppEngine's URLFetch API behind the scenes
http = AppEngineManager()
else:
# PoolManager uses a socket-level API behind the scenes
http = PoolManager()
r = http.request('GET', 'https://google.com/')
There are `limitations <https://cloud.google.com/appengine/docs/python/\
urlfetch/#Python_Quotas_and_limits>`_ to the URLFetch service and it may not be
the best choice for your application. There are three options for using
urllib3 on Google App Engine:
1. You can use :class:`AppEngineManager` with URLFetch. URLFetch is
cost-effective in many circumstances as long as your usage is within the
limitations.
2. You can use a normal :class:`~urllib3.PoolManager` by enabling sockets.
Sockets also have `limitations and restrictions
<https://cloud.google.com/appengine/docs/python/sockets/\
#limitations-and-restrictions>`_ and have a lower free quota than URLFetch.
To use sockets, be sure to specify the following in your ``app.yaml``::
env_variables:
GAE_USE_SOCKETS_HTTPLIB : 'true'
3. If you are using `App Engine Flexible
<https://cloud.google.com/appengine/docs/flexible/>`_, you can use the standard
:class:`PoolManager` without any configuration or special environment variables.
"""
from __future__ import absolute_import
import io
import logging
import warnings
from ..exceptions import (
HTTPError,
HTTPWarning,
MaxRetryError,
ProtocolError,
SSLError,
TimeoutError,
)
from ..packages.six.moves.urllib.parse import urljoin
from ..request import RequestMethods
from ..response import HTTPResponse
from ..util.retry import Retry
from ..util.timeout import Timeout
from . import _appengine_environ
try:
from google.appengine.api import urlfetch
except ImportError:
urlfetch = None
log = logging.getLogger(__name__)
class AppEnginePlatformWarning(HTTPWarning):
pass
class AppEnginePlatformError(HTTPError):
pass
class AppEngineManager(RequestMethods):
"""
Connection manager for Google App Engine sandbox applications.
This manager uses the URLFetch service directly instead of using the
emulated httplib, and is subject to URLFetch limitations as described in
the App Engine documentation `here
<https://cloud.google.com/appengine/docs/python/urlfetch>`_.
Notably it will raise an :class:`AppEnginePlatformError` if:
* URLFetch is not available.
* If you attempt to use this on App Engine Flexible, as full socket
support is available.
* If a request size is more than 10 megabytes.
* If a response size is more than 32 megabytes.
* If you use an unsupported request method such as OPTIONS.
Beyond those cases, it will raise normal urllib3 errors.
"""
def __init__(
self,
headers=None,
retries=None,
validate_certificate=True,
urlfetch_retries=True,
):
if not urlfetch:
raise AppEnginePlatformError(
"URLFetch is not available in this environment."
)
warnings.warn(
"urllib3 is using URLFetch on Google App Engine sandbox instead "
"of sockets. To use sockets directly instead of URLFetch see "
"https://urllib3.readthedocs.io/en/1.26.x/reference/urllib3.contrib.html.",
AppEnginePlatformWarning,
)
RequestMethods.__init__(self, headers)
self.validate_certificate = validate_certificate
self.urlfetch_retries = urlfetch_retries
self.retries = retries or Retry.DEFAULT
def __enter__(self):
return self
def __exit__(self, exc_type, exc_val, exc_tb):
# Return False to re-raise any potential exceptions
return False
def urlopen(
self,
method,
url,
body=None,
headers=None,
retries=None,
redirect=True,
timeout=Timeout.DEFAULT_TIMEOUT,
**response_kw
):
retries = self._get_retries(retries, redirect)
try:
follow_redirects = redirect and retries.redirect != 0 and retries.total
response = urlfetch.fetch(
url,
payload=body,
method=method,
headers=headers or {},
allow_truncated=False,
follow_redirects=self.urlfetch_retries and follow_redirects,
deadline=self._get_absolute_timeout(timeout),
validate_certificate=self.validate_certificate,
)
except urlfetch.DeadlineExceededError as e:
raise TimeoutError(self, e)
except urlfetch.InvalidURLError as e:
if "too large" in str(e):
raise AppEnginePlatformError(
"URLFetch request too large, URLFetch only "
"supports requests up to 10mb in size.",
e,
)
raise ProtocolError(e)
except urlfetch.DownloadError as e:
if "Too many redirects" in str(e):
raise MaxRetryError(self, url, reason=e)
raise ProtocolError(e)
except urlfetch.ResponseTooLargeError as e:
raise AppEnginePlatformError(
"URLFetch response too large, URLFetch only supports"
"responses up to 32mb in size.",
e,
)
except urlfetch.SSLCertificateError as e:
raise SSLError(e)
except urlfetch.InvalidMethodError as e:
raise AppEnginePlatformError(
"URLFetch does not support method: %s" % method, e
)
http_response = self._urlfetch_response_to_http_response(
response, retries=retries, **response_kw
)
