replace bundled deps with built-ins (should've done this sooner lol)

This commit is contained in:
cere 2024-02-21 03:07:30 -05:00
parent f25eecefab
commit c172a6d976
84 changed files with 0 additions and 37538 deletions

View File

@ -54,9 +54,3 @@ class LibrespotServer(Bottle):
return response
except Exception as e:
log_msg(e)
def start_thread(web_port: int) -> None:
httpd = make_server('127.0.0.1', web_port, app)
monitor = xbmc.Monitor()
while not monitor.abortRequested():
httpd.handle_request()

View File

@ -1,4 +0,0 @@
from .core import contents, where
__all__ = ["contents", "where"]
__version__ = "2024.02.02"

View File

@ -1,12 +0,0 @@
import argparse
from certifi import contents, where
parser = argparse.ArgumentParser()
parser.add_argument("-c", "--contents", action="store_true")
args = parser.parse_args()
if args.contents:
print(contents())
else:
print(where())

File diff suppressed because it is too large Load Diff

View File

@ -1,114 +0,0 @@
"""
certifi.py
~~~~~~~~~~
This module returns the installation location of cacert.pem or its contents.
"""
import sys
import atexit
def exit_cacert_ctx() -> None:
_CACERT_CTX.__exit__(None, None, None) # type: ignore[union-attr]
if sys.version_info >= (3, 11):
from importlib.resources import as_file, files
_CACERT_CTX = None
_CACERT_PATH = None
def where() -> str:
# This is slightly terrible, but we want to delay extracting the file
# in cases where we're inside of a zipimport situation until someone
# actually calls where(), but we don't want to re-extract the file
# on every call of where(), so we'll do it once then store it in a
# global variable.
global _CACERT_CTX
global _CACERT_PATH
if _CACERT_PATH is None:
# This is slightly janky, the importlib.resources API wants you to
# manage the cleanup of this file, so it doesn't actually return a
# path, it returns a context manager that will give you the path
# when you enter it and will do any cleanup when you leave it. In
# the common case of not needing a temporary file, it will just
# return the file system location and the __exit__() is a no-op.
#
# We also have to hold onto the actual context manager, because
# it will do the cleanup whenever it gets garbage collected, so
# we will also store that at the global level as well.
_CACERT_CTX = as_file(files("certifi").joinpath("cacert.pem"))
_CACERT_PATH = str(_CACERT_CTX.__enter__())
atexit.register(exit_cacert_ctx)
return _CACERT_PATH
def contents() -> str:
return files("certifi").joinpath("cacert.pem").read_text(encoding="ascii")
elif sys.version_info >= (3, 7):
from importlib.resources import path as get_path, read_text
_CACERT_CTX = None
_CACERT_PATH = None
def where() -> str:
# This is slightly terrible, but we want to delay extracting the
# file in cases where we're inside of a zipimport situation until
# someone actually calls where(), but we don't want to re-extract
# the file on every call of where(), so we'll do it once then store
# it in a global variable.
global _CACERT_CTX
global _CACERT_PATH
if _CACERT_PATH is None:
# This is slightly janky, the importlib.resources API wants you
# to manage the cleanup of this file, so it doesn't actually
# return a path, it returns a context manager that will give
# you the path when you enter it and will do any cleanup when
# you leave it. In the common case of not needing a temporary
# file, it will just return the file system location and the
# __exit__() is a no-op.
#
# We also have to hold onto the actual context manager, because
# it will do the cleanup whenever it gets garbage collected, so
# we will also store that at the global level as well.
_CACERT_CTX = get_path("certifi", "cacert.pem")
_CACERT_PATH = str(_CACERT_CTX.__enter__())
atexit.register(exit_cacert_ctx)
return _CACERT_PATH
def contents() -> str:
return read_text("certifi", "cacert.pem", encoding="ascii")
else:
import os
import types
from typing import Union
Package = Union[types.ModuleType, str]
Resource = Union[str, "os.PathLike"]
# This fallback will work for Python versions prior to 3.7 that lack the
# importlib.resources module but relies on the existing `where` function
# so won't address issues with environments like PyOxidizer that don't set
# __file__ on modules.
def read_text(
package: Package,
resource: Resource,
encoding: str = 'utf-8',
errors: str = 'strict'
) -> str:
with open(where(), encoding=encoding) as data:
return data.read()
# If we don't have importlib.resources, then we will just do the old logic
# of assuming we're on the filesystem and munge the path directly.
def where() -> str:
f = os.path.dirname(__file__)
return os.path.join(f, "cacert.pem")
def contents() -> str:
return read_text("certifi", "cacert.pem", encoding="ascii")

View File

@ -1,46 +0,0 @@
# -*- coding: utf-8 -*-
"""
Charset-Normalizer
~~~~~~~~~~~~~~
The Real First Universal Charset Detector.
A library that helps you read text from an unknown charset encoding.
Motivated by chardet, This package is trying to resolve the issue by taking a new approach.
All IANA character set names for which the Python core library provides codecs are supported.
Basic usage:
>>> from charset_normalizer import from_bytes
>>> results = from_bytes('Bсеки човек има право на образование. Oбразованието!'.encode('utf_8'))
>>> best_guess = results.best()
>>> str(best_guess)
'Bсеки човек има право на образование. Oбразованието!'
Others methods and usages are available - see the full documentation
at <https://github.com/Ousret/charset_normalizer>.
:copyright: (c) 2021 by Ahmed TAHRI
:license: MIT, see LICENSE for more details.
"""
import logging
from .api import from_bytes, from_fp, from_path, is_binary
from .legacy import detect
from .models import CharsetMatch, CharsetMatches
from .utils import set_logging_handler
from .version import VERSION, __version__
__all__ = (
"from_fp",
"from_path",
"from_bytes",
"is_binary",
"detect",
"CharsetMatch",
"CharsetMatches",
"__version__",
"VERSION",
"set_logging_handler",
)
# Attach a NullHandler to the top level logger by default
# https://docs.python.org/3.3/howto/logging.html#configuring-logging-for-a-library
logging.getLogger("charset_normalizer").addHandler(logging.NullHandler())

View File

@ -1,4 +0,0 @@
from .cli import cli_detect
if __name__ == "__main__":
cli_detect()

View File

@ -1,626 +0,0 @@
import logging
from os import PathLike
from typing import BinaryIO, List, Optional, Set, Union
from .cd import (
coherence_ratio,
encoding_languages,
mb_encoding_languages,
merge_coherence_ratios,
)
from .constant import IANA_SUPPORTED, TOO_BIG_SEQUENCE, TOO_SMALL_SEQUENCE, TRACE
from .md import mess_ratio
from .models import CharsetMatch, CharsetMatches
from .utils import (
any_specified_encoding,
cut_sequence_chunks,
iana_name,
identify_sig_or_bom,
is_cp_similar,
is_multi_byte_encoding,
should_strip_sig_or_bom,
)
# Will most likely be controversial
# logging.addLevelName(TRACE, "TRACE")
logger = logging.getLogger("charset_normalizer")
explain_handler = logging.StreamHandler()
explain_handler.setFormatter(
logging.Formatter("%(asctime)s | %(levelname)s | %(message)s")
)
def from_bytes(
sequences: Union[bytes, bytearray],
steps: int = 5,
chunk_size: int = 512,
threshold: float = 0.2,
cp_isolation: Optional[List[str]] = None,
cp_exclusion: Optional[List[str]] = None,
preemptive_behaviour: bool = True,
explain: bool = False,
language_threshold: float = 0.1,
enable_fallback: bool = True,
) -> CharsetMatches:
"""
Given a raw bytes sequence, return the best possibles charset usable to render str objects.
If there is no results, it is a strong indicator that the source is binary/not text.
By default, the process will extract 5 blocks of 512o each to assess the mess and coherence of a given sequence.
And will give up a particular code page after 20% of measured mess. Those criteria are customizable at will.
The preemptive behavior DOES NOT replace the traditional detection workflow, it prioritize a particular code page
but never take it for granted. Can improve the performance.
You may want to focus your attention to some code page or/and not others, use cp_isolation and cp_exclusion for that
purpose.
This function will strip the SIG in the payload/sequence every time except on UTF-16, UTF-32.
By default the library does not setup any handler other than the NullHandler, if you choose to set the 'explain'
toggle to True it will alter the logger configuration to add a StreamHandler that is suitable for debugging.
Custom logging format and handler can be set manually.
"""
if not isinstance(sequences, (bytearray, bytes)):
raise TypeError(
"Expected object of type bytes or bytearray, got: {0}".format(
type(sequences)
)
)
if explain:
previous_logger_level: int = logger.level
logger.addHandler(explain_handler)
logger.setLevel(TRACE)
length: int = len(sequences)
if length == 0:
logger.debug("Encoding detection on empty bytes, assuming utf_8 intention.")
if explain:
logger.removeHandler(explain_handler)
logger.setLevel(previous_logger_level or logging.WARNING)
return CharsetMatches([CharsetMatch(sequences, "utf_8", 0.0, False, [], "")])
if cp_isolation is not None:
logger.log(
TRACE,
"cp_isolation is set. use this flag for debugging purpose. "
"limited list of encoding allowed : %s.",
", ".join(cp_isolation),
)
cp_isolation = [iana_name(cp, False) for cp in cp_isolation]
else:
cp_isolation = []
if cp_exclusion is not None:
logger.log(
TRACE,
"cp_exclusion is set. use this flag for debugging purpose. "
"limited list of encoding excluded : %s.",
", ".join(cp_exclusion),
)
cp_exclusion = [iana_name(cp, False) for cp in cp_exclusion]
else:
cp_exclusion = []
if length <= (chunk_size * steps):
logger.log(
TRACE,
"override steps (%i) and chunk_size (%i) as content does not fit (%i byte(s) given) parameters.",
steps,
chunk_size,
length,
)
steps = 1
chunk_size = length
if steps > 1 and length / steps < chunk_size:
chunk_size = int(length / steps)
is_too_small_sequence: bool = len(sequences) < TOO_SMALL_SEQUENCE
is_too_large_sequence: bool = len(sequences) >= TOO_BIG_SEQUENCE
if is_too_small_sequence:
logger.log(
TRACE,
"Trying to detect encoding from a tiny portion of ({}) byte(s).".format(
length
),
)
elif is_too_large_sequence:
logger.log(
TRACE,
"Using lazy str decoding because the payload is quite large, ({}) byte(s).".format(
length
),
)
prioritized_encodings: List[str] = []
specified_encoding: Optional[str] = (
any_specified_encoding(sequences) if preemptive_behaviour else None
)
if specified_encoding is not None:
prioritized_encodings.append(specified_encoding)
logger.log(
TRACE,
"Detected declarative mark in sequence. Priority +1 given for %s.",
specified_encoding,
)
tested: Set[str] = set()
tested_but_hard_failure: List[str] = []
tested_but_soft_failure: List[str] = []
fallback_ascii: Optional[CharsetMatch] = None
fallback_u8: Optional[CharsetMatch] = None
fallback_specified: Optional[CharsetMatch] = None
results: CharsetMatches = CharsetMatches()
sig_encoding, sig_payload = identify_sig_or_bom(sequences)
if sig_encoding is not None:
prioritized_encodings.append(sig_encoding)
logger.log(
TRACE,
"Detected a SIG or BOM mark on first %i byte(s). Priority +1 given for %s.",
len(sig_payload),
sig_encoding,
)
prioritized_encodings.append("ascii")
if "utf_8" not in prioritized_encodings:
prioritized_encodings.append("utf_8")
for encoding_iana in prioritized_encodings + IANA_SUPPORTED:
if cp_isolation and encoding_iana not in cp_isolation:
continue
if cp_exclusion and encoding_iana in cp_exclusion:
continue
if encoding_iana in tested:
continue
tested.add(encoding_iana)
decoded_payload: Optional[str] = None
bom_or_sig_available: bool = sig_encoding == encoding_iana
strip_sig_or_bom: bool = bom_or_sig_available and should_strip_sig_or_bom(
encoding_iana
)
if encoding_iana in {"utf_16", "utf_32"} and not bom_or_sig_available:
logger.log(
TRACE,
"Encoding %s won't be tested as-is because it require a BOM. Will try some sub-encoder LE/BE.",
encoding_iana,
)
continue
if encoding_iana in {"utf_7"} and not bom_or_sig_available:
logger.log(
TRACE,
"Encoding %s won't be tested as-is because detection is unreliable without BOM/SIG.",
encoding_iana,
)
continue
try:
is_multi_byte_decoder: bool = is_multi_byte_encoding(encoding_iana)
except (ModuleNotFoundError, ImportError):
logger.log(
TRACE,
"Encoding %s does not provide an IncrementalDecoder",
encoding_iana,
)
continue
try:
if is_too_large_sequence and is_multi_byte_decoder is False:
str(
sequences[: int(50e4)]
if strip_sig_or_bom is False
else sequences[len(sig_payload) : int(50e4)],
encoding=encoding_iana,
)
else:
decoded_payload = str(
sequences
if strip_sig_or_bom is False
else sequences[len(sig_payload) :],
encoding=encoding_iana,
)
except (UnicodeDecodeError, LookupError) as e:
if not isinstance(e, LookupError):
logger.log(
TRACE,
"Code page %s does not fit given bytes sequence at ALL. %s",
encoding_iana,
str(e),
)
tested_but_hard_failure.append(encoding_iana)
continue
similar_soft_failure_test: bool = False
for encoding_soft_failed in tested_but_soft_failure:
if is_cp_similar(encoding_iana, encoding_soft_failed):
similar_soft_failure_test = True
break
if similar_soft_failure_test:
logger.log(
TRACE,
"%s is deemed too similar to code page %s and was consider unsuited already. Continuing!",
encoding_iana,
encoding_soft_failed,
)
continue
r_ = range(
0 if not bom_or_sig_available else len(sig_payload),
length,
int(length / steps),
)
multi_byte_bonus: bool = (
is_multi_byte_decoder
and decoded_payload is not None
and len(decoded_payload) < length
)
if multi_byte_bonus:
logger.log(
TRACE,
"Code page %s is a multi byte encoding table and it appear that at least one character "
"was encoded using n-bytes.",
encoding_iana,
)
max_chunk_gave_up: int = int(len(r_) / 4)
max_chunk_gave_up = max(max_chunk_gave_up, 2)
early_stop_count: int = 0
lazy_str_hard_failure = False
md_chunks: List[str] = []
md_ratios = []
try:
for chunk in cut_sequence_chunks(
sequences,
encoding_iana,
r_,
chunk_size,
bom_or_sig_available,
strip_sig_or_bom,
sig_payload,
is_multi_byte_decoder,
decoded_payload,
):
md_chunks.append(chunk)
md_ratios.append(
mess_ratio(
chunk,
threshold,
explain is True and 1 <= len(cp_isolation) <= 2,
)
)
if md_ratios[-1] >= threshold:
early_stop_count += 1
if (early_stop_count >= max_chunk_gave_up) or (
bom_or_sig_available and strip_sig_or_bom is False
):
break
except (
UnicodeDecodeError
) as e: # Lazy str loading may have missed something there
logger.log(
TRACE,
"LazyStr Loading: After MD chunk decode, code page %s does not fit given bytes sequence at ALL. %s",
encoding_iana,
str(e),
)
early_stop_count = max_chunk_gave_up
lazy_str_hard_failure = True
# We might want to check the sequence again with the whole content
# Only if initial MD tests passes
if (
not lazy_str_hard_failure
and is_too_large_sequence
and not is_multi_byte_decoder
):
try:
sequences[int(50e3) :].decode(encoding_iana, errors="strict")
except UnicodeDecodeError as e:
logger.log(
TRACE,
"LazyStr Loading: After final lookup, code page %s does not fit given bytes sequence at ALL. %s",
encoding_iana,
str(e),
)
tested_but_hard_failure.append(encoding_iana)
continue
mean_mess_ratio: float = sum(md_ratios) / len(md_ratios) if md_ratios else 0.0
if mean_mess_ratio >= threshold or early_stop_count >= max_chunk_gave_up:
tested_but_soft_failure.append(encoding_iana)
logger.log(
TRACE,
"%s was excluded because of initial chaos probing. Gave up %i time(s). "
"Computed mean chaos is %f %%.",
encoding_iana,
early_stop_count,
round(mean_mess_ratio * 100, ndigits=3),
)
# Preparing those fallbacks in case we got nothing.
if (
enable_fallback
and encoding_iana in ["ascii", "utf_8", specified_encoding]
and not lazy_str_hard_failure
):
fallback_entry = CharsetMatch(
sequences, encoding_iana, threshold, False, [], decoded_payload
)
if encoding_iana == specified_encoding:
fallback_specified = fallback_entry
elif encoding_iana == "ascii":
fallback_ascii = fallback_entry
else:
fallback_u8 = fallback_entry
continue
logger.log(
TRACE,
"%s passed initial chaos probing. Mean measured chaos is %f %%",
encoding_iana,
round(mean_mess_ratio * 100, ndigits=3),
)
if not is_multi_byte_decoder:
target_languages: List[str] = encoding_languages(encoding_iana)
else:
target_languages = mb_encoding_languages(encoding_iana)
if target_languages:
logger.log(
TRACE,
"{} should target any language(s) of {}".format(
encoding_iana, str(target_languages)
),
)
cd_ratios = []
# We shall skip the CD when its about ASCII
# Most of the time its not relevant to run "language-detection" on it.
if encoding_iana != "ascii":
for chunk in md_chunks:
chunk_languages = coherence_ratio(
chunk,
language_threshold,
",".join(target_languages) if target_languages else None,
)
cd_ratios.append(chunk_languages)
cd_ratios_merged = merge_coherence_ratios(cd_ratios)
if cd_ratios_merged:
logger.log(
TRACE,
"We detected language {} using {}".format(
cd_ratios_merged, encoding_iana
),
)
results.append(
CharsetMatch(
sequences,
encoding_iana,
mean_mess_ratio,
bom_or_sig_available,
cd_ratios_merged,
decoded_payload,
)
)
if (
encoding_iana in [specified_encoding, "ascii", "utf_8"]
and mean_mess_ratio < 0.1
):
logger.debug(
"Encoding detection: %s is most likely the one.", encoding_iana
)
if explain:
logger.removeHandler(explain_handler)
logger.setLevel(previous_logger_level)
return CharsetMatches([results[encoding_iana]])
if encoding_iana == sig_encoding:
logger.debug(
"Encoding detection: %s is most likely the one as we detected a BOM or SIG within "
"the beginning of the sequence.",
encoding_iana,
)
if explain:
logger.removeHandler(explain_handler)
logger.setLevel(previous_logger_level)
return CharsetMatches([results[encoding_iana]])
if len(results) == 0:
if fallback_u8 or fallback_ascii or fallback_specified:
logger.log(
TRACE,
"Nothing got out of the detection process. Using ASCII/UTF-8/Specified fallback.",
)
if fallback_specified:
logger.debug(
"Encoding detection: %s will be used as a fallback match",
fallback_specified.encoding,
)
results.append(fallback_specified)
elif (
(fallback_u8 and fallback_ascii is None)
or (
fallback_u8
and fallback_ascii
and fallback_u8.fingerprint != fallback_ascii.fingerprint
)
or (fallback_u8 is not None)
):
logger.debug("Encoding detection: utf_8 will be used as a fallback match")
results.append(fallback_u8)
elif fallback_ascii:
logger.debug("Encoding detection: ascii will be used as a fallback match")
results.append(fallback_ascii)
if results:
logger.debug(
"Encoding detection: Found %s as plausible (best-candidate) for content. With %i alternatives.",
results.best().encoding, # type: ignore
len(results) - 1,
)
else:
logger.debug("Encoding detection: Unable to determine any suitable charset.")
if explain:
logger.removeHandler(explain_handler)
logger.setLevel(previous_logger_level)
return results
def from_fp(
fp: BinaryIO,
steps: int = 5,
chunk_size: int = 512,
threshold: float = 0.20,
cp_isolation: Optional[List[str]] = None,
cp_exclusion: Optional[List[str]] = None,
preemptive_behaviour: bool = True,
explain: bool = False,
language_threshold: float = 0.1,
enable_fallback: bool = True,
) -> CharsetMatches:
"""
Same thing than the function from_bytes but using a file pointer that is already ready.
Will not close the file pointer.
"""
return from_bytes(
fp.read(),
steps,
chunk_size,
threshold,
cp_isolation,
cp_exclusion,
preemptive_behaviour,
explain,
language_threshold,
enable_fallback,
)
def from_path(
path: Union[str, bytes, PathLike], # type: ignore[type-arg]
steps: int = 5,
chunk_size: int = 512,
threshold: float = 0.20,
cp_isolation: Optional[List[str]] = None,
cp_exclusion: Optional[List[str]] = None,
preemptive_behaviour: bool = True,
explain: bool = False,
language_threshold: float = 0.1,
enable_fallback: bool = True,
) -> CharsetMatches:
"""
Same thing than the function from_bytes but with one extra step. Opening and reading given file path in binary mode.
Can raise IOError.
"""
with open(path, "rb") as fp:
return from_fp(
fp,
steps,
chunk_size,
threshold,
cp_isolation,
cp_exclusion,
preemptive_behaviour,
explain,
language_threshold,
enable_fallback,
)
def is_binary(
fp_or_path_or_payload: Union[PathLike, str, BinaryIO, bytes], # type: ignore[type-arg]
steps: int = 5,
chunk_size: int = 512,
threshold: float = 0.20,
cp_isolation: Optional[List[str]] = None,
cp_exclusion: Optional[List[str]] = None,
preemptive_behaviour: bool = True,
explain: bool = False,
language_threshold: float = 0.1,
enable_fallback: bool = False,
) -> bool:
"""
Detect if the given input (file, bytes, or path) points to a binary file. aka. not a string.
Based on the same main heuristic algorithms and default kwargs at the sole exception that fallbacks match
are disabled to be stricter around ASCII-compatible but unlikely to be a string.
"""
if isinstance(fp_or_path_or_payload, (str, PathLike)):
guesses = from_path(
fp_or_path_or_payload,
steps=steps,
chunk_size=chunk_size,
threshold=threshold,
cp_isolation=cp_isolation,
cp_exclusion=cp_exclusion,
preemptive_behaviour=preemptive_behaviour,
explain=explain,
language_threshold=language_threshold,
enable_fallback=enable_fallback,
)
elif isinstance(
fp_or_path_or_payload,
(
bytes,
bytearray,
),
):
guesses = from_bytes(
fp_or_path_or_payload,
steps=steps,
chunk_size=chunk_size,
threshold=threshold,
cp_isolation=cp_isolation,
cp_exclusion=cp_exclusion,
preemptive_behaviour=preemptive_behaviour,
explain=explain,
language_threshold=language_threshold,
enable_fallback=enable_fallback,
)
else:
guesses = from_fp(
fp_or_path_or_payload,
steps=steps,
chunk_size=chunk_size,
threshold=threshold,
cp_isolation=cp_isolation,
cp_exclusion=cp_exclusion,
preemptive_behaviour=preemptive_behaviour,
explain=explain,
language_threshold=language_threshold,
enable_fallback=enable_fallback,
)
return not guesses

View File

@ -1,395 +0,0 @@
import importlib
from codecs import IncrementalDecoder
from collections import Counter
from functools import lru_cache
from typing import Counter as TypeCounter, Dict, List, Optional, Tuple
from .constant import (
FREQUENCIES,
KO_NAMES,
LANGUAGE_SUPPORTED_COUNT,
TOO_SMALL_SEQUENCE,
ZH_NAMES,
)
from .md import is_suspiciously_successive_range
from .models import CoherenceMatches
from .utils import (
is_accentuated,
is_latin,
is_multi_byte_encoding,
is_unicode_range_secondary,
unicode_range,
)
def encoding_unicode_range(iana_name: str) -> List[str]:
"""
Return associated unicode ranges in a single byte code page.
"""
if is_multi_byte_encoding(iana_name):
raise IOError("Function not supported on multi-byte code page")
decoder = importlib.import_module(
"encodings.{}".format(iana_name)
).IncrementalDecoder
p: IncrementalDecoder = decoder(errors="ignore")
seen_ranges: Dict[str, int] = {}
character_count: int = 0
for i in range(0x40, 0xFF):
chunk: str = p.decode(bytes([i]))
if chunk:
character_range: Optional[str] = unicode_range(chunk)
if character_range is None:
continue
if is_unicode_range_secondary(character_range) is False:
if character_range not in seen_ranges:
seen_ranges[character_range] = 0
seen_ranges[character_range] += 1
character_count += 1
return sorted(
[
character_range
for character_range in seen_ranges
if seen_ranges[character_range] / character_count >= 0.15
]
)
def unicode_range_languages(primary_range: str) -> List[str]:
"""
Return inferred languages used with a unicode range.
"""
languages: List[str] = []
for language, characters in FREQUENCIES.items():
for character in characters:
if unicode_range(character) == primary_range:
languages.append(language)
break
return languages
@lru_cache()
def encoding_languages(iana_name: str) -> List[str]:
"""
Single-byte encoding language association. Some code page are heavily linked to particular language(s).
This function does the correspondence.
"""
unicode_ranges: List[str] = encoding_unicode_range(iana_name)
primary_range: Optional[str] = None
for specified_range in unicode_ranges:
if "Latin" not in specified_range:
primary_range = specified_range
break
if primary_range is None:
return ["Latin Based"]
return unicode_range_languages(primary_range)
@lru_cache()
def mb_encoding_languages(iana_name: str) -> List[str]:
"""
Multi-byte encoding language association. Some code page are heavily linked to particular language(s).
This function does the correspondence.
"""
if (
iana_name.startswith("shift_")
or iana_name.startswith("iso2022_jp")
or iana_name.startswith("euc_j")
or iana_name == "cp932"
):
return ["Japanese"]
if iana_name.startswith("gb") or iana_name in ZH_NAMES:
return ["Chinese"]
if iana_name.startswith("iso2022_kr") or iana_name in KO_NAMES:
return ["Korean"]
return []
@lru_cache(maxsize=LANGUAGE_SUPPORTED_COUNT)
def get_target_features(language: str) -> Tuple[bool, bool]:
"""
Determine main aspects from a supported language if it contains accents and if is pure Latin.
"""
target_have_accents: bool = False
target_pure_latin: bool = True
for character in FREQUENCIES[language]:
if not target_have_accents and is_accentuated(character):
target_have_accents = True
if target_pure_latin and is_latin(character) is False:
target_pure_latin = False
return target_have_accents, target_pure_latin
def alphabet_languages(
characters: List[str], ignore_non_latin: bool = False
) -> List[str]:
"""
Return associated languages associated to given characters.
"""
languages: List[Tuple[str, float]] = []
source_have_accents = any(is_accentuated(character) for character in characters)
for language, language_characters in FREQUENCIES.items():
target_have_accents, target_pure_latin = get_target_features(language)
if ignore_non_latin and target_pure_latin is False:
continue
if target_have_accents is False and source_have_accents:
continue
character_count: int = len(language_characters)
character_match_count: int = len(
[c for c in language_characters if c in characters]
)
ratio: float = character_match_count / character_count
if ratio >= 0.2:
languages.append((language, ratio))
languages = sorted(languages, key=lambda x: x[1], reverse=True)
return [compatible_language[0] for compatible_language in languages]
def characters_popularity_compare(
language: str, ordered_characters: List[str]
) -> float:
"""
Determine if a ordered characters list (by occurrence from most appearance to rarest) match a particular language.
The result is a ratio between 0. (absolutely no correspondence) and 1. (near perfect fit).
Beware that is function is not strict on the match in order to ease the detection. (Meaning close match is 1.)
"""
if language not in FREQUENCIES:
raise ValueError("{} not available".format(language))
character_approved_count: int = 0
FREQUENCIES_language_set = set(FREQUENCIES[language])
ordered_characters_count: int = len(ordered_characters)
target_language_characters_count: int = len(FREQUENCIES[language])
large_alphabet: bool = target_language_characters_count > 26
for character, character_rank in zip(
ordered_characters, range(0, ordered_characters_count)
):
if character not in FREQUENCIES_language_set:
continue
character_rank_in_language: int = FREQUENCIES[language].index(character)
expected_projection_ratio: float = (
target_language_characters_count / ordered_characters_count
)
character_rank_projection: int = int(character_rank * expected_projection_ratio)
if (
large_alphabet is False
and abs(character_rank_projection - character_rank_in_language) > 4
):
continue
if (
large_alphabet is True
and abs(character_rank_projection - character_rank_in_language)
< target_language_characters_count / 3
):
character_approved_count += 1
continue
characters_before_source: List[str] = FREQUENCIES[language][
0:character_rank_in_language
]
characters_after_source: List[str] = FREQUENCIES[language][
character_rank_in_language:
]
characters_before: List[str] = ordered_characters[0:character_rank]
characters_after: List[str] = ordered_characters[character_rank:]
before_match_count: int = len(
set(characters_before) & set(characters_before_source)
)
after_match_count: int = len(
set(characters_after) & set(characters_after_source)
)
if len(characters_before_source) == 0 and before_match_count <= 4:
character_approved_count += 1
continue
if len(characters_after_source) == 0 and after_match_count <= 4:
character_approved_count += 1
continue
if (
before_match_count / len(characters_before_source) >= 0.4
or after_match_count / len(characters_after_source) >= 0.4
):
character_approved_count += 1
continue
return character_approved_count / len(ordered_characters)
def alpha_unicode_split(decoded_sequence: str) -> List[str]:
"""
Given a decoded text sequence, return a list of str. Unicode range / alphabet separation.
Ex. a text containing English/Latin with a bit a Hebrew will return two items in the resulting list;
One containing the latin letters and the other hebrew.
"""
layers: Dict[str, str] = {}
for character in decoded_sequence:
if character.isalpha() is False:
continue
character_range: Optional[str] = unicode_range(character)
if character_range is None:
continue
layer_target_range: Optional[str] = None
for discovered_range in layers:
if (
is_suspiciously_successive_range(discovered_range, character_range)
is False
):
layer_target_range = discovered_range
break
if layer_target_range is None:
layer_target_range = character_range
if layer_target_range not in layers:
layers[layer_target_range] = character.lower()
continue
layers[layer_target_range] += character.lower()
return list(layers.values())
def merge_coherence_ratios(results: List[CoherenceMatches]) -> CoherenceMatches:
"""
This function merge results previously given by the function coherence_ratio.
The return type is the same as coherence_ratio.
"""
per_language_ratios: Dict[str, List[float]] = {}
for result in results:
for sub_result in result:
language, ratio = sub_result
if language not in per_language_ratios:
per_language_ratios[language] = [ratio]
continue
per_language_ratios[language].append(ratio)
merge = [
(
language,
round(
sum(per_language_ratios[language]) / len(per_language_ratios[language]),
4,
),
)
for language in per_language_ratios
]
return sorted(merge, key=lambda x: x[1], reverse=True)
def filter_alt_coherence_matches(results: CoherenceMatches) -> CoherenceMatches:
"""
We shall NOT return "English—" in CoherenceMatches because it is an alternative
of "English". This function only keeps the best match and remove the em-dash in it.
"""
index_results: Dict[str, List[float]] = dict()
for result in results:
language, ratio = result
no_em_name: str = language.replace("", "")
if no_em_name not in index_results:
index_results[no_em_name] = []
index_results[no_em_name].append(ratio)
if any(len(index_results[e]) > 1 for e in index_results):
filtered_results: CoherenceMatches = []
for language in index_results:
filtered_results.append((language, max(index_results[language])))
return filtered_results
return results
@lru_cache(maxsize=2048)
def coherence_ratio(
decoded_sequence: str, threshold: float = 0.1, lg_inclusion: Optional[str] = None
) -> CoherenceMatches:
"""
Detect ANY language that can be identified in given sequence. The sequence will be analysed by layers.
A layer = Character extraction by alphabets/ranges.
"""
results: List[Tuple[str, float]] = []
ignore_non_latin: bool = False
sufficient_match_count: int = 0
lg_inclusion_list = lg_inclusion.split(",") if lg_inclusion is not None else []
if "Latin Based" in lg_inclusion_list:
ignore_non_latin = True
lg_inclusion_list.remove("Latin Based")
for layer in alpha_unicode_split(decoded_sequence):
sequence_frequencies: TypeCounter[str] = Counter(layer)
most_common = sequence_frequencies.most_common()
character_count: int = sum(o for c, o in most_common)
if character_count <= TOO_SMALL_SEQUENCE:
continue
popular_character_ordered: List[str] = [c for c, o in most_common]
for language in lg_inclusion_list or alphabet_languages(
popular_character_ordered, ignore_non_latin
):
ratio: float = characters_popularity_compare(
language, popular_character_ordered
)
if ratio < threshold:
continue
elif ratio >= 0.8:
sufficient_match_count += 1
results.append((language, round(ratio, 4)))
if sufficient_match_count >= 3:
break
return sorted(
filter_alt_coherence_matches(results), key=lambda x: x[1], reverse=True
)

