scripts/python3/bot/lib/python3.11/site-packages/urllib3/connectionpool.py

1187 lines
42 KiB
Python
Raw Normal View History

2024-08-29 12:07:22 +03:00
from __future__ import annotations
import errno
import logging
import queue
import sys
import typing
import warnings
import weakref
from socket import timeout as SocketTimeout
from types import TracebackType
from ._base_connection import _TYPE_BODY
from ._collections import HTTPHeaderDict
from ._request_methods import RequestMethods
from .connection import (
BaseSSLError,
BrokenPipeError,
DummyConnection,
HTTPConnection,
HTTPException,
HTTPSConnection,
ProxyConfig,
_wrap_proxy_error,
)
from .connection import port_by_scheme as port_by_scheme
from .exceptions import (
ClosedPoolError,
EmptyPoolError,
FullPoolError,
HostChangedError,
InsecureRequestWarning,
LocationValueError,
MaxRetryError,
NewConnectionError,
ProtocolError,
ProxyError,
ReadTimeoutError,
SSLError,
TimeoutError,
)
from .response import BaseHTTPResponse
from .util.connection import is_connection_dropped
from .util.proxy import connection_requires_http_tunnel
from .util.request import _TYPE_BODY_POSITION, set_file_position
from .util.retry import Retry
from .util.ssl_match_hostname import CertificateError
from .util.timeout import _DEFAULT_TIMEOUT, _TYPE_DEFAULT, Timeout
from .util.url import Url, _encode_target
from .util.url import _normalize_host as normalize_host
from .util.url import parse_url
from .util.util import to_str
if typing.TYPE_CHECKING:
import ssl
from typing import Literal
from ._base_connection import BaseHTTPConnection, BaseHTTPSConnection
log = logging.getLogger(__name__)
_TYPE_TIMEOUT = typing.Union[Timeout, float, _TYPE_DEFAULT, None]
_SelfT = typing.TypeVar("_SelfT")
# Pool objects
class ConnectionPool:
"""
Base class for all connection pools, such as
:class:`.HTTPConnectionPool` and :class:`.HTTPSConnectionPool`.
.. note::
ConnectionPool.urlopen() does not normalize or percent-encode target URIs
which is useful if your target server doesn't support percent-encoded
target URIs.
"""
scheme: str | None = None
QueueCls = queue.LifoQueue
def __init__(self, host: str, port: int | None = None) -> None:
if not host:
raise LocationValueError("No host specified.")
self.host = _normalize_host(host, scheme=self.scheme)
self.port = port
# This property uses 'normalize_host()' (not '_normalize_host()')
# to avoid removing square braces around IPv6 addresses.
# This value is sent to `HTTPConnection.set_tunnel()` if called
# because square braces are required for HTTP CONNECT tunneling.
self._tunnel_host = normalize_host(host, scheme=self.scheme).lower()
def __str__(self) -> str:
return f"{type(self).__name__}(host={self.host!r}, port={self.port!r})"
def __enter__(self: _SelfT) -> _SelfT:
return self
def __exit__(
self,
exc_type: type[BaseException] | None,
exc_val: BaseException | None,
exc_tb: TracebackType | None,
) -> Literal[False]:
self.close()
# Return False to re-raise any potential exceptions
return False
def close(self) -> None:
"""
Close all pooled connections and disable the pool.
"""
# This is taken from http://hg.python.org/cpython/file/7aaba721ebc0/Lib/socket.py#l252
_blocking_errnos = {errno.EAGAIN, errno.EWOULDBLOCK}
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:`http.client.HTTPConnection`.
:param port:
Port used for this HTTP Connection (None is equivalent to 80), passed
into :class:`http.client.HTTPConnection`.
:param timeout:
Socket timeout in seconds for each individual connection. This can
be a float or integer, which sets the timeout for the HTTP request,
or an instance of :class:`urllib3.util.Timeout` which gives you more
fine-grained control over request timeouts. After the constructor has
been parsed, this is always a `urllib3.util.Timeout` object.