# Handle redirect?
redirect_location = redirect and http_response.get_redirect_location()
if redirect_location:
# Check for redirect response
if self.urlfetch_retries and retries.raise_on_redirect:
raise MaxRetryError(self, url, "too many redirects")
else:
if http_response.status == 303:
method = "GET"
try:
retries = retries.increment(
method, url, response=http_response, _pool=self
)
except MaxRetryError:
if retries.raise_on_redirect:
raise MaxRetryError(self, url, "too many redirects")
return http_response
retries.sleep_for_retry(http_response)
log.debug("Redirecting %s -> %s", url, redirect_location)
redirect_url = urljoin(url, redirect_location)
return self.urlopen(
method,
redirect_url,
body,
headers,
retries=retries,
redirect=redirect,
timeout=timeout,
**response_kw
)
# Check if we should retry the HTTP response.
has_retry_after = bool(http_response.headers.get("Retry-After"))
if retries.is_retry(method, http_response.status, has_retry_after):
retries = retries.increment(method, url, response=http_response, _pool=self)
log.debug("Retry: %s", url)
retries.sleep(http_response)
return self.urlopen(
method,
url,
body=body,
headers=headers,
retries=retries,
redirect=redirect,
timeout=timeout,
**response_kw
)
return http_response
def _urlfetch_response_to_http_response(self, urlfetch_resp, **response_kw):
if is_prod_appengine():
# Production GAE handles deflate encoding automatically, but does
# not remove the encoding header.
content_encoding = urlfetch_resp.headers.get("content-encoding")
if content_encoding == "deflate":
del urlfetch_resp.headers["content-encoding"]
transfer_encoding = urlfetch_resp.headers.get("transfer-encoding")
# We have a full response's content,
# so let's make sure we don't report ourselves as chunked data.
if transfer_encoding == "chunked":
encodings = transfer_encoding.split(",")
encodings.remove("chunked")
urlfetch_resp.headers["transfer-encoding"] = ",".join(encodings)
original_response = HTTPResponse(
# In order for decoding to work, we must present the content as
# a file-like object.
body=io.BytesIO(urlfetch_resp.content),
msg=urlfetch_resp.header_msg,
headers=urlfetch_resp.headers,
status=urlfetch_resp.status_code,
**response_kw
)
return HTTPResponse(
body=io.BytesIO(urlfetch_resp.content),
headers=urlfetch_resp.headers,
status=urlfetch_resp.status_code,
original_response=original_response,
**response_kw
)
def _get_absolute_timeout(self, timeout):
if timeout is Timeout.DEFAULT_TIMEOUT:
return None # Defer to URLFetch's default.
if isinstance(timeout, Timeout):
if timeout._read is not None or timeout._connect is not None:
warnings.warn(
"URLFetch does not support granular timeout settings, "
"reverting to total or default URLFetch timeout.",
AppEnginePlatformWarning,
)
return timeout.total
return timeout
def _get_retries(self, retries, redirect):
if not isinstance(retries, Retry):
retries = Retry.from_int(retries, redirect=redirect, default=self.retries)
if retries.connect or retries.read or retries.redirect:
warnings.warn(
"URLFetch only supports total retries and does not "
"recognize connect, read, or redirect retry parameters.",
AppEnginePlatformWarning,
)
return retries
# Alias methods from _appengine_environ to maintain public API interface.
is_appengine = _appengine_environ.is_appengine
is_appengine_sandbox = _appengine_environ.is_appengine_sandbox
is_local_appengine = _appengine_environ.is_local_appengine
is_prod_appengine = _appengine_environ.is_prod_appengine
is_prod_appengine_mvms = _appengine_environ.is_prod_appengine_mvms

View File

@@ -0,0 +1,130 @@
"""
NTLM authenticating pool, contributed by erikcederstran
Issue #10, see: http://code.google.com/p/urllib3/issues/detail?id=10
"""
from __future__ import absolute_import
import warnings
from logging import getLogger
from ntlm import ntlm
from .. import HTTPSConnectionPool
from ..packages.six.moves.http_client import HTTPSConnection
warnings.warn(
"The 'urllib3.contrib.ntlmpool' module is deprecated and will be removed "
"in urllib3 v2.0 release, urllib3 is not able to support it properly due "
"to reasons listed in issue: https://github.com/urllib3/urllib3/issues/2282. "
"If you are a user of this module please comment in the mentioned issue.",
DeprecationWarning,
)
log = getLogger(__name__)
class NTLMConnectionPool(HTTPSConnectionPool):
"""
Implements an NTLM authentication version of an urllib3 connection pool
"""
scheme = "https"
def __init__(self, user, pw, authurl, *args, **kwargs):
"""
authurl is a random URL on the server that is protected by NTLM.
user is the Windows user, probably in the DOMAIN\\username format.
pw is the password for the user.
