view urllib3/contrib/socks.py @ 9:f9f1d0a0599a

planemo upload for repository https://toolrepo.galaxytrakr.org/view/jpayne/bioproject_to_srr_2/556cac4fb538
author jpayne
date Sun, 05 May 2024 23:58:19 -0400
parents 5eb2d5e3bf22
children
line wrap: on
line source
"""
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 annotations

try:
    import socks  # type: ignore[import-not-found]
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/latest/advanced-usage.html#socks-proxies"
        ),
        DependencyWarning,
    )
    raise

import typing
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  # type: ignore[assignment]

from typing import TypedDict


class _TYPE_SOCKS_OPTIONS(TypedDict):
    socks_version: int
    proxy_host: str | None
    proxy_port: str | None
    username: str | None
    password: str | None
    rdns: bool


class SOCKSConnection(HTTPConnection):
    """
    A plain-text HTTP connection that connects via a SOCKS proxy.
    """

    def __init__(
        self,
        _socks_options: _TYPE_SOCKS_OPTIONS,
        *args: typing.Any,
        **kwargs: typing.Any,
    ) -> None:
        self._socks_options = _socks_options
        super().__init__(*args, **kwargs)

    def _new_conn(self) -> socks.socksocket:
        """
        Establish a new connection via the SOCKS proxy.
        """
        extra_kw: dict[str, typing.Any] = {}
        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 as e:
            raise ConnectTimeoutError(
                self,
                f"Connection to {self.host} timed out. (connect timeout={self.timeout})",
            ) from e

        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,
                        f"Connection to {self.host} timed out. (connect timeout={self.timeout})",
                    ) from e
                else:
                    # Adding `from e` messes with coverage somehow, so it's omitted.
                    # See #2386.
                    raise NewConnectionError(
                        self, f"Failed to establish a new connection: {error}"
                    )
            else:
                raise NewConnectionError(
                    self, f"Failed to establish a new connection: {e}"
                ) from e

        except OSError as e:  # Defensive: PySocks should catch all these.
            raise NewConnectionError(
                self, f"Failed to establish a new connection: {e}"
            ) from 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: str,
        username: str | None = None,
        password: str | None = None,
        num_pools: int = 10,
        headers: typing.Mapping[str, str] | None = None,
        **connection_pool_kw: typing.Any,
    ):
        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(f"Unable to determine SOCKS version from {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().__init__(num_pools, headers, **connection_pool_kw)

        self.pool_classes_by_scheme = SOCKSProxyManager.pool_classes_by_scheme