From 7c271a466fe557f7c883f6105728f73f29cdb293 Mon Sep 17 00:00:00 2001 From: Kenneth Reitz Date: Sun, 23 Oct 2011 16:18:52 -0400 Subject: [PATCH] Import urllib3 --- requests/packages/urllib3/__init__.py | 48 +++ requests/packages/urllib3/_collections.py | 120 ++++++ requests/packages/urllib3/connectionpool.py | 525 ++++++++++++++++++++++++++ requests/packages/urllib3/contrib/__init__.py | 0 requests/packages/urllib3/contrib/ntlmpool.py | 117 ++++++ requests/packages/urllib3/exceptions.py | 35 ++ requests/packages/urllib3/filepost.py | 71 ++++ requests/packages/urllib3/poolmanager.py | 128 +++++++ requests/packages/urllib3/request.py | 145 +++++++ requests/packages/urllib3/response.py | 181 +++++++++ 10 files changed, 1370 insertions(+) create mode 100644 requests/packages/urllib3/__init__.py create mode 100644 requests/packages/urllib3/_collections.py create mode 100644 requests/packages/urllib3/connectionpool.py create mode 100644 requests/packages/urllib3/contrib/__init__.py create mode 100644 requests/packages/urllib3/contrib/ntlmpool.py create mode 100644 requests/packages/urllib3/exceptions.py create mode 100644 requests/packages/urllib3/filepost.py create mode 100644 requests/packages/urllib3/poolmanager.py create mode 100644 requests/packages/urllib3/request.py create mode 100644 requests/packages/urllib3/response.py diff --git a/requests/packages/urllib3/__init__.py b/requests/packages/urllib3/__init__.py new file mode 100644 index 0000000..7a2d018 --- /dev/null +++ b/requests/packages/urllib3/__init__.py @@ -0,0 +1,48 @@ +# urllib3/__init__.py +# Copyright 2008-2011 Andrey Petrov and contributors (see CONTRIBUTORS.txt) +# +# This module is part of urllib3 and is released under +# the MIT License: http://www.opensource.org/licenses/mit-license.php + +""" +urllib3 - Thread-safe connection pooling and re-using. +""" + +__author__ = 'Andrey Petrov (andrey.petrov@shazow.net)' +__license__ = 'MIT' +__version__ = '1.0.1' + + +from .connectionpool import ( + HTTPConnectionPool, + HTTPSConnectionPool, + connection_from_url, + get_host, + make_headers) + + +from .exceptions import ( + HTTPError, + MaxRetryError, + SSLError, + TimeoutError) + +from .poolmanager import PoolManager, ProxyManager, proxy_from_url +from .response import HTTPResponse +from .filepost import encode_multipart_formdata + + +# Set default logging handler to avoid "No handler found" warnings. +import logging +try: + from logging import NullHandler +except ImportError: + class NullHandler(logging.Handler): + def emit(self, record): + pass + +logging.getLogger(__name__).addHandler(NullHandler()) + +# ... Clean up. +del logging +del NullHandler diff --git a/requests/packages/urllib3/_collections.py b/requests/packages/urllib3/_collections.py new file mode 100644 index 0000000..bd01da3 --- /dev/null +++ b/requests/packages/urllib3/_collections.py @@ -0,0 +1,120 @@ +# urllib3/_collections.py +# Copyright 2008-2011 Andrey Petrov and contributors (see CONTRIBUTORS.txt) +# +# This module is part of urllib3 and is released under +# the MIT License: http://www.opensource.org/licenses/mit-license.php + +from collections import MutableMapping, deque + + +__all__ = ['RecentlyUsedContainer'] + + +class AccessEntry(object): + __slots__ = ('key', 'is_valid') + + def __init__(self, key, is_valid=True): + self.key = key + self.is_valid = is_valid + + +class RecentlyUsedContainer(MutableMapping): + """ + Provides a dict-like that maintains up to ``maxsize`` keys while throwing + away the least-recently-used keys beyond ``maxsize``. + """ + + # TODO: Make this threadsafe. _prune_invalidated_entries should be the + # only real pain-point for this. + + # If len(self.access_log) exceeds self._maxsize * CLEANUP_FACTOR, then we + # will attempt to cleanup the invalidated entries in the access_log + # datastructure during the next 'get' operation. + CLEANUP_FACTOR = 10 + + def __init__(self, maxsize=10): + self._maxsize = maxsize + + self._container = {} + + # We use a deque to to store our keys ordered by the last access. + self.access_log = deque() + + # We look up the access log entry by the key to invalidate it so we can + # insert a new authorative entry at the head without having to dig and + # find the old entry for removal immediately. + self.access_lookup = {} + + # Trigger a heap cleanup when we get past this size + self.access_log_limit = maxsize * self.CLEANUP_FACTOR + + def _push_entry(self, key): + "Push entry onto our access log, invalidate the old entry if exists." + # Invalidate old entry if it exists + old_entry = self.access_lookup.get(key) + if old_entry: + old_entry.is_valid = False + + new_entry = AccessEntry(key) + + self.access_lookup[key] = new_entry + self.access_log.appendleft(new_entry) + + def _prune_entries(self, num): + "Pop entries from our access log until we popped ``num`` valid ones." + while num > 0: + p = self.access_log.pop() + + if not p.is_valid: + continue # Invalidated entry, skip + + del self._container[p.key] + del self.access_lookup[p.key] + num -= 1 + + def _prune_invalidated_entries(self): + "Rebuild our access_log without the invalidated entries." + self.access_log = deque(e for e in self.access_log if e.is_valid) + + def _get_ordered_access_keys(self): + # Used for testing + return [e.key for e in self.access_log if e.is_valid] + + def __getitem__(self, key): + item = self._container.get(key) + + if not item: + return + + # Insert new entry with new high priority, also implicitly invalidates + # the old entry. + self._push_entry(key) + + if len(self.access_log) > self.access_log_limit: + # Heap is getting too big, try to clean up any tailing invalidated + # entries. + self._prune_invalidated_entries() + + return item + + def __setitem__(self, key, item): + # Add item to our container and access log + self._container[key] = item + self._push_entry(key) + + # Discard invalid and excess entries + self._prune_entries(len(self._container) - self._maxsize) + + def __delitem__(self, key): + self._invalidate_entry(key) + del self._container[key] + del self._access_lookup[key] + + def __len__(self): + return self._container.__len__() + + def __iter__(self): + return self._container.__iter__() + + def __contains__(self, key): + return self._container.__contains__(key) diff --git a/requests/packages/urllib3/connectionpool.py b/requests/packages/urllib3/connectionpool.py new file mode 100644 index 0000000..4c2c6b5 --- /dev/null +++ b/requests/packages/urllib3/connectionpool.py @@ -0,0 +1,525 @@ +# urllib3/connectionpool.py +# Copyright 2008-2011 Andrey Petrov and contributors (see CONTRIBUTORS.txt) +# +# This module is part of urllib3 and is released under +# the MIT License: http://www.opensource.org/licenses/mit-license.php + +import logging +import socket + + +from httplib import HTTPConnection, HTTPSConnection, HTTPException +from Queue import Queue, Empty, Full +from select import select +from socket import error as SocketError, timeout as SocketTimeout + + +try: + import ssl + BaseSSLError = ssl.SSLError +except ImportError: + ssl = None + BaseSSLError = None + + +from .request import RequestMethods +from .response import HTTPResponse +from .exceptions import ( + SSLError, + MaxRetryError, + TimeoutError, + HostChangedError, + EmptyPoolError, +) + + +log = logging.getLogger(__name__) + +_Default = object() + + +## Connection objects (extension of httplib) + +class VerifiedHTTPSConnection(HTTPSConnection): + """ + Based on httplib.HTTPSConnection but wraps the socket with + SSL certification. + """ + cert_reqs = None + ca_certs = None + + def set_cert(self, key_file=None, cert_file=None, + cert_reqs='CERT_NONE', ca_certs=None): + ssl_req_scheme = { + 'CERT_NONE': ssl.CERT_NONE, + 'CERT_OPTIONAL': ssl.CERT_OPTIONAL, + 'CERT_REQUIRED': ssl.CERT_REQUIRED + } + + self.key_file = key_file + self.cert_file = cert_file + self.cert_reqs = ssl_req_scheme.get(cert_reqs) or ssl.CERT_NONE + self.ca_certs = ca_certs + + def connect(self): + # Add certificate verification + sock = socket.create_connection((self.host, self.port), self.timeout) + + # Wrap socket using verification with the root certs in + # trusted_root_certs + self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, + cert_reqs=self.cert_reqs, + ca_certs=self.ca_certs) + + +## Pool objects + +class ConnectionPool(object): + """ + Base class for all connection pools, such as + :class:`.HTTPConnectionPool` and :class:`.HTTPSConnectionPool`. + """ + pass + + +class HTTPConnectionPool(ConnectionPool, RequestMethods): + """ + Thread-safe connection pool for one host. + + :param host: + Host used for this HTTP Connection (e.g. "localhost"), passed into + :class:`httplib.HTTPConnection`. + + :param port: + Port used for this HTTP Connection (None is equivalent to 80), passed + into :class:`httplib.HTTPConnection`. + + :param strict: + Causes BadStatusLine to be raised if the status line can't be parsed + as a valid HTTP/1.0 or 1.1 status line, passed into + :class:`httplib.HTTPConnection`. + + :param timeout: + Socket timeout for each individual connection, can be a float. None + disables timeout. + + :param maxsize: + Number of connections to save that can be reused. More than 1 is useful + in multithreaded situations. If ``block`` is set to false, more + connections will be created but they will not be saved once they've + been used. + + :param block: + If set to True, no more than ``maxsize`` connections will be used at + a time. When no free connections are available, the call will block + until a connection has been released. This is a useful side effect for + particular multithreaded situations where one does not want to use more + than maxsize connections per host to prevent flooding. + + :param headers: + Headers to include with all requests, unless other headers are given + explicitly. + """ + + scheme = 'http' + + def __init__(self, host, port=None, strict=False, timeout=None, maxsize=1, + block=False, headers=None): + self.host = host + self.port = port + self.strict = strict + self.timeout = timeout + self.pool = Queue(maxsize) + self.block = block + self.headers = headers