View File

@ -1,6 +0,0 @@
from .__main__ import cli_detect, query_yes_no
__all__ = (
"cli_detect",
"query_yes_no",
)

View File

@ -1,296 +0,0 @@
import argparse
import sys
from json import dumps
from os.path import abspath, basename, dirname, join, realpath
from platform import python_version
from typing import List, Optional
from unicodedata import unidata_version
import charset_normalizer.md as md_module
from charset_normalizer import from_fp
from charset_normalizer.models import CliDetectionResult
from charset_normalizer.version import __version__
def query_yes_no(question: str, default: str = "yes") -> bool:
"""Ask a yes/no question via input() and return their answer.
"question" is a string that is presented to the user.
"default" is the presumed answer if the user just hits <Enter>.
It must be "yes" (the default), "no" or None (meaning
an answer is required of the user).
The "answer" return value is True for "yes" or False for "no".
Credit goes to (c) https://stackoverflow.com/questions/3041986/apt-command-line-interface-like-yes-no-input
"""
valid = {"yes": True, "y": True, "ye": True, "no": False, "n": False}
if default is None:
prompt = " [y/n] "
elif default == "yes":
prompt = " [Y/n] "
elif default == "no":
prompt = " [y/N] "
else:
raise ValueError("invalid default answer: '%s'" % default)
while True:
sys.stdout.write(question + prompt)
choice = input().lower()
if default is not None and choice == "":
return valid[default]
elif choice in valid:
return valid[choice]
else:
sys.stdout.write("Please respond with 'yes' or 'no' " "(or 'y' or 'n').\n")
def cli_detect(argv: Optional[List[str]] = None) -> int:
"""
CLI assistant using ARGV and ArgumentParser
:param argv:
:return: 0 if everything is fine, anything else equal trouble
"""
parser = argparse.ArgumentParser(
description="The Real First Universal Charset Detector. "
"Discover originating encoding used on text file. "
"Normalize text to unicode."
)
parser.add_argument(
"files", type=argparse.FileType("rb"), nargs="+", help="File(s) to be analysed"
)
parser.add_argument(
"-v",
"--verbose",
action="store_true",
default=False,
dest="verbose",
help="Display complementary information about file if any. "
"Stdout will contain logs about the detection process.",
)
parser.add_argument(
"-a",
"--with-alternative",
action="store_true",
default=False,
dest="alternatives",
help="Output complementary possibilities if any. Top-level JSON WILL be a list.",
)
parser.add_argument(
"-n",
"--normalize",
action="store_true",
default=False,
dest="normalize",
help="Permit to normalize input file. If not set, program does not write anything.",
)
parser.add_argument(
"-m",
"--minimal",
action="store_true",
default=False,
dest="minimal",
help="Only output the charset detected to STDOUT. Disabling JSON output.",
)
parser.add_argument(
"-r",
"--replace",
action="store_true",
default=False,
dest="replace",
help="Replace file when trying to normalize it instead of creating a new one.",
)
parser.add_argument(
"-f",
"--force",
action="store_true",
default=False,
dest="force",
help="Replace file without asking if you are sure, use this flag with caution.",
)
parser.add_argument(
"-t",
"--threshold",
action="store",
default=0.2,
type=float,
dest="threshold",
help="Define a custom maximum amount of chaos allowed in decoded content. 0. <= chaos <= 1.",
)
parser.add_argument(
"--version",
action="version",
version="Charset-Normalizer {} - Python {} - Unicode {} - SpeedUp {}".format(
__version__,
python_version(),
unidata_version,
"OFF" if md_module.__file__.lower().endswith(".py") else "ON",
),
help="Show version information and exit.",
)
args = parser.parse_args(argv)
if args.replace is True and args.normalize is False:
print("Use --replace in addition of --normalize only.", file=sys.stderr)
return 1
if args.force is True and args.replace is False:
print("Use --force in addition of --replace only.", file=sys.stderr)
return 1
if args.threshold < 0.0 or args.threshold > 1.0:
print("--threshold VALUE should be between 0. AND 1.", file=sys.stderr)
return 1
x_ = []
for my_file in args.files:
matches = from_fp(my_file, threshold=args.threshold, explain=args.verbose)
best_guess = matches.best()
if best_guess is None:
print(
'Unable to identify originating encoding for "{}". {}'.format(
my_file.name,
"Maybe try increasing maximum amount of chaos."
if args.threshold < 1.0
else "",
),
file=sys.stderr,
)
x_.append(
CliDetectionResult(
abspath(my_file.name),
None,
[],
[],
"Unknown",
[],
False,
1.0,
0.0,
None,
True,
)
)
else:
x_.append(
CliDetectionResult(
abspath(my_file.name),
best_guess.encoding,
best_guess.encoding_aliases,
[
cp
for cp in best_guess.could_be_from_charset
if cp != best_guess.encoding
],
best_guess.language,
best_guess.alphabets,
best_guess.bom,
best_guess.percent_chaos,
best_guess.percent_coherence,
None,
True,
)
)
if len(matches) > 1 and args.alternatives:
for el in matches:
if el != best_guess:
x_.append(
CliDetectionResult(
abspath(my_file.name),
el.encoding,
el.encoding_aliases,
[
cp
for cp in el.could_be_from_charset
if cp != el.encoding
],
el.language,
el.alphabets,
el.bom,
el.percent_chaos,
el.percent_coherence,
None,
False,
)
)
if args.normalize is True:
if best_guess.encoding.startswith("utf") is True:
print(
'"{}" file does not need to be normalized, as it already came from unicode.'.format(
my_file.name
),
file=sys.stderr,
)
if my_file.closed is False:
my_file.close()
continue
dir_path = dirname(realpath(my_file.name))
file_name = basename(realpath(my_file.name))
o_: List[str] = file_name.split(".")
if args.replace is False:
o_.insert(-1, best_guess.encoding)
if my_file.closed is False:
my_file.close()
elif (
args.force is False
and query_yes_no(
'Are you sure to normalize "{}" by replacing it ?'.format(
my_file.name
),
"no",
)
is False
):
if my_file.closed is False:
my_file.close()
continue
try:
x_[0].unicode_path = join(dir_path, ".".join(o_))
with open(x_[0].unicode_path, "w", encoding="utf-8") as fp:
fp.write(str(best_guess))
except IOError as e:
print(str(e), file=sys.stderr)
if my_file.closed is False:
my_file.close()
return 2
if my_file.closed is False:
my_file.close()
if args.minimal is False:
print(
dumps(
[el.__dict__ for el in x_] if len(x_) > 1 else x_[0].__dict__,
ensure_ascii=True,
indent=4,
)
)
else:
for my_file in args.files:
print(
", ".join(
[
el.encoding or "undefined"
for el in x_
if el.path == abspath(my_file.name)
]
)
)
return 0
if __name__ == "__main__":
cli_detect()

File diff suppressed because it is too large Load Diff

View File

@ -1,54 +0,0 @@
from typing import Any, Dict, Optional, Union
from warnings import warn
from .api import from_bytes
from .constant import CHARDET_CORRESPONDENCE
def detect(
byte_str: bytes, should_rename_legacy: bool = False, **kwargs: Any
) -> Dict[str, Optional[Union[str, float]]]:
"""
chardet legacy method
Detect the encoding of the given byte string. It should be mostly backward-compatible.
Encoding name will match Chardet own writing whenever possible. (Not on encoding name unsupported by it)
This function is deprecated and should be used to migrate your project easily, consult the documentation for
further information. Not planned for removal.
:param byte_str: The byte sequence to examine.
:param should_rename_legacy: Should we rename legacy encodings
to their more modern equivalents?
"""
if len(kwargs):
warn(
f"charset-normalizer disregard arguments '{','.join(list(kwargs.keys()))}' in legacy function detect()"
)
if not isinstance(byte_str, (bytearray, bytes)):
raise TypeError( # pragma: nocover
"Expected object of type bytes or bytearray, got: "
"{0}".format(type(byte_str))
)
if isinstance(byte_str, bytearray):
byte_str = bytes(byte_str)
r = from_bytes(byte_str).best()
encoding = r.encoding if r is not None else None
language = r.language if r is not None and r.language != "Unknown" else ""
confidence = 1.0 - r.chaos if r is not None else None
# Note: CharsetNormalizer does not return 'UTF-8-SIG' as the sig get stripped in the detection/normalization process
# but chardet does return 'utf-8-sig' and it is a valid codec name.
if r is not None and encoding == "utf_8" and r.bom:
encoding += "_sig"
if should_rename_legacy is False and encoding in CHARDET_CORRESPONDENCE:
encoding = CHARDET_CORRESPONDENCE[encoding]
return {
"encoding": encoding,
"language": language,
"confidence": confidence,
}

View File

@ -1,615 +0,0 @@
from functools import lru_cache
from logging import getLogger
from typing import List, Optional
from .constant import (
COMMON_SAFE_ASCII_CHARACTERS,
TRACE,
UNICODE_SECONDARY_RANGE_KEYWORD,
)
from .utils import (
is_accentuated,
is_arabic,
is_arabic_isolated_form,
is_case_variable,
is_cjk,
is_emoticon,
is_hangul,
is_hiragana,
is_katakana,
is_latin,
is_punctuation,
is_separator,
is_symbol,
is_thai,
is_unprintable,
remove_accent,
unicode_range,
)
class MessDetectorPlugin:
"""
Base abstract class used for mess detection plugins.
All detectors MUST extend and implement given methods.
"""
def eligible(self, character: str) -> bool:
"""
Determine if given character should be fed in.
"""
raise NotImplementedError # pragma: nocover
def feed(self, character: str) -> None:
"""
The main routine to be executed upon character.
Insert the logic in witch the text would be considered chaotic.
"""
raise NotImplementedError # pragma: nocover
def reset(self) -> None: # pragma: no cover
"""
Permit to reset the plugin to the initial state.
"""
raise NotImplementedError
@property
def ratio(self) -> float:
"""
Compute the chaos ratio based on what your feed() has seen.
Must NOT be lower than 0.; No restriction gt 0.
"""
raise NotImplementedError # pragma: nocover
class TooManySymbolOrPunctuationPlugin(MessDetectorPlugin):
def __init__(self) -> None:
self._punctuation_count: int = 0
self._symbol_count: int = 0
self._character_count: int = 0
self._last_printable_char: Optional[str] = None
self._frenzy_symbol_in_word: bool = False
def eligible(self, character: str) -> bool:
return character.isprintable()
def feed(self, character: str) -> None:
self._character_count += 1
if (
character != self._last_printable_char
and character not in COMMON_SAFE_ASCII_CHARACTERS
):
if is_punctuation(character):
self._punctuation_count += 1
elif (
character.isdigit() is False
and is_symbol(character)
and is_emoticon(character) is False
):
self._symbol_count += 2
self._last_printable_char = character
def reset(self) -> None: # pragma: no cover
self._punctuation_count = 0
self._character_count = 0
self._symbol_count = 0
@property
def ratio(self) -> float:
if self._character_count == 0:
return 0.0
ratio_of_punctuation: float = (
self._punctuation_count + self._symbol_count
) / self._character_count
return ratio_of_punctuation if ratio_of_punctuation >= 0.3 else 0.0
class TooManyAccentuatedPlugin(MessDetectorPlugin):
def __init__(self) -> None:
self._character_count: int = 0
self._accentuated_count: int = 0
def eligible(self, character: str) -> bool:
return character.isalpha()
def feed(self, character: str) -> None:
self._character_count += 1
if is_accentuated(character):
self._accentuated_count += 1
def reset(self) -> None: # pragma: no cover
self._character_count = 0
self._accentuated_count = 0
@property
def ratio(self) -> float:
if self._character_count < 8:
return 0.0
ratio_of_accentuation: float = self._accentuated_count / self._character_count
return ratio_of_accentuation if ratio_of_accentuation >= 0.35 else 0.0
class UnprintablePlugin(MessDetectorPlugin):
def __init__(self) -> None:
self._unprintable_count: int = 0
self._character_count: int = 0
def eligible(self, character: str) -> bool:
return True
def feed(self, character: str) -> None:
if is_unprintable(character):
self._unprintable_count += 1
self._character_count += 1
def reset(self) -> None: # pragma: no cover
self._unprintable_count = 0
@property
def ratio(self) -> float:
if self._character_count == 0:
return 0.0
return (self._unprintable_count * 8) / self._character_count
class SuspiciousDuplicateAccentPlugin(MessDetectorPlugin):
def __init__(self) -> None:
self._successive_count: int = 0
self._character_count: int = 0
self._last_latin_character: Optional[str] = None
def eligible(self, character: str) -> bool:
return character.isalpha() and is_latin(character)
def feed(self, character: str) -> None:
self._character_count += 1
if (
self._last_latin_character is not None
and is_accentuated(character)
and is_accentuated(self._last_latin_character)
):
if character.isupper() and self._last_latin_character.isupper():
self._successive_count += 1
# Worse if its the same char duplicated with different accent.
if remove_accent(character) == remove_accent(self._last_latin_character):
self._successive_count += 1
self._last_latin_character = character
def reset(self) -> None: # pragma: no cover
self._successive_count = 0
self._character_count = 0
self._last_latin_character = None
@property
def ratio(self) -> float:
if self._character_count == 0:
return 0.0
return (self._successive_count * 2) / self._character_count
class SuspiciousRange(MessDetectorPlugin):
def __init__(self) -> None:
self._suspicious_successive_range_count: int = 0
self._character_count: int = 0
self._last_printable_seen: Optional[str] = None
def eligible(self, character: str) -> bool:
return character.isprintable()
def feed(self, character: str) -> None:
self._character_count += 1
if (
character.isspace()
or is_punctuation(character)
or character in COMMON_SAFE_ASCII_CHARACTERS
):
self._last_printable_seen = None
return
if self._last_printable_seen is None:
self._last_printable_seen = character
return
unicode_range_a: Optional[str] = unicode_range(self._last_printable_seen)
unicode_range_b: Optional[str] = unicode_range(character)
if is_suspiciously_successive_range(unicode_range_a, unicode_range_b):
self._suspicious_successive_range_count += 1
self._last_printable_seen = character
def reset(self) -> None: # pragma: no cover
self._character_count = 0
self._suspicious_successive_range_count = 0
self._last_printable_seen = None
@property
def ratio(self) -> float:
if self._character_count <= 24:
return 0.0
ratio_of_suspicious_range_usage: float = (
self._suspicious_successive_range_count * 2
) / self._character_count
return ratio_of_suspicious_range_usage
class SuperWeirdWordPlugin(MessDetectorPlugin):
def __init__(self) -> None:
self._word_count: int = 0
self._bad_word_count: int = 0
self._foreign_long_count: int = 0
self._is_current_word_bad: bool = False
self._foreign_long_watch: bool = False
self._character_count: int = 0
self._bad_character_count: int = 0
self._buffer: str = ""
self._buffer_accent_count: int = 0
def eligible(self, character: str) -> bool:
return True
def feed(self, character: str) -> None:
if character.isalpha():
self._buffer += character
if is_accentuated(character):
self._buffer_accent_count += 1
if (
self._foreign_long_watch is False
and (is_latin(character) is False or is_accentuated(character))
and is_cjk(character) is False
and is_hangul(character) is False
and is_katakana(character) is False
and is_hiragana(character) is False
and is_thai(character) is False
):
self._foreign_long_watch = True
return
if not self._buffer:
return
if (
character.isspace() or is_punctuation(character) or is_separator(character)
) and self._buffer:
self._word_count += 1
buffer_length: int = len(self._buffer)
self._character_count += buffer_length
if buffer_length >= 4:
if self._buffer_accent_count / buffer_length > 0.34:
self._is_current_word_bad = True
# Word/Buffer ending with an upper case accentuated letter are so rare,
# that we will consider them all as suspicious. Same weight as foreign_long suspicious.
if (
is_accentuated(self._buffer[-1])
and self._buffer[-1].isupper()
and all(_.isupper() for _ in self._buffer) is False
):
self._foreign_long_count += 1
self._is_current_word_bad = True
if buffer_length >= 24 and self._foreign_long_watch:
camel_case_dst = [
i
for c, i in zip(self._buffer, range(0, buffer_length))
if c.isupper()
]
probable_camel_cased: bool = False
if camel_case_dst and (len(camel_case_dst) / buffer_length <= 0.3):
probable_camel_cased = True
if not probable_camel_cased:
self._foreign_long_count += 1
self._is_current_word_bad = True
if self._is_current_word_bad:
self._bad_word_count += 1
self._bad_character_count += len(self._buffer)
self._is_current_word_bad = False
self._foreign_long_watch = False
self._buffer = ""
self._buffer_accent_count = 0
elif (
character not in {"<", ">", "-", "=", "~", "|", "_"}
and character.isdigit() is False
and is_symbol(character)
):
self._is_current_word_bad = True
self._buffer += character
def reset(self) -> None: # pragma: no cover
self._buffer = ""
self._is_current_word_bad = False
self._foreign_long_watch = False
self._bad_word_count = 0
self._word_count = 0
self._character_count = 0
self._bad_character_count = 0
self._foreign_long_count = 0
@property
def ratio(self) -> float:
if self._word_count <= 10 and self._foreign_long_count == 0:
return 0.0
return self._bad_character_count / self._character_count
class CjkInvalidStopPlugin(MessDetectorPlugin):
"""
GB(Chinese) based encoding often render the stop incorrectly when the content does not fit and
can be easily detected. Searching for the overuse of '' and ''.
"""
def __init__(self) -> None:
self._wrong_stop_count: int = 0
self._cjk_character_count: int = 0
def eligible(self, character: str) -> bool:
return True
def feed(self, character: str) -> None:
if character in {"", ""}:
self._wrong_stop_count += 1
return
if is_cjk(character):
self._cjk_character_count += 1
def reset(self) -> None: # pragma: no cover
self._wrong_stop_count = 0
self._cjk_character_count = 0
@property
def ratio(self) -> float:
if self._cjk_character_count < 16:
return 0.0
return self._wrong_stop_count / self._cjk_character_count
class ArchaicUpperLowerPlugin(MessDetectorPlugin):
def __init__(self) -> None:
self._buf: bool = False
self._character_count_since_last_sep: int = 0
self._successive_upper_lower_count: int = 0
self._successive_upper_lower_count_final: int = 0
self._character_count: int = 0
self._last_alpha_seen: Optional[str] = None
self._current_ascii_only: bool = True
def eligible(self, character: str) -> bool:
return True
def feed(self, character: str) -> None:
is_concerned = character.isalpha() and is_case_variable(character)
chunk_sep = is_concerned is False
if chunk_sep and self._character_count_since_last_sep > 0:
if (
self._character_count_since_last_sep <= 64
and character.isdigit() is False
and self._current_ascii_only is False
):
self._successive_upper_lower_count_final += (
self._successive_upper_lower_count
)
self._successive_upper_lower_count = 0
self._character_count_since_last_sep = 0
self._last_alpha_seen = None
self._buf = False
self._character_count += 1
self._current_ascii_only = True
return
if self._current_ascii_only is True and character.isascii() is False:
self._current_ascii_only = False
if self._last_alpha_seen is not None:
if (character.isupper() and self._last_alpha_seen.islower()) or (
character.islower() and self._last_alpha_seen.isupper()
):
if self._buf is True:
self._successive_upper_lower_count += 2
self._buf = False
else:
self._buf = True
else:
self._buf = False
self._character_count += 1
self._character_count_since_last_sep += 1
self._last_alpha_seen = character
def reset(self) -> None: # pragma: no cover
self._character_count = 0
self._character_count_since_last_sep = 0
self._successive_upper_lower_count = 0
self._successive_upper_lower_count_final = 0
self._last_alpha_seen = None
self._buf = False
self._current_ascii_only = True
@property
def ratio(self) -> float:
if self._character_count == 0:
return 0.0
return self._successive_upper_lower_count_final / self._character_count
class ArabicIsolatedFormPlugin(MessDetectorPlugin):
def __init__(self) -> None:
self._character_count: int = 0
self._isolated_form_count: int = 0
def reset(self) -> None: # pragma: no cover
self._character_count = 0
self._isolated_form_count = 0
def eligible(self, character: str) -> bool:
return is_arabic(character)
def feed(self, character: str) -> None:
self._character_count += 1
if is_arabic_isolated_form(character):
self._isolated_form_count += 1
@property
def ratio(self) -> float:
if self._character_count < 8:
return 0.0
isolated_form_usage: float = self._isolated_form_count / self._character_count
return isolated_form_usage
@lru_cache(maxsize=1024)
def is_suspiciously_successive_range(
unicode_range_a: Optional[str], unicode_range_b: Optional[str]
) -> bool:
"""
Determine if two Unicode range seen next to each other can be considered as suspicious.
"""
if unicode_range_a is None or unicode_range_b is None:
return True
if unicode_range_a == unicode_range_b:
return False
if "Latin" in unicode_range_a and "Latin" in unicode_range_b:
return False
if "Emoticons" in unicode_range_a or "Emoticons" in unicode_range_b:
return False
# Latin characters can be accompanied with a combining diacritical mark
# eg. Vietnamese.
if ("Latin" in unicode_range_a or "Latin" in unicode_range_b) and (
"Combining" in unicode_range_a or "Combining" in unicode_range_b
):
return False
keywords_range_a, keywords_range_b = unicode_range_a.split(
" "
), unicode_range_b.split(" ")
for el in keywords_range_a:
if el in UNICODE_SECONDARY_RANGE_KEYWORD:
continue
if el in keywords_range_b:
return False
# Japanese Exception
range_a_jp_chars, range_b_jp_chars = (
unicode_range_a
in (
"Hiragana",
"Katakana",
),
unicode_range_b in ("Hiragana", "Katakana"),
)
if (range_a_jp_chars or range_b_jp_chars) and (
"CJK" in unicode_range_a or "CJK" in unicode_range_b
):
return False
if range_a_jp_chars and range_b_jp_chars:
return False
if "Hangul" in unicode_range_a or "Hangul" in unicode_range_b:
if "CJK" in unicode_range_a or "CJK" in unicode_range_b:
return False
if unicode_range_a == "Basic Latin" or unicode_range_b == "Basic Latin":
return False
# Chinese/Japanese use dedicated range for punctuation and/or separators.
if ("CJK" in unicode_range_a or "CJK" in unicode_range_b) or (
unicode_range_a in ["Katakana", "Hiragana"]
and unicode_range_b in ["Katakana", "Hiragana"]
):
if "Punctuation" in unicode_range_a or "Punctuation" in unicode_range_b:
return False
if "Forms" in unicode_range_a or "Forms" in unicode_range_b:
return False
if unicode_range_a == "Basic Latin" or unicode_range_b == "Basic Latin":
return False
return True
@lru_cache(maxsize=2048)
def mess_ratio(
decoded_sequence: str, maximum_threshold: float = 0.2, debug: bool = False
) -> float:
"""
Compute a mess ratio given a decoded bytes sequence. The maximum threshold does stop the computation earlier.
"""
detectors: List[MessDetectorPlugin] = [
md_class() for md_class in MessDetectorPlugin.__subclasses__()
]
length: int = len(decoded_sequence) + 1
mean_mess_ratio: float = 0.0
if length < 512:
intermediary_mean_mess_ratio_calc: int = 32
elif length <= 1024:
intermediary_mean_mess_ratio_calc = 64
else:
intermediary_mean_mess_ratio_calc = 128
for character, index in zip(decoded_sequence + "\n", range(length)):
for detector in detectors:
if detector.eligible(character):
detector.feed(character)
if (
index > 0 and index % intermediary_mean_mess_ratio_calc == 0
) or index == length - 1:
mean_mess_ratio = sum(dt.ratio for dt in detectors)
if mean_mess_ratio >= maximum_threshold:
break
if debug:
logger = getLogger("charset_normalizer")
logger.log(
TRACE,
"Mess-detector extended-analysis start. "
f"intermediary_mean_mess_ratio_calc={intermediary_mean_mess_ratio_calc} mean_mess_ratio={mean_mess_ratio} "
f"maximum_threshold={maximum_threshold}",
)
if len(decoded_sequence) > 16:
logger.log(TRACE, f"Starting with: {decoded_sequence[:16]}")
logger.log(TRACE, f"Ending with: {decoded_sequence[-16::]}")
for dt in detectors: # pragma: nocover
logger.log(TRACE, f"{dt.__class__}: {dt.ratio}")
return round(mean_mess_ratio, 3)