"""
super(NTLMConnectionPool, self).__init__(*args, **kwargs)
self.authurl = authurl
self.rawuser = user
user_parts = user.split("\\", 1)
self.domain = user_parts[0].upper()
self.user = user_parts[1]
self.pw = pw
def _new_conn(self):
# Performs the NTLM handshake that secures the connection. The socket
# must be kept open while requests are performed.
self.num_connections += 1
log.debug(
"Starting NTLM HTTPS connection no. %d: https://%s%s",
self.num_connections,
self.host,
self.authurl,
)
headers = {"Connection": "Keep-Alive"}
req_header = "Authorization"
resp_header = "www-authenticate"
conn = HTTPSConnection(host=self.host, port=self.port)
# Send negotiation message
headers[req_header] = "NTLM %s" % ntlm.create_NTLM_NEGOTIATE_MESSAGE(
self.rawuser
)
log.debug("Request headers: %s", headers)
conn.request("GET", self.authurl, None, headers)
res = conn.getresponse()
reshdr = dict(res.headers)
log.debug("Response status: %s %s", res.status, res.reason)
log.debug("Response headers: %s", reshdr)
log.debug("Response data: %s [...]", res.read(100))
# Remove the reference to the socket, so that it can not be closed by
# the response object (we want to keep the socket open)
res.fp = None
# Server should respond with a challenge message
auth_header_values = reshdr[resp_header].split(", ")
auth_header_value = None
for s in auth_header_values:
if s[:5] == "NTLM ":
auth_header_value = s[5:]
if auth_header_value is None:
raise Exception(
"Unexpected %s response header: %s" % (resp_header, reshdr[resp_header])
)
# Send authentication message
ServerChallenge, NegotiateFlags = ntlm.parse_NTLM_CHALLENGE_MESSAGE(
auth_header_value
)
auth_msg = ntlm.create_NTLM_AUTHENTICATE_MESSAGE(
ServerChallenge, self.user, self.domain, self.pw, NegotiateFlags
)
headers[req_header] = "NTLM %s" % auth_msg
log.debug("Request headers: %s", headers)
conn.request("GET", self.authurl, None, headers)
res = conn.getresponse()
log.debug("Response status: %s %s", res.status, res.reason)
log.debug("Response headers: %s", dict(res.headers))
log.debug("Response data: %s [...]", res.read()[:100])
if res.status != 200:
if res.status == 401:
raise Exception("Server rejected request: wrong username or password")
raise Exception("Wrong server response: %s %s" % (res.status, res.reason))
res.fp = None
log.debug("Connection established")
return conn
def urlopen(
self,
method,
url,
body=None,
headers=None,
retries=3,
redirect=True,
assert_same_host=True,
):
if headers is None:
headers = {}
headers["Connection"] = "Keep-Alive"
return super(NTLMConnectionPool, self).urlopen(
method, url, body, headers, retries, redirect, assert_same_host
)

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,216 @@
# -*- coding: utf-8 -*-
"""
This module contains provisional support for SOCKS proxies from within
urllib3. This module supports SOCKS4, SOCKS4A (an extension of SOCKS4), and
SOCKS5. To enable its functionality, either install PySocks or install this
module with the ``socks`` extra.
The SOCKS implementation supports the full range of urllib3 features. It also
supports the following SOCKS features:
- SOCKS4A (``proxy_url='socks4a://...``)
- SOCKS4 (``proxy_url='socks4://...``)
- SOCKS5 with remote DNS (``proxy_url='socks5h://...``)
- SOCKS5 with local DNS (``proxy_url='socks5://...``)
- Usernames and passwords for the SOCKS proxy
.. note::
It is recommended to use ``socks5h://`` or ``socks4a://`` schemes in
your ``proxy_url`` to ensure that DNS resolution is done from the remote
server instead of client-side when connecting to a domain name.
SOCKS4 supports IPv4 and domain names with the SOCKS4A extension. SOCKS5
supports IPv4, IPv6, and domain names.
When connecting to a SOCKS4 proxy the ``username`` portion of the ``proxy_url``
will be sent as the ``userid`` section of the SOCKS request:
.. code-block:: python
proxy_url="socks4a://<userid>@proxy-host"
When connecting to a SOCKS5 proxy the ``username`` and ``password`` portion
of the ``proxy_url`` will be sent as the username/password to authenticate
with the proxy:
.. code-block:: python
proxy_url="socks5h://<username>:<password>@proxy-host"
"""
from __future__ import absolute_import
try:
import socks
except ImportError:
import warnings
from ..exceptions import DependencyWarning
warnings.warn(
(
"SOCKS support in urllib3 requires the installation of optional "
"dependencies: specifically, PySocks. For more information, see "
"https://urllib3.readthedocs.io/en/1.26.x/contrib.html#socks-proxies"
),
DependencyWarning,
)
raise
from socket import error as SocketError
from socket import timeout as SocketTimeout
from ..connection import HTTPConnection, HTTPSConnection
from ..connectionpool import HTTPConnectionPool, HTTPSConnectionPool
from ..exceptions import ConnectTimeoutError, NewConnectionError
from ..poolmanager import PoolManager
from ..util.url import parse_url
try:
import ssl
except ImportError:
ssl = None
class SOCKSConnection(HTTPConnection):
"""
A plain-text HTTP connection that connects via a SOCKS proxy.
"""
def __init__(self, *args, **kwargs):
self._socks_options = kwargs.pop("_socks_options")
super(SOCKSConnection, self).__init__(*args, **kwargs)
def _new_conn(self):
"""
Establish a new connection via the SOCKS proxy.