or {} + + # Fill the queue up so that doing get() on it will block properly + for _ in xrange(maxsize): + self.pool.put(None) + + # These are mostly for testing and debugging purposes. + self.num_connections = 0 + self.num_requests = 0 + + def _new_conn(self): + """ + Return a fresh :class:`httplib.HTTPConnection`. + """ + self.num_connections += 1 + log.info("Starting new HTTP connection (%d): %s" % + (self.num_connections, self.host)) + return HTTPConnection(host=self.host, port=self.port) + + def _get_conn(self, timeout=None): + """ + Get a connection. Will return a pooled connection if one is available. + + If no connections are available and :prop:`.block` is ``False``, then a + fresh connection is returned. + + :param timeout: + Seconds to wait before giving up and raising + :class:`urllib3.exceptions.EmptyPoolError` if the pool is empty and + :prop:`.block` is ``True``. + """ + conn = None + try: + conn = self.pool.get(block=self.block, timeout=timeout) + + # If this is a persistent connection, check if it got disconnected + if conn and conn.sock and select([conn.sock], [], [], 0.0)[0]: + # Either data is buffered (bad), or the connection is dropped. + log.info("Resetting dropped connection: %s" % self.host) + conn.close() + + except Empty: + if self.block: + raise EmptyPoolError("Pool reached maximum size and no more " + "connections are allowed.") + pass # Oh well, we'll create a new connection then + + return conn or self._new_conn() + + def _put_conn(self, conn): + """ + Put a connection back into the pool. + + :param conn: + Connection object for the current host and port as returned by + :meth:`._new_conn` or :meth:`._get_conn`. + + If the pool is already full, the connection is discarded because we + exceeded maxsize. If connections are discarded frequently, then maxsize + should be increased. + """ + try: + self.pool.put(conn, block=False) + except Full: + # This should never happen if self.block == True + log.warning("HttpConnectionPool is full, discarding connection: %s" + % self.host) + + def _make_request(self, conn, method, url, timeout=_Default, + **httplib_request_kw): + """ + Perform a request on a given httplib connection object taken from our + pool. + """ + self.num_requests += 1 + + if timeout is _Default: + timeout = self.timeout + + conn.request(method, url, **httplib_request_kw) + conn.sock.settimeout(timeout) + httplib_response = conn.getresponse() + + log.debug("\"%s %s %s\" %s %s" % + (method, url, + conn._http_vsn_str, # pylint: disable-msg=W0212 + httplib_response.status, httplib_response.length)) + + return httplib_response + + + def is_same_host(self, url): + """ + Check if the given ``url`` is a member of the same host as this + conncetion pool. + """ + # TODO: Add optional support for socket.gethostbyname checking. + return (url.startswith('/') or + get_host(url) == (self.scheme, self.host, self.port)) + + def urlopen(self, method, url, body=None, headers=None, retries=3, + redirect=True, assert_same_host=True, timeout=_Default, + pool_timeout=None, release_conn=None, **response_kw): + """ + Get a connection from the pool and perform an HTTP request. This is the + lowest level call for making a request, so you'll need to specify all + the raw details. + + .. note:: + + More commonly, it's appropriate to use a convenience method provided + by :class:`.RequestMethods`, such as :meth:`.request`. + + :param method: + HTTP request method (such as GET, POST, PUT, etc.) + + :param body: + Data to send in the request body (useful for creating + POST requests, see HTTPConnectionPool.post_url for + more convenience). + + :param headers: + Dictionary of custom headers to send, such as User-Agent, + If-None-Match, etc. If None, pool headers are used. If provided, + these headers completely replace any pool-specific headers. + + :param retries: + Number of retries to allow before raising a MaxRetryError exception. + + :param redirect: + Automatically handle redirects (status codes 301, 302, 303, 307), + each redirect counts as a retry. + + :param assert_same_host: + If ``True``, will make sure that the host of the pool requests is + consistent else will raise HostChangedError. When False, you can + use the pool on an HTTP proxy and request foreign hosts. + + :param timeout: + If specified, overrides the default timeout for this one request. + + :param pool_timeout: + If set and the pool is set to block=True, then this method will + block for ``pool_timeout`` seconds and raise EmptyPoolError if no + connection is available within the time period. + + :param release_conn: + If False, then the urlopen call will not release the connection + back into the pool once a response is received. This is useful if + you're not preloading the response's content immediately. You will + need to call ``r.release_conn()`` on the response ``r`` to return + the connection back into the pool. If None, it takes the value of + ``response_kw.get('preload_content', True)``. + + :param \**response_kw: + Additional parameters are passed to + :meth:`urllib3.response.HTTPResponse.from_httplib` + """ + if headers is None: + headers = self.headers + + if retries < 0: + raise MaxRetryError("Max retries exceeded for url: %s" % url) + + if release_conn is None: + release_conn = response_kw.get('preload_content', True) + + # Check host + if assert_same_host and not self.is_same_host(url): + host = "%s://%s" % (self.scheme, self.host) + if self.port: + host = "%s:%d" % (host, self.port) + + raise HostChangedError("Connection pool with host '%s' tried to " + "open a foreign host: %s" % (host, url)) + + conn = None + + try: + # Request a connection from the queue + # (Could raise SocketError: Bad file descriptor) + conn = self._get_conn(timeout=pool_timeout) + + # Make the request on the httplib connection object + httplib_response = self._make_request(conn, method, url, + timeout=timeout, + body=body, headers=headers) + + # If we're going to release the connection in ``finally:``, then + # the request doesn't need to know about the connection. Otherwise + # it will also try to release it and we'll have a double-release + # mess. + response_conn = not release_conn and conn + + # Import httplib's response into our own wrapper object + response = HTTPResponse.from_httplib(httplib_response, + pool=self, + connection=response_conn, + **response_kw) + + # else: + # The connection will be put back into the pool when + # ``response.release_conn()`` is called (implicitly by + # ``response.read()``) + + except (SocketTimeout, Empty), e: + # Timed out either by socket or queue + raise TimeoutError("Request timed out after %f seconds" % + self.timeout) + + except (BaseSSLError), e: + # SSL certificate error + raise SSLError(e) + + except (HTTPException, SocketError), e: + # Connection broken, discard. It will be replaced next _get_conn(). + conn = None + + finally: + if conn and release_conn: + # Put the connection back to be reused + self._put_conn(conn) + + if not conn: + log.warn("Retrying (%d attempts remain) after connection " + "broken by '%r': %s" % (retries, e, url)) + return self.urlopen(method, url, body, headers, retries - 1, + redirect, assert_same_host) # Try again + + # Handle redirection + if (redirect and + response.status in [301, 302, 303, 307] and + 'location' in response.headers): # Redirect, retry + log.info("Redirecting %s -> %s" % + (url, response.headers.get('location'))) + return self.urlopen(method, response.headers.get('location'), body, + headers, retries - 1, redirect, + assert_same_host) + + return response + + +class HTTPSConnectionPool(HTTPConnectionPool): + """ + Same as :class:`.HTTPConnectionPool`, but HTTPS. + + When Python is compiled with the :mod:`ssl` module, then + :class:`.VerifiedHTTPSConnection` is used, which *can* verify certificates, + instead of :class:httplib.HTTPSConnection`. + + The ``key_file``, ``cert_file``, ``cert_reqs``, and ``ca_certs`` parameters + are only used if :mod:`ssl` is available and are fed into + :meth:`ssl.wrap_socket` to upgrade the connection socket into an SSL socket. + """ + + scheme = 'https' + + def __init__(self, host, port=None, + strict=False, timeout=None, maxsize=1, + block=False, headers=None, + key_file=None, cert_file=None, + cert_reqs='CERT_NONE', ca_certs=None): + + super(HTTPSConnectionPool, self).__init__(host, port, + strict, timeout, maxsize, + block, headers) + self.key_file = key_file + self.cert_file = cert_file + self.cert_reqs = cert_reqs + self.ca_certs = ca_certs + + def _new_conn(self): + """ + Return a fresh :class:`httplib.HTTPSConnection`. + """ + self.num_connections += 1 + log.info("Starting new HTTPS connection (%d): %s" + % (self.num_connections, self.host)) + + if not ssl: + return HTTPSConnection(host=self.host, port=self.port) + + connection = VerifiedHTTPSConnection(host=self.host, port=self.port) + connection.set_cert(key_file=self.key_file, cert_file=self.cert_file, + cert_reqs=self.cert_reqs, ca_certs=self.ca_certs) + return connection + + +## Helpers + +def make_headers(keep_alive=None, accept_encoding=None, user_agent=None, + basic_auth=None): + """ + Shortcuts for generating request headers. + + :param keep_alive: + If ``True``, adds 'connection: keep-alive' header. + + :param accept_encoding: + Can be a boolean, list, or string. + ``True`` translates to 'gzip,deflate'. + List will get joined by comma. + String will be used as provided. + + :param user_agent: + String representing the user-agent you want, such as + "python-urllib3/0.6" + + :param basic_auth: + Colon-separated username:password string for 'authorization: basic ...' + auth header. + + Example: :: + + >>> make_headers(keep_alive=True, user_agent="Batman/1.0") + {'connection': 'keep-alive', 'user-agent': 'Batman/1.0'} + >>> make_headers(accept_encoding=True) + {'accept-encoding': 'gzip,deflate'} + """ + headers = {} + if accept_encoding: + if isinstance(accept_encoding, str): + pass + elif isinstance(accept_encoding, list): + accept_encoding = ','.join(accept_encoding) + else: + accept_encoding = 'gzip,deflate' + headers['accept-encoding'] = accept_encoding + + if user_agent: + headers['user-agent'] = user_agent + + if keep_alive: + headers['connection'] = 'keep-alive' + + if basic_auth: + headers['authorization'] = 'Basic ' + \ + basic_auth.encode('base64').strip() + + return headers + + +def get_host(url): + """ + Given a url, return its scheme, host and port (None if it's not there). + + For example: :: + + >>> get_host('http://google.com/mail/') + ('http', 'google.com', None) + >>> get_host('google.com:80') + ('http', 'google.com', 80) + """ + # This code is actually similar to urlparse.urlsplit, but much + # simplified for our needs. + port = None + scheme = 'http' + if '//' in url: + scheme, url = url.split('://', 1) + if '/' in url: + url, _path = url.split('/', 1) + if ':' in url: + url, port = url.split(':', 1) + port = int(port) + return scheme, url, port + + +def connection_from_url(url, **kw): + """ + Given a url, return an :class:`.ConnectionPool` instance of its host. + + This is a shortcut for not having to parse out the scheme, host, and port + of the url before creating an :class:`.ConnectionPool` instance. + + :param url: + Absolute URL string that must include the scheme. Port is optional. + + :param \**kw: + Passes additional parameters to the constructor of the appropriate + :class:`.ConnectionPool`. Useful for specifying things like + timeout, maxsize, headers, etc. + + Example: :: + + >>> conn = connection_from_url('http://google.com/') + >>> r = conn.request('GET', '/') + """ + scheme, host, port = get_host(url) + if scheme == 'https': + return HTTPSConnectionPool(host, port=port, **kw) + else: + return HTTPConnectionPool(host, port=port, **kw) diff --git a/requests/packages/urllib3/contrib/__init__.py b/requests/packages/urllib3/contrib/__init__.py new file mode 100644 index 0000000..e69de29 diff --git a/requests/packages/urllib3/contrib/ntlmpool.py b/requests/packages/urllib3/contrib/ntlmpool.py new file mode 100644 index 0000000..c5f010e --- /dev/null +++ b/requests/packages/urllib3/contrib/ntlmpool.py @@ -0,0 +1,117 @@ +# urllib3/contrib/ntlmpool.py +# Copyright 2008-2011 Andrey Petrov and contributors (see CONTRIBUTORS.txt) +# +# This module is part of urllib3 and is released under +# the MIT License: http://www.opensource.org/licenses/mit-license.php + +""" +NTLM authenticating pool, contributed by erikcederstran + +Issue #10, see: http://code.google.com/p/urllib3/issues/detail?id=10 +""" + +import httplib +from logging import getLogger +from ntlm import ntlm + +from urllib3 import HTTPSConnectionPool + + +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 = {} + headers['Connection'] = 'Keep-Alive' + req_header = 'Authorization' + resp_header = 'www-authenticate' + + conn = httplib.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.getheaders()) + 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.getheaders())) + 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) diff --git a/requests/packages/urllib3/exceptions.py b/requests/packages/urllib3/exceptions.py new file mode 100644 index 0000000..69f459b --- /dev/null +++ b/requests/packages/urllib3/exceptions.py @@ -0,0 +1,35 @@ +# urllib3/exceptions.py +# Copyright 2008-2011 Andrey Petrov and contributors (see CONTRIBUTORS.txt) +# +# This module is part of urllib3 and is released under +# the MIT License: http://www.opensource.org/licenses/mit-license.php + +## Exceptions + +class HTTPError(Exception): + "Base exception used by this module." + pass + + +class SSLError(Exception): + "Raised when SSL certificate fails in an HTTPS connection." + pass + + +class MaxRetryError(HTTPError): + "Raised when the maximum number of retries is exceeded." + pass + + +class TimeoutError(HTTPError): + "Raised when a socket timeout occurs." + pass + + +class HostChangedError(HTTPError): + "Raised when an existing pool gets a request for a foreign host." + pass + +class EmptyPoolError(HTTPError): + "Raised when a pool runs out of connections and no more are allowed." + pass diff --git a/requests/packages/urllib3/filepost.py b/requests/packages/urllib3/filepost.py new file mode 100644 index 0000000..2ffea8b --- /dev/null +++ b/requests/packages/urllib3/filepost.py @@ -0,0 +1,71 @@ +# urllib3/filepost.py +# Copyright 2008-2011 Andrey Petrov and contributors (see CONTRIBUTORS.txt) +# +# This module is part of urllib3 and is released under +# the MIT License: http://www.opensource.org/licenses/mit-license.php + +import codecs +import mimetools +import mimetypes + +try: + from cStringIO import StringIO +except ImportError: + from StringIO import StringIO # pylint: disable-msg=W0404 + + +writer = codecs.lookup('utf-8')[3] + + +def get_content_type(filename): + return mimetypes.guess_type(filename)[0] or 'application/octet-stream' + + +def encode_multipart_formdata(fields, boundary=None): + """ + Encode a dictionary of ``fields`` using the multipart/form-data mime format. + + :param fields: + Dictionary of fields. The key is treated as the field name, and the + value as the body of the form-data. If the value is a tuple of two + elements, then the