View File

@ -1,340 +0,0 @@
from encodings.aliases import aliases
from hashlib import sha256
from json import dumps
from typing import Any, Dict, Iterator, List, Optional, Tuple, Union
from .constant import TOO_BIG_SEQUENCE
from .utils import iana_name, is_multi_byte_encoding, unicode_range
class CharsetMatch:
def __init__(
self,
payload: bytes,
guessed_encoding: str,
mean_mess_ratio: float,
has_sig_or_bom: bool,
languages: "CoherenceMatches",
decoded_payload: Optional[str] = None,
):
self._payload: bytes = payload
self._encoding: str = guessed_encoding
self._mean_mess_ratio: float = mean_mess_ratio
self._languages: CoherenceMatches = languages
self._has_sig_or_bom: bool = has_sig_or_bom
self._unicode_ranges: Optional[List[str]] = None
self._leaves: List[CharsetMatch] = []
self._mean_coherence_ratio: float = 0.0
self._output_payload: Optional[bytes] = None
self._output_encoding: Optional[str] = None
self._string: Optional[str] = decoded_payload
def __eq__(self, other: object) -> bool:
if not isinstance(other, CharsetMatch):
raise TypeError(
"__eq__ cannot be invoked on {} and {}.".format(
str(other.__class__), str(self.__class__)
)
)
return self.encoding == other.encoding and self.fingerprint == other.fingerprint
def __lt__(self, other: object) -> bool:
"""
Implemented to make sorted available upon CharsetMatches items.
"""
if not isinstance(other, CharsetMatch):
raise ValueError
chaos_difference: float = abs(self.chaos - other.chaos)
coherence_difference: float = abs(self.coherence - other.coherence)
# Below 1% difference --> Use Coherence
if chaos_difference < 0.01 and coherence_difference > 0.02:
return self.coherence > other.coherence
elif chaos_difference < 0.01 and coherence_difference <= 0.02:
# When having a difficult decision, use the result that decoded as many multi-byte as possible.
# preserve RAM usage!
if len(self._payload) >= TOO_BIG_SEQUENCE:
return self.chaos < other.chaos
return self.multi_byte_usage > other.multi_byte_usage
return self.chaos < other.chaos
@property
def multi_byte_usage(self) -> float:
return 1.0 - (len(str(self)) / len(self.raw))
def __str__(self) -> str:
# Lazy Str Loading
if self._string is None:
self._string = str(self._payload, self._encoding, "strict")
return self._string
def __repr__(self) -> str:
return "<CharsetMatch '{}' bytes({})>".format(self.encoding, self.fingerprint)
def add_submatch(self, other: "CharsetMatch") -> None:
if not isinstance(other, CharsetMatch) or other == self:
raise ValueError(
"Unable to add instance <{}> as a submatch of a CharsetMatch".format(
other.__class__
)
)
other._string = None # Unload RAM usage; dirty trick.
self._leaves.append(other)
@property
def encoding(self) -> str:
return self._encoding
@property
def encoding_aliases(self) -> List[str]:
"""
Encoding name are known by many name, using this could help when searching for IBM855 when it's listed as CP855.
"""
also_known_as: List[str] = []
for u, p in aliases.items():
if self.encoding == u:
also_known_as.append(p)
elif self.encoding == p:
also_known_as.append(u)
return also_known_as
@property
def bom(self) -> bool:
return self._has_sig_or_bom
@property
def byte_order_mark(self) -> bool:
return self._has_sig_or_bom
@property
def languages(self) -> List[str]:
"""
Return the complete list of possible languages found in decoded sequence.
Usually not really useful. Returned list may be empty even if 'language' property return something != 'Unknown'.
"""
return [e[0] for e in self._languages]
@property
def language(self) -> str:
"""
Most probable language found in decoded sequence. If none were detected or inferred, the property will return
"Unknown".
"""
if not self._languages:
# Trying to infer the language based on the given encoding
# Its either English or we should not pronounce ourselves in certain cases.
if "ascii" in self.could_be_from_charset:
return "English"
# doing it there to avoid circular import
from charset_normalizer.cd import encoding_languages, mb_encoding_languages
languages = (
mb_encoding_languages(self.encoding)
if is_multi_byte_encoding(self.encoding)
else encoding_languages(self.encoding)
)
if len(languages) == 0 or "Latin Based" in languages:
return "Unknown"
return languages[0]
return self._languages[0][0]
@property
def chaos(self) -> float:
return self._mean_mess_ratio
@property
def coherence(self) -> float:
if not self._languages:
return 0.0
return self._languages[0][1]
@property
def percent_chaos(self) -> float:
return round(self.chaos * 100, ndigits=3)
@property
def percent_coherence(self) -> float:
return round(self.coherence * 100, ndigits=3)
@property
def raw(self) -> bytes:
"""
Original untouched bytes.
"""
return self._payload
@property
def submatch(self) -> List["CharsetMatch"]:
return self._leaves
@property
def has_submatch(self) -> bool:
return len(self._leaves) > 0
@property
def alphabets(self) -> List[str]:
if self._unicode_ranges is not None:
return self._unicode_ranges
# list detected ranges
detected_ranges: List[Optional[str]] = [
unicode_range(char) for char in str(self)
]
# filter and sort
self._unicode_ranges = sorted(list({r for r in detected_ranges if r}))
return self._unicode_ranges
@property
def could_be_from_charset(self) -> List[str]:
"""
The complete list of encoding that output the exact SAME str result and therefore could be the originating
encoding.
This list does include the encoding available in property 'encoding'.
"""
return [self._encoding] + [m.encoding for m in self._leaves]
def output(self, encoding: str = "utf_8") -> bytes:
"""
Method to get re-encoded bytes payload using given target encoding. Default to UTF-8.
Any errors will be simply ignored by the encoder NOT replaced.
"""
if self._output_encoding is None or self._output_encoding != encoding:
self._output_encoding = encoding
self._output_payload = str(self).encode(encoding, "replace")
return self._output_payload # type: ignore
@property
def fingerprint(self) -> str:
"""
Retrieve the unique SHA256 computed using the transformed (re-encoded) payload. Not the original one.
"""
return sha256(self.output()).hexdigest()
class CharsetMatches:
"""
Container with every CharsetMatch items ordered by default from most probable to the less one.
Act like a list(iterable) but does not implements all related methods.
"""
def __init__(self, results: Optional[List[CharsetMatch]] = None):
self._results: List[CharsetMatch] = sorted(results) if results else []
def __iter__(self) -> Iterator[CharsetMatch]:
yield from self._results
def __getitem__(self, item: Union[int, str]) -> CharsetMatch:
"""
Retrieve a single item either by its position or encoding name (alias may be used here).
Raise KeyError upon invalid index or encoding not present in results.
"""
if isinstance(item, int):
return self._results[item]
if isinstance(item, str):
item = iana_name(item, False)
for result in self._results:
if item in result.could_be_from_charset:
return result
raise KeyError
def __len__(self) -> int:
return len(self._results)
def __bool__(self) -> bool:
return len(self._results) > 0
def append(self, item: CharsetMatch) -> None:
"""
Insert a single match. Will be inserted accordingly to preserve sort.
Can be inserted as a submatch.
"""
if not isinstance(item, CharsetMatch):
raise ValueError(
"Cannot append instance '{}' to CharsetMatches".format(
str(item.__class__)
)
)
# We should disable the submatch factoring when the input file is too heavy (conserve RAM usage)
if len(item.raw) <= TOO_BIG_SEQUENCE:
for match in self._results:
if match.fingerprint == item.fingerprint and match.chaos == item.chaos:
match.add_submatch(item)
return
self._results.append(item)
self._results = sorted(self._results)
def best(self) -> Optional["CharsetMatch"]:
"""
Simply return the first match. Strict equivalent to matches[0].
"""
if not self._results:
return None
return self._results[0]
def first(self) -> Optional["CharsetMatch"]:
"""
Redundant method, call the method best(). Kept for BC reasons.
"""
return self.best()
CoherenceMatch = Tuple[str, float]
CoherenceMatches = List[CoherenceMatch]
class CliDetectionResult:
def __init__(
self,
path: str,
encoding: Optional[str],
encoding_aliases: List[str],
alternative_encodings: List[str],
language: str,
alphabets: List[str],
has_sig_or_bom: bool,
chaos: float,
coherence: float,
unicode_path: Optional[str],
is_preferred: bool,
):
self.path: str = path
self.unicode_path: Optional[str] = unicode_path
self.encoding: Optional[str] = encoding
self.encoding_aliases: List[str] = encoding_aliases
self.alternative_encodings: List[str] = alternative_encodings
self.language: str = language
self.alphabets: List[str] = alphabets
self.has_sig_or_bom: bool = has_sig_or_bom
self.chaos: float = chaos
self.coherence: float = coherence
self.is_preferred: bool = is_preferred
@property
def __dict__(self) -> Dict[str, Any]: # type: ignore
return {
"path": self.path,
"encoding": self.encoding,
"encoding_aliases": self.encoding_aliases,
"alternative_encodings": self.alternative_encodings,
"language": self.language,
"alphabets": self.alphabets,
"has_sig_or_bom": self.has_sig_or_bom,
"chaos": self.chaos,
"coherence": self.coherence,
"unicode_path": self.unicode_path,
"is_preferred": self.is_preferred,
}
def to_json(self) -> str:
return dumps(self.__dict__, ensure_ascii=True, indent=4)

View File

@ -1,421 +0,0 @@
import importlib
import logging
import unicodedata
from codecs import IncrementalDecoder
from encodings.aliases import aliases
from functools import lru_cache
from re import findall
from typing import Generator, List, Optional, Set, Tuple, Union
from _multibytecodec import MultibyteIncrementalDecoder
from .constant import (
ENCODING_MARKS,
IANA_SUPPORTED_SIMILAR,
RE_POSSIBLE_ENCODING_INDICATION,
UNICODE_RANGES_COMBINED,
UNICODE_SECONDARY_RANGE_KEYWORD,
UTF8_MAXIMAL_ALLOCATION,
)
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_accentuated(character: str) -> bool:
try:
description: str = unicodedata.name(character)
except ValueError:
return False
return (
"WITH GRAVE" in description
or "WITH ACUTE" in description
or "WITH CEDILLA" in description
or "WITH DIAERESIS" in description
or "WITH CIRCUMFLEX" in description
or "WITH TILDE" in description
or "WITH MACRON" in description
or "WITH RING ABOVE" in description
)
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def remove_accent(character: str) -> str:
decomposed: str = unicodedata.decomposition(character)
if not decomposed:
return character
codes: List[str] = decomposed.split(" ")
return chr(int(codes[0], 16))
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def unicode_range(character: str) -> Optional[str]:
"""
Retrieve the Unicode range official name from a single character.
"""
character_ord: int = ord(character)
for range_name, ord_range in UNICODE_RANGES_COMBINED.items():
if character_ord in ord_range:
return range_name
return None
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_latin(character: str) -> bool:
try:
description: str = unicodedata.name(character)
except ValueError:
return False
return "LATIN" in description
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_punctuation(character: str) -> bool:
character_category: str = unicodedata.category(character)
if "P" in character_category:
return True
character_range: Optional[str] = unicode_range(character)
if character_range is None:
return False
return "Punctuation" in character_range
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_symbol(character: str) -> bool:
character_category: str = unicodedata.category(character)
if "S" in character_category or "N" in character_category:
return True
character_range: Optional[str] = unicode_range(character)
if character_range is None:
return False
return "Forms" in character_range and character_category != "Lo"
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_emoticon(character: str) -> bool:
character_range: Optional[str] = unicode_range(character)
if character_range is None:
return False
return "Emoticons" in character_range or "Pictographs" in character_range
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_separator(character: str) -> bool:
if character.isspace() or character in {"", "+", "<", ">"}:
return True
character_category: str = unicodedata.category(character)
return "Z" in character_category or character_category in {"Po", "Pd", "Pc"}
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_case_variable(character: str) -> bool:
return character.islower() != character.isupper()
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_cjk(character: str) -> bool:
try:
character_name = unicodedata.name(character)
except ValueError:
return False
return "CJK" in character_name
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_hiragana(character: str) -> bool:
try:
character_name = unicodedata.name(character)
except ValueError:
return False
return "HIRAGANA" in character_name
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_katakana(character: str) -> bool:
try:
character_name = unicodedata.name(character)
except ValueError:
return False
return "KATAKANA" in character_name
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_hangul(character: str) -> bool:
try:
character_name = unicodedata.name(character)
except ValueError:
return False
return "HANGUL" in character_name
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_thai(character: str) -> bool:
try:
character_name = unicodedata.name(character)
except ValueError:
return False
return "THAI" in character_name
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_arabic(character: str) -> bool:
try:
character_name = unicodedata.name(character)
except ValueError:
return False
return "ARABIC" in character_name
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_arabic_isolated_form(character: str) -> bool:
try:
character_name = unicodedata.name(character)
except ValueError:
return False
return "ARABIC" in character_name and "ISOLATED FORM" in character_name
@lru_cache(maxsize=len(UNICODE_RANGES_COMBINED))
def is_unicode_range_secondary(range_name: str) -> bool:
return any(keyword in range_name for keyword in UNICODE_SECONDARY_RANGE_KEYWORD)
@lru_cache(maxsize=UTF8_MAXIMAL_ALLOCATION)
def is_unprintable(character: str) -> bool:
return (
character.isspace() is False # includes \n \t \r \v
and character.isprintable() is False
and character != "\x1A" # Why? Its the ASCII substitute character.
and character != "\ufeff" # bug discovered in Python,
# Zero Width No-Break Space located in Arabic Presentation Forms-B, Unicode 1.1 not acknowledged as space.
)
def any_specified_encoding(sequence: bytes, search_zone: int = 8192) -> Optional[str]:
"""
Extract using ASCII-only decoder any specified encoding in the first n-bytes.
"""
if not isinstance(sequence, bytes):
raise TypeError
seq_len: int = len(sequence)
results: List[str] = findall(
RE_POSSIBLE_ENCODING_INDICATION,
sequence[: min(seq_len, search_zone)].decode("ascii", errors="ignore"),
)
if len(results) == 0:
return None
for specified_encoding in results:
specified_encoding = specified_encoding.lower().replace("-", "_")
encoding_alias: str
encoding_iana: str
for encoding_alias, encoding_iana in aliases.items():
if encoding_alias == specified_encoding:
return encoding_iana
if encoding_iana == specified_encoding:
return encoding_iana
return None
@lru_cache(maxsize=128)
def is_multi_byte_encoding(name: str) -> bool:
"""
Verify is a specific encoding is a multi byte one based on it IANA name
"""
return name in {
"utf_8",
"utf_8_sig",
"utf_16",
"utf_16_be",
"utf_16_le",
"utf_32",
"utf_32_le",
"utf_32_be",
"utf_7",
} or issubclass(
importlib.import_module("encodings.{}".format(name)).IncrementalDecoder,
MultibyteIncrementalDecoder,
)
def identify_sig_or_bom(sequence: bytes) -> Tuple[Optional[str], bytes]:
"""
Identify and extract SIG/BOM in given sequence.
"""
for iana_encoding in ENCODING_MARKS:
marks: Union[bytes, List[bytes]] = ENCODING_MARKS[iana_encoding]
if isinstance(marks, bytes):
marks = [marks]
for mark in marks:
if sequence.startswith(mark):
return iana_encoding, mark
return None, b""
def should_strip_sig_or_bom(iana_encoding: str) -> bool:
return iana_encoding not in {"utf_16", "utf_32"}
def iana_name(cp_name: str, strict: bool = True) -> str:
cp_name = cp_name.lower().replace("-", "_")
encoding_alias: str
encoding_iana: str
for encoding_alias, encoding_iana in aliases.items():
if cp_name in [encoding_alias, encoding_iana]:
return encoding_iana
if strict:
raise ValueError("Unable to retrieve IANA for '{}'".format(cp_name))
return cp_name
def range_scan(decoded_sequence: str) -> List[str]:
ranges: Set[str] = set()
for character in decoded_sequence:
character_range: Optional[str] = unicode_range(character)
if character_range is None:
continue
ranges.add(character_range)
return list(ranges)
def cp_similarity(iana_name_a: str, iana_name_b: str) -> float:
if is_multi_byte_encoding(iana_name_a) or is_multi_byte_encoding(iana_name_b):
return 0.0
decoder_a = importlib.import_module(
"encodings.{}".format(iana_name_a)
).IncrementalDecoder
decoder_b = importlib.import_module(
"encodings.{}".format(iana_name_b)
).IncrementalDecoder
id_a: IncrementalDecoder = decoder_a(errors="ignore")
id_b: IncrementalDecoder = decoder_b(errors="ignore")
character_match_count: int = 0
for i in range(255):
to_be_decoded: bytes = bytes([i])
if id_a.decode(to_be_decoded) == id_b.decode(to_be_decoded):
character_match_count += 1
return character_match_count / 254
def is_cp_similar(iana_name_a: str, iana_name_b: str) -> bool:
"""
Determine if two code page are at least 80% similar. IANA_SUPPORTED_SIMILAR dict was generated using
the function cp_similarity.
"""
return (
iana_name_a in IANA_SUPPORTED_SIMILAR
and iana_name_b in IANA_SUPPORTED_SIMILAR[iana_name_a]
)
def set_logging_handler(
name: str = "charset_normalizer",
level: int = logging.INFO,
format_string: str = "%(asctime)s | %(levelname)s | %(message)s",
) -> None:
logger = logging.getLogger(name)
logger.setLevel(level)
handler = logging.StreamHandler()
handler.setFormatter(logging.Formatter(format_string))
logger.addHandler(handler)
def cut_sequence_chunks(
sequences: bytes,
encoding_iana: str,
offsets: range,
chunk_size: int,
bom_or_sig_available: bool,
strip_sig_or_bom: bool,
sig_payload: bytes,
is_multi_byte_decoder: bool,
decoded_payload: Optional[str] = None,
) -> Generator[str, None, None]:
if decoded_payload and is_multi_byte_decoder is False:
for i in offsets:
chunk = decoded_payload[i : i + chunk_size]
if not chunk:
break
yield chunk
else:
for i in offsets:
chunk_end = i + chunk_size
if chunk_end > len(sequences) + 8:
continue
cut_sequence = sequences[i : i + chunk_size]
if bom_or_sig_available and strip_sig_or_bom is False:
cut_sequence = sig_payload + cut_sequence
chunk = cut_sequence.decode(
encoding_iana,
errors="ignore" if is_multi_byte_decoder else "strict",
)
# multi-byte bad cutting detector and adjustment
# not the cleanest way to perform that fix but clever enough for now.
if is_multi_byte_decoder and i > 0:
chunk_partial_size_chk: int = min(chunk_size, 16)
if (
decoded_payload
and chunk[:chunk_partial_size_chk] not in decoded_payload
):
for j in range(i, i - 4, -1):
cut_sequence = sequences[j:chunk_end]
if bom_or_sig_available and strip_sig_or_bom is False:
cut_sequence = sig_payload + cut_sequence
chunk = cut_sequence.decode(encoding_iana, errors="ignore")
if chunk[:chunk_partial_size_chk] in decoded_payload:
break
yield chunk

View File

@ -1,6 +0,0 @@
"""
Expose version
"""
__version__ = "3.3.2"
VERSION = __version__.split(".")

View File

@ -1,44 +0,0 @@
from .package_data import __version__
from .core import (
IDNABidiError,
IDNAError,
InvalidCodepoint,
InvalidCodepointContext,
alabel,
check_bidi,
check_hyphen_ok,
check_initial_combiner,
check_label,
check_nfc,
decode,
encode,
ulabel,
uts46_remap,
valid_contextj,
valid_contexto,
valid_label_length,
valid_string_length,
)
from .intranges import intranges_contain
__all__ = [
"IDNABidiError",
"IDNAError",
"InvalidCodepoint",
"InvalidCodepointContext",
"alabel",
"check_bidi",
"check_hyphen_ok",
"check_initial_combiner",
"check_label",
"check_nfc",
"decode",
"encode",
"intranges_contain",
"ulabel",
"uts46_remap",
"valid_contextj",
"valid_contexto",
"valid_label_length",
"valid_string_length",
]

View File

@ -1,118 +0,0 @@
from .core import encode, decode, alabel, ulabel, IDNAError
import codecs
import re
from typing import Any, Tuple, Optional
_unicode_dots_re = re.compile('[\u002e\u3002\uff0e\uff61]')
class Codec(codecs.Codec):
def encode(self, data: str, errors: str = 'strict') -> Tuple[bytes, int]:
if errors != 'strict':
raise IDNAError('Unsupported error handling \"{}\"'.format(errors))
if not data:
return b"", 0
return encode(data), len(data)
def decode(self, data: bytes, errors: str = 'strict') -> Tuple[str, int]:
if errors != 'strict':
raise IDNAError('Unsupported error handling \"{}\"'.format(errors))
if not data:
return '', 0
return decode(data), len(data)
class IncrementalEncoder(codecs.BufferedIncrementalEncoder):
def _buffer_encode(self, data: str, errors: str, final: bool) -> Tuple[bytes, int]:
if errors != 'strict':
raise IDNAError('Unsupported error handling \"{}\"'.format(errors))
if not data:
return b'', 0
labels = _unicode_dots_re.split(data)
trailing_dot = b''
if labels:
if not labels[-1]:
trailing_dot = b'.'
del labels[-1]
elif not final:
# Keep potentially unfinished label until the next call
del labels[-1]
if labels:
trailing_dot = b'.'
result = []
size = 0
for label in labels:
result.append(alabel(label))
if size:
size += 1
size += len(label)
# Join with U+002E
result_bytes = b'.'.join(result) + trailing_dot
size += len(trailing_dot)
return result_bytes, size
class IncrementalDecoder(codecs.BufferedIncrementalDecoder):
def _buffer_decode(self, data: Any, errors: str, final: bool) -> Tuple[str, int]:
if errors != 'strict':
raise IDNAError('Unsupported error handling \"{}\"'.format(errors))
if not data:
return ('', 0)
if not isinstance(data, str):
data = str(data, 'ascii')
labels = _unicode_dots_re.split(data)
trailing_dot = ''
if labels:
if not labels[-1]:
trailing_dot = '.'
del labels[-1]
elif not final:
# Keep potentially unfinished label until the next call
del labels[-1]
if labels:
trailing_dot = '.'
result = []
size = 0
for label in labels:
result.append(ulabel(label))
if size:
size += 1
size += len(label)
result_str = '.'.join(result) + trailing_dot
size += len(trailing_dot)
return (result_str, size)
class StreamWriter(Codec, codecs.StreamWriter):
pass
class StreamReader(Codec, codecs.StreamReader):
pass
def search_function(name: str) -> Optional[codecs.CodecInfo]:
if name != 'idna2008':
return None
return codecs.CodecInfo(
name=name,
encode=Codec().encode,
decode=Codec().decode,
incrementalencoder=IncrementalEncoder,
incrementaldecoder=IncrementalDecoder,
streamwriter=StreamWriter,
streamreader=StreamReader,
)
codecs.register(search_function)

View File

@ -1,13 +0,0 @@
from .core import *
from .codec import *
from typing import Any, Union
def ToASCII(label: str) -> bytes:
return encode(label)
def ToUnicode(label: Union[bytes, bytearray]) -> str:
return decode(label)
def nameprep(s: Any) -> None:
raise NotImplementedError('IDNA 2008 does not utilise nameprep protocol')