"""
extra_kw = {}
if self.source_address:
extra_kw["source_address"] = self.source_address
if self.socket_options:
extra_kw["socket_options"] = self.socket_options
try:
conn = socks.create_connection(
(self.host, self.port),
proxy_type=self._socks_options["socks_version"],
proxy_addr=self._socks_options["proxy_host"],
proxy_port=self._socks_options["proxy_port"],
proxy_username=self._socks_options["username"],
proxy_password=self._socks_options["password"],
proxy_rdns=self._socks_options["rdns"],
timeout=self.timeout,
**extra_kw
)
except SocketTimeout:
raise ConnectTimeoutError(
self,
"Connection to %s timed out. (connect timeout=%s)"
% (self.host, self.timeout),
)
except socks.ProxyError as e:
# This is fragile as hell, but it seems to be the only way to raise
# useful errors here.
if e.socket_err:
error = e.socket_err
if isinstance(error, SocketTimeout):
raise ConnectTimeoutError(
self,
"Connection to %s timed out. (connect timeout=%s)"
% (self.host, self.timeout),
)
else:
raise NewConnectionError(
self, "Failed to establish a new connection: %s" % error
)
else:
raise NewConnectionError(
self, "Failed to establish a new connection: %s" % e
)
except SocketError as e: # Defensive: PySocks should catch all these.
raise NewConnectionError(
self, "Failed to establish a new connection: %s" % e
)
return conn
# We don't need to duplicate the Verified/Unverified distinction from
# urllib3/connection.py here because the HTTPSConnection will already have been
# correctly set to either the Verified or Unverified form by that module. This
# means the SOCKSHTTPSConnection will automatically be the correct type.
class SOCKSHTTPSConnection(SOCKSConnection, HTTPSConnection):
pass
class SOCKSHTTPConnectionPool(HTTPConnectionPool):
ConnectionCls = SOCKSConnection
class SOCKSHTTPSConnectionPool(HTTPSConnectionPool):
ConnectionCls = SOCKSHTTPSConnection
class SOCKSProxyManager(PoolManager):
"""
A version of the urllib3 ProxyManager that routes connections via the
defined SOCKS proxy.
"""
pool_classes_by_scheme = {
"http": SOCKSHTTPConnectionPool,
"https": SOCKSHTTPSConnectionPool,
}
def __init__(
self,
proxy_url,
username=None,
password=None,
num_pools=10,
headers=None,
**connection_pool_kw
):
parsed = parse_url(proxy_url)
if username is None and password is None and parsed.auth is not None:
split = parsed.auth.split(":")
if len(split) == 2:
username, password = split
if parsed.scheme == "socks5":
socks_version = socks.PROXY_TYPE_SOCKS5
rdns = False
elif parsed.scheme == "socks5h":
socks_version = socks.PROXY_TYPE_SOCKS5
rdns = True
elif parsed.scheme == "socks4":
socks_version = socks.PROXY_TYPE_SOCKS4
rdns = False
elif parsed.scheme == "socks4a":
socks_version = socks.PROXY_TYPE_SOCKS4
rdns = True
else:
raise ValueError("Unable to determine SOCKS version from %s" % proxy_url)
self.proxy_url = proxy_url
socks_options = {
"socks_version": socks_version,
"proxy_host": parsed.host,
"proxy_port": parsed.port,
"username": username,
"password": password,
"rdns": rdns,
}
connection_pool_kw["_socks_options"] = socks_options
super(SOCKSProxyManager, self).__init__(
num_pools, headers, **connection_pool_kw
)
self.pool_classes_by_scheme = SOCKSProxyManager.pool_classes_by_scheme

View File

@@ -0,0 +1,323 @@
from __future__ import absolute_import
from .packages.six.moves.http_client import IncompleteRead as httplib_IncompleteRead
# Base Exceptions
class HTTPError(Exception):
"""Base exception used by this module."""
pass
class HTTPWarning(Warning):
"""Base warning used by this module."""
pass
class PoolError(HTTPError):
"""Base exception for errors caused within a pool."""
def __init__(self, pool, message):
self.pool = pool
HTTPError.__init__(self, "%s: %s" % (pool, message))
def __reduce__(self):
# For pickling purposes.
return self.__class__, (None, None)
class RequestError(PoolError):
"""Base exception for PoolErrors that have associated URLs."""
def __init__(self, pool, url, message):
self.url = url
PoolError.__init__(self, pool, message)
def __reduce__(self):
# For pickling purposes.
return self.__class__, (None, self.url, None)
class SSLError(HTTPError):
"""Raised when SSL certificate fails in an HTTPS connection."""
pass
class ProxyError(HTTPError):
"""Raised when the connection to a proxy fails."""
def __init__(self, message, error, *args):
super(ProxyError, self).__init__(message, error, *args)
self.original_error = error
class DecodeError(HTTPError):
"""Raised when automatic decoding based on Content-Type fails."""
pass
class ProtocolError(HTTPError):
"""Raised when something unexpected happens mid-request/response."""
pass
#: Renamed to ProtocolError but aliased for backwards compatibility.
ConnectionError = ProtocolError
# Leaf Exceptions
class MaxRetryError(RequestError):
"""Raised when the maximum number of retries is exceeded.