: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.
:param retries:
Retry configuration to use by default with requests in this pool.
:param _proxy:
Parsed proxy URL, should not be used directly, instead, see
:class:`urllib3.ProxyManager`
:param _proxy_headers:
A dictionary with proxy headers, should not be used directly,
instead, see :class:`urllib3.ProxyManager`
:param \\**conn_kw:
Additional parameters are used to create fresh :class:`urllib3.connection.HTTPConnection`,
:class:`urllib3.connection.HTTPSConnection` instances.
"""
scheme = "http"
ConnectionCls: (
type[BaseHTTPConnection] | type[BaseHTTPSConnection]
) = HTTPConnection
def __init__(
self,
host: str,
port: int | None = None,
timeout: _TYPE_TIMEOUT | None = _DEFAULT_TIMEOUT,
maxsize: int = 1,
block: bool = False,
headers: typing.Mapping[str, str] | None = None,
retries: Retry | bool | int | None = None,
_proxy: Url | None = None,
_proxy_headers: typing.Mapping[str, str] | None = None,
_proxy_config: ProxyConfig | None = None,
**conn_kw: typing.Any,
):
ConnectionPool.__init__(self, host, port)
RequestMethods.__init__(self, headers)
if not isinstance(timeout, Timeout):
timeout = Timeout.from_float(timeout)
if retries is None:
retries = Retry.DEFAULT
self.timeout = timeout
self.retries = retries
self.pool: queue.LifoQueue[typing.Any] | None = self.QueueCls(maxsize)
self.block = block
self.proxy = _proxy
self.proxy_headers = _proxy_headers or {}
self.proxy_config = _proxy_config
# Fill the queue up so that doing get() on it will block properly
for _ in range(maxsize):
self.pool.put(None)
# These are mostly for testing and debugging purposes.
self.num_connections = 0
self.num_requests = 0
self.conn_kw = conn_kw
if self.proxy:
# Enable Nagle's algorithm for proxies, to avoid packet fragmentation.
# We cannot know if the user has added default socket options, so we cannot replace the
# list.
self.conn_kw.setdefault("socket_options", [])
self.conn_kw["proxy"] = self.proxy
self.conn_kw["proxy_config"] = self.proxy_config
# Do not pass 'self' as callback to 'finalize'.
# Then the 'finalize' would keep an endless living (leak) to self.
# By just passing a reference to the pool allows the garbage collector
# to free self if nobody else has a reference to it.
pool = self.pool
# Close all the HTTPConnections in the pool before the
# HTTPConnectionPool object is garbage collected.
weakref.finalize(self, _close_pool_connections, pool)
def _new_conn(self) -> BaseHTTPConnection:
"""
Return a fresh :class:`HTTPConnection`.
"""
self.num_connections += 1
log.debug(
"Starting new HTTP connection (%d): %s:%s",
self.num_connections,
self.host,
self.port or "80",
)
conn = self.ConnectionCls(
host=self.host,
port=self.port,
timeout=self.timeout.connect_timeout,
**self.conn_kw,
)
return conn
def _get_conn(self, timeout: float | None = None) -> BaseHTTPConnection:
"""
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
if self.pool is None:
raise ClosedPoolError(self, "Pool is closed.")
try:
conn = self.pool.get(block=self.block, timeout=timeout)
except AttributeError: # self.pool is None
raise ClosedPoolError(self, "Pool is closed.") from None # Defensive:
except queue.Empty:
if self.block:
raise EmptyPoolError(
self,
"Pool is empty and a new connection can't be opened due to blocking mode.",
) from None
pass # Oh well, we'll create a new connection then
# If this is a persistent connection, check if it got disconnected
if conn and is_connection_dropped(conn):
log.debug("Resetting dropped connection: %s", self.host)
conn.close()
return conn or self._new_conn()
def _put_conn(self, conn: BaseHTTPConnection | None) -> None:
"""
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 closed and discarded
because we exceeded maxsize. If connections are discarded frequently,
then maxsize should be increased.