View File

@ -1,400 +0,0 @@
from . import idnadata
import bisect
import unicodedata
import re
from typing import Union, Optional
from .intranges import intranges_contain
_virama_combining_class = 9
_alabel_prefix = b'xn--'
_unicode_dots_re = re.compile('[\u002e\u3002\uff0e\uff61]')
class IDNAError(UnicodeError):
""" Base exception for all IDNA-encoding related problems """
pass
class IDNABidiError(IDNAError):
""" Exception when bidirectional requirements are not satisfied """
pass
class InvalidCodepoint(IDNAError):
""" Exception when a disallowed or unallocated codepoint is used """
pass
class InvalidCodepointContext(IDNAError):
""" Exception when the codepoint is not valid in the context it is used """
pass
def _combining_class(cp: int) -> int:
v = unicodedata.combining(chr(cp))
if v == 0:
if not unicodedata.name(chr(cp)):
raise ValueError('Unknown character in unicodedata')
return v
def _is_script(cp: str, script: str) -> bool:
return intranges_contain(ord(cp), idnadata.scripts[script])
def _punycode(s: str) -> bytes:
return s.encode('punycode')
def _unot(s: int) -> str:
return 'U+{:04X}'.format(s)
def valid_label_length(label: Union[bytes, str]) -> bool:
if len(label) > 63:
return False
return True
def valid_string_length(label: Union[bytes, str], trailing_dot: bool) -> bool:
if len(label) > (254 if trailing_dot else 253):
return False
return True
def check_bidi(label: str, check_ltr: bool = False) -> bool:
# Bidi rules should only be applied if string contains RTL characters
bidi_label = False
for (idx, cp) in enumerate(label, 1):
direction = unicodedata.bidirectional(cp)
if direction == '':
# String likely comes from a newer version of Unicode
raise IDNABidiError('Unknown directionality in label {} at position {}'.format(repr(label), idx))
if direction in ['R', 'AL', 'AN']:
bidi_label = True
if not bidi_label and not check_ltr:
return True
# Bidi rule 1
direction = unicodedata.bidirectional(label[0])
if direction in ['R', 'AL']:
rtl = True
elif direction == 'L':
rtl = False
else:
raise IDNABidiError('First codepoint in label {} must be directionality L, R or AL'.format(repr(label)))
valid_ending = False
number_type = None # type: Optional[str]
for (idx, cp) in enumerate(label, 1):
direction = unicodedata.bidirectional(cp)
if rtl:
# Bidi rule 2
if not direction in ['R', 'AL', 'AN', 'EN', 'ES', 'CS', 'ET', 'ON', 'BN', 'NSM']:
raise IDNABidiError('Invalid direction for codepoint at position {} in a right-to-left label'.format(idx))
# Bidi rule 3
if direction in ['R', 'AL', 'EN', 'AN']:
valid_ending = True
elif direction != 'NSM':
valid_ending = False
# Bidi rule 4
if direction in ['AN', 'EN']:
if not number_type:
number_type = direction
else:
if number_type != direction:
raise IDNABidiError('Can not mix numeral types in a right-to-left label')
else:
# Bidi rule 5
if not direction in ['L', 'EN', 'ES', 'CS', 'ET', 'ON', 'BN', 'NSM']:
raise IDNABidiError('Invalid direction for codepoint at position {} in a left-to-right label'.format(idx))
# Bidi rule 6
if direction in ['L', 'EN']:
valid_ending = True
elif direction != 'NSM':
valid_ending = False
if not valid_ending:
raise IDNABidiError('Label ends with illegal codepoint directionality')
return True
def check_initial_combiner(label: str) -> bool:
if unicodedata.category(label[0])[0] == 'M':
raise IDNAError('Label begins with an illegal combining character')
return True
def check_hyphen_ok(label: str) -> bool:
if label[2:4] == '--':
raise IDNAError('Label has disallowed hyphens in 3rd and 4th position')
if label[0] == '-' or label[-1] == '-':
raise IDNAError('Label must not start or end with a hyphen')
return True
def check_nfc(label: str) -> None:
if unicodedata.normalize('NFC', label) != label:
raise IDNAError('Label must be in Normalization Form C')
def valid_contextj(label: str, pos: int) -> bool:
cp_value = ord(label[pos])
if cp_value == 0x200c:
if pos > 0:
if _combining_class(ord(label[pos - 1])) == _virama_combining_class:
return True
ok = False
for i in range(pos-1, -1, -1):
joining_type = idnadata.joining_types.get(ord(label[i]))
if joining_type == ord('T'):
continue
if joining_type in [ord('L'), ord('D')]:
ok = True
break
if not ok:
return False
ok = False
for i in range(pos+1, len(label)):
joining_type = idnadata.joining_types.get(ord(label[i]))
if joining_type == ord('T'):
continue
if joining_type in [ord('R'), ord('D')]:
ok = True
break
return ok
if cp_value == 0x200d:
if pos > 0:
if _combining_class(ord(label[pos - 1])) == _virama_combining_class:
return True
return False
else:
return False
def valid_contexto(label: str, pos: int, exception: bool = False) -> bool:
cp_value = ord(label[pos])
if cp_value == 0x00b7:
if 0 < pos < len(label)-1:
if ord(label[pos - 1]) == 0x006c and ord(label[pos + 1]) == 0x006c:
return True
return False
elif cp_value == 0x0375:
if pos < len(label)-1 and len(label) > 1:
return _is_script(label[pos + 1], 'Greek')
return False
elif cp_value == 0x05f3 or cp_value == 0x05f4:
if pos > 0:
return _is_script(label[pos - 1], 'Hebrew')
return False
elif cp_value == 0x30fb:
for cp in label:
if cp == '\u30fb':
continue
if _is_script(cp, 'Hiragana') or _is_script(cp, 'Katakana') or _is_script(cp, 'Han'):
return True
return False
elif 0x660 <= cp_value <= 0x669:
for cp in label:
if 0x6f0 <= ord(cp) <= 0x06f9:
return False
return True
elif 0x6f0 <= cp_value <= 0x6f9:
for cp in label:
if 0x660 <= ord(cp) <= 0x0669:
return False
return True
return False
def check_label(label: Union[str, bytes, bytearray]) -> None:
if isinstance(label, (bytes, bytearray)):
label = label.decode('utf-8')
if len(label) == 0:
raise IDNAError('Empty Label')
check_nfc(label)
check_hyphen_ok(label)
check_initial_combiner(label)
for (pos, cp) in enumerate(label):
cp_value = ord(cp)
if intranges_contain(cp_value, idnadata.codepoint_classes['PVALID']):
continue
elif intranges_contain(cp_value, idnadata.codepoint_classes['CONTEXTJ']):
try:
if not valid_contextj(label, pos):
raise InvalidCodepointContext('Joiner {} not allowed at position {} in {}'.format(
_unot(cp_value), pos+1, repr(label)))
except ValueError:
raise IDNAError('Unknown codepoint adjacent to joiner {} at position {} in {}'.format(
_unot(cp_value), pos+1, repr(label)))
elif intranges_contain(cp_value, idnadata.codepoint_classes['CONTEXTO']):
if not valid_contexto(label, pos):
raise InvalidCodepointContext('Codepoint {} not allowed at position {} in {}'.format(_unot(cp_value), pos+1, repr(label)))
else:
raise InvalidCodepoint('Codepoint {} at position {} of {} not allowed'.format(_unot(cp_value), pos+1, repr(label)))
check_bidi(label)
def alabel(label: str) -> bytes:
try:
label_bytes = label.encode('ascii')
ulabel(label_bytes)
if not valid_label_length(label_bytes):
raise IDNAError('Label too long')
return label_bytes
except UnicodeEncodeError:
pass
if not label:
raise IDNAError('No Input')
label = str(label)
check_label(label)
label_bytes = _punycode(label)
label_bytes = _alabel_prefix + label_bytes
if not valid_label_length(label_bytes):
raise IDNAError('Label too long')
return label_bytes
def ulabel(label: Union[str, bytes, bytearray]) -> str:
if not isinstance(label, (bytes, bytearray)):
try:
label_bytes = label.encode('ascii')
except UnicodeEncodeError:
check_label(label)
return label
else:
label_bytes = label
label_bytes = label_bytes.lower()
if label_bytes.startswith(_alabel_prefix):
label_bytes = label_bytes[len(_alabel_prefix):]
if not label_bytes:
raise IDNAError('Malformed A-label, no Punycode eligible content found')
if label_bytes.decode('ascii')[-1] == '-':
raise IDNAError('A-label must not end with a hyphen')
else:
check_label(label_bytes)
return label_bytes.decode('ascii')
try:
label = label_bytes.decode('punycode')
except UnicodeError:
raise IDNAError('Invalid A-label')
check_label(label)
return label
def uts46_remap(domain: str, std3_rules: bool = True, transitional: bool = False) -> str:
"""Re-map the characters in the string according to UTS46 processing."""
from .uts46data import uts46data
output = ''
for pos, char in enumerate(domain):
code_point = ord(char)
try:
uts46row = uts46data[code_point if code_point < 256 else
bisect.bisect_left(uts46data, (code_point, 'Z')) - 1]
status = uts46row[1]
replacement = None # type: Optional[str]
if len(uts46row) == 3:
replacement = uts46row[2]
if (status == 'V' or
(status == 'D' and not transitional) or
(status == '3' and not std3_rules and replacement is None)):
output += char
elif replacement is not None and (status == 'M' or
(status == '3' and not std3_rules) or
(status == 'D' and transitional)):
output += replacement
elif status != 'I':
raise IndexError()
except IndexError:
raise InvalidCodepoint(
'Codepoint {} not allowed at position {} in {}'.format(
_unot(code_point), pos + 1, repr(domain)))
return unicodedata.normalize('NFC', output)
def encode(s: Union[str, bytes, bytearray], strict: bool = False, uts46: bool = False, std3_rules: bool = False, transitional: bool = False) -> bytes:
if not isinstance(s, str):
try:
s = str(s, 'ascii')
except UnicodeDecodeError:
raise IDNAError('should pass a unicode string to the function rather than a byte string.')
if uts46:
s = uts46_remap(s, std3_rules, transitional)
trailing_dot = False
result = []
if strict:
labels = s.split('.')
else:
labels = _unicode_dots_re.split(s)
if not labels or labels == ['']:
raise IDNAError('Empty domain')
if labels[-1] == '':
del labels[-1]
trailing_dot = True
for label in labels:
s = alabel(label)
if s:
result.append(s)
else:
raise IDNAError('Empty label')
if trailing_dot:
result.append(b'')
s = b'.'.join(result)
if not valid_string_length(s, trailing_dot):
raise IDNAError('Domain too long')
return s
def decode(s: Union[str, bytes, bytearray], strict: bool = False, uts46: bool = False, std3_rules: bool = False) -> str:
try:
if not isinstance(s, str):
s = str(s, 'ascii')
except UnicodeDecodeError:
raise IDNAError('Invalid ASCII in A-label')
if uts46:
s = uts46_remap(s, std3_rules, False)
trailing_dot = False
result = []
if not strict:
labels = _unicode_dots_re.split(s)
else:
labels = s.split('.')
if not labels or labels == ['']:
raise IDNAError('Empty domain')
if not labels[-1]:
del labels[-1]
trailing_dot = True
for label in labels:
s = ulabel(label)
if s:
result.append(s)
else:
raise IDNAError('Empty label')
if trailing_dot:
result.append('')
return '.'.join(result)

File diff suppressed because it is too large Load Diff

View File

@ -1,54 +0,0 @@
"""
Given a list of integers, made up of (hopefully) a small number of long runs
of consecutive integers, compute a representation of the form
((start1, end1), (start2, end2) ...). Then answer the question "was x present
in the original list?" in time O(log(# runs)).
"""
import bisect
from typing import List, Tuple
def intranges_from_list(list_: List[int]) -> Tuple[int, ...]:
"""Represent a list of integers as a sequence of ranges:
((start_0, end_0), (start_1, end_1), ...), such that the original
integers are exactly those x such that start_i <= x < end_i for some i.
Ranges are encoded as single integers (start << 32 | end), not as tuples.
"""
sorted_list = sorted(list_)
ranges = []
last_write = -1
for i in range(len(sorted_list)):
if i+1 < len(sorted_list):
if sorted_list[i] == sorted_list[i+1]-1:
continue
current_range = sorted_list[last_write+1:i+1]
ranges.append(_encode_range(current_range[0], current_range[-1] + 1))
last_write = i
return tuple(ranges)
def _encode_range(start: int, end: int) -> int:
return (start << 32) | end
def _decode_range(r: int) -> Tuple[int, int]:
return (r >> 32), (r & ((1 << 32) - 1))
def intranges_contain(int_: int, ranges: Tuple[int, ...]) -> bool:
"""Determine if `int_` falls into one of the ranges in `ranges`."""
tuple_ = _encode_range(int_, 0)
pos = bisect.bisect_left(ranges, tuple_)
# we could be immediately ahead of a tuple (start, end)
# with start < int_ <= end
if pos > 0:
left, right = _decode_range(ranges[pos-1])
if left <= int_ < right:
return True
# or we could be immediately behind a tuple (int_, end)
if pos < len(ranges):
left, _ = _decode_range(ranges[pos])
if left == int_:
return True
return False

View File

@ -1,2 +0,0 @@
__version__ = '3.6'

File diff suppressed because it is too large Load Diff

View File

@ -1,180 +0,0 @@
# __
# /__) _ _ _ _ _/ _
# / ( (- (/ (/ (- _) / _)
# /
"""
Requests HTTP Library
~~~~~~~~~~~~~~~~~~~~~
Requests is an HTTP library, written in Python, for human beings.
Basic GET usage:
>>> import requests
>>> r = requests.get('https://www.python.org')
>>> r.status_code
200
>>> b'Python is a programming language' in r.content
True
... or POST:
>>> payload = dict(key1='value1', key2='value2')
>>> r = requests.post('https://httpbin.org/post', data=payload)
>>> print(r.text)
{
...
"form": {
"key1": "value1",
"key2": "value2"
},
...
}
The other HTTP methods are supported - see `requests.api`. Full documentation
is at <https://requests.readthedocs.io>.
:copyright: (c) 2017 by Kenneth Reitz.
:license: Apache 2.0, see LICENSE for more details.
"""
import warnings
import urllib3
from .exceptions import RequestsDependencyWarning
try:
from charset_normalizer import __version__ as charset_normalizer_version
except ImportError:
charset_normalizer_version = None
try:
from chardet import __version__ as chardet_version
except ImportError:
chardet_version = None
def check_compatibility(urllib3_version, chardet_version, charset_normalizer_version):
urllib3_version = urllib3_version.split(".")
assert urllib3_version != ["dev"] # Verify urllib3 isn't installed from git.
# Sometimes, urllib3 only reports its version as 16.1.
if len(urllib3_version) == 2:
urllib3_version.append("0")
# Check urllib3 for compatibility.
major, minor, patch = urllib3_version # noqa: F811
major, minor, patch = int(major), int(minor), int(patch)
# urllib3 >= 1.21.1
assert major >= 1
if major == 1:
assert minor >= 21
# Check charset_normalizer for compatibility.
if chardet_version:
major, minor, patch = chardet_version.split(".")[:3]
major, minor, patch = int(major), int(minor), int(patch)
# chardet_version >= 3.0.2, < 6.0.0
assert (3, 0, 2) <= (major, minor, patch) < (6, 0, 0)
elif charset_normalizer_version:
major, minor, patch = charset_normalizer_version.split(".")[:3]
major, minor, patch = int(major), int(minor), int(patch)
# charset_normalizer >= 2.0.0 < 4.0.0
assert (2, 0, 0) <= (major, minor, patch) < (4, 0, 0)
else:
raise Exception("You need either charset_normalizer or chardet installed")
def _check_cryptography(cryptography_version):
# cryptography < 1.3.4
try:
cryptography_version = list(map(int, cryptography_version.split(".")))
except ValueError:
return
if cryptography_version < [1, 3, 4]:
warning = "Old version of cryptography ({}) may cause slowdown.".format(
cryptography_version
)
warnings.warn(warning, RequestsDependencyWarning)
# Check imported dependencies for compatibility.
try:
check_compatibility(
urllib3.__version__, chardet_version, charset_normalizer_version
)
except (AssertionError, ValueError):
warnings.warn(
"urllib3 ({}) or chardet ({})/charset_normalizer ({}) doesn't match a supported "
"version!".format(
urllib3.__version__, chardet_version, charset_normalizer_version
),
RequestsDependencyWarning,
)
# Attempt to enable urllib3's fallback for SNI support
# if the standard library doesn't support SNI or the
# 'ssl' library isn't available.
try:
try:
import ssl
except ImportError:
ssl = None
if not getattr(ssl, "HAS_SNI", False):
from urllib3.contrib import pyopenssl
pyopenssl.inject_into_urllib3()
# Check cryptography version
from cryptography import __version__ as cryptography_version
_check_cryptography(cryptography_version)
except ImportError:
pass
# urllib3's DependencyWarnings should be silenced.
from urllib3.exceptions import DependencyWarning
warnings.simplefilter("ignore", DependencyWarning)
# Set default logging handler to avoid "No handler found" warnings.
import logging
from logging import NullHandler
from . import packages, utils
from .__version__ import (
__author__,
__author_email__,
__build__,
__cake__,
__copyright__,
__description__,
__license__,
__title__,
__url__,
__version__,
)
from .api import delete, get, head, options, patch, post, put, request
from .exceptions import (
ConnectionError,
ConnectTimeout,
FileModeWarning,
HTTPError,
JSONDecodeError,
ReadTimeout,
RequestException,
Timeout,
TooManyRedirects,
URLRequired,
)
from .models import PreparedRequest, Request, Response
from .sessions import Session, session
from .status_codes import codes
logging.getLogger(__name__).addHandler(NullHandler())
# FileModeWarnings go off per the default.
warnings.simplefilter("default", FileModeWarning, append=True)

View File

@ -1,14 +0,0 @@
# .-. .-. .-. . . .-. .-. .-. .-.
# |( |- |.| | | |- `-. | `-.
# ' ' `-' `-`.`-' `-' `-' ' `-'
__title__ = "requests"
__description__ = "Python HTTP for Humans."
__url__ = "https://requests.readthedocs.io"
__version__ = "2.31.0"
__build__ = 0x023100
__author__ = "Kenneth Reitz"
__author_email__ = "me@kennethreitz.org"
__license__ = "Apache 2.0"
__copyright__ = "Copyright Kenneth Reitz"
__cake__ = "\u2728 \U0001f370 \u2728"

View File

@ -1,50 +0,0 @@
"""
requests._internal_utils
~~~~~~~~~~~~~~
Provides utility functions that are consumed internally by Requests
which depend on extremely few external helpers (such as compat)
"""
import re
from .compat import builtin_str
_VALID_HEADER_NAME_RE_BYTE = re.compile(rb"^[^:\s][^:\r\n]*$")
_VALID_HEADER_NAME_RE_STR = re.compile(r"^[^:\s][^:\r\n]*$")
_VALID_HEADER_VALUE_RE_BYTE = re.compile(rb"^\S[^\r\n]*$|^$")
_VALID_HEADER_VALUE_RE_STR = re.compile(r"^\S[^\r\n]*$|^$")
_HEADER_VALIDATORS_STR = (_VALID_HEADER_NAME_RE_STR, _VALID_HEADER_VALUE_RE_STR)
_HEADER_VALIDATORS_BYTE = (_VALID_HEADER_NAME_RE_BYTE, _VALID_HEADER_VALUE_RE_BYTE)
HEADER_VALIDATORS = {
bytes: _HEADER_VALIDATORS_BYTE,
str: _HEADER_VALIDATORS_STR,
}
def to_native_string(string, encoding="ascii"):
"""Given a string object, regardless of type, returns a representation of
that string in the native string type, encoding and decoding where
necessary. This assumes ASCII unless told otherwise.
"""
if isinstance(string, builtin_str):
out = string
else:
out = string.decode(encoding)
return out
def unicode_is_ascii(u_string):
"""Determine if unicode string only contains ASCII characters.
:param str u_string: unicode string to check. Must be unicode
and not Python 2 `str`.
:rtype: bool
"""
assert isinstance(u_string, str)
try:
u_string.encode("ascii")
return True
except UnicodeEncodeError:
return False

View File

@ -1,538 +0,0 @@
"""
requests.adapters
~~~~~~~~~~~~~~~~~
This module contains the transport adapters that Requests uses to define
and maintain connections.
"""
import os.path
import socket # noqa: F401
from urllib3.exceptions import ClosedPoolError, ConnectTimeoutError
from urllib3.exceptions import HTTPError as _HTTPError
from urllib3.exceptions import InvalidHeader as _InvalidHeader
from urllib3.exceptions import (
LocationValueError,
MaxRetryError,
NewConnectionError,
ProtocolError,
)
from urllib3.exceptions import ProxyError as _ProxyError
from urllib3.exceptions import ReadTimeoutError, ResponseError
from urllib3.exceptions import SSLError as _SSLError
from urllib3.poolmanager import PoolManager, proxy_from_url
from urllib3.util import Timeout as TimeoutSauce
from urllib3.util import parse_url
from urllib3.util.retry import Retry
from .auth import _basic_auth_str
from .compat import basestring, urlparse
from .cookies import extract_cookies_to_jar
from .exceptions import (
ConnectionError,
ConnectTimeout,
InvalidHeader,
InvalidProxyURL,
InvalidSchema,
InvalidURL,
ProxyError,
ReadTimeout,
RetryError,
SSLError,
)
from .models import Response
from .structures import CaseInsensitiveDict
from .utils import (
DEFAULT_CA_BUNDLE_PATH,
extract_zipped_paths,
get_auth_from_url,
get_encoding_from_headers,
prepend_scheme_if_needed,
select_proxy,
urldefragauth,
)
try:
from urllib3.contrib.socks import SOCKSProxyManager
except ImportError:
def SOCKSProxyManager(*args, **kwargs):
raise InvalidSchema("Missing dependencies for SOCKS support.")
DEFAULT_POOLBLOCK = False
DEFAULT_POOLSIZE = 10
DEFAULT_RETRIES = 0
DEFAULT_POOL_TIMEOUT = None
class BaseAdapter:
"""The Base Transport Adapter"""
def __init__(self):
super().__init__()
def send(
self, request, stream=False, timeout=None, verify=True, cert=None, proxies=None
):
"""Sends PreparedRequest object. Returns Response object.
:param request: The :class:`PreparedRequest <PreparedRequest>` being sent.
:param stream: (optional) Whether to stream the request content.
:param timeout: (optional) How long to wait for the server to send
data before giving up, as a float, or a :ref:`(connect timeout,
read timeout) <timeouts>` tuple.
:type timeout: float or tuple
:param verify: (optional) Either a boolean, in which case it controls whether we verify
the server's TLS certificate, or a string, in which case it must be a path
to a CA bundle to use
:param cert: (optional) Any user-provided SSL certificate to be trusted.
:param proxies: (optional) The proxies dictionary to apply to the request.
"""
raise NotImplementedError
def close(self):
"""Cleans up adapter specific items."""
raise NotImplementedError
class HTTPAdapter(BaseAdapter):
"""The built-in HTTP Adapter for urllib3.
Provides a general-case interface for Requests sessions to contact HTTP and
HTTPS urls by implementing the Transport Adapter interface. This class will
usually be created by the :class:`Session <Session>` class under the
covers.
:param pool_connections: The number of urllib3 connection pools to cache.
:param pool_maxsize: The maximum number of connections to save in the pool.
:param max_retries: The maximum number of retries each connection
should attempt. Note, this applies only to failed DNS lookups, socket
connections and connection timeouts, never to requests where data has
made it to the server. By default, Requests does not retry failed
connections. If you need granular control over the conditions under
which we retry a request, import urllib3's ``Retry`` class and pass
that instead.
:param pool_block: Whether the connection pool should block for connections.
Usage::
>>> import requests
>>> s = requests.Session()
>>> a = requests.adapters.HTTPAdapter(max_retries=3)
>>> s.mount('http://', a)
"""
__attrs__ = [
"max_retries",
"config",
"_pool_connections",
"_pool_maxsize",
"_pool_block",
]
def __init__(
self,
pool_connections=DEFAULT_POOLSIZE,
pool_maxsize=DEFAULT_POOLSIZE,
max_retries=DEFAULT_RETRIES,
pool_block=DEFAULT_POOLBLOCK,
):
if max_retries == DEFAULT_RETRIES:
self.max_retries = Retry(0, read=False)
else:
self.max_retries = Retry.from_int(max_retries)
self.config = {}
self.proxy_manager = {}
super().__init__()
self._pool_connections = pool_connections
self._pool_maxsize = pool_maxsize
self._pool_block = pool_block
self.init_poolmanager(pool_connections, pool_maxsize, block=pool_block)
def __getstate__(self):
return {attr: getattr(self, attr, None) for attr in self.__attrs__}
def __setstate__(self, state):
# Can't handle by adding 'proxy_manager' to self.__attrs__ because
# self.poolmanager uses a lambda function, which isn't pickleable.
self.proxy_manager = {}
self.config = {}
for attr, value in state.items():
setattr(self, attr, value)
self.init_poolmanager(
self._pool_connections, self._pool_maxsize, block=self._pool_block
)
def init_poolmanager(
self, connections, maxsize, block=DEFAULT_POOLBLOCK, **pool_kwargs
):
"""Initializes a urllib3 PoolManager.
This method should not be called from user code, and is only
exposed for use when subclassing the
:class:`HTTPAdapter <requests.adapters.HTTPAdapter>`.
:param connections: The number of urllib3 connection pools to cache.
:param maxsize: The maximum number of connections to save in the pool.
:param block: Block when no free connections are available.
:param pool_kwargs: Extra keyword arguments used to initialize the Pool Manager.
"""
# save these values for pickling
self._pool_connections = connections
self._pool_maxsize = maxsize
self._pool_block = block
self.poolmanager = PoolManager(
num_pools=connections,
maxsize=maxsize,
block=block,
**pool_kwargs,
)
def proxy_manager_for(self, proxy, **proxy_kwargs):
"""Return urllib3 ProxyManager for the given proxy.
This method should not be called from user code, and is only
exposed for use when subclassing the
:class:`HTTPAdapter <requests.adapters.HTTPAdapter>`.
:param proxy: The proxy to return a urllib3 ProxyManager for.
:param proxy_kwargs: Extra keyword arguments used to configure the Proxy Manager.
:returns: ProxyManager
:rtype: urllib3.ProxyManager
"""
if proxy in self.proxy_manager:
manager = self.proxy_manager[proxy]
elif proxy.lower().startswith("socks"):
username, password = get_auth_from_url(proxy)
manager = self.proxy_manager[proxy] = SOCKSProxyManager(
proxy,
username=username,
password=password,
num_pools=self._pool_connections,
maxsize=self._pool_maxsize,
block=self._pool_block,
**proxy_kwargs,
)
else:
proxy_headers = self.proxy_headers(proxy)
manager = self.proxy_manager[proxy] = proxy_from_url(
proxy,
proxy_headers=proxy_headers,
num_pools=self._pool_connections,
maxsize=self._pool_maxsize,
block=self._pool_block,
**proxy_kwargs,
)
return manager
def cert_verify(self, conn, url, verify, cert):
"""Verify a SSL certificate. This method should not be called from user
code, and is only exposed for use when subclassing the
:class:`HTTPAdapter <requests.adapters.HTTPAdapter>`.
:param conn: The urllib3 connection object associated with the cert.
:param url: The requested URL.
:param verify: Either a boolean, in which case it controls whether we verify
the server's TLS certificate, or a string, in which case it must be a path
to a CA bundle to use
:param cert: The SSL certificate to verify.
"""
if url.lower().startswith("https") and verify:
cert_loc = None
# Allow self-specified cert location.
if verify is not True:
cert_loc = verify
if not cert_loc:
cert_loc = extract_zipped_paths(DEFAULT_CA_BUNDLE_PATH)
if not cert_loc or not os.path.exists(cert_loc):
raise OSError(
f"Could not find a suitable TLS CA certificate bundle, "
f"invalid path: {cert_loc}"
)
conn.cert_reqs = "CERT_REQUIRED"
if not os.path.isdir(cert_loc):
conn.ca_certs = cert_loc
else:
conn.ca_cert_dir = cert_loc
else:
conn.cert_reqs = "CERT_NONE"
conn.ca_certs = None
conn.ca_cert_dir = None
if cert:
if not isinstance(cert, basestring):
conn.cert_file = cert[0]
conn.key_file = cert[1]
else:
conn.cert_file = cert
conn.key_file = None
if conn.cert_file and not os.path.exists(conn.cert_file):
raise OSError(
f"Could not find the TLS certificate file, "
f"invalid path: {conn.cert_file}"
)
if conn.key_file and not os.path.exists(conn.key_file):
raise OSError(
f"Could not find the TLS key file, invalid path: {conn.key_file}"
)
def build_response(self, req, resp):
"""Builds a :class:`Response <requests.Response>` object from a urllib3
response. This should not be called from user code, and is only exposed
for use when subclassing the
:class:`HTTPAdapter <requests.adapters.HTTPAdapter>`
:param req: The :class:`PreparedRequest <PreparedRequest>` used to generate the response.
:param resp: The urllib3 response object.
:rtype: requests.Response
"""
response = Response()
# Fallback to None if there's no status_code, for whatever reason.
response.status_code = getattr(resp, "status", None)
# Make headers case-insensitive.
response.headers = CaseInsensitiveDict(getattr(resp, "headers", {}))
# Set encoding.
response.encoding = get_encoding_from_headers(response.headers)
response.raw = resp
response.reason = response.raw.reason
if isinstance(req.url, bytes):
response.url = req.url.decode("utf-8")
else:
response.url = req.url
# Add new cookies from the server.
extract_cookies_to_jar(response.cookies, req, resp)
# Give the Response some context.
response.request = req
response.connection = self
return response
def get_connection(self, url, proxies=None):
"""Returns a urllib3 connection for the given URL. This should not be
called from user code, and is only exposed for use when subclassing the
:class:`HTTPAdapter <requests.adapters.HTTPAdapter>`.
:param url: The URL to connect to.
:param proxies: (optional) A Requests-style dictionary of proxies used on this request.
:rtype: urllib3.ConnectionPool
"""
proxy = select_proxy(url, proxies)
if proxy:
proxy = prepend_scheme_if_needed(proxy, "http")
proxy_url = parse_url(proxy)
if not proxy_url.host:
raise InvalidProxyURL(
"Please check proxy URL. It is malformed "
"and could be missing the host."
)
proxy_manager = self.proxy_manager_for(proxy)
conn = proxy_manager.connection_from_url(url)
else:
# Only scheme should be lower case
parsed = urlparse(url)
url = parsed.geturl()
conn = self.poolmanager.connection_from_url(url)
return conn
def close(self):
"""Disposes of any internal state.
Currently, this closes the PoolManager and any active ProxyManager,
which closes any pooled connections.
"""
self.poolmanager.clear()
for proxy in self.proxy_manager.values():
proxy.clear()
def request_url(self, request, proxies):
"""Obtain the url to use when making the final request.
If the message is being sent through a HTTP proxy, the full URL has to
be used. Otherwise, we should only use the path portion of the URL.
This should not be called from user code, and is only exposed for use
when subclassing the
:class:`HTTPAdapter <requests.adapters.HTTPAdapter>`.
:param request: The :class:`PreparedRequest <PreparedRequest>` being sent.
:param proxies: A dictionary of schemes or schemes and hosts to proxy URLs.
:rtype: str
"""
proxy = select_proxy(request.url, proxies)
scheme = urlparse(request.url).scheme
is_proxied_http_request = proxy and scheme != "https"
using_socks_proxy = False
if proxy:
proxy_scheme = urlparse(proxy).scheme.lower()
using_socks_proxy = proxy_scheme.startswith("socks")
url = request.path_url
if is_proxied_http_request and not using_socks_proxy:
url = urldefragauth(request.url)
return url
def add_headers(self, request, **kwargs):
"""Add any headers needed by the connection. As of v2.0 this does
nothing by default, but is left for overriding by users that subclass
the :class:`HTTPAdapter <requests.adapters.HTTPAdapter>`.
This should not be called from user code, and is only exposed for use
when subclassing the
:class:`HTTPAdapter <requests.adapters.HTTPAdapter>`.
:param request: The :class:`PreparedRequest <PreparedRequest>` to add headers to.
:param kwargs: The keyword arguments from the call to send().
"""
pass
def proxy_headers(self, proxy):
"""Returns a dictionary of the headers to add to any request sent
through a proxy. This works with urllib3 magic to ensure that they are
correctly sent to the proxy, rather than in a tunnelled request if
CONNECT is being used.
This should not be called from user code, and is only exposed for use
when subclassing the
:class:`HTTPAdapter <requests.adapters.HTTPAdapter>`.
:param proxy: The url of the proxy being used for this request.
:rtype: dict
"""
headers = {}
username, password = get_auth_from_url(proxy)
if username:
headers["Proxy-Authorization"] = _basic_auth_str(username, password)
return headers
def send(
self, request, stream=False, timeout=None, verify=True, cert=None, proxies=None
):
"""Sends PreparedRequest object. Returns Response object.
:param request: The :class:`PreparedRequest <PreparedRequest>` being sent.
:param stream: (optional) Whether to stream the request content.
:param timeout: (optional) How long to wait for the server to send
data before giving up, as a float, or a :ref:`(connect timeout,
read timeout) <timeouts>` tuple.
:type timeout: float or tuple or urllib3 Timeout object
:param verify: (optional) Either a boolean, in which case it controls whether
we verify the server's TLS certificate, or a string, in which case it
must be a path to a CA bundle to use
:param cert: (optional) Any user-provided SSL certificate to be trusted.
:param proxies: (optional) The proxies dictionary to apply to the request.
:rtype: requests.Response
"""
try:
conn = self.get_connection(request.url, proxies)
except LocationValueError as e:
raise InvalidURL(e, request=request)
self.cert_verify(conn, request.url, verify, cert)
url = self.request_url(request, proxies)
self.add_headers(
request,
stream=stream,
timeout=timeout,
verify=verify,
cert=cert,
proxies=proxies,
)
chunked = not (request.body is None or "Content-Length" in request.headers)
if isinstance(timeout, tuple):
try:
connect, read = timeout
timeout = TimeoutSauce(connect=connect, read=read)
except ValueError:
raise ValueError(
f"Invalid timeout {timeout}. Pass a (connect, read) timeout tuple, "
f"or a single float to set both timeouts to the same value."
)
elif isinstance(timeout, TimeoutSauce):
pass
else:
timeout = TimeoutSauce(connect=timeout, read=timeout)
try:
resp = conn.urlopen(
method=request.method,
url=url,
body=request.body,
headers=request.headers,
redirect=False,
assert_same_host=False,
preload_content=False,
decode_content=False,
retries=self.max_retries,
timeout=timeout,
chunked=chunked,
)
except (ProtocolError, OSError) as err:
raise ConnectionError(err, request=request)
except MaxRetryError as e:
if isinstance(e.reason, ConnectTimeoutError):
# TODO: Remove this in 3.0.0: see #2811
if not isinstance(e.reason, NewConnectionError):
raise ConnectTimeout(e, request=request)
if isinstance(e.reason, ResponseError):
raise RetryError(e, request=request)
if isinstance(e.reason, _ProxyError):
raise ProxyError(e, request=request)
if isinstance(e.reason, _SSLError):
# This branch is for urllib3 v1.22 and later.
raise SSLError(e, request=request)
raise ConnectionError(e, request=request)
except ClosedPoolError as e:
raise ConnectionError(e, request=request)
except _ProxyError as e:
raise ProxyError(e)
except (_SSLError, _HTTPError) as e:
if isinstance(e, _SSLError):
# This branch is for urllib3 versions earlier than v1.22
raise SSLError(e, request=request)
elif isinstance(e, ReadTimeoutError):
raise ReadTimeout(e, request=request)
elif isinstance(e, _InvalidHeader):
raise InvalidHeader(e, request=request)
else:
raise
return self.build_response(request, resp)

View File

@ -1,157 +0,0 @@
"""
requests.api
~~~~~~~~~~~~
This module implements the Requests API.
:copyright: (c) 2012 by Kenneth Reitz.
:license: Apache2, see LICENSE for more details.
"""
from . import sessions
def request(method, url, **kwargs):
"""Constructs and sends a :class:`Request <Request>`.
:param method: method for the new :class:`Request` object: ``GET``, ``OPTIONS``, ``HEAD``, ``POST``, ``PUT``, ``PATCH``, or ``DELETE``.
:param url: URL for the new :class:`Request` object.
:param params: (optional) Dictionary, list of tuples or bytes to send
in the query string for the :class:`Request`.
:param data: (optional) Dictionary, list of tuples, bytes, or file-like
object to send in the body of the :class:`Request`.
:param json: (optional) A JSON serializable Python object to send in the body of the :class:`Request`.
:param headers: (optional) Dictionary of HTTP Headers to send with the :class:`Request`.
:param cookies: (optional) Dict or CookieJar object to send with the :class:`Request`.
:param files: (optional) Dictionary of ``'name': file-like-objects`` (or ``{'name': file-tuple}``) for multipart encoding upload.
``file-tuple`` can be a 2-tuple ``('filename', fileobj)``, 3-tuple ``('filename', fileobj, 'content_type')``
or a 4-tuple ``('filename', fileobj, 'content_type', custom_headers)``, where ``'content-type'`` is a string
defining the content type of the given file and ``custom_headers`` a dict-like object containing additional headers
to add for the file.
:param auth: (optional) Auth tuple to enable Basic/Digest/Custom HTTP Auth.
:param timeout: (optional) How many seconds to wait for the server to send data
before giving up, as a float, or a :ref:`(connect timeout, read
timeout) <timeouts>` tuple.
:type timeout: float or tuple
:param allow_redirects: (optional) Boolean. Enable/disable GET/OPTIONS/POST/PUT/PATCH/DELETE/HEAD redirection. Defaults to ``True``.
:type allow_redirects: bool
:param proxies: (optional) Dictionary mapping protocol to the URL of the proxy.
:param verify: (optional) Either a boolean, in which case it controls whether we verify
the server's TLS certificate, or a string, in which case it must be a path
to a CA bundle to use. Defaults to ``True``.
:param stream: (optional) if ``False``, the response content will be immediately downloaded.
:param cert: (optional) if String, path to ssl client cert file (.pem). If Tuple, ('cert', 'key') pair.
:return: :class:`Response <Response>` object
:rtype: requests.Response
Usage::
>>> import requests
>>> req = requests.request('GET', 'https://httpbin.org/get')
>>> req
<Response [200]>
"""
# By using the 'with' statement we are sure the session is closed, thus we
# avoid leaving sockets open which can trigger a ResourceWarning in some
# cases, and look like a memory leak in others.
with sessions.Session() as session:
return session.request(method=method, url=url, **kwargs)
def get(url, params=None, **kwargs):
r"""Sends a GET request.
:param url: URL for the new :class:`Request` object.
:param params: (optional) Dictionary, list of tuples or bytes to send
in the query string for the :class:`Request`.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:return: :class:`Response <Response>` object
:rtype: requests.Response
"""
return request("get", url, params=params, **kwargs)
def options(url, **kwargs):
r"""Sends an OPTIONS request.
:param url: URL for the new :class:`Request` object.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:return: :class:`Response <Response>` object
:rtype: requests.Response
"""
return request("options", url, **kwargs)
def head(url, **kwargs):
r"""Sends a HEAD request.
:param url: URL for the new :class:`Request` object.
:param \*\*kwargs: Optional arguments that ``request`` takes. If
`allow_redirects` is not provided, it will be set to `False` (as
opposed to the default :meth:`request` behavior).
:return: :class:`Response <Response>` object
:rtype: requests.Response
"""
kwargs.setdefault("allow_redirects", False)
return request("head", url, **kwargs)
def post(url, data=None, json=None, **kwargs):
r"""Sends a POST request.
:param url: URL for the new :class:`Request` object.
:param data: (optional) Dictionary, list of tuples, bytes, or file-like
object to send in the body of the :class:`Request`.
:param json: (optional) A JSON serializable Python object to send in the body of the :class:`Request`.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:return: :class:`Response <Response>` object
:rtype: requests.Response
"""
return request("post", url, data=data, json=json, **kwargs)
def put(url, data=None, **kwargs):
r"""Sends a PUT request.
:param url: URL for the new :class:`Request` object.
:param data: (optional) Dictionary, list of tuples, bytes, or file-like
object to send in the body of the :class:`Request`.
:param json: (optional) A JSON serializable Python object to send in the body of the :class:`Request`.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:return: :class:`Response <Response>` object
:rtype: requests.Response
"""
return request("put", url, data=data, **kwargs)
def patch(url, data=None, **kwargs):
r"""Sends a PATCH request.
:param url: URL for the new :class:`Request` object.
:param data: (optional) Dictionary, list of tuples, bytes, or file-like
object to send in the body of the :class:`Request`.
:param json: (optional) A JSON serializable Python object to send in the body of the :class:`Request`.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:return: :class:`Response <Response>` object
:rtype: requests.Response
"""
return request("patch", url, data=data, **kwargs)
def delete(url, **kwargs):
r"""Sends a DELETE request.
:param url: URL for the new :class:`Request` object.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:return: :class:`Response <Response>` object
:rtype: requests.Response
"""
return request("delete", url, **kwargs)