:param pool: The connection pool
:type pool: :class:`~urllib3.connectionpool.HTTPConnectionPool`
:param string url: The requested Url
:param exceptions.Exception reason: The underlying error
"""
def __init__(self, pool, url, reason=None):
self.reason = reason
message = "Max retries exceeded with url: %s (Caused by %r)" % (url, reason)
RequestError.__init__(self, pool, url, message)
class HostChangedError(RequestError):
"""Raised when an existing pool gets a request for a foreign host."""
def __init__(self, pool, url, retries=3):
message = "Tried to open a foreign host with url: %s" % url
RequestError.__init__(self, pool, url, message)
self.retries = retries
class TimeoutStateError(HTTPError):
"""Raised when passing an invalid state to a timeout"""
pass
class TimeoutError(HTTPError):
"""Raised when a socket timeout error occurs.
Catching this error will catch both :exc:`ReadTimeoutErrors
<ReadTimeoutError>` and :exc:`ConnectTimeoutErrors <ConnectTimeoutError>`.
"""
pass
class ReadTimeoutError(TimeoutError, RequestError):
"""Raised when a socket timeout occurs while receiving data from a server"""
pass
# This timeout error does not have a URL attached and needs to inherit from the
# base HTTPError
class ConnectTimeoutError(TimeoutError):
"""Raised when a socket timeout occurs while connecting to a server"""
pass
class NewConnectionError(ConnectTimeoutError, PoolError):
"""Raised when we fail to establish a new connection. Usually ECONNREFUSED."""
pass
class EmptyPoolError(PoolError):
"""Raised when a pool runs out of connections and no more are allowed."""
pass
class ClosedPoolError(PoolError):
"""Raised when a request enters a pool after the pool has been closed."""
pass
class LocationValueError(ValueError, HTTPError):
"""Raised when there is something wrong with a given URL input."""
pass
class LocationParseError(LocationValueError):
"""Raised when get_host or similar fails to parse the URL input."""
def __init__(self, location):
message = "Failed to parse: %s" % location
HTTPError.__init__(self, message)
self.location = location
class URLSchemeUnknown(LocationValueError):
"""Raised when a URL input has an unsupported scheme."""
def __init__(self, scheme):
message = "Not supported URL scheme %s" % scheme
super(URLSchemeUnknown, self).__init__(message)
self.scheme = scheme
class ResponseError(HTTPError):
"""Used as a container for an error reason supplied in a MaxRetryError."""
GENERIC_ERROR = "too many error responses"
SPECIFIC_ERROR = "too many {status_code} error responses"
class SecurityWarning(HTTPWarning):
"""Warned when performing security reducing actions"""
pass
class SubjectAltNameWarning(SecurityWarning):
"""Warned when connecting to a host with a certificate missing a SAN."""
pass
class InsecureRequestWarning(SecurityWarning):
"""Warned when making an unverified HTTPS request."""
pass
class SystemTimeWarning(SecurityWarning):
"""Warned when system time is suspected to be wrong"""
pass
class InsecurePlatformWarning(SecurityWarning):
"""Warned when certain TLS/SSL configuration is not available on a platform."""
pass
class SNIMissingWarning(HTTPWarning):
"""Warned when making a HTTPS request without SNI available."""
pass
class DependencyWarning(HTTPWarning):
"""
Warned when an attempt is made to import a module with missing optional
dependencies.
"""
pass
class ResponseNotChunked(ProtocolError, ValueError):
"""Response needs to be chunked in order to read it as chunks."""
pass
class BodyNotHttplibCompatible(HTTPError):
"""
Body should be :class:`http.client.HTTPResponse` like
(have an fp attribute which returns raw chunks) for read_chunked().
"""
pass
class IncompleteRead(HTTPError, httplib_IncompleteRead):
"""
Response length doesn't match expected Content-Length
Subclass of :class:`http.client.IncompleteRead` to allow int value
for ``partial`` to avoid creating large objects on streamed reads.
"""
def __init__(self, partial, expected):
super(IncompleteRead, self).__init__(partial, expected)
def __repr__(self):
return "IncompleteRead(%i bytes read, %i more expected)" % (
self.partial,
self.expected,
)
class InvalidChunkLength(HTTPError, httplib_IncompleteRead):
"""Invalid chunk length in a chunked response."""
def __init__(self, response, length):
super(InvalidChunkLength, self).__init__(
response.tell(), response.length_remaining
)
self.response = response
self.length = length
def __repr__(self):
return "InvalidChunkLength(got length %r, %i bytes read)" % (
self.length,
self.partial,
)
class InvalidHeader(HTTPError):
"""The header provided was somehow invalid."""
pass
class ProxySchemeUnknown(AssertionError, URLSchemeUnknown):
"""ProxyManager does not support the supplied scheme"""
# TODO(t-8ch): Stop inheriting from AssertionError in v2.0.
def __init__(self, scheme):
# 'localhost' is here because our URL parser parses
# localhost:8080 -> scheme=localhost, remove if we fix this.
if scheme == "localhost":
scheme = None
if scheme is None:
message = "Proxy URL had no scheme, should start with http:// or https://"
else:
message = (
"Proxy URL had unsupported scheme %s, should use http:// or https://"
% scheme
)
super(ProxySchemeUnknown, self).__init__(message)
class ProxySchemeUnsupported(ValueError):
"""Fetching HTTPS resources through HTTPS proxies is unsupported"""
pass
class HeaderParsingError(HTTPError):
"""Raised by assert_header_parsing, but we convert it to a log.warning statement."""
def __init__(self, defects, unparsed_data):
message = "%s, unparsed data: %r" % (defects or "Unknown", unparsed_data)
super(HeaderParsingError, self).__init__(message)
class UnrewindableBodyError(HTTPError):
"""urllib3 encountered an error when trying to rewind a body"""
pass

View File

@@ -0,0 +1,274 @@
from __future__ import absolute_import
import email.utils
import mimetypes
import re
from .packages import six
def guess_content_type(filename, default="application/octet-stream"):
"""
Guess the "Content-Type" of a file.