If the pool is closed, then the connection will be closed and discarded.
"""
if self.pool is not None:
try:
self.pool.put(conn, block=False)
return # Everything is dandy, done.
except AttributeError:
# self.pool is None.
pass
except queue.Full:
# Connection never got put back into the pool, close it.
if conn:
conn.close()
if self.block:
# This should never happen if you got the conn from self._get_conn
raise FullPoolError(
self,
"Pool reached maximum size and no more connections are allowed.",
) from None
log.warning(
"Connection pool is full, discarding connection: %s. Connection pool size: %s",
self.host,
self.pool.qsize(),
)
# Connection never got put back into the pool, close it.
if conn:
conn.close()
def _validate_conn(self, conn: BaseHTTPConnection) -> None:
"""
Called right before a request is made, after the socket is created.
"""
def _prepare_proxy(self, conn: BaseHTTPConnection) -> None:
# Nothing to do for HTTP connections.
pass
def _get_timeout(self, timeout: _TYPE_TIMEOUT) -> Timeout:
"""Helper that always returns a :class:`urllib3.util.Timeout`"""
if timeout is _DEFAULT_TIMEOUT:
return self.timeout.clone()
if isinstance(timeout, Timeout):
return timeout.clone()
else:
# User passed us an int/float. This is for backwards compatibility,
# can be removed later
return Timeout.from_float(timeout)
def _raise_timeout(
self,
err: BaseSSLError | OSError | SocketTimeout,
url: str,
timeout_value: _TYPE_TIMEOUT | None,
) -> None:
"""Is the error actually a timeout? Will raise a ReadTimeout or pass"""
if isinstance(err, SocketTimeout):
raise ReadTimeoutError(
self, url, f"Read timed out. (read timeout={timeout_value})"
) from err
# See the above comment about EAGAIN in Python 3.
if hasattr(err, "errno") and err.errno in _blocking_errnos:
raise ReadTimeoutError(
self, url, f"Read timed out. (read timeout={timeout_value})"
) from err
def _make_request(
self,
conn: BaseHTTPConnection,
method: str,
url: str,
body: _TYPE_BODY | None = None,
headers: typing.Mapping[str, str] | None = None,
retries: Retry | None = None,
timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT,
chunked: bool = False,
response_conn: BaseHTTPConnection | None = None,
preload_content: bool = True,
decode_content: bool = True,
enforce_content_length: bool = True,
) -> BaseHTTPResponse:
"""
Perform a request on a given urllib connection object taken from our
pool.
:param conn:
a connection from one of our connection pools
:param method:
HTTP request method (such as GET, POST, PUT, etc.)
:param url:
The URL to perform the request on.
:param body:
Data to send in the request body, either :class:`str`, :class:`bytes`,
an iterable of :class:`str`/:class:`bytes`, or a file-like object.
: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:
Configure the number of retries to allow before raising a
:class:`~urllib3.exceptions.MaxRetryError` exception.
Pass ``None`` to retry until you receive a response. Pass a
:class:`~urllib3.util.retry.Retry` object for fine-grained control
over different types of retries.
Pass an integer number to retry connection errors that many times,
but no other types of errors. Pass zero to never retry.
If ``False``, then retries are disabled and any exception is raised
immediately. Also, instead of raising a MaxRetryError on redirects,
the redirect response will be returned.
:type retries: :class:`~urllib3.util.retry.Retry`, False, or an int.
:param timeout:
If specified, overrides the default timeout for this one
request. It may be a float (in seconds) or an instance of
:class:`urllib3.util.Timeout`.
:param chunked:
If True, urllib3 will send the body using chunked transfer
encoding. Otherwise, urllib3 will send the body using the standard
content-length form. Defaults to False.
:param response_conn:
Set this to ``None`` if you will handle releasing the connection or
set the connection to have the response release it.
:param preload_content:
If True, the response's body will be preloaded during construction.
:param decode_content:
If True, will attempt to decode the body based on the
'content-encoding' header.