View File

@ -1,315 +0,0 @@
"""
requests.auth
~~~~~~~~~~~~~
This module contains the authentication handlers for Requests.
"""
import hashlib
import os
import re
import threading
import time
import warnings
from base64 import b64encode
from ._internal_utils import to_native_string
from .compat import basestring, str, urlparse
from .cookies import extract_cookies_to_jar
from .utils import parse_dict_header
CONTENT_TYPE_FORM_URLENCODED = "application/x-www-form-urlencoded"
CONTENT_TYPE_MULTI_PART = "multipart/form-data"
def _basic_auth_str(username, password):
"""Returns a Basic Auth string."""
# "I want us to put a big-ol' comment on top of it that
# says that this behaviour is dumb but we need to preserve
# it because people are relying on it."
# - Lukasa
#
# These are here solely to maintain backwards compatibility
# for things like ints. This will be removed in 3.0.0.
if not isinstance(username, basestring):
warnings.warn(
"Non-string usernames will no longer be supported in Requests "
"3.0.0. Please convert the object you've passed in ({!r}) to "
"a string or bytes object in the near future to avoid "
"problems.".format(username),
category=DeprecationWarning,
)
username = str(username)
if not isinstance(password, basestring):
warnings.warn(
"Non-string passwords will no longer be supported in Requests "
"3.0.0. Please convert the object you've passed in ({!r}) to "
"a string or bytes object in the near future to avoid "
"problems.".format(type(password)),
category=DeprecationWarning,
)
password = str(password)
# -- End Removal --
if isinstance(username, str):
username = username.encode("latin1")
if isinstance(password, str):
password = password.encode("latin1")
authstr = "Basic " + to_native_string(
b64encode(b":".join((username, password))).strip()
)
return authstr
class AuthBase:
"""Base class that all auth implementations derive from"""
def __call__(self, r):
raise NotImplementedError("Auth hooks must be callable.")
class HTTPBasicAuth(AuthBase):
"""Attaches HTTP Basic Authentication to the given Request object."""
def __init__(self, username, password):
self.username = username
self.password = password
def __eq__(self, other):
return all(
[
self.username == getattr(other, "username", None),
self.password == getattr(other, "password", None),
]
)
def __ne__(self, other):
return not self == other
def __call__(self, r):
r.headers["Authorization"] = _basic_auth_str(self.username, self.password)
return r
class HTTPProxyAuth(HTTPBasicAuth):
"""Attaches HTTP Proxy Authentication to a given Request object."""
def __call__(self, r):
r.headers["Proxy-Authorization"] = _basic_auth_str(self.username, self.password)
return r
class HTTPDigestAuth(AuthBase):
"""Attaches HTTP Digest Authentication to the given Request object."""
def __init__(self, username, password):
self.username = username
self.password = password
# Keep state in per-thread local storage
self._thread_local = threading.local()
def init_per_thread_state(self):
# Ensure state is initialized just once per-thread
if not hasattr(self._thread_local, "init"):
self._thread_local.init = True
self._thread_local.last_nonce = ""
self._thread_local.nonce_count = 0
self._thread_local.chal = {}
self._thread_local.pos = None
self._thread_local.num_401_calls = None
def build_digest_header(self, method, url):
"""
:rtype: str
"""
realm = self._thread_local.chal["realm"]
nonce = self._thread_local.chal["nonce"]
qop = self._thread_local.chal.get("qop")
algorithm = self._thread_local.chal.get("algorithm")
opaque = self._thread_local.chal.get("opaque")
hash_utf8 = None
if algorithm is None:
_algorithm = "MD5"
else:
_algorithm = algorithm.upper()
# lambdas assume digest modules are imported at the top level
if _algorithm == "MD5" or _algorithm == "MD5-SESS":
def md5_utf8(x):
if isinstance(x, str):
x = x.encode("utf-8")
return hashlib.md5(x).hexdigest()
hash_utf8 = md5_utf8
elif _algorithm == "SHA":
def sha_utf8(x):
if isinstance(x, str):
x = x.encode("utf-8")
return hashlib.sha1(x).hexdigest()
hash_utf8 = sha_utf8
elif _algorithm == "SHA-256":
def sha256_utf8(x):
if isinstance(x, str):
x = x.encode("utf-8")
return hashlib.sha256(x).hexdigest()
hash_utf8 = sha256_utf8
elif _algorithm == "SHA-512":
def sha512_utf8(x):
if isinstance(x, str):
x = x.encode("utf-8")
return hashlib.sha512(x).hexdigest()
hash_utf8 = sha512_utf8
KD = lambda s, d: hash_utf8(f"{s}:{d}") # noqa:E731
if hash_utf8 is None:
return None
# XXX not implemented yet
entdig = None
p_parsed = urlparse(url)
#: path is request-uri defined in RFC 2616 which should not be empty
path = p_parsed.path or "/"
if p_parsed.query:
path += f"?{p_parsed.query}"
A1 = f"{self.username}:{realm}:{self.password}"
A2 = f"{method}:{path}"
HA1 = hash_utf8(A1)
HA2 = hash_utf8(A2)
if nonce == self._thread_local.last_nonce:
self._thread_local.nonce_count += 1
else:
self._thread_local.nonce_count = 1
ncvalue = f"{self._thread_local.nonce_count:08x}"
s = str(self._thread_local.nonce_count).encode("utf-8")
s += nonce.encode("utf-8")
s += time.ctime().encode("utf-8")
s += os.urandom(8)
cnonce = hashlib.sha1(s).hexdigest()[:16]
if _algorithm == "MD5-SESS":
HA1 = hash_utf8(f"{HA1}:{nonce}:{cnonce}")
if not qop:
respdig = KD(HA1, f"{nonce}:{HA2}")
elif qop == "auth" or "auth" in qop.split(","):
noncebit = f"{nonce}:{ncvalue}:{cnonce}:auth:{HA2}"
respdig = KD(HA1, noncebit)
else:
# XXX handle auth-int.
return None
self._thread_local.last_nonce = nonce
# XXX should the partial digests be encoded too?
base = (
f'username="{self.username}", realm="{realm}", nonce="{nonce}", '
f'uri="{path}", response="{respdig}"'
)
if opaque:
base += f', opaque="{opaque}"'
if algorithm:
base += f', algorithm="{algorithm}"'
if entdig:
base += f', digest="{entdig}"'
if qop:
base += f', qop="auth", nc={ncvalue}, cnonce="{cnonce}"'
return f"Digest {base}"
def handle_redirect(self, r, **kwargs):
"""Reset num_401_calls counter on redirects."""
if r.is_redirect:
self._thread_local.num_401_calls = 1
def handle_401(self, r, **kwargs):
"""
Takes the given response and tries digest-auth, if needed.
:rtype: requests.Response
"""
# If response is not 4xx, do not auth
# See https://github.com/psf/requests/issues/3772
if not 400 <= r.status_code < 500:
self._thread_local.num_401_calls = 1
return r
if self._thread_local.pos is not None:
# Rewind the file position indicator of the body to where
# it was to resend the request.
r.request.body.seek(self._thread_local.pos)
s_auth = r.headers.get("www-authenticate", "")
if "digest" in s_auth.lower() and self._thread_local.num_401_calls < 2:
self._thread_local.num_401_calls += 1
pat = re.compile(r"digest ", flags=re.IGNORECASE)
self._thread_local.chal = parse_dict_header(pat.sub("", s_auth, count=1))
# Consume content and release the original connection
# to allow our new request to reuse the same one.
r.content
r.close()
prep = r.request.copy()
extract_cookies_to_jar(prep._cookies, r.request, r.raw)
prep.prepare_cookies(prep._cookies)
prep.headers["Authorization"] = self.build_digest_header(
prep.method, prep.url
)
_r = r.connection.send(prep, **kwargs)
_r.history.append(r)
_r.request = prep
return _r
self._thread_local.num_401_calls = 1
return r
def __call__(self, r):
# Initialize per-thread state, if needed
self.init_per_thread_state()
# If we have a saved nonce, skip the 401
if self._thread_local.last_nonce:
r.headers["Authorization"] = self.build_digest_header(r.method, r.url)
try:
self._thread_local.pos = r.body.tell()
except AttributeError:
# In the case of HTTPDigestAuth being reused and the body of
# the previous request was a file-like object, pos has the
# file position of the previous body. Ensure it's set to
# None.
self._thread_local.pos = None
r.register_hook("response", self.handle_401)
r.register_hook("response", self.handle_redirect)
self._thread_local.num_401_calls = 1
return r
def __eq__(self, other):
return all(
[
self.username == getattr(other, "username", None),
self.password == getattr(other, "password", None),
]
)
def __ne__(self, other):
return not self == other

View File

@ -1,17 +0,0 @@
#!/usr/bin/env python
"""
requests.certs
~~~~~~~~~~~~~~
This module returns the preferred default CA certificate bundle. There is
only one the one from the certifi package.
If you are packaging Requests, e.g., for a Linux distribution or a managed
environment, you can change the definition of where() to return a separately
packaged CA bundle.
"""
from certifi import where
if __name__ == "__main__":
print(where())

View File

@ -1,79 +0,0 @@
"""
requests.compat
~~~~~~~~~~~~~~~
This module previously handled import compatibility issues
between Python 2 and Python 3. It remains for backwards
compatibility until the next major version.
"""
try:
import chardet
except ImportError:
import charset_normalizer as chardet
import sys
# -------
# Pythons
# -------
# Syntax sugar.
_ver = sys.version_info
#: Python 2.x?
is_py2 = _ver[0] == 2
#: Python 3.x?
is_py3 = _ver[0] == 3
# json/simplejson module import resolution
has_simplejson = False
try:
import simplejson as json
has_simplejson = True
except ImportError:
import json
if has_simplejson:
from simplejson import JSONDecodeError
else:
from json import JSONDecodeError
# Keep OrderedDict for backwards compatibility.
from collections import OrderedDict
from collections.abc import Callable, Mapping, MutableMapping
from http import cookiejar as cookielib
from http.cookies import Morsel
from io import StringIO
# --------------
# Legacy Imports
# --------------
from urllib.parse import (
quote,
quote_plus,
unquote,
unquote_plus,
urldefrag,
urlencode,
urljoin,
urlparse,
urlsplit,
urlunparse,
)
from urllib.request import (
getproxies,
getproxies_environment,
parse_http_list,
proxy_bypass,
proxy_bypass_environment,
)
builtin_str = str
str = str
bytes = bytes
basestring = (str, bytes)
numeric_types = (int, float)
integer_types = (int,)

View File

@ -1,561 +0,0 @@
"""
requests.cookies
~~~~~~~~~~~~~~~~
Compatibility code to be able to use `cookielib.CookieJar` with requests.
requests.utils imports from here, so be careful with imports.
"""
import calendar
import copy
import time
from ._internal_utils import to_native_string
from .compat import Morsel, MutableMapping, cookielib, urlparse, urlunparse
try:
import threading
except ImportError:
import dummy_threading as threading
class MockRequest:
"""Wraps a `requests.Request` to mimic a `urllib2.Request`.
The code in `cookielib.CookieJar` expects this interface in order to correctly
manage cookie policies, i.e., determine whether a cookie can be set, given the
domains of the request and the cookie.
The original request object is read-only. The client is responsible for collecting
the new headers via `get_new_headers()` and interpreting them appropriately. You
probably want `get_cookie_header`, defined below.
"""
def __init__(self, request):
self._r = request
self._new_headers = {}
self.type = urlparse(self._r.url).scheme
def get_type(self):
return self.type
def get_host(self):
return urlparse(self._r.url).netloc
def get_origin_req_host(self):
return self.get_host()
def get_full_url(self):
# Only return the response's URL if the user hadn't set the Host
# header
if not self._r.headers.get("Host"):
return self._r.url
# If they did set it, retrieve it and reconstruct the expected domain
host = to_native_string(self._r.headers["Host"], encoding="utf-8")
parsed = urlparse(self._r.url)
# Reconstruct the URL as we expect it
return urlunparse(
[
parsed.scheme,
host,
parsed.path,
parsed.params,
parsed.query,
parsed.fragment,
]
)
def is_unverifiable(self):
return True
def has_header(self, name):
return name in self._r.headers or name in self._new_headers
def get_header(self, name, default=None):
return self._r.headers.get(name, self._new_headers.get(name, default))
def add_header(self, key, val):
"""cookielib has no legitimate use for this method; add it back if you find one."""
raise NotImplementedError(
"Cookie headers should be added with add_unredirected_header()"
)
def add_unredirected_header(self, name, value):
self._new_headers[name] = value
def get_new_headers(self):
return self._new_headers
@property
def unverifiable(self):
return self.is_unverifiable()
@property
def origin_req_host(self):
return self.get_origin_req_host()
@property
def host(self):
return self.get_host()
class MockResponse:
"""Wraps a `httplib.HTTPMessage` to mimic a `urllib.addinfourl`.
...what? Basically, expose the parsed HTTP headers from the server response
the way `cookielib` expects to see them.
"""
def __init__(self, headers):
"""Make a MockResponse for `cookielib` to read.
:param headers: a httplib.HTTPMessage or analogous carrying the headers
"""
self._headers = headers
def info(self):
return self._headers
def getheaders(self, name):
self._headers.getheaders(name)
def extract_cookies_to_jar(jar, request, response):
"""Extract the cookies from the response into a CookieJar.
:param jar: cookielib.CookieJar (not necessarily a RequestsCookieJar)
:param request: our own requests.Request object
:param response: urllib3.HTTPResponse object
"""
if not (hasattr(response, "_original_response") and response._original_response):
return
# the _original_response field is the wrapped httplib.HTTPResponse object,
req = MockRequest(request)
# pull out the HTTPMessage with the headers and put it in the mock:
res = MockResponse(response._original_response.msg)
jar.extract_cookies(res, req)
def get_cookie_header(jar, request):
"""
Produce an appropriate Cookie header string to be sent with `request`, or None.
:rtype: str
"""
r = MockRequest(request)
jar.add_cookie_header(r)
return r.get_new_headers().get("Cookie")
def remove_cookie_by_name(cookiejar, name, domain=None, path=None):
"""Unsets a cookie by name, by default over all domains and paths.
Wraps CookieJar.clear(), is O(n).
"""
clearables = []
for cookie in cookiejar:
if cookie.name != name:
continue
if domain is not None and domain != cookie.domain:
continue
if path is not None and path != cookie.path:
continue
clearables.append((cookie.domain, cookie.path, cookie.name))
for domain, path, name in clearables:
cookiejar.clear(domain, path, name)
class CookieConflictError(RuntimeError):
"""There are two cookies that meet the criteria specified in the cookie jar.
Use .get and .set and include domain and path args in order to be more specific.
"""
class RequestsCookieJar(cookielib.CookieJar, MutableMapping):
"""Compatibility class; is a cookielib.CookieJar, but exposes a dict
interface.
This is the CookieJar we create by default for requests and sessions that
don't specify one, since some clients may expect response.cookies and
session.cookies to support dict operations.
Requests does not use the dict interface internally; it's just for
compatibility with external client code. All requests code should work
out of the box with externally provided instances of ``CookieJar``, e.g.
``LWPCookieJar`` and ``FileCookieJar``.
Unlike a regular CookieJar, this class is pickleable.
.. warning:: dictionary operations that are normally O(1) may be O(n).
"""
def get(self, name, default=None, domain=None, path=None):
"""Dict-like get() that also supports optional domain and path args in
order to resolve naming collisions from using one cookie jar over
multiple domains.
.. warning:: operation is O(n), not O(1).
"""
try:
return self._find_no_duplicates(name, domain, path)
except KeyError:
return default
def set(self, name, value, **kwargs):
"""Dict-like set() that also supports optional domain and path args in
order to resolve naming collisions from using one cookie jar over
multiple domains.
"""
# support client code that unsets cookies by assignment of a None value:
if value is None:
remove_cookie_by_name(
self, name, domain=kwargs.get("domain"), path=kwargs.get("path")
)
return
if isinstance(value, Morsel):
c = morsel_to_cookie(value)
else:
c = create_cookie(name, value, **kwargs)
self.set_cookie(c)
return c
def iterkeys(self):
"""Dict-like iterkeys() that returns an iterator of names of cookies
from the jar.
.. seealso:: itervalues() and iteritems().
"""
for cookie in iter(self):
yield cookie.name
def keys(self):
"""Dict-like keys() that returns a list of names of cookies from the
jar.
.. seealso:: values() and items().
"""
return list(self.iterkeys())
def itervalues(self):
"""Dict-like itervalues() that returns an iterator of values of cookies
from the jar.
.. seealso:: iterkeys() and iteritems().
"""
for cookie in iter(self):
yield cookie.value
def values(self):
"""Dict-like values() that returns a list of values of cookies from the
jar.
.. seealso:: keys() and items().
"""
return list(self.itervalues())
def iteritems(self):
"""Dict-like iteritems() that returns an iterator of name-value tuples
from the jar.
.. seealso:: iterkeys() and itervalues().
"""
for cookie in iter(self):
yield cookie.name, cookie.value
def items(self):
"""Dict-like items() that returns a list of name-value tuples from the
jar. Allows client-code to call ``dict(RequestsCookieJar)`` and get a
vanilla python dict of key value pairs.
.. seealso:: keys() and values().
"""
return list(self.iteritems())
def list_domains(self):
"""Utility method to list all the domains in the jar."""
domains = []
for cookie in iter(self):
if cookie.domain not in domains:
domains.append(cookie.domain)
return domains
def list_paths(self):
"""Utility method to list all the paths in the jar."""
paths = []
for cookie in iter(self):
if cookie.path not in paths:
paths.append(cookie.path)
return paths
def multiple_domains(self):
"""Returns True if there are multiple domains in the jar.
Returns False otherwise.
:rtype: bool
"""
domains = []
for cookie in iter(self):
if cookie.domain is not None and cookie.domain in domains:
return True
domains.append(cookie.domain)
return False # there is only one domain in jar
def get_dict(self, domain=None, path=None):
"""Takes as an argument an optional domain and path and returns a plain
old Python dict of name-value pairs of cookies that meet the
requirements.
:rtype: dict
"""
dictionary = {}
for cookie in iter(self):
if (domain is None or cookie.domain == domain) and (
path is None or cookie.path == path
):
dictionary[cookie.name] = cookie.value
return dictionary
def __contains__(self, name):
try:
return super().__contains__(name)
except CookieConflictError:
return True
def __getitem__(self, name):
"""Dict-like __getitem__() for compatibility with client code. Throws
exception if there are more than one cookie with name. In that case,
use the more explicit get() method instead.
.. warning:: operation is O(n), not O(1).
"""
return self._find_no_duplicates(name)
def __setitem__(self, name, value):
"""Dict-like __setitem__ for compatibility with client code. Throws
exception if there is already a cookie of that name in the jar. In that
case, use the more explicit set() method instead.
"""
self.set(name, value)
def __delitem__(self, name):
"""Deletes a cookie given a name. Wraps ``cookielib.CookieJar``'s
``remove_cookie_by_name()``.
"""
remove_cookie_by_name(self, name)
def set_cookie(self, cookie, *args, **kwargs):
if (
hasattr(cookie.value, "startswith")
and cookie.value.startswith('"')
and cookie.value.endswith('"')
):
cookie.value = cookie.value.replace('\\"', "")
return super().set_cookie(cookie, *args, **kwargs)
def update(self, other):
"""Updates this jar with cookies from another CookieJar or dict-like"""
if isinstance(other, cookielib.CookieJar):
for cookie in other:
self.set_cookie(copy.copy(cookie))
else:
super().update(other)
def _find(self, name, domain=None, path=None):
"""Requests uses this method internally to get cookie values.
If there are conflicting cookies, _find arbitrarily chooses one.
See _find_no_duplicates if you want an exception thrown if there are
conflicting cookies.
:param name: a string containing name of cookie
:param domain: (optional) string containing domain of cookie
:param path: (optional) string containing path of cookie
:return: cookie.value
"""
for cookie in iter(self):
if cookie.name == name:
if domain is None or cookie.domain == domain:
if path is None or cookie.path == path:
return cookie.value
raise KeyError(f"name={name!r}, domain={domain!r}, path={path!r}")
def _find_no_duplicates(self, name, domain=None, path=None):
"""Both ``__get_item__`` and ``get`` call this function: it's never
used elsewhere in Requests.
:param name: a string containing name of cookie
:param domain: (optional) string containing domain of cookie
:param path: (optional) string containing path of cookie
:raises KeyError: if cookie is not found
:raises CookieConflictError: if there are multiple cookies
that match name and optionally domain and path
:return: cookie.value
"""
toReturn = None
for cookie in iter(self):
if cookie.name == name:
if domain is None or cookie.domain == domain:
if path is None or cookie.path == path:
if toReturn is not None:
# if there are multiple cookies that meet passed in criteria
raise CookieConflictError(
f"There are multiple cookies with name, {name!r}"
)
# we will eventually return this as long as no cookie conflict
toReturn = cookie.value
if toReturn:
return toReturn
raise KeyError(f"name={name!r}, domain={domain!r}, path={path!r}")
def __getstate__(self):
"""Unlike a normal CookieJar, this class is pickleable."""
state = self.__dict__.copy()
# remove the unpickleable RLock object
state.pop("_cookies_lock")
return state
def __setstate__(self, state):
"""Unlike a normal CookieJar, this class is pickleable."""
self.__dict__.update(state)
if "_cookies_lock" not in self.__dict__:
self._cookies_lock = threading.RLock()
def copy(self):
"""Return a copy of this RequestsCookieJar."""
new_cj = RequestsCookieJar()
new_cj.set_policy(self.get_policy())
new_cj.update(self)
return new_cj
def get_policy(self):
"""Return the CookiePolicy instance used."""
return self._policy
def _copy_cookie_jar(jar):
if jar is None:
return None
if hasattr(jar, "copy"):
# We're dealing with an instance of RequestsCookieJar
return jar.copy()
# We're dealing with a generic CookieJar instance
new_jar = copy.copy(jar)
new_jar.clear()
for cookie in jar:
new_jar.set_cookie(copy.copy(cookie))
return new_jar
def create_cookie(name, value, **kwargs):
"""Make a cookie from underspecified parameters.
By default, the pair of `name` and `value` will be set for the domain ''
and sent on every request (this is sometimes called a "supercookie").
"""
result = {
"version": 0,
"name": name,
"value": value,
"port": None,
"domain": "",
"path": "/",
"secure": False,
"expires": None,
"discard": True,
"comment": None,
"comment_url": None,
"rest": {"HttpOnly": None},
"rfc2109": False,
}
badargs = set(kwargs) - set(result)
if badargs:
raise TypeError(
f"create_cookie() got unexpected keyword arguments: {list(badargs)}"
)
result.update(kwargs)
result["port_specified"] = bool(result["port"])
result["domain_specified"] = bool(result["domain"])
result["domain_initial_dot"] = result["domain"].startswith(".")
result["path_specified"] = bool(result["path"])
return cookielib.Cookie(**result)
def morsel_to_cookie(morsel):
"""Convert a Morsel object into a Cookie containing the one k/v pair."""
expires = None
if morsel["max-age"]:
try:
expires = int(time.time() + int(morsel["max-age"]))
except ValueError:
raise TypeError(f"max-age: {morsel['max-age']} must be integer")
elif morsel["expires"]:
time_template = "%a, %d-%b-%Y %H:%M:%S GMT"
expires = calendar.timegm(time.strptime(morsel["expires"], time_template))
return create_cookie(
comment=morsel["comment"],
comment_url=bool(morsel["comment"]),
discard=False,
domain=morsel["domain"],
expires=expires,
name=morsel.key,
path=morsel["path"],
port=None,
rest={"HttpOnly": morsel["httponly"]},
rfc2109=False,
secure=bool(morsel["secure"]),
value=morsel.value,
version=morsel["version"] or 0,
)
def cookiejar_from_dict(cookie_dict, cookiejar=None, overwrite=True):
"""Returns a CookieJar from a key/value dictionary.
:param cookie_dict: Dict of key/values to insert into CookieJar.
:param cookiejar: (optional) A cookiejar to add the cookies to.
:param overwrite: (optional) If False, will not replace cookies
already in the jar with new ones.
:rtype: CookieJar
"""
if cookiejar is None:
cookiejar = RequestsCookieJar()
if cookie_dict is not None:
names_from_jar = [cookie.name for cookie in cookiejar]
for name in cookie_dict:
if overwrite or (name not in names_from_jar):
cookiejar.set_cookie(create_cookie(name, cookie_dict[name]))
return cookiejar
def merge_cookies(cookiejar, cookies):
"""Add cookies to cookiejar and returns a merged CookieJar.
:param cookiejar: CookieJar object to add the cookies to.
:param cookies: Dictionary or CookieJar object to be added.
:rtype: CookieJar
"""
if not isinstance(cookiejar, cookielib.CookieJar):
raise ValueError("You can only merge into CookieJar")
if isinstance(cookies, dict):
cookiejar = cookiejar_from_dict(cookies, cookiejar=cookiejar, overwrite=False)
elif isinstance(cookies, cookielib.CookieJar):
try:
cookiejar.update(cookies)
except AttributeError:
for cookie_in_jar in cookies:
cookiejar.set_cookie(cookie_in_jar)
return cookiejar