:param filename:
The filename to guess the "Content-Type" of using :mod:`mimetypes`.
:param default:
If no "Content-Type" can be guessed, default to `default`.
"""
if filename:
return mimetypes.guess_type(filename)[0] or default
return default
def format_header_param_rfc2231(name, value):
"""
Helper function to format and quote a single header parameter using the
strategy defined in RFC 2231.
Particularly useful for header parameters which might contain
non-ASCII values, like file names. This follows
`RFC 2388 Section 4.4 <https://tools.ietf.org/html/rfc2388#section-4.4>`_.
:param name:
The name of the parameter, a string expected to be ASCII only.
:param value:
The value of the parameter, provided as ``bytes`` or `str``.
:ret:
An RFC-2231-formatted unicode string.
"""
if isinstance(value, six.binary_type):
value = value.decode("utf-8")
if not any(ch in value for ch in '"\\\r\n'):
result = u'%s="%s"' % (name, value)
try:
result.encode("ascii")
except (UnicodeEncodeError, UnicodeDecodeError):
pass
else:
return result
if six.PY2: # Python 2:
value = value.encode("utf-8")
# encode_rfc2231 accepts an encoded string and returns an ascii-encoded
# string in Python 2 but accepts and returns unicode strings in Python 3
value = email.utils.encode_rfc2231(value, "utf-8")
value = "%s*=%s" % (name, value)
if six.PY2: # Python 2:
value = value.decode("utf-8")
return value
_HTML5_REPLACEMENTS = {
u"\u0022": u"%22",
# Replace "\" with "\\".
u"\u005C": u"\u005C\u005C",
}
# All control characters from 0x00 to 0x1F *except* 0x1B.
_HTML5_REPLACEMENTS.update(
{
six.unichr(cc): u"%{:02X}".format(cc)
for cc in range(0x00, 0x1F + 1)
if cc not in (0x1B,)
}
)
def _replace_multiple(value, needles_and_replacements):
def replacer(match):
return needles_and_replacements[match.group(0)]
pattern = re.compile(
r"|".join([re.escape(needle) for needle in needles_and_replacements.keys()])
)
result = pattern.sub(replacer, value)
return result
def format_header_param_html5(name, value):
"""
Helper function to format and quote a single header parameter using the
HTML5 strategy.
Particularly useful for header parameters which might contain
non-ASCII values, like file names. This follows the `HTML5 Working Draft
Section 4.10.22.7`_ and matches the behavior of curl and modern browsers.
.. _HTML5 Working Draft Section 4.10.22.7:
https://w3c.github.io/html/sec-forms.html#multipart-form-data
:param name:
The name of the parameter, a string expected to be ASCII only.
:param value:
The value of the parameter, provided as ``bytes`` or `str``.
:ret:
A unicode string, stripped of troublesome characters.
"""
if isinstance(value, six.binary_type):
value = value.decode("utf-8")
value = _replace_multiple(value, _HTML5_REPLACEMENTS)
return u'%s="%s"' % (name, value)
# For backwards-compatibility.
format_header_param = format_header_param_html5
class RequestField(object):
"""
A data container for request body parameters.
:param name:
The name of this request field. Must be unicode.
:param data:
The data/value body.
:param filename:
An optional filename of the request field. Must be unicode.
:param headers:
An optional dict-like object of headers to initially use for the field.
:param header_formatter:
An optional callable that is used to encode and format the headers. By
default, this is :func:`format_header_param_html5`.
"""
def __init__(
self,
name,
data,
filename=None,
headers=None,
header_formatter=format_header_param_html5,
):
self._name = name
self._filename = filename
self.data = data
self.headers = {}
if headers:
self.headers = dict(headers)
self.header_formatter = header_formatter
@classmethod
def from_tuples(cls, fieldname, value, header_formatter=format_header_param_html5):
"""
A :class:`~urllib3.fields.RequestField` factory from old-style tuple parameters.
Supports constructing :class:`~urllib3.fields.RequestField` from
parameter of key/value strings AND key/filetuple. A filetuple is a
(filename, data, MIME type) tuple where the MIME type is optional.
For example::
'foo': 'bar',
'fakefile': ('foofile.txt', 'contents of foofile'),
'realfile': ('barfile.txt', open('realfile').read()),
'typedfile': ('bazfile.bin', open('bazfile').read(), 'image/jpeg'),
'nonamefile': 'contents of nonamefile field',
Field names and filenames must be unicode.