:param enforce_content_length:
Enforce content length checking. Body returned by server must match
value of Content-Length header, if present. Otherwise, raise error.
"""
self.num_requests += 1
timeout_obj = self._get_timeout(timeout)
timeout_obj.start_connect()
conn.timeout = Timeout.resolve_default_timeout(timeout_obj.connect_timeout)
try:
# Trigger any extra validation we need to do.
try:
self._validate_conn(conn)
except (SocketTimeout, BaseSSLError) as e:
self._raise_timeout(err=e, url=url, timeout_value=conn.timeout)
raise
# _validate_conn() starts the connection to an HTTPS proxy
# so we need to wrap errors with 'ProxyError' here too.
except (
OSError,
NewConnectionError,
TimeoutError,
BaseSSLError,
CertificateError,
SSLError,
) as e:
new_e: Exception = e
if isinstance(e, (BaseSSLError, CertificateError)):
new_e = SSLError(e)
# If the connection didn't successfully connect to it's proxy
# then there
if isinstance(
new_e, (OSError, NewConnectionError, TimeoutError, SSLError)
) and (conn and conn.proxy and not conn.has_connected_to_proxy):
new_e = _wrap_proxy_error(new_e, conn.proxy.scheme)
raise new_e
# conn.request() calls http.client.*.request, not the method in
# urllib3.request. It also calls makefile (recv) on the socket.
try:
conn.request(
method,
url,
body=body,
headers=headers,
chunked=chunked,
preload_content=preload_content,
decode_content=decode_content,
enforce_content_length=enforce_content_length,
)
# We are swallowing BrokenPipeError (errno.EPIPE) since the server is
# legitimately able to close the connection after sending a valid response.
# With this behaviour, the received response is still readable.
except BrokenPipeError:
pass
except OSError as e:
# MacOS/Linux
# EPROTOTYPE and ECONNRESET are needed on macOS
# https://erickt.github.io/blog/2014/11/19/adventures-in-debugging-a-potential-osx-kernel-bug/
# Condition changed later to emit ECONNRESET instead of only EPROTOTYPE.
if e.errno != errno.EPROTOTYPE and e.errno != errno.ECONNRESET:
raise
# Reset the timeout for the recv() on the socket
read_timeout = timeout_obj.read_timeout
if not conn.is_closed:
# In Python 3 socket.py will catch EAGAIN and return None when you
# try and read into the file pointer created by http.client, which
# instead raises a BadStatusLine exception. Instead of catching
# the exception and assuming all BadStatusLine exceptions are read
# timeouts, check for a zero timeout before making the request.
if read_timeout == 0:
raise ReadTimeoutError(
self, url, f"Read timed out. (read timeout={read_timeout})"
)
conn.timeout = read_timeout
# Receive the response from the server
try:
response = conn.getresponse()
except (BaseSSLError, OSError) as e:
self._raise_timeout(err=e, url=url, timeout_value=read_timeout)
raise
# Set properties that are used by the pooling layer.
response.retries = retries
response._connection = response_conn # type: ignore[attr-defined]
response._pool = self # type: ignore[attr-defined]
# emscripten connection doesn't have _http_vsn_str
http_version = getattr(conn, "_http_vsn_str", "HTTP/?")
log.debug(
'%s://%s:%s "%s %s %s" %s %s',
self.scheme,
self.host,
self.port,
method,
url,
# HTTP version
http_version,
response.status,
response.length_remaining,
)
return response
def close(self) -> None:
"""
Close all pooled connections and disable the pool.
"""
if self.pool is None:
return
# Disable access to the pool
old_pool, self.pool = self.pool, None
# Close all the HTTPConnections in the pool.
_close_pool_connections(old_pool)
def is_same_host(self, url: str) -> bool:
"""
Check if the given ``url`` is a member of the same host as this
connection pool.