View File

@ -1,141 +0,0 @@
"""
requests.exceptions
~~~~~~~~~~~~~~~~~~~
This module contains the set of Requests' exceptions.
"""
from urllib3.exceptions import HTTPError as BaseHTTPError
from .compat import JSONDecodeError as CompatJSONDecodeError
class RequestException(IOError):
"""There was an ambiguous exception that occurred while handling your
request.
"""
def __init__(self, *args, **kwargs):
"""Initialize RequestException with `request` and `response` objects."""
response = kwargs.pop("response", None)
self.response = response
self.request = kwargs.pop("request", None)
if response is not None and not self.request and hasattr(response, "request"):
self.request = self.response.request
super().__init__(*args, **kwargs)
class InvalidJSONError(RequestException):
"""A JSON error occurred."""
class JSONDecodeError(InvalidJSONError, CompatJSONDecodeError):
"""Couldn't decode the text into json"""
def __init__(self, *args, **kwargs):
"""
Construct the JSONDecodeError instance first with all
args. Then use it's args to construct the IOError so that
the json specific args aren't used as IOError specific args
and the error message from JSONDecodeError is preserved.
"""
CompatJSONDecodeError.__init__(self, *args)
InvalidJSONError.__init__(self, *self.args, **kwargs)
class HTTPError(RequestException):
"""An HTTP error occurred."""
class ConnectionError(RequestException):
"""A Connection error occurred."""
class ProxyError(ConnectionError):
"""A proxy error occurred."""
class SSLError(ConnectionError):
"""An SSL error occurred."""
class Timeout(RequestException):
"""The request timed out.
Catching this error will catch both
:exc:`~requests.exceptions.ConnectTimeout` and
:exc:`~requests.exceptions.ReadTimeout` errors.
"""
class ConnectTimeout(ConnectionError, Timeout):
"""The request timed out while trying to connect to the remote server.
Requests that produced this error are safe to retry.
"""
class ReadTimeout(Timeout):
"""The server did not send any data in the allotted amount of time."""
class URLRequired(RequestException):
"""A valid URL is required to make a request."""
class TooManyRedirects(RequestException):
"""Too many redirects."""
class MissingSchema(RequestException, ValueError):
"""The URL scheme (e.g. http or https) is missing."""
class InvalidSchema(RequestException, ValueError):
"""The URL scheme provided is either invalid or unsupported."""
class InvalidURL(RequestException, ValueError):
"""The URL provided was somehow invalid."""
class InvalidHeader(RequestException, ValueError):
"""The header value provided was somehow invalid."""
class InvalidProxyURL(InvalidURL):
"""The proxy URL provided is invalid."""
class ChunkedEncodingError(RequestException):
"""The server declared chunked encoding but sent an invalid chunk."""
class ContentDecodingError(RequestException, BaseHTTPError):
"""Failed to decode response content."""
class StreamConsumedError(RequestException, TypeError):
"""The content for this response was already consumed."""
class RetryError(RequestException):
"""Custom retries logic failed"""
class UnrewindableBodyError(RequestException):
"""Requests encountered an error when trying to rewind a body."""
# Warnings
class RequestsWarning(Warning):
"""Base warning for Requests."""
class FileModeWarning(RequestsWarning, DeprecationWarning):
"""A file was opened in text mode, but Requests determined its binary length."""
class RequestsDependencyWarning(RequestsWarning):
"""An imported dependency doesn't match the expected version range."""

View File

@ -1,134 +0,0 @@
"""Module containing bug report helper(s)."""
import json
import platform
import ssl
import sys
import idna
import urllib3
from . import __version__ as requests_version
try:
import charset_normalizer
except ImportError:
charset_normalizer = None
try:
import chardet
except ImportError:
chardet = None
try:
from urllib3.contrib import pyopenssl
except ImportError:
pyopenssl = None
OpenSSL = None
cryptography = None
else:
import cryptography
import OpenSSL
def _implementation():
"""Return a dict with the Python implementation and version.
Provide both the name and the version of the Python implementation
currently running. For example, on CPython 3.10.3 it will return
{'name': 'CPython', 'version': '3.10.3'}.
This function works best on CPython and PyPy: in particular, it probably
doesn't work for Jython or IronPython. Future investigation should be done
to work out the correct shape of the code for those platforms.
"""
implementation = platform.python_implementation()
if implementation == "CPython":
implementation_version = platform.python_version()
elif implementation == "PyPy":
implementation_version = "{}.{}.{}".format(
sys.pypy_version_info.major,
sys.pypy_version_info.minor,
sys.pypy_version_info.micro,
)
if sys.pypy_version_info.releaselevel != "final":
implementation_version = "".join(
[implementation_version, sys.pypy_version_info.releaselevel]
)
elif implementation == "Jython":
implementation_version = platform.python_version() # Complete Guess
elif implementation == "IronPython":
implementation_version = platform.python_version() # Complete Guess
else:
implementation_version = "Unknown"
return {"name": implementation, "version": implementation_version}
def info():
"""Generate information for a bug report."""
try:
platform_info = {
"system": platform.system(),
"release": platform.release(),
}
except OSError:
platform_info = {
"system": "Unknown",
"release": "Unknown",
}
implementation_info = _implementation()
urllib3_info = {"version": urllib3.__version__}
charset_normalizer_info = {"version": None}
chardet_info = {"version": None}
if charset_normalizer:
charset_normalizer_info = {"version": charset_normalizer.__version__}
if chardet:
chardet_info = {"version": chardet.__version__}
pyopenssl_info = {
"version": None,
"openssl_version": "",
}
if OpenSSL:
pyopenssl_info = {
"version": OpenSSL.__version__,
"openssl_version": f"{OpenSSL.SSL.OPENSSL_VERSION_NUMBER:x}",
}
cryptography_info = {
"version": getattr(cryptography, "__version__", ""),
}
idna_info = {
"version": getattr(idna, "__version__", ""),
}
system_ssl = ssl.OPENSSL_VERSION_NUMBER
system_ssl_info = {"version": f"{system_ssl:x}" if system_ssl is not None else ""}
return {
"platform": platform_info,
"implementation": implementation_info,
"system_ssl": system_ssl_info,
"using_pyopenssl": pyopenssl is not None,
"using_charset_normalizer": chardet is None,
"pyOpenSSL": pyopenssl_info,
"urllib3": urllib3_info,
"chardet": chardet_info,
"charset_normalizer": charset_normalizer_info,
"cryptography": cryptography_info,
"idna": idna_info,
"requests": {
"version": requests_version,
},
}
def main():
"""Pretty-print the bug information as JSON."""
print(json.dumps(info(), sort_keys=True, indent=2))
if __name__ == "__main__":
main()

View File

@ -1,33 +0,0 @@
"""
requests.hooks
~~~~~~~~~~~~~~
This module provides the capabilities for the Requests hooks system.
Available hooks:
``response``:
The response generated from a Request.
"""
HOOKS = ["response"]
def default_hooks():
return {event: [] for event in HOOKS}
# TODO: response is the only one
def dispatch_hook(key, hooks, hook_data, **kwargs):
"""Dispatches a hook dictionary on a given piece of data."""
hooks = hooks or {}
hooks = hooks.get(key)
if hooks:
if hasattr(hooks, "__call__"):
hooks = [hooks]
for hook in hooks:
_hook_data = hook(hook_data, **kwargs)
if _hook_data is not None:
hook_data = _hook_data
return hook_data

File diff suppressed because it is too large Load Diff

View File

@ -1,28 +0,0 @@
import sys
try:
import chardet
except ImportError:
import warnings
import charset_normalizer as chardet
warnings.filterwarnings("ignore", "Trying to detect", module="charset_normalizer")
# This code exists for backwards compatibility reasons.
# I don't like it either. Just look the other way. :)
for package in ("urllib3", "idna"):
locals()[package] = __import__(package)
# This traversal is apparently necessary such that the identities are
# preserved (requests.packages.urllib3.* is urllib3.*)
for mod in list(sys.modules):
if mod == package or mod.startswith(f"{package}."):
sys.modules[f"requests.packages.{mod}"] = sys.modules[mod]
target = chardet.__name__
for mod in list(sys.modules):
if mod == target or mod.startswith(f"{target}."):
target = target.replace(target, "chardet")
sys.modules[f"requests.packages.{target}"] = sys.modules[mod]
# Kinda cool, though, right?

View File

@ -1,833 +0,0 @@
"""
requests.sessions
~~~~~~~~~~~~~~~~~
This module provides a Session object to manage and persist settings across
requests (cookies, auth, proxies).
"""
import os
import sys
import time
from collections import OrderedDict
from datetime import timedelta
from ._internal_utils import to_native_string
from .adapters import HTTPAdapter
from .auth import _basic_auth_str
from .compat import Mapping, cookielib, urljoin, urlparse
from .cookies import (
RequestsCookieJar,
cookiejar_from_dict,
extract_cookies_to_jar,
merge_cookies,
)
from .exceptions import (
ChunkedEncodingError,
ContentDecodingError,
InvalidSchema,
TooManyRedirects,
)
from .hooks import default_hooks, dispatch_hook
# formerly defined here, reexposed here for backward compatibility
from .models import ( # noqa: F401
DEFAULT_REDIRECT_LIMIT,
REDIRECT_STATI,
PreparedRequest,
Request,
)
from .status_codes import codes
from .structures import CaseInsensitiveDict
from .utils import ( # noqa: F401
DEFAULT_PORTS,
default_headers,
get_auth_from_url,
get_environ_proxies,
get_netrc_auth,
requote_uri,
resolve_proxies,
rewind_body,
should_bypass_proxies,
to_key_val_list,
)
# Preferred clock, based on which one is more accurate on a given system.
if sys.platform == "win32":
preferred_clock = time.perf_counter
else:
preferred_clock = time.time
def merge_setting(request_setting, session_setting, dict_class=OrderedDict):
"""Determines appropriate setting for a given request, taking into account
the explicit setting on that request, and the setting in the session. If a
setting is a dictionary, they will be merged together using `dict_class`
"""
if session_setting is None:
return request_setting
if request_setting is None:
return session_setting
# Bypass if not a dictionary (e.g. verify)
if not (
isinstance(session_setting, Mapping) and isinstance(request_setting, Mapping)
):
return request_setting
merged_setting = dict_class(to_key_val_list(session_setting))
merged_setting.update(to_key_val_list(request_setting))
# Remove keys that are set to None. Extract keys first to avoid altering
# the dictionary during iteration.
none_keys = [k for (k, v) in merged_setting.items() if v is None]
for key in none_keys:
del merged_setting[key]
return merged_setting
def merge_hooks(request_hooks, session_hooks, dict_class=OrderedDict):
"""Properly merges both requests and session hooks.
This is necessary because when request_hooks == {'response': []}, the
merge breaks Session hooks entirely.
"""
if session_hooks is None or session_hooks.get("response") == []:
return request_hooks
if request_hooks is None or request_hooks.get("response") == []:
return session_hooks
return merge_setting(request_hooks, session_hooks, dict_class)
class SessionRedirectMixin:
def get_redirect_target(self, resp):
"""Receives a Response. Returns a redirect URI or ``None``"""
# Due to the nature of how requests processes redirects this method will
# be called at least once upon the original response and at least twice
# on each subsequent redirect response (if any).
# If a custom mixin is used to handle this logic, it may be advantageous
# to cache the redirect location onto the response object as a private
# attribute.
if resp.is_redirect:
location = resp.headers["location"]
# Currently the underlying http module on py3 decode headers
# in latin1, but empirical evidence suggests that latin1 is very
# rarely used with non-ASCII characters in HTTP headers.
# It is more likely to get UTF8 header rather than latin1.
# This causes incorrect handling of UTF8 encoded location headers.
# To solve this, we re-encode the location in latin1.
location = location.encode("latin1")
return to_native_string(location, "utf8")
return None
def should_strip_auth(self, old_url, new_url):
"""Decide whether Authorization header should be removed when redirecting"""
old_parsed = urlparse(old_url)
new_parsed = urlparse(new_url)
if old_parsed.hostname != new_parsed.hostname:
return True
# Special case: allow http -> https redirect when using the standard
# ports. This isn't specified by RFC 7235, but is kept to avoid
# breaking backwards compatibility with older versions of requests
# that allowed any redirects on the same host.
if (
old_parsed.scheme == "http"
and old_parsed.port in (80, None)
and new_parsed.scheme == "https"
and new_parsed.port in (443, None)
):
return False
# Handle default port usage corresponding to scheme.
changed_port = old_parsed.port != new_parsed.port
changed_scheme = old_parsed.scheme != new_parsed.scheme
default_port = (DEFAULT_PORTS.get(old_parsed.scheme, None), None)
if (
not changed_scheme
and old_parsed.port in default_port
and new_parsed.port in default_port
):
return False
# Standard case: root URI must match
return changed_port or changed_scheme
def resolve_redirects(
self,
resp,
req,
stream=False,
timeout=None,
verify=True,
cert=None,
proxies=None,
yield_requests=False,
**adapter_kwargs,
):
"""Receives a Response. Returns a generator of Responses or Requests."""
hist = [] # keep track of history
url = self.get_redirect_target(resp)
previous_fragment = urlparse(req.url).fragment
while url:
prepared_request = req.copy()
# Update history and keep track of redirects.
# resp.history must ignore the original request in this loop
hist.append(resp)
resp.history = hist[1:]
try:
resp.content # Consume socket so it can be released
except (ChunkedEncodingError, ContentDecodingError, RuntimeError):
resp.raw.read(decode_content=False)
if len(resp.history) >= self.max_redirects:
raise TooManyRedirects(
f"Exceeded {self.max_redirects} redirects.", response=resp
)
# Release the connection back into the pool.
resp.close()
# Handle redirection without scheme (see: RFC 1808 Section 4)
if url.startswith("//"):
parsed_rurl = urlparse(resp.url)
url = ":".join([to_native_string(parsed_rurl.scheme), url])
# Normalize url case and attach previous fragment if needed (RFC 7231 7.1.2)
parsed = urlparse(url)
if parsed.fragment == "" and previous_fragment:
parsed = parsed._replace(fragment=previous_fragment)
elif parsed.fragment:
previous_fragment = parsed.fragment
url = parsed.geturl()
# Facilitate relative 'location' headers, as allowed by RFC 7231.
# (e.g. '/path/to/resource' instead of 'http://domain.tld/path/to/resource')
# Compliant with RFC3986, we percent encode the url.
if not parsed.netloc:
url = urljoin(resp.url, requote_uri(url))
else:
url = requote_uri(url)
prepared_request.url = to_native_string(url)
self.rebuild_method(prepared_request, resp)
# https://github.com/psf/requests/issues/1084
if resp.status_code not in (
codes.temporary_redirect,
codes.permanent_redirect,
):
# https://github.com/psf/requests/issues/3490
purged_headers = ("Content-Length", "Content-Type", "Transfer-Encoding")
for header in purged_headers:
prepared_request.headers.pop(header, None)
prepared_request.body = None
headers = prepared_request.headers
headers.pop("Cookie", None)
# Extract any cookies sent on the response to the cookiejar
# in the new request. Because we've mutated our copied prepared
# request, use the old one that we haven't yet touched.
extract_cookies_to_jar(prepared_request._cookies, req, resp.raw)
merge_cookies(prepared_request._cookies, self.cookies)
prepared_request.prepare_cookies(prepared_request._cookies)
# Rebuild auth and proxy information.
proxies = self.rebuild_proxies(prepared_request, proxies)
self.rebuild_auth(prepared_request, resp)
# A failed tell() sets `_body_position` to `object()`. This non-None
# value ensures `rewindable` will be True, allowing us to raise an
# UnrewindableBodyError, instead of hanging the connection.
rewindable = prepared_request._body_position is not None and (
"Content-Length" in headers or "Transfer-Encoding" in headers
)
# Attempt to rewind consumed file-like object.
if rewindable:
rewind_body(prepared_request)
# Override the original request.
req = prepared_request
if yield_requests:
yield req
else:
resp = self.send(
req,
stream=stream,
timeout=timeout,
verify=verify,
cert=cert,
proxies=proxies,
allow_redirects=False,
**adapter_kwargs,
)
extract_cookies_to_jar(self.cookies, prepared_request, resp.raw)
# extract redirect url, if any, for the next loop
url = self.get_redirect_target(resp)
yield resp
def rebuild_auth(self, prepared_request, response):
"""When being redirected we may want to strip authentication from the
request to avoid leaking credentials. This method intelligently removes
and reapplies authentication where possible to avoid credential loss.
"""
headers = prepared_request.headers
url = prepared_request.url
if "Authorization" in headers and self.should_strip_auth(
response.request.url, url
):
# If we get redirected to a new host, we should strip out any
# authentication headers.
del headers["Authorization"]
# .netrc might have more auth for us on our new host.
new_auth = get_netrc_auth(url) if self.trust_env else None
if new_auth is not None:
prepared_request.prepare_auth(new_auth)
def rebuild_proxies(self, prepared_request, proxies):
"""This method re-evaluates the proxy configuration by considering the
environment variables. If we are redirected to a URL covered by
NO_PROXY, we strip the proxy configuration. Otherwise, we set missing
proxy keys for this URL (in case they were stripped by a previous
redirect).
This method also replaces the Proxy-Authorization header where
necessary.
:rtype: dict
"""
headers = prepared_request.headers
scheme = urlparse(prepared_request.url).scheme
new_proxies = resolve_proxies(prepared_request, proxies, self.trust_env)
if "Proxy-Authorization" in headers:
del headers["Proxy-Authorization"]
try:
username, password = get_auth_from_url(new_proxies[scheme])
except KeyError:
username, password = None, None
# urllib3 handles proxy authorization for us in the standard adapter.
# Avoid appending this to TLS tunneled requests where it may be leaked.
if not scheme.startswith('https') and username and password:
headers["Proxy-Authorization"] = _basic_auth_str(username, password)
return new_proxies
def rebuild_method(self, prepared_request, response):
"""When being redirected we may want to change the method of the request
based on certain specs or browser behavior.
"""
method = prepared_request.method
# https://tools.ietf.org/html/rfc7231#section-6.4.4
if response.status_code == codes.see_other and method != "HEAD":
method = "GET"
# Do what the browsers do, despite standards...
# First, turn 302s into GETs.
if response.status_code == codes.found and method != "HEAD":
method = "GET"
# Second, if a POST is responded to with a 301, turn it into a GET.
# This bizarre behaviour is explained in Issue 1704.
if response.status_code == codes.moved and method == "POST":
method = "GET"
prepared_request.method = method
class Session(SessionRedirectMixin):
"""A Requests session.
Provides cookie persistence, connection-pooling, and configuration.
Basic Usage::
>>> import requests
>>> s = requests.Session()
>>> s.get('https://httpbin.org/get')
<Response [200]>
Or as a context manager::
>>> with requests.Session() as s:
... s.get('https://httpbin.org/get')
<Response [200]>
"""
__attrs__ = [
"headers",
"cookies",
"auth",
"proxies",
"hooks",
"params",
"verify",
"cert",
"adapters",
"stream",
"trust_env",
"max_redirects",
]
def __init__(self):
#: A case-insensitive dictionary of headers to be sent on each
#: :class:`Request <Request>` sent from this
#: :class:`Session <Session>`.
self.headers = default_headers()
#: Default Authentication tuple or object to attach to
#: :class:`Request <Request>`.
self.auth = None
#: Dictionary mapping protocol or protocol and host to the URL of the proxy
#: (e.g. {'http': 'foo.bar:3128', 'http://host.name': 'foo.bar:4012'}) to
#: be used on each :class:`Request <Request>`.
self.proxies = {}
#: Event-handling hooks.
self.hooks = default_hooks()
#: Dictionary of querystring data to attach to each
#: :class:`Request <Request>`. The dictionary values may be lists for
#: representing multivalued query parameters.
self.params = {}
#: Stream response content default.
self.stream = False
#: SSL Verification default.
#: Defaults to `True`, requiring requests to verify the TLS certificate at the
#: remote end.
#: If verify is set to `False`, requests will accept any TLS certificate
#: presented by the server, and will ignore hostname mismatches and/or
#: expired certificates, which will make your application vulnerable to
#: man-in-the-middle (MitM) attacks.
#: Only set this to `False` for testing.
self.verify = True
#: SSL client certificate default, if String, path to ssl client
#: cert file (.pem). If Tuple, ('cert', 'key') pair.
self.cert = None
#: Maximum number of redirects allowed. If the request exceeds this
#: limit, a :class:`TooManyRedirects` exception is raised.
#: This defaults to requests.models.DEFAULT_REDIRECT_LIMIT, which is
#: 30.
self.max_redirects = DEFAULT_REDIRECT_LIMIT
#: Trust environment settings for proxy configuration, default
#: authentication and similar.
self.trust_env = True
#: A CookieJar containing all currently outstanding cookies set on this
#: session. By default it is a
#: :class:`RequestsCookieJar <requests.cookies.RequestsCookieJar>`, but
#: may be any other ``cookielib.CookieJar`` compatible object.
self.cookies = cookiejar_from_dict({})
# Default connection adapters.
self.adapters = OrderedDict()
self.mount("https://", HTTPAdapter())
self.mount("http://", HTTPAdapter())
def __enter__(self):
return self
def __exit__(self, *args):
self.close()
def prepare_request(self, request):
"""Constructs a :class:`PreparedRequest <PreparedRequest>` for
transmission and returns it. The :class:`PreparedRequest` has settings
merged from the :class:`Request <Request>` instance and those of the
:class:`Session`.
:param request: :class:`Request` instance to prepare with this
session's settings.
:rtype: requests.PreparedRequest
"""
cookies = request.cookies or {}
# Bootstrap CookieJar.
if not isinstance(cookies, cookielib.CookieJar):
cookies = cookiejar_from_dict(cookies)
# Merge with session cookies
merged_cookies = merge_cookies(
merge_cookies(RequestsCookieJar(), self.cookies), cookies
)
# Set environment's basic authentication if not explicitly set.
auth = request.auth
if self.trust_env and not auth and not self.auth:
auth = get_netrc_auth(request.url)
p = PreparedRequest()
p.prepare(
method=request.method.upper(),
url=request.url,
files=request.files,
data=request.data,
json=request.json,
headers=merge_setting(
request.headers, self.headers, dict_class=CaseInsensitiveDict
),
params=merge_setting(request.params, self.params),
auth=merge_setting(auth, self.auth),
cookies=merged_cookies,
hooks=merge_hooks(request.hooks, self.hooks),
)
return p
def request(
self,
method,
url,
params=None,
data=None,
headers=None,
cookies=None,
files=None,
auth=None,
timeout=None,
allow_redirects=True,
proxies=None,
hooks=None,
stream=None,
verify=None,
cert=None,
json=None,
):
"""Constructs a :class:`Request <Request>`, prepares it and sends it.
Returns :class:`Response <Response>` object.
:param method: method for the new :class:`Request` object.
:param url: URL for the new :class:`Request` object.
:param params: (optional) Dictionary or bytes to be sent in the query
string for the :class:`Request`.
:param data: (optional) Dictionary, list of tuples, bytes, or file-like
object to send in the body of the :class:`Request`.
:param json: (optional) json to send in the body of the
:class:`Request`.
:param headers: (optional) Dictionary of HTTP Headers to send with the
:class:`Request`.
:param cookies: (optional) Dict or CookieJar object to send with the
:class:`Request`.
:param files: (optional) Dictionary of ``'filename': file-like-objects``
for multipart encoding upload.
:param auth: (optional) Auth tuple or callable to enable
Basic/Digest/Custom HTTP Auth.
:param timeout: (optional) How long to wait for the server to send
data before giving up, as a float, or a :ref:`(connect timeout,
read timeout) <timeouts>` tuple.
:type timeout: float or tuple
:param allow_redirects: (optional) Set to True by default.
:type allow_redirects: bool
:param proxies: (optional) Dictionary mapping protocol or protocol and
hostname to the URL of the proxy.
:param stream: (optional) whether to immediately download the response
content. Defaults to ``False``.
:param verify: (optional) Either a boolean, in which case it controls whether we verify
the server's TLS certificate, or a string, in which case it must be a path
to a CA bundle to use. Defaults to ``True``. When set to
``False``, requests will accept any TLS certificate presented by
the server, and will ignore hostname mismatches and/or expired
certificates, which will make your application vulnerable to
man-in-the-middle (MitM) attacks. Setting verify to ``False``
may be useful during local development or testing.
:param cert: (optional) if String, path to ssl client cert file (.pem).
If Tuple, ('cert', 'key') pair.
:rtype: requests.Response
"""
# Create the Request.
req = Request(
method=method.upper(),
url=url,
headers=headers,
files=files,
data=data or {},
json=json,
params=params or {},
auth=auth,
cookies=cookies,
hooks=hooks,
)
prep = self.prepare_request(req)
proxies = proxies or {}
settings = self.merge_environment_settings(
prep.url, proxies, stream, verify, cert
)
# Send the request.
send_kwargs = {
"timeout": timeout,
"allow_redirects": allow_redirects,
}
send_kwargs.update(settings)
resp = self.send(prep, **send_kwargs)
return resp
def get(self, url, **kwargs):
r"""Sends a GET request. Returns :class:`Response` object.
:param url: URL for the new :class:`Request` object.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:rtype: requests.Response
"""
kwargs.setdefault("allow_redirects", True)
return self.request("GET", url, **kwargs)
def options(self, url, **kwargs):
r"""Sends a OPTIONS request. Returns :class:`Response` object.
:param url: URL for the new :class:`Request` object.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:rtype: requests.Response
"""
kwargs.setdefault("allow_redirects", True)
return self.request("OPTIONS", url, **kwargs)
def head(self, url, **kwargs):
r"""Sends a HEAD request. Returns :class:`Response` object.
:param url: URL for the new :class:`Request` object.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:rtype: requests.Response
"""
kwargs.setdefault("allow_redirects", False)
return self.request("HEAD", url, **kwargs)
def post(self, url, data=None, json=None, **kwargs):
r"""Sends a POST request. Returns :class:`Response` object.
:param url: URL for the new :class:`Request` object.
:param data: (optional) Dictionary, list of tuples, bytes, or file-like
object to send in the body of the :class:`Request`.
:param json: (optional) json to send in the body of the :class:`Request`.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:rtype: requests.Response
"""
return self.request("POST", url, data=data, json=json, **kwargs)
def put(self, url, data=None, **kwargs):
r"""Sends a PUT request. Returns :class:`Response` object.
:param url: URL for the new :class:`Request` object.
:param data: (optional) Dictionary, list of tuples, bytes, or file-like
object to send in the body of the :class:`Request`.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:rtype: requests.Response
"""
return self.request("PUT", url, data=data, **kwargs)
def patch(self, url, data=None, **kwargs):
r"""Sends a PATCH request. Returns :class:`Response` object.
:param url: URL for the new :class:`Request` object.
:param data: (optional) Dictionary, list of tuples, bytes, or file-like
object to send in the body of the :class:`Request`.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:rtype: requests.Response
"""
return self.request("PATCH", url, data=data, **kwargs)
def delete(self, url, **kwargs):
r"""Sends a DELETE request. Returns :class:`Response` object.
:param url: URL for the new :class:`Request` object.
:param \*\*kwargs: Optional arguments that ``request`` takes.
:rtype: requests.Response
"""
return self.request("DELETE", url, **kwargs)
def send(self, request, **kwargs):
"""Send a given PreparedRequest.
:rtype: requests.Response
"""
# Set defaults that the hooks can utilize to ensure they always have
# the correct parameters to reproduce the previous request.
kwargs.setdefault("stream", self.stream)
kwargs.setdefault("verify", self.verify)
kwargs.setdefault("cert", self.cert)
if "proxies" not in kwargs:
kwargs["proxies"] = resolve_proxies(request, self.proxies, self.trust_env)
# It's possible that users might accidentally send a Request object.
# Guard against that specific failure case.
if isinstance(request, Request):
raise ValueError("You can only send PreparedRequests.")
# Set up variables needed for resolve_redirects and dispatching of hooks
allow_redirects = kwargs.pop("allow_redirects", True)
stream = kwargs.get("stream")
hooks = request.hooks
# Get the appropriate adapter to use
adapter = self.get_adapter(url=request.url)
# Start time (approximately) of the request
start = preferred_clock()
# Send the request
r = adapter.send(request, **kwargs)
# Total elapsed time of the request (approximately)
elapsed = preferred_clock() - start
r.elapsed = timedelta(seconds=elapsed)
# Response manipulation hooks
r = dispatch_hook("response", hooks, r, **kwargs)
# Persist cookies
if r.history:
# If the hooks create history then we want those cookies too
for resp in r.history:
extract_cookies_to_jar(self.cookies, resp.request, resp.raw)
extract_cookies_to_jar(self.cookies, request, r.raw)
# Resolve redirects if allowed.
if allow_redirects:
# Redirect resolving generator.
gen = self.resolve_redirects(r, request, **kwargs)
history = [resp for resp in gen]
else:
history = []
# Shuffle things around if there's history.
if history:
# Insert the first (original) request at the start
history.insert(0, r)
# Get the last request made
r = history.pop()
r.history = history
# If redirects aren't being followed, store the response on the Request for Response.next().
if not allow_redirects:
try:
r._next = next(
self.resolve_redirects(r, request, yield_requests=True, **kwargs)
)
except StopIteration:
pass
if not stream:
r.content
return r
def merge_environment_settings(self, url, proxies, stream, verify, cert):
"""
Check the environment and merge it with some settings.
:rtype: dict
"""
# Gather clues from the surrounding environment.
if self.trust_env:
# Set environment's proxies.
no_proxy = proxies.get("no_proxy") if proxies is not None else None
env_proxies = get_environ_proxies(url, no_proxy=no_proxy)
for (k, v) in env_proxies.items():
proxies.setdefault(k, v)
# Look for requests environment configuration
# and be compatible with cURL.
if verify is True or verify is None:
verify = (
os.environ.get("REQUESTS_CA_BUNDLE")
or os.environ.get("CURL_CA_BUNDLE")
or verify
)
# Merge all the kwargs.
proxies = merge_setting(proxies, self.proxies)
stream = merge_setting(stream, self.stream)
verify = merge_setting(verify, self.verify)
cert = merge_setting(cert, self.cert)
return {"proxies": proxies, "stream": stream, "verify": verify, "cert": cert}
def get_adapter(self, url):
"""
Returns the appropriate connection adapter for the given URL.
:rtype: requests.adapters.BaseAdapter
"""
for (prefix, adapter) in self.adapters.items():
if url.lower().startswith(prefix.lower()):
return adapter
# Nothing matches :-/
raise InvalidSchema(f"No connection adapters were found for {url!r}")
def close(self):
"""Closes all adapters and as such the session"""
for v in self.adapters.values():
v.close()
def mount(self, prefix, adapter):
"""Registers a connection adapter to a prefix.
Adapters are sorted in descending order by prefix length.
"""
self.adapters[prefix] = adapter
keys_to_move = [k for k in self.adapters if len(k) < len(prefix)]
for key in keys_to_move:
self.adapters[key] = self.adapters.pop(key)
def __getstate__(self):
state = {attr: getattr(self, attr, None) for attr in self.__attrs__}
return state
def __setstate__(self, state):
for attr, value in state.items():
setattr(self, attr, value)
def session():
"""
Returns a :class:`Session` for context-management.
.. deprecated:: 1.0.0
This method has been deprecated since version 1.0.0 and is only kept for
backwards compatibility. New code should use :class:`~requests.sessions.Session`
to create a session. This may be removed at a future date.
:rtype: Session
"""
return Session()

View File

@ -1,128 +0,0 @@
r"""
The ``codes`` object defines a mapping from common names for HTTP statuses
to their numerical codes, accessible either as attributes or as dictionary
items.
Example::
>>> import requests
>>> requests.codes['temporary_redirect']
307
>>> requests.codes.teapot
418
>>> requests.codes['\o/']
200
Some codes have multiple names, and both upper- and lower-case versions of
the names are allowed. For example, ``codes.ok``, ``codes.OK``, and
``codes.okay`` all correspond to the HTTP status code 200.
"""
from .structures import LookupDict
_codes = {
# Informational.
100: ("continue",),
101: ("switching_protocols",),
102: ("processing",),
103: ("checkpoint",),
122: ("uri_too_long", "request_uri_too_long"),
200: ("ok", "okay", "all_ok", "all_okay", "all_good", "\\o/", ""),
201: ("created",),
202: ("accepted",),
203: ("non_authoritative_info", "non_authoritative_information"),
204: ("no_content",),
205: ("reset_content", "reset"),
206: ("partial_content", "partial"),
207: ("multi_status", "multiple_status", "multi_stati", "multiple_stati"),
208: ("already_reported",),
226: ("im_used",),
# Redirection.
300: ("multiple_choices",),
301: ("moved_permanently", "moved", "\\o-"),
302: ("found",),
303: ("see_other", "other"),
304: ("not_modified",),
305: ("use_proxy",),
306: ("switch_proxy",),
307: ("temporary_redirect", "temporary_moved", "temporary"),
308: (
"permanent_redirect",
"resume_incomplete",
"resume",
), # "resume" and "resume_incomplete" to be removed in 3.0
# Client Error.
400: ("bad_request", "bad"),
401: ("unauthorized",),
402: ("payment_required", "payment"),
403: ("forbidden",),
404: ("not_found", "-o-"),
405: ("method_not_allowed", "not_allowed"),
406: ("not_acceptable",),
407: ("proxy_authentication_required", "proxy_auth", "proxy_authentication"),
408: ("request_timeout", "timeout"),
409: ("conflict",),
410: ("gone",),
411: ("length_required",),
412: ("precondition_failed", "precondition"),
413: ("request_entity_too_large",),
414: ("request_uri_too_large",),
415: ("unsupported_media_type", "unsupported_media", "media_type"),
416: (
"requested_range_not_satisfiable",
"requested_range",
"range_not_satisfiable",
),
417: ("expectation_failed",),
418: ("im_a_teapot", "teapot", "i_am_a_teapot"),
421: ("misdirected_request",),
422: ("unprocessable_entity", "unprocessable"),
423: ("locked",),
424: ("failed_dependency", "dependency"),
425: ("unordered_collection", "unordered"),
426: ("upgrade_required", "upgrade"),
428: ("precondition_required", "precondition"),
429: ("too_many_requests", "too_many"),
431: ("header_fields_too_large", "fields_too_large"),
444: ("no_response", "none"),
449: ("retry_with", "retry"),
450: ("blocked_by_windows_parental_controls", "parental_controls"),
451: ("unavailable_for_legal_reasons", "legal_reasons"),
499: ("client_closed_request",),
# Server Error.
500: ("internal_server_error", "server_error", "/o\\", ""),
501: ("not_implemented",),
502: ("bad_gateway",),
503: ("service_unavailable", "unavailable"),
504: ("gateway_timeout",),
505: ("http_version_not_supported", "http_version"),
506: ("variant_also_negotiates",),
507: ("insufficient_storage",),
509: ("bandwidth_limit_exceeded", "bandwidth"),
510: ("not_extended",),
511: ("network_authentication_required", "network_auth", "network_authentication"),
}
codes = LookupDict(name="status_codes")
def _init():
for code, titles in _codes.items():
for title in titles:
setattr(codes, title, code)
if not title.startswith(("\\", "/")):
setattr(codes, title.upper(), code)
def doc(code):
names = ", ".join(f"``{n}``" for n in _codes[code])
return "* %d: %s" % (code, names)
global __doc__
__doc__ = (
__doc__ + "\n" + "\n".join(doc(code) for code in sorted(_codes))
if __doc__ is not None
else None
)
_init()

View File

@ -1,99 +0,0 @@
"""
requests.structures
~~~~~~~~~~~~~~~~~~~
Data structures that power Requests.
"""
from collections import OrderedDict
from .compat import Mapping, MutableMapping
class CaseInsensitiveDict(MutableMapping):
"""A case-insensitive ``dict``-like object.
Implements all methods and operations of
``MutableMapping`` as well as dict's ``copy``. Also
provides ``lower_items``.
All keys are expected to be strings. The structure remembers the
case of the last key to be set, and ``iter(instance)``,
``keys()``, ``items()``, ``iterkeys()``, and ``iteritems()``
will contain case-sensitive keys. However, querying and contains
testing is case insensitive::
cid = CaseInsensitiveDict()
cid['Accept'] = 'application/json'
cid['aCCEPT'] == 'application/json' # True
list(cid) == ['Accept'] # True
For example, ``headers['content-encoding']`` will return the
value of a ``'Content-Encoding'`` response header, regardless
of how the header name was originally stored.
If the constructor, ``.update``, or equality comparison
operations are given keys that have equal ``.lower()``s, the
behavior is undefined.
"""
def __init__(self, data=None, **kwargs):
self._store = OrderedDict()
if data is None:
data = {}
self.update(data, **kwargs)
def __setitem__(self, key, value):
# Use the lowercased key for lookups, but store the actual
# key alongside the value.
self._store[key.lower()] = (key, value)
def __getitem__(self, key):
return self._store[key.lower()][1]
def __delitem__(self, key):
del self._store[key.lower()]
def __iter__(self):
return (casedkey for casedkey, mappedvalue in self._store.values())
def __len__(self):
return len(self._store)
def lower_items(self):
"""Like iteritems(), but with all lowercase keys."""
return ((lowerkey, keyval[1]) for (lowerkey, keyval) in self._store.items())
def __eq__(self, other):
if isinstance(other, Mapping):
other = CaseInsensitiveDict(other)
else:
return NotImplemented
# Compare insensitively
return dict(self.lower_items()) == dict(other.lower_items())
# Copy is required
def copy(self):
return CaseInsensitiveDict(self._store.values())
def __repr__(self):
return str(dict(self.items()))
class LookupDict(dict):
"""Dictionary lookup object."""
def __init__(self, name=None):
self.name = name
super().__init__()
def __repr__(self):
return f"<lookup '{self.name}'>"
def __getitem__(self, key):
# We allow fall-through here, so values default to None
return self.__dict__.get(key, None)
def get(self, key, default=None):
return self.__dict__.get(key, default)