first element is treated as the filename of the + form-data section. + + :param boundary: + If not specified, then a random boundary will be generated using + :func:`mimetools.choose_boundary`. + """ + body = StringIO() + if boundary is None: + boundary = mimetools.choose_boundary() + + for fieldname, value in fields.iteritems(): + body.write('--%s\r\n' % (boundary)) + + if isinstance(value, tuple): + filename, data = value + writer(body).write('Content-Disposition: form-data; name="%s"; ' + 'filename="%s"\r\n' % (fieldname, filename)) + body.write('Content-Type: %s\r\n\r\n' % + (get_content_type(filename))) + else: + data = value + writer(body).write('Content-Disposition: form-data; name="%s"\r\n' + % (fieldname)) + body.write('Content-Type: text/plain\r\n\r\n') + + if isinstance(data, int): + data = str(data) # Backwards compatibility + + if isinstance(data, unicode): + writer(body).write(data) + else: + body.write(data) + + body.write('\r\n') + + body.write('--%s--\r\n' % (boundary)) + + content_type = 'multipart/form-data; boundary=%s' % boundary + + return body.getvalue(), content_type diff --git a/requests/packages/urllib3/poolmanager.py b/requests/packages/urllib3/poolmanager.py new file mode 100644 index 0000000..622789e --- /dev/null +++ b/requests/packages/urllib3/poolmanager.py @@ -0,0 +1,128 @@ +# urllib3/poolmanager.py +# Copyright 2008-2011 Andrey Petrov and contributors (see CONTRIBUTORS.txt) +# +# This module is part of urllib3 and is released under +# the MIT License: http://www.opensource.org/licenses/mit-license.php + +from ._collections import RecentlyUsedContainer +from .connectionpool import ( + HTTPConnectionPool, HTTPSConnectionPool, + get_host, connection_from_url, +) + + +__all__ = ['PoolManager', 'ProxyManager', 'proxy_from_url'] + + +from .request import RequestMethods +from .connectionpool import HTTPConnectionPool, HTTPSConnectionPool + + +pool_classes_by_scheme = { + 'http': HTTPConnectionPool, + 'https': HTTPSConnectionPool, +} + +port_by_scheme = { + 'http': 80, + 'https': 443, +} + + +class PoolManager(RequestMethods): + """ + Allows for arbitrary requests while transparently keeping track of + necessary connection pools for you. + + :param num_pools: + Number of connection pools to cache before discarding the least recently + used pool. + + :param \**connection_pool_kw: + Additional parameters are used to create fresh + :class:`urllib3.connectionpool.ConnectionPool` instances. + + Example: :: + + >>> manager = PoolManager() + >>> r = manager.urlopen("http://google.com/") + >>> r = manager.urlopen("http://google.com/mail") + >>> r = manager.urlopen("http://yahoo.com/") + >>> len(r.pools) + 2 + + """ + + # TODO: Make sure there are no memory leaks here. + + def __init__(self, num_pools=10, **connection_pool_kw): + self.connection_pool_kw = connection_pool_kw + self.pools = RecentlyUsedContainer(num_pools) + + def connection_from_host(self, host, port=80, scheme='http'): + """ + Get a :class:`ConnectionPool` based on the host, port, and scheme. + + Note that an appropriate ``port`` value is required here to normalize + connection pools in our container most effectively. + """ + pool_key = (scheme, host, port) + + # If the scheme, host, or port doesn't match existing open connections, + # open a new ConnectionPool. + pool = self.pools.get(pool_key) + if pool: + return pool + + # Make a fresh ConnectionPool of the desired type + pool_cls = pool_classes_by_scheme[scheme] + pool = pool_cls(host, port, **self.connection_pool_kw) + + self.pools[pool_key] = pool + + return pool + + def connection_from_url(self, url): + """ + Similar to :func:`urllib3.connectionpool.connection_from_url` but + doesn't pass any additional parameters to the + :class:`urllib3.connectionpool.ConnectionPool` constructor. + + Additional parameters are taken from the :class:`.PoolManager` + constructor. + """ + scheme, host, port = get_host(url) + + port = port or port_by_scheme.get(scheme, 80) + + return self.connection_from_host(host, port=port, scheme=scheme) + + def urlopen(self, method, url, **kw): + """ + Same as :meth:`urllib3.connectionpool.HTTPConnectionPool.urlopen`. + + ``url`` must be absolute, such that an appropriate + :class:`urllib3.connectionpool.ConnectionPool` can be chosen for it. + """ + conn = self.connection_from_url(url) + return conn.urlopen(method, url, assert_same_host=False, **kw) + + +class ProxyManager(object): + """ + Given a ConnectionPool to a proxy, the ProxyManager's ``urlopen`` method + will make requests to any url through the defined proxy. + """ + + def __init__(self, proxy_pool): + self.proxy_pool = proxy_pool + + def urlopen(self, method, url, **kw): + "Same as HTTP(S)ConnectionPool.urlopen, ``url`` must be absolute." + kw['assert_same_host'] = False + return self.proxy_pool.urlopen(method, url, **kw) + + +def proxy_from_url(url, **pool_kw): + proxy_pool = connection_from_url(url, **pool_kw) + return ProxyManager(proxy_pool) diff --git a/requests/packages/urllib3/request.py b/requests/packages/urllib3/request.py new file