"""
if url.startswith("/"):
return True
# TODO: Add optional support for socket.gethostbyname checking.
scheme, _, host, port, *_ = parse_url(url)
scheme = scheme or "http"
if host is not None:
host = _normalize_host(host, scheme=scheme)
# Use explicit default port for comparison when none is given
if self.port and not port:
port = port_by_scheme.get(scheme)
elif not self.port and port == port_by_scheme.get(scheme):
port = None
return (scheme, host, port) == (self.scheme, self.host, self.port)
def urlopen( # type: ignore[override]
self,
method: str,
url: str,
body: _TYPE_BODY | None = None,
headers: typing.Mapping[str, str] | None = None,
retries: Retry | bool | int | None = None,
redirect: bool = True,
assert_same_host: bool = True,
timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT,
pool_timeout: int | None = None,
release_conn: bool | None = None,
chunked: bool = False,
body_pos: _TYPE_BODY_POSITION | None = None,
preload_content: bool = True,
decode_content: bool = True,
**response_kw: typing.Any,
) -> BaseHTTPResponse:
"""
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
such as :meth:`request`.
.. note::
`release_conn` will only behave as expected if
`preload_content=False` because we want to make
`preload_content=False` the default behaviour someday soon without
breaking backwards compatibility.
:param method:
HTTP request method (such as GET, POST, PUT, etc.)
:param url:
The URL to perform the request on.
:param body:
Data to send in the request body, either :class:`str`, :class:`bytes`,
an iterable of :class:`str`/:class:`bytes`, or a file-like object.
: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:
Configure the number of retries to allow before raising a
:class:`~urllib3.exceptions.MaxRetryError` exception.
If ``None`` (default) will retry 3 times, see ``Retry.DEFAULT``. Pass a
:class:`~urllib3.util.retry.Retry` object for fine-grained control
over different types of retries.
Pass an integer number to retry connection errors that many times,
but no other types of errors. Pass zero to never retry.
If ``False``, then retries are disabled and any exception is raised
immediately. Also, instead of raising a MaxRetryError on redirects,
the redirect response will be returned.
:type retries: :class:`~urllib3.util.retry.Retry`, False, or an int.
:param redirect:
If True, automatically handle redirects (status codes 301, 302,
303, 307, 308). Each redirect counts as a retry. Disabling retries
will disable redirect, too.
: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. It may be a float (in seconds) or an instance of
:class:`urllib3.util.Timeout`.
: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 bool preload_content:
If True, the response's body will be preloaded into memory.
:param bool decode_content:
If True, will attempt to decode the body based on the
'content-encoding' header.
:param release_conn:
If False, then the urlopen call will not release the connection
back into the pool once a response is received (but will release if
you read the entire contents of the response such as when
`preload_content=True`). 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 ``preload_content``
which defaults to ``True``.
:param bool chunked:
If True, urllib3 will send the body using chunked transfer
encoding. Otherwise, urllib3 will send the body using the standard
content-length form. Defaults to False.
:param int body_pos:
Position to seek to in file-like body in the event of a retry or
redirect. Typically this won't need to be set because urllib3 will
auto-populate the value when needed.
"""
parsed_url = parse_url(url)
destination_scheme = parsed_url.scheme
if headers is None:
headers = self.headers
if not isinstance(retries, Retry):
retries = Retry.from_int(retries, redirect=redirect, default=self.retries)
if release_conn is None:
release_conn = preload_content
# Check host
if assert_same_host and not self.is_same_host(url):
raise HostChangedError(self, url, retries)
# Ensure that the URL we're connecting to is properly encoded
if url.startswith("/"):
url = to_str(_encode_target(url))
else:
url = to_str(parsed_url.url)