File diff suppressed because it is too large Load Diff

View File

@ -1,211 +0,0 @@
"""
Python HTTP library with thread-safe connection pooling, file post support, user friendly, and more
"""
from __future__ import annotations
# Set default logging handler to avoid "No handler found" warnings.
import logging
import sys
import typing
import warnings
from logging import NullHandler
from . import exceptions
from ._base_connection import _TYPE_BODY
from ._collections import HTTPHeaderDict
from ._version import __version__
from .connectionpool import HTTPConnectionPool, HTTPSConnectionPool, connection_from_url
from .filepost import _TYPE_FIELDS, encode_multipart_formdata
from .poolmanager import PoolManager, ProxyManager, proxy_from_url
from .response import BaseHTTPResponse, HTTPResponse
from .util.request import make_headers
from .util.retry import Retry
from .util.timeout import Timeout
# Ensure that Python is compiled with OpenSSL 1.1.1+
# If the 'ssl' module isn't available at all that's
# fine, we only care if the module is available.
try:
import ssl
except ImportError:
pass
else:
if not ssl.OPENSSL_VERSION.startswith("OpenSSL "): # Defensive:
warnings.warn(
"urllib3 v2 only supports OpenSSL 1.1.1+, currently "
f"the 'ssl' module is compiled with {ssl.OPENSSL_VERSION!r}. "
"See: https://github.com/urllib3/urllib3/issues/3020",
exceptions.NotOpenSSLWarning,
)
elif ssl.OPENSSL_VERSION_INFO < (1, 1, 1): # Defensive:
raise ImportError(
"urllib3 v2 only supports OpenSSL 1.1.1+, currently "
f"the 'ssl' module is compiled with {ssl.OPENSSL_VERSION!r}. "
"See: https://github.com/urllib3/urllib3/issues/2168"
)
__author__ = "Andrey Petrov (andrey.petrov@shazow.net)"
__license__ = "MIT"
__version__ = __version__
__all__ = (
"HTTPConnectionPool",
"HTTPHeaderDict",
"HTTPSConnectionPool",
"PoolManager",
"ProxyManager",
"HTTPResponse",
"Retry",
"Timeout",
"add_stderr_logger",
"connection_from_url",
"disable_warnings",
"encode_multipart_formdata",
"make_headers",
"proxy_from_url",
"request",
"BaseHTTPResponse",
)
logging.getLogger(__name__).addHandler(NullHandler())
def add_stderr_logger(
level: int = logging.DEBUG,
) -> logging.StreamHandler[typing.TextIO]:
"""
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)
# InsecurePlatformWarning's don't vary between requests, so we keep it default.
warnings.simplefilter("default", exceptions.InsecurePlatformWarning, append=True)
def disable_warnings(category: type[Warning] = exceptions.HTTPWarning) -> None:
"""
Helper for quickly disabling all urllib3 warnings.
"""
warnings.simplefilter("ignore", category)
_DEFAULT_POOL = PoolManager()
def request(
method: str,
url: str,
*,
body: _TYPE_BODY | None = None,
fields: _TYPE_FIELDS | None = None,
headers: typing.Mapping[str, str] | None = None,
preload_content: bool | None = True,
decode_content: bool | None = True,
redirect: bool | None = True,
retries: Retry | bool | int | None = None,
timeout: Timeout | float | int | None = 3,
json: typing.Any | None = None,
) -> BaseHTTPResponse:
"""
A convenience, top-level request method. It uses a module-global ``PoolManager`` instance.
Therefore, its side effects could be shared across dependencies relying on it.
To avoid side effects create a new ``PoolManager`` instance and use it instead.
The method does not accept low-level ``**urlopen_kw`` keyword arguments.
: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 fields:
Data to encode and send in the request body.
:param headers:
Dictionary of custom headers to send, such as User-Agent,
If-None-Match, etc.
: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 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 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 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 json:
Data to encode and send as JSON with UTF-encoded in the request body.
The ``"Content-Type"`` header will be set to ``"application/json"``
unless specified otherwise.
"""
return _DEFAULT_POOL.request(
method,
url,
body=body,
fields=fields,
headers=headers,
preload_content=preload_content,
decode_content=decode_content,
redirect=redirect,
retries=retries,
timeout=timeout,
json=json,
)
if sys.platform == "emscripten":
from .contrib.emscripten import inject_into_urllib3 # noqa: 401
inject_into_urllib3()

View File

@ -1,172 +0,0 @@
from __future__ import annotations
import typing
from .util.connection import _TYPE_SOCKET_OPTIONS
from .util.timeout import _DEFAULT_TIMEOUT, _TYPE_TIMEOUT
from .util.url import Url
_TYPE_BODY = typing.Union[bytes, typing.IO[typing.Any], typing.Iterable[bytes], str]
class ProxyConfig(typing.NamedTuple):
ssl_context: ssl.SSLContext | None
use_forwarding_for_https: bool
assert_hostname: None | str | Literal[False]
assert_fingerprint: str | None
class _ResponseOptions(typing.NamedTuple):
# TODO: Remove this in favor of a better
# HTTP request/response lifecycle tracking.
request_method: str
request_url: str
preload_content: bool
decode_content: bool
enforce_content_length: bool
if typing.TYPE_CHECKING:
import ssl
from typing import Literal, Protocol
from .response import BaseHTTPResponse
class BaseHTTPConnection(Protocol):
default_port: typing.ClassVar[int]
default_socket_options: typing.ClassVar[_TYPE_SOCKET_OPTIONS]
host: str
port: int
timeout: None | (
float
) # Instance doesn't store _DEFAULT_TIMEOUT, must be resolved.
blocksize: int
source_address: tuple[str, int] | None
socket_options: _TYPE_SOCKET_OPTIONS | None
proxy: Url | None
proxy_config: ProxyConfig | None
is_verified: bool
proxy_is_verified: bool | None
def __init__(
self,
host: str,
port: int | None = None,
*,
timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT,
source_address: tuple[str, int] | None = None,
blocksize: int = 8192,
socket_options: _TYPE_SOCKET_OPTIONS | None = ...,
proxy: Url | None = None,
proxy_config: ProxyConfig | None = None,
) -> None:
...
def set_tunnel(
self,
host: str,
port: int | None = None,
headers: typing.Mapping[str, str] | None = None,
scheme: str = "http",
) -> None:
...
def connect(self) -> None:
...
def request(
self,
method: str,
url: str,
body: _TYPE_BODY | None = None,
headers: typing.Mapping[str, str] | None = None,
# We know *at least* botocore is depending on the order of the
# first 3 parameters so to be safe we only mark the later ones
# as keyword-only to ensure we have space to extend.
*,
chunked: bool = False,
preload_content: bool = True,
decode_content: bool = True,
enforce_content_length: bool = True,
) -> None:
...
def getresponse(self) -> BaseHTTPResponse:
...
def close(self) -> None:
...
@property
def is_closed(self) -> bool:
"""Whether the connection either is brand new or has been previously closed.
If this property is True then both ``is_connected`` and ``has_connected_to_proxy``
properties must be False.
"""
@property
def is_connected(self) -> bool:
"""Whether the connection is actively connected to any origin (proxy or target)"""
@property
def has_connected_to_proxy(self) -> bool:
"""Whether the connection has successfully connected to its proxy.
This returns False if no proxy is in use. Used to determine whether
errors are coming from the proxy layer or from tunnelling to the target origin.
"""
class BaseHTTPSConnection(BaseHTTPConnection, Protocol):
default_port: typing.ClassVar[int]
default_socket_options: typing.ClassVar[_TYPE_SOCKET_OPTIONS]
# Certificate verification methods
cert_reqs: int | str | None
assert_hostname: None | str | Literal[False]
assert_fingerprint: str | None
ssl_context: ssl.SSLContext | None
# Trusted CAs
ca_certs: str | None
ca_cert_dir: str | None
ca_cert_data: None | str | bytes
# TLS version
ssl_minimum_version: int | None
ssl_maximum_version: int | None
ssl_version: int | str | None # Deprecated
# Client certificates
cert_file: str | None
key_file: str | None
key_password: str | None
def __init__(
self,
host: str,
port: int | None = None,
*,
timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT,
source_address: tuple[str, int] | None = None,
blocksize: int = 16384,
socket_options: _TYPE_SOCKET_OPTIONS | None = ...,
proxy: Url | None = None,
proxy_config: ProxyConfig | None = None,
cert_reqs: int | str | None = None,
assert_hostname: None | str | Literal[False] = None,
assert_fingerprint: str | None = None,
server_hostname: str | None = None,
ssl_context: ssl.SSLContext | None = None,
ca_certs: str | None = None,
ca_cert_dir: str | None = None,
ca_cert_data: None | str | bytes = None,
ssl_minimum_version: int | None = None,
ssl_maximum_version: int | None = None,
ssl_version: int | str | None = None, # Deprecated
cert_file: str | None = None,
key_file: str | None = None,
key_password: str | None = None,
) -> None:
...

View File

@ -1,483 +0,0 @@
from __future__ import annotations
import typing
from collections import OrderedDict
from enum import Enum, auto
from threading import RLock
if typing.TYPE_CHECKING:
# We can only import Protocol if TYPE_CHECKING because it's a development
# dependency, and is not available at runtime.
from typing import Protocol
from typing_extensions import Self
class HasGettableStringKeys(Protocol):
def keys(self) -> typing.Iterator[str]:
...
def __getitem__(self, key: str) -> str:
...
__all__ = ["RecentlyUsedContainer", "HTTPHeaderDict"]
# Key type
_KT = typing.TypeVar("_KT")
# Value type
_VT = typing.TypeVar("_VT")
# Default type
_DT = typing.TypeVar("_DT")
ValidHTTPHeaderSource = typing.Union[
"HTTPHeaderDict",
typing.Mapping[str, str],
typing.Iterable[typing.Tuple[str, str]],
"HasGettableStringKeys",
]
class _Sentinel(Enum):
not_passed = auto()
def ensure_can_construct_http_header_dict(
potential: object,
) -> ValidHTTPHeaderSource | None:
if isinstance(potential, HTTPHeaderDict):
return potential
elif isinstance(potential, typing.Mapping):
# Full runtime checking of the contents of a Mapping is expensive, so for the
# purposes of typechecking, we assume that any Mapping is the right shape.
return typing.cast(typing.Mapping[str, str], potential)
elif isinstance(potential, typing.Iterable):
# Similarly to Mapping, full runtime checking of the contents of an Iterable is
# expensive, so for the purposes of typechecking, we assume that any Iterable
# is the right shape.
return typing.cast(typing.Iterable[typing.Tuple[str, str]], potential)
elif hasattr(potential, "keys") and hasattr(potential, "__getitem__"):
return typing.cast("HasGettableStringKeys", potential)
else:
return None
class RecentlyUsedContainer(typing.Generic[_KT, _VT], typing.MutableMapping[_KT, _VT]):
"""
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
"""
_container: typing.OrderedDict[_KT, _VT]
_maxsize: int
dispose_func: typing.Callable[[_VT], None] | None
lock: RLock
def __init__(
self,
maxsize: int = 10,
dispose_func: typing.Callable[[_VT], None] | None = None,
) -> None:
super().__init__()
self._maxsize = maxsize
self.dispose_func = dispose_func
self._container = OrderedDict()
self.lock = RLock()
def __getitem__(self, key: _KT) -> _VT:
# 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: _KT, value: _VT) -> None:
evicted_item = None
with self.lock:
# Possibly evict the existing value of 'key'
try:
# If the key exists, we'll overwrite it, which won't change the
# size of the pool. Because accessing a key should move it to
# the end of the eviction line, we pop it out first.
evicted_item = key, self._container.pop(key)
self._container[key] = value
except KeyError:
# When the key does not exist, we insert the value first so that
# evicting works in all cases, including when self._maxsize is 0
self._container[key] = value
if len(self._container) > self._maxsize:
# If we didn't evict an existing value, and we've hit our maximum
# size, then we have to evict the least recently used item from
# the beginning of the container.
evicted_item = self._container.popitem(last=False)
# After releasing the lock on the pool, dispose of any evicted value.
if evicted_item is not None and self.dispose_func:
_, evicted_value = evicted_item
self.dispose_func(evicted_value)
def __delitem__(self, key: _KT) -> None:
with self.lock:
value = self._container.pop(key)
if self.dispose_func:
self.dispose_func(value)
def __len__(self) -> int:
with self.lock:
return len(self._container)
def __iter__(self) -> typing.NoReturn:
raise NotImplementedError(
"Iteration over this class is unlikely to be threadsafe."
)
def clear(self) -> None:
with self.lock:
# Copy pointers to all values, then wipe the mapping
values = list(self._container.values())
self._container.clear()
if self.dispose_func:
for value in values:
self.dispose_func(value)
def keys(self) -> set[_KT]: # type: ignore[override]
with self.lock:
return set(self._container.keys())
class HTTPHeaderDictItemView(typing.Set[typing.Tuple[str, str]]):
"""
HTTPHeaderDict is unusual for a Mapping[str, str] in that it has two modes of
address.
If we directly try to get an item with a particular name, we will get a string
back that is the concatenated version of all the values:
>>> d['X-Header-Name']
'Value1, Value2, Value3'
However, if we iterate over an HTTPHeaderDict's items, we will optionally combine
these values based on whether combine=True was called when building up the dictionary
>>> d = HTTPHeaderDict({"A": "1", "B": "foo"})
>>> d.add("A", "2", combine=True)
>>> d.add("B", "bar")
>>> list(d.items())
[
('A', '1, 2'),
('B', 'foo'),
('B', 'bar'),
]
This class conforms to the interface required by the MutableMapping ABC while
also giving us the nonstandard iteration behavior we want; items with duplicate
keys, ordered by time of first insertion.
"""
_headers: HTTPHeaderDict
def __init__(self, headers: HTTPHeaderDict) -> None:
self._headers = headers
def __len__(self) -> int:
return len(list(self._headers.iteritems()))
def __iter__(self) -> typing.Iterator[tuple[str, str]]:
return self._headers.iteritems()
def __contains__(self, item: object) -> bool:
if isinstance(item, tuple) and len(item) == 2:
passed_key, passed_val = item
if isinstance(passed_key, str) and isinstance(passed_val, str):
return self._headers._has_value_for_header(passed_key, passed_val)
return False
class HTTPHeaderDict(typing.MutableMapping[str, str]):
"""
: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'
"""
_container: typing.MutableMapping[str, list[str]]
def __init__(self, headers: ValidHTTPHeaderSource | None = None, **kwargs: str):
super().__init__()
self._container = {} # 'dict' is insert-ordered
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: str, val: str) -> None:
# avoid a bytes/str comparison by decoding before httplib
if isinstance(key, bytes):
key = key.decode("latin-1")
self._container[key.lower()] = [key, val]
def __getitem__(self, key: str) -> str:
val = self._container[key.lower()]
return ", ".join(val[1:])
def __delitem__(self, key: str) -> None:
del self._container[key.lower()]
def __contains__(self, key: object) -> bool:
if isinstance(key, str):
return key.lower() in self._container
return False
def setdefault(self, key: str, default: str = "") -> str:
return super().setdefault(key, default)
def __eq__(self, other: object) -> bool:
maybe_constructable = ensure_can_construct_http_header_dict(other)
if maybe_constructable is None:
return False
else:
other_as_http_header_dict = type(self)(maybe_constructable)
return {k.lower(): v for k, v in self.itermerged()} == {
k.lower(): v for k, v in other_as_http_header_dict.itermerged()
}
def __ne__(self, other: object) -> bool:
return not self.__eq__(other)
def __len__(self) -> int:
return len(self._container)
def __iter__(self) -> typing.Iterator[str]:
# Only provide the originally cased names
for vals in self._container.values():
yield vals[0]
def discard(self, key: str) -> None:
try:
del self[key]
except KeyError:
pass
def add(self, key: str, val: str, *, combine: bool = False) -> None:
"""Adds a (name, value) pair, doesn't overwrite the value if it already
exists.
If this is called with combine=True, instead of adding a new header value
as a distinct item during iteration, this will instead append the value to
any existing header value with a comma. If no existing header value exists
for the key, then the value will simply be added, ignoring the combine parameter.
>>> headers = HTTPHeaderDict(foo='bar')
>>> headers.add('Foo', 'baz')
>>> headers['foo']
'bar, baz'
>>> list(headers.items())
[('foo', 'bar'), ('foo', 'baz')]
>>> headers.add('foo', 'quz', combine=True)
>>> list(headers.items())
[('foo', 'bar, baz, quz')]
"""
# avoid a bytes/str comparison by decoding before httplib
if isinstance(key, bytes):
key = key.decode("latin-1")
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:
# if there are values here, then there is at least the initial
# key/value pair
assert len(vals) >= 2
if combine:
vals[-1] = vals[-1] + ", " + val
else:
vals.append(val)
def extend(self, *args: ValidHTTPHeaderSource, **kwargs: str) -> None:
"""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(
f"extend() takes at most 1 positional arguments ({len(args)} given)"
)
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, typing.Mapping):
for key, val in other.items():
self.add(key, val)
elif isinstance(other, typing.Iterable):
other = typing.cast(typing.Iterable[typing.Tuple[str, str]], other)
for key, value in other:
self.add(key, value)
elif hasattr(other, "keys") and hasattr(other, "__getitem__"):
# THIS IS NOT A TYPESAFE BRANCH
# In this branch, the object has a `keys` attr but is not a Mapping or any of
# the other types indicated in the method signature. We do some stuff with
# it as though it partially implements the Mapping interface, but we're not
# doing that stuff safely AT ALL.
for key in other.keys():
self.add(key, other[key])
for key, value in kwargs.items():
self.add(key, value)
@typing.overload
def getlist(self, key: str) -> list[str]:
...
@typing.overload
def getlist(self, key: str, default: _DT) -> list[str] | _DT:
...
def getlist(
self, key: str, default: _Sentinel | _DT = _Sentinel.not_passed
) -> list[str] | _DT:
"""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 _Sentinel.not_passed:
# _DT is unbound; empty list is instance of List[str]
return []
# _DT is bound; default is instance of _DT
return default
else:
# _DT may or may not be bound; vals[1:] is instance of List[str], which
# meets our external interface requirement of `Union[List[str], _DT]`.
return vals[1:]
def _prepare_for_method_change(self) -> Self:
"""
Remove content-specific header fields before changing the request
method to GET or HEAD according to RFC 9110, Section 15.4.
"""
content_specific_headers = [
"Content-Encoding",
"Content-Language",
"Content-Location",
"Content-Type",
"Content-Length",
"Digest",
"Last-Modified",
]
for header in content_specific_headers:
self.discard(header)
return self
# Backwards compatibility for httplib
getheaders = getlist
getallmatchingheaders = getlist
iget = getlist
# Backwards compatibility for http.cookiejar
get_all = getlist
def __repr__(self) -> str:
return f"{type(self).__name__}({dict(self.itermerged())})"
def _copy_from(self, other: HTTPHeaderDict) -> None:
for key in other:
val = other.getlist(key)
self._container[key.lower()] = [key, *val]
def copy(self) -> HTTPHeaderDict:
clone = type(self)()
clone._copy_from(self)
return clone
def iteritems(self) -> typing.Iterator[tuple[str, str]]:
"""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) -> typing.Iterator[tuple[str, str]]:
"""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) -> HTTPHeaderDictItemView: # type: ignore[override]
return HTTPHeaderDictItemView(self)
def _has_value_for_header(self, header_name: str, potential_value: str) -> bool:
if header_name in self:
return potential_value in self._container[header_name.lower()][1:]
return False
def __ior__(self, other: object) -> HTTPHeaderDict:
# Supports extending a header dict in-place using operator |=
# combining items with add instead of __setitem__
maybe_constructable = ensure_can_construct_http_header_dict(other)
if maybe_constructable is None:
return NotImplemented
self.extend(maybe_constructable)
return self
def __or__(self, other: object) -> HTTPHeaderDict:
# Supports merging header dicts using operator |
# combining items with add instead of __setitem__
maybe_constructable = ensure_can_construct_http_header_dict(other)
if maybe_constructable is None:
return NotImplemented
result = self.copy()
result.extend(maybe_constructable)
return result
def __ror__(self, other: object) -> HTTPHeaderDict:
# Supports merging header dicts using operator | when other is on left side
# combining items with add instead of __setitem__
maybe_constructable = ensure_can_construct_http_header_dict(other)
if maybe_constructable is None:
return NotImplemented
result = type(self)(maybe_constructable)
result.extend(self)
return result