"""
if isinstance(value, tuple):
if len(value) == 3:
filename, data, content_type = value
else:
filename, data = value
content_type = guess_content_type(filename)
else:
filename = None
content_type = None
data = value
request_param = cls(
fieldname, data, filename=filename, header_formatter=header_formatter
)
request_param.make_multipart(content_type=content_type)
return request_param
def _render_part(self, name, value):
"""
Overridable helper function to format a single header parameter. By
default, this calls ``self.header_formatter``.
:param name:
The name of the parameter, a string expected to be ASCII only.
:param value:
The value of the parameter, provided as a unicode string.
"""
return self.header_formatter(name, value)
def _render_parts(self, header_parts):
"""
Helper function to format and quote a single header.
Useful for single headers that are composed of multiple items. E.g.,
'Content-Disposition' fields.
:param header_parts:
A sequence of (k, v) tuples or a :class:`dict` of (k, v) to format
as `k1="v1"; k2="v2"; ...`.
"""
parts = []
iterable = header_parts
if isinstance(header_parts, dict):
iterable = header_parts.items()
for name, value in iterable:
if value is not None:
parts.append(self._render_part(name, value))
return u"; ".join(parts)
def render_headers(self):
"""
Renders the headers for this request field.
"""
lines = []
sort_keys = ["Content-Disposition", "Content-Type", "Content-Location"]
for sort_key in sort_keys:
if self.headers.get(sort_key, False):
lines.append(u"%s: %s" % (sort_key, self.headers[sort_key]))
for header_name, header_value in self.headers.items():
if header_name not in sort_keys:
if header_value:
lines.append(u"%s: %s" % (header_name, header_value))
lines.append(u"\r\n")
return u"\r\n".join(lines)
def make_multipart(
self, content_disposition=None, content_type=None, content_location=None
):
"""
Makes this request field into a multipart request field.
This method overrides "Content-Disposition", "Content-Type" and
"Content-Location" headers to the request parameter.
:param content_type:
The 'Content-Type' of the request body.
:param content_location:
The 'Content-Location' of the request body.
"""
self.headers["Content-Disposition"] = content_disposition or u"form-data"
self.headers["Content-Disposition"] += u"; ".join(
[
u"",
self._render_parts(
((u"name", self._name), (u"filename", self._filename))
),
]
)
self.headers["Content-Type"] = content_type
self.headers["Content-Location"] = content_location

View File

@@ -0,0 +1,98 @@
from __future__ import absolute_import
import binascii
import codecs
import os
from io import BytesIO
from .fields import RequestField
from .packages import six
from .packages.six import b
writer = codecs.lookup("utf-8")[3]
def choose_boundary():
"""
Our embarrassingly-simple replacement for mimetools.choose_boundary.
"""
boundary = binascii.hexlify(os.urandom(16))
if not six.PY2:
boundary = boundary.decode("ascii")
return boundary
def iter_field_objects(fields):
"""
Iterate over fields.
Supports list of (k, v) tuples and dicts, and lists of
:class:`~urllib3.fields.RequestField`.
"""
if isinstance(fields, dict):
i = six.iteritems(fields)
else:
i = iter(fields)
for field in i:
if isinstance(field, RequestField):
yield field
else:
yield RequestField.from_tuples(*field)
def iter_fields(fields):
"""
.. deprecated:: 1.6
Iterate over fields.
The addition of :class:`~urllib3.fields.RequestField` makes this function
obsolete. Instead, use :func:`iter_field_objects`, which returns
:class:`~urllib3.fields.RequestField` objects.
Supports list of (k, v) tuples and dicts.
"""
if isinstance(fields, dict):
return ((k, v) for k, v in six.iteritems(fields))
return ((k, v) for k, v in fields)
def encode_multipart_formdata(fields, boundary=None):
"""
Encode a dictionary of ``fields`` using the multipart/form-data MIME format.
:param fields:
Dictionary of fields or list of (key, :class:`~urllib3.fields.RequestField`).
:param boundary:
If not specified, then a random boundary will be generated using
:func:`urllib3.filepost.choose_boundary`.
"""
body = BytesIO()
if boundary is None:
boundary = choose_boundary()
for field in iter_field_objects(fields):
body.write(b("--%s\r\n" % (boundary)))
writer(body).write(field.render_headers())
data = field.data
if isinstance(data, int):
data = str(data) # Backwards compatibility
if isinstance(data, six.text_type):
writer(body).write(data)
else:
body.write(data)
body.write(b"\r\n")
body.write(b("--%s--\r\n" % (boundary)))
content_type = str("multipart/form-data; boundary=%s" % boundary)
return body.getvalue(), content_type

View File

@@ -0,0 +1,51 @@
# -*- coding: utf-8 -*-
"""
backports.makefile
~~~~~~~~~~~~~~~~~~
Backports the Python 3 ``socket.makefile`` method for use with anything that
wants to create a "fake" socket object.
"""
import io
from socket import SocketIO
def backport_makefile(
self, mode="r", buffering=None, encoding=None, errors=None, newline=None
):
"""
Backport of ``socket.makefile`` from Python 3.5.