conn = None
# Track whether `conn` needs to be released before
# returning/raising/recursing. Update this variable if necessary, and
# leave `release_conn` constant throughout the function. That way, if
# the function recurses, the original value of `release_conn` will be
# passed down into the recursive call, and its value will be respected.
#
# See issue #651 [1] for details.
#
# [1] <https://github.com/urllib3/urllib3/issues/651>
release_this_conn = release_conn
http_tunnel_required = connection_requires_http_tunnel(
self.proxy, self.proxy_config, destination_scheme
)
# Merge the proxy headers. Only done when not using HTTP CONNECT. We
# have to copy the headers dict so we can safely change it without those
# changes being reflected in anyone else's copy.
if not http_tunnel_required:
headers = headers.copy() # type: ignore[attr-defined]
headers.update(self.proxy_headers) # type: ignore[union-attr]
# Must keep the exception bound to a separate variable or else Python 3
# complains about UnboundLocalError.
err = None
# Keep track of whether we cleanly exited the except block. This
# ensures we do proper cleanup in finally.
clean_exit = False
# Rewind body position, if needed. Record current position
# for future rewinds in the event of a redirect/retry.
body_pos = set_file_position(body, body_pos)
try:
# Request a connection from the queue.
timeout_obj = self._get_timeout(timeout)
conn = self._get_conn(timeout=pool_timeout)
conn.timeout = timeout_obj.connect_timeout # type: ignore[assignment]
# Is this a closed/new connection that requires CONNECT tunnelling?
if self.proxy is not None and http_tunnel_required and conn.is_closed:
try:
self._prepare_proxy(conn)
except (BaseSSLError, OSError, SocketTimeout) as e:
self._raise_timeout(
err=e, url=self.proxy.url, timeout_value=conn.timeout
)
raise
# If we're going to release the connection in ``finally:``, then
# the response 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 = conn if not release_conn else None
# Make the request on the HTTPConnection object
response = self._make_request(
conn,
method,
url,
timeout=timeout_obj,
body=body,
headers=headers,
chunked=chunked,
retries=retries,
response_conn=response_conn,
preload_content=preload_content,
decode_content=decode_content,
**response_kw,
)
# Everything went great!
clean_exit = True
except EmptyPoolError:
# Didn't get a connection from the pool, no need to clean up
clean_exit = True
release_this_conn = False
raise
except (
TimeoutError,
HTTPException,
OSError,
ProtocolError,
BaseSSLError,
SSLError,
CertificateError,
ProxyError,
) as e:
# Discard the connection for these exceptions. It will be
# replaced during the next _get_conn() call.
clean_exit = False
new_e: Exception = e
if isinstance(e, (BaseSSLError, CertificateError)):
new_e = SSLError(e)
if isinstance(
new_e,
(
OSError,
NewConnectionError,
TimeoutError,
SSLError,
HTTPException,
),
) and (conn and conn.proxy and not conn.has_connected_to_proxy):
new_e = _wrap_proxy_error(new_e, conn.proxy.scheme)
elif isinstance(new_e, (OSError, HTTPException)):
new_e = ProtocolError("Connection aborted.", new_e)
retries = retries.increment(
method, url, error=new_e, _pool=self, _stacktrace=sys.exc_info()[2]
)
retries.sleep()
# Keep track of the error for the retry warning.
err = e
finally:
if not clean_exit:
# We hit some kind of exception, handled or otherwise. We need
# to throw the connection away unless explicitly told not to.
# Close the connection, set the variable to None, and make sure
# we put the None back in the pool to avoid leaking it.
if conn:
conn.close()
conn = None
release_this_conn = True
if release_this_conn:
# Put the connection back to be reused. If the connection is
# expired then it will be None, which will get replaced with a
# fresh connection during _get_conn.
self._put_conn(conn)
if not conn:
# Try again
log.warning(
"Retrying (%r) after connection broken by '%r': %s", retries, err, url
)
return self.urlopen(
method,
url,
body,
headers,
retries,
redirect,
assert_same_host,
timeout=timeout,
pool_timeout=pool_timeout,
release_conn=release_conn,
chunked=chunked,
body_pos=body_pos,
preload_content=preload_content,
decode_content=decode_content,
**response_kw,
)
# Handle redirect?
redirect_location = redirect and response.get_redirect_location()
if redirect_location:
if response.status == 303:
# Change the method according to RFC 9110, Section 15.4.4.
method = "GET"
# And lose the body not to transfer anything sensitive.
body = None
headers = HTTPHeaderDict(headers)._prepare_for_method_change()
try:
retries = retries.increment(method, url, response=response, _pool=self)
except MaxRetryError:
if retries.raise_on_redirect:
response.drain_conn()
raise
return response
response.drain_conn()
retries.sleep_for_retry(response)
log.debug("Redirecting %s -> %s", url, redirect_location)
return self.urlopen(
method,
redirect_location,
body,
headers,
retries=retries,
redirect=redirect,
assert_same_host=assert_same_host,
timeout=timeout,
pool_timeout=pool_timeout,
release_conn=release_conn,
chunked=chunked,
body_pos=body_pos,
preload_content=preload_content,
decode_content=decode_content,
**response_kw,
)
# Check if we should retry the HTTP response.
has_retry_after = bool(response.headers.get("Retry-After"))
if retries.is_retry(method, response.status, has_retry_after):
try:
retries = retries.increment(method, url, response=response, _pool=self)
except MaxRetryError:
if retries.raise_on_status:
response.drain_conn()
raise
return response
response.drain_conn()
retries.sleep(response)
log.debug("Retry: %s", url)
return self.urlopen(
method,
url,
body,
headers,
retries=retries,
redirect=redirect,
assert_same_host=assert_same_host,
timeout=timeout,
pool_timeout=pool_timeout,
release_conn=release_conn,
chunked=chunked,
body_pos=body_pos,
preload_content=preload_content,
decode_content=decode_content,
**response_kw,
)
return response
class HTTPSConnectionPool(HTTPConnectionPool):
"""
Same as :class:`.HTTPConnectionPool`, but HTTPS.
:class:`.HTTPSConnection` uses one of ``assert_fingerprint``,
``assert_hostname`` and ``host`` in this order to verify connections.
If ``assert_hostname`` is False, no verification is done.
The ``key_file``, ``cert_file``, ``cert_reqs``, ``ca_certs``,
``ca_cert_dir``, ``ssl_version``, ``key_password`` are only used if :mod:`ssl`
is available and are fed into :meth:`urllib3.util.ssl_wrap_socket` to upgrade
the connection socket into an SSL socket.
"""
scheme = "https"
ConnectionCls: type[BaseHTTPSConnection] = HTTPSConnection
def __init__(
self,
host: str,
port: int | None = None,
timeout: _TYPE_TIMEOUT | None = _DEFAULT_TIMEOUT,
maxsize: int = 1,
block: bool = False,
headers: typing.Mapping[str, str] | None = None,
retries: Retry | bool | int | None = None,
_proxy: Url | None = None,
_proxy_headers: typing.Mapping[str, str] | None = None,
key_file: str | None = None,
cert_file: str | None = None,
cert_reqs: int | str | None = None,
key_password: str | None = None,
ca_certs: str | None = None,
ssl_version: int | str | None = None,
ssl_minimum_version: ssl.TLSVersion | None = None,
ssl_maximum_version: ssl.TLSVersion | None = None,
assert_hostname: str | Literal[False] | None = None,
assert_fingerprint: str | None = None,
ca_cert_dir: str | None = None,
**conn_kw: typing.Any,
) -> None:
super().__init__(
host,
port,
timeout,
maxsize,
block,
headers,
retries,
_proxy,
_proxy_headers,
**conn_kw,
)
self.key_file = key_file
self.cert_file = cert_file
self.cert_reqs = cert_reqs
self.key_password = key_password
self.ca_certs = ca_certs
self.ca_cert_dir = ca_cert_dir
self.ssl_version = ssl_version
self.ssl_minimum_version = ssl_minimum_version
self.ssl_maximum_version = ssl_maximum_version
self.assert_hostname = assert_hostname
self.assert_fingerprint = assert_fingerprint
def _prepare_proxy(self, conn: HTTPSConnection) -> None: # type: ignore[override]
"""Establishes a tunnel connection through HTTP CONNECT."""
if self.proxy and self.proxy.scheme == "https":
tunnel_scheme = "https"
else:
tunnel_scheme = "http"
conn.set_tunnel(
scheme=tunnel_scheme,
host=self._tunnel_host,
port=self.port,
headers=self.proxy_headers,
)
conn.connect()
def _new_conn(self) -> BaseHTTPSConnection:
"""
Return a fresh :class:`urllib3.connection.HTTPConnection`.