mode 100644 index 0000000..a7e0b5d --- /dev/null +++ b/requests/packages/urllib3/request.py @@ -0,0 +1,145 @@ +# urllib3/request.py +# Copyright 2008-2011 Andrey Petrov and contributors (see CONTRIBUTORS.txt) +# +# This module is part of urllib3 and is released under +# the MIT License: http://www.opensource.org/licenses/mit-license.php + + +from urllib import urlencode + +from .filepost import encode_multipart_formdata + + +__all__ = ['RequestMethods'] + + +class RequestMethods(object): + """ + Convenience mixin for classes who implement a :meth:`urlopen` method, such + as :class:`~urllib3.connectionpool.HTTPConnectionPool` and + :class:`~urllib3.poolmanager.PoolManager`. + + Provides behavior for making common types of HTTP request methods and + decides which type of request field encoding to use. + + Specifically, + + :meth:`.request_encode_url` is for sending requests whose fields are encoded + in the URL (such as GET, HEAD, DELETE). + + :meth:`.request_encode_body` is for sending requests whose fields are + encoded in the *body* of the request using multipart or www-orm-urlencoded + (such as for POST, PUT, PATCH). + + :meth:`.request` is for making any kind of request, it will look up the + appropriate encoding format and use one of the above two methods to make + the request. + """ + + _encode_url_methods = set(['DELETE', 'GET', 'HEAD', 'OPTIONS']) + + _encode_body_methods = set(['PATCH', 'POST', 'PUT', 'TRACE']) + + def urlopen(self, method, url, body=None, headers=None, + encode_multipart=True, multipart_boundary=None, + **kw): + raise NotImplemented("Classes extending RequestMethods must implement " + "their own ``urlopen`` method.") + + def request(self, method, url, fields=None, headers=None, **urlopen_kw): + """ + Make a request using :meth:`urlopen` with the appropriate encoding of + ``fields`` based on the ``method`` used. + + This is a convenience method that requires the least amount of manual + effort. It can be used in most situations, while still having the option + to drop down to more specific methods when necessary, such as + :meth:`request_encode_url`, :meth:`request_encode_body`, + or even the lowest level :meth:`urlopen`. + """ + method = method.upper() + + if method in self._encode_url_methods: + return self.request_encode_url(method, url, fields=fields, + headers=headers, + **urlopen_kw) + else: + return self.request_encode_body(method, url, fields=fields, + headers=headers, + **urlopen_kw) + + def request_encode_url(self, method, url, fields=None, **urlopen_kw): + """ + Make a request using :meth:`urlopen` with the ``fields`` encoded in + the url. This is useful for request methods like GET, HEAD, DELETE, etc. + """ + if fields: + url += '?' + urlencode(fields) + return self.urlopen(method, url, **urlopen_kw) + + def request_encode_body(self, method, url, fields=None, headers=None, + encode_multipart=True, multipart_boundary=None, + **urlopen_kw): + """ + Make a request using :meth:`urlopen` with the ``fields`` encoded in + the body. This is useful for request methods like POST, PUT, PATCH, etc. + + When ``encode_multipart=True`` (default), then + :meth:`urllib3.filepost.encode_multipart_formdata` is used to encode the + payload with the appropriate content type. Otherwise + :meth:`urllib.urlencode` is used with the + 'application/x-www-form-urlencoded' content type. + + Multipart encoding must be used when posting files, and it's reasonably + safe to use it in other times too. However, it may break request signing, + such as with OAuth. + + Supports an optional ``fields`` parameter of key/value strings AND + key/filetuple. A filetuple is a (filename, data) tuple. For example: :: + + fields = { + 'foo': 'bar', + 'fakefile': ('foofile.txt', 'contents of foofile'), + 'realfile': ('barfile.txt', open('realfile').read()), + 'nonamefile': ('contents of nonamefile field'), + } + + When uploading a file, providing a filename (the first parameter of the + tuple) is optional but recommended to best mimick behavior of browsers. + + Note that if ``headers`` are supplied, the 'Content-Type' header will be + overwritten because it depends on the dynamic random boundary string + which is used to compose the body of the request. The random boundary + string can be explicitly set with the ``multipart_boundary`` parameter. + """ + if encode_multipart: + body, content_type = encode_multipart_formdata(fields or {}, + boundary=multipart_boundary) + else: + body, content_type = (urlencode(fields or {}), + 'application/x-www-form-urlencoded') + + headers = headers or {} + headers.update({'Content-Type': content_type}) + + return self.urlopen(method, url, body=body, headers=headers, + **urlopen_kw) + + # Deprecated: + + def get_url(self, url, fields=None, **urlopen_kw): + """ + .. deprecated:: 1.0 + Use :meth:`request` instead. + """ + return self.request_encode_url('GET', url, fields=fields, + **urlopen_kw) + + def post_url(self, url, fields=None, headers=None, **urlopen_kw): + """ + .. deprecated:: 1.0 + Use :meth:`request` instead. + """ + return