"""
if not set(mode) <= {"r", "w", "b"}:
raise ValueError("invalid mode %r (only r, w, b allowed)" % (mode,))
writing = "w" in mode
reading = "r" in mode or not writing
assert reading or writing
binary = "b" in mode
rawmode = ""
if reading:
rawmode += "r"
if writing:
rawmode += "w"
raw = SocketIO(self, rawmode)
self._makefile_refs += 1
if buffering is None:
buffering = -1
if buffering < 0:
buffering = io.DEFAULT_BUFFER_SIZE
if buffering == 0:
if not binary:
raise ValueError("unbuffered streams must be binary")
return raw
if reading and writing:
buffer = io.BufferedRWPair(raw, raw, buffering)
elif reading:
buffer = io.BufferedReader(raw, buffering)
else:
assert writing
buffer = io.BufferedWriter(raw, buffering)
if binary:
return buffer
text = io.TextIOWrapper(buffer, encoding, errors, newline)
text.mode = mode
return text

View File

@@ -0,0 +1,155 @@
# -*- coding: utf-8 -*-
"""
backports.weakref_finalize
~~~~~~~~~~~~~~~~~~
Backports the Python 3 ``weakref.finalize`` method.
"""
from __future__ import absolute_import
import itertools
import sys
from weakref import ref
__all__ = ["weakref_finalize"]
class weakref_finalize(object):
"""Class for finalization of weakrefable objects
finalize(obj, func, *args, **kwargs) returns a callable finalizer
object which will be called when obj is garbage collected. The
first time the finalizer is called it evaluates func(*arg, **kwargs)
and returns the result. After this the finalizer is dead, and
calling it just returns None.
When the program exits any remaining finalizers for which the
atexit attribute is true will be run in reverse order of creation.
By default atexit is true.
"""
# Finalizer objects don't have any state of their own. They are
# just used as keys to lookup _Info objects in the registry. This
# ensures that they cannot be part of a ref-cycle.
__slots__ = ()
_registry = {}
_shutdown = False
_index_iter = itertools.count()
_dirty = False
_registered_with_atexit = False
class _Info(object):
__slots__ = ("weakref", "func", "args", "kwargs", "atexit", "index")
def __init__(self, obj, func, *args, **kwargs):
if not self._registered_with_atexit:
# We may register the exit function more than once because
# of a thread race, but that is harmless
import atexit
atexit.register(self._exitfunc)
weakref_finalize._registered_with_atexit = True
info = self._Info()
info.weakref = ref(obj, self)
info.func = func
info.args = args
info.kwargs = kwargs or None
info.atexit = True
info.index = next(self._index_iter)
self._registry[self] = info
weakref_finalize._dirty = True
def __call__(self, _=None):
"""If alive then mark as dead and return func(*args, **kwargs);
otherwise return None"""
info = self._registry.pop(self, None)
if info and not self._shutdown:
return info.func(*info.args, **(info.kwargs or {}))
def detach(self):
"""If alive then mark as dead and return (obj, func, args, kwargs);
otherwise return None"""
info = self._registry.get(self)
obj = info and info.weakref()
if obj is not None and self._registry.pop(self, None):
return (obj, info.func, info.args, info.kwargs or {})
def peek(self):
"""If alive then return (obj, func, args, kwargs);
otherwise return None"""
info = self._registry.get(self)
obj = info and info.weakref()
if obj is not None:
return (obj, info.func, info.args, info.kwargs or {})
@property
def alive(self):
"""Whether finalizer is alive"""
return self in self._registry
@property
def atexit(self):
"""Whether finalizer should be called at exit"""
info = self._registry.get(self)
return bool(info) and info.atexit
@atexit.setter
def atexit(self, value):
info = self._registry.get(self)
if info:
info.atexit = bool(value)
def __repr__(self):
info = self._registry.get(self)
obj = info and info.weakref()
if obj is None:
return "<%s object at %#x; dead>" % (type(self).__name__, id(self))
else:
return "<%s object at %#x; for %r at %#x>" % (
type(self).__name__,
id(self),
type(obj).__name__,
id(obj),
)
@classmethod
def _select_for_exit(cls):
# Return live finalizers marked for exit, oldest first
L = [(f, i) for (f, i) in cls._registry.items() if i.atexit]
L.sort(key=lambda item: item[1].index)
return [f for (f, i) in L]
@classmethod
def _exitfunc(cls):
# At shutdown invoke finalizers for which atexit is true.
# This is called once all other non-daemonic threads have been
# joined.
reenable_gc = False
try:
if cls._registry:
import gc
if gc.isenabled():
reenable_gc = True
gc.disable()
pending = None
while True:
if pending is None or weakref_finalize._dirty:
pending = cls._select_for_exit()
weakref_finalize._dirty = False
if not pending:
break
f = pending.pop()
try:
# gc is disabled, so (assuming no daemonic
# threads) the following is the only line in
# this function which might trigger creation
# of a new finalizer
f()
except Exception:
sys.excepthook(*sys.exc_info())
assert f not in cls._registry
finally:
# prevent any more finalizers from executing during shutdown
weakref_finalize._shutdown = True
if reenable_gc:
gc.enable()

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

Some files were not shown because too many files have changed in this diff Show More