"""
self.num_connections += 1
log.debug(
"Starting new HTTPS connection (%d): %s:%s",
self.num_connections,
self.host,
self.port or "443",
)
if not self.ConnectionCls or self.ConnectionCls is DummyConnection: # type: ignore[comparison-overlap]
raise ImportError(
"Can't connect to HTTPS URL because the SSL module is not available."
)
actual_host: str = self.host
actual_port = self.port
if self.proxy is not None and self.proxy.host is not None:
actual_host = self.proxy.host
actual_port = self.proxy.port
return self.ConnectionCls(
host=actual_host,
port=actual_port,
timeout=self.timeout.connect_timeout,
cert_file=self.cert_file,
key_file=self.key_file,
key_password=self.key_password,
cert_reqs=self.cert_reqs,
ca_certs=self.ca_certs,
ca_cert_dir=self.ca_cert_dir,
assert_hostname=self.assert_hostname,
assert_fingerprint=self.assert_fingerprint,
ssl_version=self.ssl_version,
ssl_minimum_version=self.ssl_minimum_version,
ssl_maximum_version=self.ssl_maximum_version,
**self.conn_kw,
)
def _validate_conn(self, conn: BaseHTTPConnection) -> None:
"""
Called right before a request is made, after the socket is created.
"""
super()._validate_conn(conn)
# Force connect early to allow us to validate the connection.
if conn.is_closed:
conn.connect()
# TODO revise this, see https://github.com/urllib3/urllib3/issues/2791
if not conn.is_verified and not conn.proxy_is_verified:
warnings.warn(
(
f"Unverified HTTPS request is being made to host '{conn.host}'. "
"Adding certificate verification is strongly advised. See: "
"https://urllib3.readthedocs.io/en/latest/advanced-usage.html"
"#tls-warnings"
),
InsecureRequestWarning,
)
def connection_from_url(url: str, **kw: typing.Any) -> HTTPConnectionPool:
"""
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, *_ = parse_url(url)
scheme = scheme or "http"
port = port or port_by_scheme.get(scheme, 80)
if scheme == "https":
return HTTPSConnectionPool(host, port=port, **kw) # type: ignore[arg-type]
else:
return HTTPConnectionPool(host, port=port, **kw) # type: ignore[arg-type]
@typing.overload
def _normalize_host(host: None, scheme: str | None) -> None:
...
@typing.overload
def _normalize_host(host: str, scheme: str | None) -> str:
...
def _normalize_host(host: str | None, scheme: str | None) -> str | None:
"""
Normalize hosts for comparisons and use with sockets.
"""
host = normalize_host(host, scheme)
# httplib doesn't like it when we include brackets in IPv6 addresses
# Specifically, if we include brackets but also pass the port then
# httplib crazily doubles up the square brackets on the Host header.
# Instead, we need to make sure we never pass ``None`` as the port.
# However, for backward compatibility reasons we can't actually
# *assert* that. See http://bugs.python.org/issue28539
if host and host.startswith("[") and host.endswith("]"):
host = host[1:-1]
return host
def _url_from_pool(
pool: HTTPConnectionPool | HTTPSConnectionPool, path: str | None = None
) -> str:
"""Returns the URL from a given connection pool. This is mainly used for testing and logging."""
return Url(scheme=pool.scheme, host=pool.host, port=pool.port, path=path).url
def _close_pool_connections(pool: queue.LifoQueue[typing.Any]) -> None:
"""Drains a queue of connections and closes each one."""
try:
while True:
conn = pool.get(block=False)
if conn:
conn.close()
except queue.Empty:
pass # Done.