self.request_encode_body('POST', url, fields=fields, + headers=headers, + **urlopen_kw) diff --git a/requests/packages/urllib3/response.py b/requests/packages/urllib3/response.py new file mode 100644 index 0000000..4cd15c1 --- /dev/null +++ b/requests/packages/urllib3/response.py @@ -0,0 +1,181 @@ +# urllib3/response.py +# Copyright 2008-2011 Andrey Petrov and contributors (see CONTRIBUTORS.txt) +# +# This module is part of urllib3 and is released under +# the MIT License: http://www.opensource.org/licenses/mit-license.php + +import gzip +import logging +import zlib + + +try: + from cStringIO import StringIO +except ImportError: + from StringIO import StringIO # pylint: disable-msg=W0404 + + +from .exceptions import HTTPError + + +log = logging.getLogger(__name__) + + +def decode_gzip(data): + gzipper = gzip.GzipFile(fileobj=StringIO(data)) + return gzipper.read() + + +def decode_deflate(data): + try: + return zlib.decompress(data) + except zlib.error: + return zlib.decompress(data, -zlib.MAX_WBITS) + + +class HTTPResponse(object): + """ + HTTP Response container. + + Backwards-compatible to httplib's HTTPResponse but the response ``body`` is + loaded and decoded on-demand when the ``data`` property is accessed. + + Extra parameters for behaviour not present in httplib.HTTPResponse: + + :param preload_content: + If True, the response's body will be preloaded during construction. + + :param decode_content: + If True, attempts to decode specific content-encoding's based on headers + (like 'gzip' and 'deflate') will be skipped and raw data will be used + instead. + + :param original_response: + When this HTTPResponse wrapper is generated from an httplib.HTTPResponse + object, it's convenient to include the original for debug purposes. It's + otherwise unused. + """ + + CONTENT_DECODERS = { + 'gzip': decode_gzip, + 'deflate': decode_deflate, + } + + def __init__(self, body='', headers=None, status=0, version=0, reason=None, + strict=0, preload_content=True, decode_content=True, + original_response=None, pool=None, connection=None): + self.headers = headers or {} + self.status = status + self.version = version + self.reason = reason + self.strict = strict + + self._decode_content = decode_content + self._body = None + self._fp = None + self._original_response = original_response + + self._pool = pool + self._connection = connection + + if hasattr(body, 'read'): + self._fp = body + + if preload_content: + self._body = self.read(decode_content=decode_content) + + def release_conn(self): + if not self._pool or not self._connection: + return + + self._pool._put_conn(self._connection) + self._connection = None + + @property + def data(self): + # For backwords-compat with earlier urllib3 0.4 and earlier. + if self._body: + return self._body + + if self._fp: + return self.read(decode_content=self._decode_content, + cache_content=True) + + def read(self, amt=None, decode_content=True, cache_content=False): + """ + Similar to :meth:`httplib.HTTPResponse.read`, but with two additional + parameters: ``decode_content`` and ``cache_content``. + + :param amt: + How much of the content to read. If specified, decoding and caching + is skipped because we can't decode partial content nor does it make + sense to cache partial content as the full response. + + :param decode_content: + If True, will attempt to decode the body based on the + 'content-encoding' header. (Overridden if ``amt`` is set.) + + :param cache_content: + If True, will save the returned data such that the same result is + returned despite of the state of the underlying file object. This + is useful if you want the ``.data`` property to continue working + after having ``.read()`` the file object. (Overridden if ``amt`` is + set.) + """ + content_encoding = self.headers.get('content-encoding') + decoder = self.CONTENT_DECODERS.get(content_encoding) + + data = self._fp and self._fp.read(amt) + + try: + + if amt: + return data + + if not decode_content or not decoder: + if cache_content: + self._body = data + + return data + + try: + data = decoder(data) + except IOError: + raise HTTPError("Received response with content-encoding: %s, but " + "failed to decode it." % content_encoding) + + if cache_content: + self._body = data + + return data + + finally: + + if self._original_response and self._original_response.isclosed(): + self.release_conn() + + @staticmethod + def from_httplib(r, **response_kw): + """ + Given an :class:`httplib.HTTPResponse` instance ``r``, return a + corresponding :class:`urllib3.response.HTTPResponse` object. + + Remaining parameters are passed to the HTTPResponse constructor, along + with ``original_response=r``. + """ + + return HTTPResponse(body=r, + headers=dict(r.getheaders()), + status=r.status, + version=r.version, + reason=r.reason, + strict=r.strict, + original_response=r, + **response_kw) + + # Backwards-compatibility methods for httplib.HTTPResponse + def getheaders(self): + return self.headers + + def getheader(self, name, default=None): + return self.headers.get(name, default) -- 2.7.4