View File

@ -1,279 +0,0 @@
from __future__ import annotations
import json as _json
import typing
from urllib.parse import urlencode
from ._base_connection import _TYPE_BODY
from ._collections import HTTPHeaderDict
from .filepost import _TYPE_FIELDS, encode_multipart_formdata
from .response import BaseHTTPResponse
__all__ = ["RequestMethods"]
_TYPE_ENCODE_URL_FIELDS = typing.Union[
typing.Sequence[typing.Tuple[str, typing.Union[str, bytes]]],
typing.Mapping[str, typing.Union[str, bytes]],
]
class RequestMethods:
"""
Convenience mixin for classes who implement a :meth:`urlopen` method, such
as :class:`urllib3.HTTPConnectionPool` and
:class:`urllib3.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-form-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.
Initializer parameters:
:param headers:
Headers to include with all requests, unless other headers are given
explicitly.
"""
_encode_url_methods = {"DELETE", "GET", "HEAD", "OPTIONS"}
def __init__(self, headers: typing.Mapping[str, str] | None = None) -> None:
self.headers = headers or {}
def urlopen(
self,
method: str,
url: str,
body: _TYPE_BODY | None = None,
headers: typing.Mapping[str, str] | None = None,
encode_multipart: bool = True,
multipart_boundary: str | None = None,
**kw: typing.Any,
) -> BaseHTTPResponse: # Abstract
raise NotImplementedError(
"Classes extending RequestMethods must implement "
"their own ``urlopen`` method."
)
def request(
self,
method: str,
url: str,
body: _TYPE_BODY | None = None,
fields: _TYPE_FIELDS | None = None,
headers: typing.Mapping[str, str] | None = None,
json: typing.Any | None = None,
**urlopen_kw: typing.Any,
) -> BaseHTTPResponse:
"""
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`.
: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 fields:
Data to encode and send in the request body. Values are processed
by :func:`urllib.parse.urlencode`.
: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 json:
Data to encode and send as JSON with UTF-encoded in the request body.
The ``"Content-Type"`` header will be set to ``"application/json"``
unless specified otherwise.
"""
method = method.upper()
if json is not None and body is not None:
raise TypeError(
"request got values for both 'body' and 'json' parameters which are mutually exclusive"
)
if json is not None:
if headers is None:
headers = self.headers
if not ("content-type" in map(str.lower, headers.keys())):
headers = HTTPHeaderDict(headers)
headers["Content-Type"] = "application/json"
body = _json.dumps(json, separators=(",", ":"), ensure_ascii=False).encode(
"utf-8"
)
if body is not None:
urlopen_kw["body"] = body
if method in self._encode_url_methods:
return self.request_encode_url(
method,
url,
fields=fields, # type: ignore[arg-type]
headers=headers,
**urlopen_kw,
)
else:
return self.request_encode_body(
method, url, fields=fields, headers=headers, **urlopen_kw
)
def request_encode_url(
self,
method: str,
url: str,
fields: _TYPE_ENCODE_URL_FIELDS | None = None,
headers: typing.Mapping[str, str] | None = None,
**urlopen_kw: str,
) -> BaseHTTPResponse:
"""
Make a request using :meth:`urlopen` with the ``fields`` encoded in
the url. This is useful for request methods like GET, HEAD, DELETE, etc.
:param method:
HTTP request method (such as GET, POST, PUT, etc.)
:param url:
The URL to perform the request on.
:param fields:
Data to encode and send in the request body.
: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.
"""
if headers is None:
headers = self.headers
extra_kw: dict[str, typing.Any] = {"headers": headers}
extra_kw.update(urlopen_kw)
if fields:
url += "?" + urlencode(fields)
return self.urlopen(method, url, **extra_kw)
def request_encode_body(
self,
method: str,
url: str,
fields: _TYPE_FIELDS | None = None,
headers: typing.Mapping[str, str] | None = None,
encode_multipart: bool = True,
multipart_boundary: str | None = None,
**urlopen_kw: str,
) -> BaseHTTPResponse:
"""
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
:func:`urllib3.encode_multipart_formdata` is used to encode
the payload with the appropriate content type. Otherwise
:func:`urllib.parse.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, MIME type) tuple where
the MIME type is optional. For example::
fields = {
'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',
}
When uploading a file, providing a filename (the first parameter of the
tuple) is optional but recommended to best mimic 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.
:param method:
HTTP request method (such as GET, POST, PUT, etc.)
:param url:
The URL to perform the request on.
:param fields:
Data to encode and send in the request body.
: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 encode_multipart:
If True, encode the ``fields`` using the multipart/form-data MIME
format.
:param multipart_boundary:
If not specified, then a random boundary will be generated using
:func:`urllib3.filepost.choose_boundary`.
"""
if headers is None:
headers = self.headers
extra_kw: dict[str, typing.Any] = {"headers": HTTPHeaderDict(headers)}
body: bytes | str
if fields:
if "body" in urlopen_kw:
raise TypeError(
"request got values for both 'fields' and 'body', can only specify one."
)
if encode_multipart:
body, content_type = encode_multipart_formdata(
fields, boundary=multipart_boundary
)
else:
body, content_type = (
urlencode(fields), # type: ignore[arg-type]
"application/x-www-form-urlencoded",
)
extra_kw["body"] = body
extra_kw["headers"].setdefault("Content-Type", content_type)
extra_kw.update(urlopen_kw)
return self.urlopen(method, url, **extra_kw)

View File

@ -1,4 +0,0 @@
# This file is protected via CODEOWNERS
from __future__ import annotations
__version__ = "2.2.1"

View File

@ -1,930 +0,0 @@
from __future__ import annotations
import datetime
import logging
import os
import re
import socket
import sys
import typing
import warnings
from http.client import HTTPConnection as _HTTPConnection
from http.client import HTTPException as HTTPException # noqa: F401
from http.client import ResponseNotReady
from socket import timeout as SocketTimeout
if typing.TYPE_CHECKING:
from typing import Literal
from .response import HTTPResponse
from .util.ssl_ import _TYPE_PEER_CERT_RET_DICT
from .util.ssltransport import SSLTransport
from ._collections import HTTPHeaderDict
from .util.response import assert_header_parsing
from .util.timeout import _DEFAULT_TIMEOUT, _TYPE_TIMEOUT, Timeout
from .util.util import to_str
from .util.wait import wait_for_read
try: # Compiled with SSL?
import ssl
BaseSSLError = ssl.SSLError
except (ImportError, AttributeError):
ssl = None # type: ignore[assignment]
class BaseSSLError(BaseException): # type: ignore[no-redef]
pass
from ._base_connection import _TYPE_BODY
from ._base_connection import ProxyConfig as ProxyConfig
from ._base_connection import _ResponseOptions as _ResponseOptions
from ._version import __version__
from .exceptions import (
ConnectTimeoutError,
HeaderParsingError,
NameResolutionError,
NewConnectionError,
ProxyError,
SystemTimeWarning,
)
from .util import SKIP_HEADER, SKIPPABLE_HEADERS, connection, ssl_
from .util.request import body_to_chunks
from .util.ssl_ import assert_fingerprint as _assert_fingerprint
from .util.ssl_ import (
create_urllib3_context,
is_ipaddress,
resolve_cert_reqs,
resolve_ssl_version,
ssl_wrap_socket,
)
from .util.ssl_match_hostname import CertificateError, match_hostname
from .util.url import Url
# Not a no-op, we're adding this to the namespace so it can be imported.
ConnectionError = ConnectionError
BrokenPipeError = BrokenPipeError
log = logging.getLogger(__name__)
port_by_scheme = {"http": 80, "https": 443}
# When it comes time to update this value as a part of regular maintenance
# (ie test_recent_date is failing) update it to ~6 months before the current date.
RECENT_DATE = datetime.date(2023, 6, 1)
_CONTAINS_CONTROL_CHAR_RE = re.compile(r"[^-!#$%&'*+.^_`|~0-9a-zA-Z]")
_HAS_SYS_AUDIT = hasattr(sys, "audit")
class HTTPConnection(_HTTPConnection):
"""
Based on :class:`http.client.HTTPConnection` but provides an extra constructor
backwards-compatibility layer between older and newer Pythons.
Additional keyword parameters are used to configure attributes of the connection.
Accepted parameters include:
- ``source_address``: Set the source address for the current connection.
- ``socket_options``: Set specific options on the underlying socket. If not specified, then
defaults are loaded from ``HTTPConnection.default_socket_options`` which includes disabling
Nagle's algorithm (sets TCP_NODELAY to 1) unless the connection is behind a proxy.
For example, if you wish to enable TCP Keep Alive in addition to the defaults,
you might pass:
.. code-block:: python
HTTPConnection.default_socket_options + [
(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1),
]
Or you may want to disable the defaults by passing an empty list (e.g., ``[]``).
"""
default_port: typing.ClassVar[int] = port_by_scheme["http"] # type: ignore[misc]
#: Disable Nagle's algorithm by default.
#: ``[(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)]``
default_socket_options: typing.ClassVar[connection._TYPE_SOCKET_OPTIONS] = [
(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)
]
#: Whether this connection verifies the host's certificate.
is_verified: bool = False
#: Whether this proxy connection verified the proxy host's certificate.
# If no proxy is currently connected to the value will be ``None``.
proxy_is_verified: bool | None = None
blocksize: int
source_address: tuple[str, int] | None
socket_options: connection._TYPE_SOCKET_OPTIONS | None
_has_connected_to_proxy: bool
_response_options: _ResponseOptions | None
_tunnel_host: str | None
_tunnel_port: int | None
_tunnel_scheme: str | None
def __init__(
self,
host: str,
port: int | None = None,
*,
timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT,
source_address: tuple[str, int] | None = None,
blocksize: int = 16384,
socket_options: None
| (connection._TYPE_SOCKET_OPTIONS) = default_socket_options,
proxy: Url | None = None,
proxy_config: ProxyConfig | None = None,
) -> None:
super().__init__(
host=host,
port=port,
timeout=Timeout.resolve_default_timeout(timeout),
source_address=source_address,
blocksize=blocksize,
)
self.socket_options = socket_options
self.proxy = proxy
self.proxy_config = proxy_config
self._has_connected_to_proxy = False
self._response_options = None
self._tunnel_host: str | None = None
self._tunnel_port: int | None = None
self._tunnel_scheme: str | None = None
@property
def host(self) -> str:
"""
Getter method to remove any trailing dots that indicate the hostname is an FQDN.
In general, SSL certificates don't include the trailing dot indicating a
fully-qualified domain name, and thus, they don't validate properly when
checked against a domain name that includes the dot. In addition, some
servers may not expect to receive the trailing dot when provided.
However, the hostname with trailing dot is critical to DNS resolution; doing a
lookup with the trailing dot will properly only resolve the appropriate FQDN,
whereas a lookup without a trailing dot will search the system's search domain
list. Thus, it's important to keep the original host around for use only in
those cases where it's appropriate (i.e., when doing DNS lookup to establish the
actual TCP connection across which we're going to send HTTP requests).
"""
return self._dns_host.rstrip(".")
@host.setter
def host(self, value: str) -> None:
"""
Setter for the `host` property.
We assume that only urllib3 uses the _dns_host attribute; httplib itself
only uses `host`, and it seems reasonable that other libraries follow suit.
"""
self._dns_host = value
def _new_conn(self) -> socket.socket:
"""Establish a socket connection and set nodelay settings on it.
:return: New socket connection.
"""
try:
sock = connection.create_connection(
(self._dns_host, self.port),
self.timeout,
source_address=self.source_address,
socket_options=self.socket_options,
)
except socket.gaierror as e:
raise NameResolutionError(self.host, self, e) from e
except SocketTimeout as e:
raise ConnectTimeoutError(
self,
f"Connection to {self.host} timed out. (connect timeout={self.timeout})",
) from e
except OSError as e:
raise NewConnectionError(
self, f"Failed to establish a new connection: {e}"
) from e
# Audit hooks are only available in Python 3.8+
if _HAS_SYS_AUDIT:
sys.audit("http.client.connect", self, self.host, self.port)
return sock
def set_tunnel(
self,
host: str,
port: int | None = None,
headers: typing.Mapping[str, str] | None = None,
scheme: str = "http",
) -> None:
if scheme not in ("http", "https"):
raise ValueError(
f"Invalid proxy scheme for tunneling: {scheme!r}, must be either 'http' or 'https'"
)
super().set_tunnel(host, port=port, headers=headers)
self._tunnel_scheme = scheme
def connect(self) -> None:
self.sock = self._new_conn()
if self._tunnel_host:
# If we're tunneling it means we're connected to our proxy.
self._has_connected_to_proxy = True
# TODO: Fix tunnel so it doesn't depend on self.sock state.
self._tunnel() # type: ignore[attr-defined]
# If there's a proxy to be connected to we are fully connected.
# This is set twice (once above and here) due to forwarding proxies
# not using tunnelling.
self._has_connected_to_proxy = bool(self.proxy)
if self._has_connected_to_proxy:
self.proxy_is_verified = False
@property
def is_closed(self) -> bool:
return self.sock is None
@property
def is_connected(self) -> bool:
if self.sock is None:
return False
return not wait_for_read(self.sock, timeout=0.0)
@property
def has_connected_to_proxy(self) -> bool:
return self._has_connected_to_proxy
@property
def proxy_is_forwarding(self) -> bool:
"""
Return True if a forwarding proxy is configured, else return False
"""
return bool(self.proxy) and self._tunnel_host is None
def close(self) -> None:
try:
super().close()
finally:
# Reset all stateful properties so connection
# can be re-used without leaking prior configs.
self.sock = None
self.is_verified = False
self.proxy_is_verified = None
self._has_connected_to_proxy = False
self._response_options = None
self._tunnel_host = None
self._tunnel_port = None
self._tunnel_scheme = None
def putrequest(
self,
method: str,
url: str,
skip_host: bool = False,
skip_accept_encoding: bool = False,
) -> None:
""""""
# Empty docstring because the indentation of CPython's implementation
# is broken but we don't want this method in our documentation.
match = _CONTAINS_CONTROL_CHAR_RE.search(method)
if match:
raise ValueError(
f"Method cannot contain non-token characters {method!r} (found at least {match.group()!r})"
)
return super().putrequest(
method, url, skip_host=skip_host, skip_accept_encoding=skip_accept_encoding
)
def putheader(self, header: str, *values: str) -> None: # type: ignore[override]
""""""
if not any(isinstance(v, str) and v == SKIP_HEADER for v in values):
super().putheader(header, *values)
elif to_str(header.lower()) not in SKIPPABLE_HEADERS:
skippable_headers = "', '".join(
[str.title(header) for header in sorted(SKIPPABLE_HEADERS)]
)
raise ValueError(
f"urllib3.util.SKIP_HEADER only supports '{skippable_headers}'"
)
# `request` method's signature intentionally violates LSP.
# urllib3's API is different from `http.client.HTTPConnection` and the subclassing is only incidental.
def request( # type: ignore[override]
self,
method: str,
url: str,
body: _TYPE_BODY | None = None,
headers: typing.Mapping[str, str] | None = None,
*,
chunked: bool = False,
preload_content: bool = True,
decode_content: bool = True,
enforce_content_length: bool = True,
) -> None:
# Update the inner socket's timeout value to send the request.
# This only triggers if the connection is re-used.
if self.sock is not None:
self.sock.settimeout(self.timeout)
# Store these values to be fed into the HTTPResponse
# object later. TODO: Remove this in favor of a real
# HTTP lifecycle mechanism.
# We have to store these before we call .request()
# because sometimes we can still salvage a response
# off the wire even if we aren't able to completely
# send the request body.
self._response_options = _ResponseOptions(
request_method=method,
request_url=url,
preload_content=preload_content,
decode_content=decode_content,
enforce_content_length=enforce_content_length,
)
if headers is None:
headers = {}
header_keys = frozenset(to_str(k.lower()) for k in headers)
skip_accept_encoding = "accept-encoding" in header_keys
skip_host = "host" in header_keys
self.putrequest(
method, url, skip_accept_encoding=skip_accept_encoding, skip_host=skip_host
)
# Transform the body into an iterable of sendall()-able chunks
# and detect if an explicit Content-Length is doable.
chunks_and_cl = body_to_chunks(body, method=method, blocksize=self.blocksize)
chunks = chunks_and_cl.chunks
content_length = chunks_and_cl.content_length
# When chunked is explicit set to 'True' we respect that.
if chunked:
if "transfer-encoding" not in header_keys:
self.putheader("Transfer-Encoding", "chunked")
else:
# Detect whether a framing mechanism is already in use. If so
# we respect that value, otherwise we pick chunked vs content-length
# depending on the type of 'body'.
if "content-length" in header_keys:
chunked = False
elif "transfer-encoding" in header_keys:
chunked = True
# Otherwise we go off the recommendation of 'body_to_chunks()'.
else:
chunked = False
if content_length is None:
if chunks is not None:
chunked = True
self.putheader("Transfer-Encoding", "chunked")
else:
self.putheader("Content-Length", str(content_length))
# Now that framing headers are out of the way we send all the other headers.
if "user-agent" not in header_keys:
self.putheader("User-Agent", _get_default_user_agent())
for header, value in headers.items():
self.putheader(header, value)
self.endheaders()
# If we're given a body we start sending that in chunks.
if chunks is not None:
for chunk in chunks:
# Sending empty chunks isn't allowed for TE: chunked
# as it indicates the end of the body.
if not chunk:
continue
if isinstance(chunk, str):
chunk = chunk.encode("utf-8")
if chunked:
self.send(b"%x\r\n%b\r\n" % (len(chunk), chunk))
else:
self.send(chunk)
# Regardless of whether we have a body or not, if we're in
# chunked mode we want to send an explicit empty chunk.
if chunked:
self.send(b"0\r\n\r\n")
def request_chunked(
self,
method: str,
url: str,
body: _TYPE_BODY | None = None,
headers: typing.Mapping[str, str] | None = None,
) -> None:
"""
Alternative to the common request method, which sends the
body with chunked encoding and not as one block
"""
warnings.warn(
"HTTPConnection.request_chunked() is deprecated and will be removed "
"in urllib3 v2.1.0. Instead use HTTPConnection.request(..., chunked=True).",
category=DeprecationWarning,
stacklevel=2,
)
self.request(method, url, body=body, headers=headers, chunked=True)
def getresponse( # type: ignore[override]
self,
) -> HTTPResponse:
"""
Get the response from the server.
If the HTTPConnection is in the correct state, returns an instance of HTTPResponse or of whatever object is returned by the response_class variable.
If a request has not been sent or if a previous response has not be handled, ResponseNotReady is raised. If the HTTP response indicates that the connection should be closed, then it will be closed before the response is returned. When the connection is closed, the underlying socket is closed.
"""
# Raise the same error as http.client.HTTPConnection
if self._response_options is None:
raise ResponseNotReady()
# Reset this attribute for being used again.
resp_options = self._response_options
self._response_options = None
# Since the connection's timeout value may have been updated
# we need to set the timeout on the socket.
self.sock.settimeout(self.timeout)
# This is needed here to avoid circular import errors
from .response import HTTPResponse
# Get the response from http.client.HTTPConnection
httplib_response = super().getresponse()
try:
assert_header_parsing(httplib_response.msg)
except (HeaderParsingError, TypeError) as hpe:
log.warning(
"Failed to parse headers (url=%s): %s",
_url_from_connection(self, resp_options.request_url),
hpe,
exc_info=True,
)
headers = HTTPHeaderDict(httplib_response.msg.items())
response = HTTPResponse(
body=httplib_response,
headers=headers,
status=httplib_response.status,
version=httplib_response.version,
reason=httplib_response.reason,
preload_content=resp_options.preload_content,
decode_content=resp_options.decode_content,
original_response=httplib_response,
enforce_content_length=resp_options.enforce_content_length,
request_method=resp_options.request_method,
request_url=resp_options.request_url,
)
return response
class HTTPSConnection(HTTPConnection):
"""
Many of the parameters to this constructor are passed to the underlying SSL
socket by means of :py:func:`urllib3.util.ssl_wrap_socket`.
"""
default_port = port_by_scheme["https"] # type: ignore[misc]
cert_reqs: int | str | None = None
ca_certs: str | None = None
ca_cert_dir: str | None = None
ca_cert_data: None | str | bytes = None
ssl_version: int | str | None = None
ssl_minimum_version: int | None = None
ssl_maximum_version: int | None = None
assert_fingerprint: str | None = None
def __init__(
self,
host: str,
port: int | None = None,
*,
timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT,
source_address: tuple[str, int] | None = None,
blocksize: int = 16384,
socket_options: None
| (connection._TYPE_SOCKET_OPTIONS) = HTTPConnection.default_socket_options,
proxy: Url | None = None,
proxy_config: ProxyConfig | None = None,
cert_reqs: int | str | None = None,
assert_hostname: None | str | Literal[False] = None,
assert_fingerprint: str | None = None,
server_hostname: str | None = None,
ssl_context: ssl.SSLContext | None = None,
ca_certs: str | None = None,
ca_cert_dir: str | None = None,
ca_cert_data: None | str | bytes = None,
ssl_minimum_version: int | None = None,
ssl_maximum_version: int | None = None,
ssl_version: int | str | None = None, # Deprecated
cert_file: str | None = None,
key_file: str | None = None,
key_password: str | None = None,
) -> None:
super().__init__(
host,
port=port,
timeout=timeout,
source_address=source_address,
blocksize=blocksize,
socket_options=socket_options,
proxy=proxy,
proxy_config=proxy_config,
)
self.key_file = key_file
self.cert_file = cert_file
self.key_password = key_password
self.ssl_context = ssl_context
self.server_hostname = server_hostname
self.assert_hostname = assert_hostname
self.assert_fingerprint = assert_fingerprint
self.ssl_version = ssl_version
self.ssl_minimum_version = ssl_minimum_version
self.ssl_maximum_version = ssl_maximum_version
self.ca_certs = ca_certs and os.path.expanduser(ca_certs)
self.ca_cert_dir = ca_cert_dir and os.path.expanduser(ca_cert_dir)
self.ca_cert_data = ca_cert_data
# cert_reqs depends on ssl_context so calculate last.
if cert_reqs is None:
if self.ssl_context is not None:
cert_reqs = self.ssl_context.verify_mode
else:
cert_reqs = resolve_cert_reqs(None)
self.cert_reqs = cert_reqs
def set_cert(
self,
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,
assert_hostname: None | str | Literal[False] = None,
assert_fingerprint: str | None = None,
ca_cert_dir: str | None = None,
ca_cert_data: None | str | bytes = None,
) -> None:
"""
This method should only be called once, before the connection is used.
"""
warnings.warn(
"HTTPSConnection.set_cert() is deprecated and will be removed "
"in urllib3 v2.1.0. Instead provide the parameters to the "
"HTTPSConnection constructor.",
category=DeprecationWarning,
stacklevel=2,
)
# If cert_reqs is not provided we'll assume CERT_REQUIRED unless we also
# have an SSLContext object in which case we'll use its verify_mode.
if cert_reqs is None:
if self.ssl_context is not None:
cert_reqs = self.ssl_context.verify_mode
else:
cert_reqs = resolve_cert_reqs(None)
self.key_file = key_file
self.cert_file = cert_file
self.cert_reqs = cert_reqs
self.key_password = key_password
self.assert_hostname = assert_hostname
self.assert_fingerprint = assert_fingerprint
self.ca_certs = ca_certs and os.path.expanduser(ca_certs)
self.ca_cert_dir = ca_cert_dir and os.path.expanduser(ca_cert_dir)
self.ca_cert_data = ca_cert_data
def connect(self) -> None:
sock: socket.socket | ssl.SSLSocket
self.sock = sock = self._new_conn()
server_hostname: str = self.host
tls_in_tls = False
# Do we need to establish a tunnel?
if self._tunnel_host is not None:
# We're tunneling to an HTTPS origin so need to do TLS-in-TLS.
if self._tunnel_scheme == "https":
# _connect_tls_proxy will verify and assign proxy_is_verified
self.sock = sock = self._connect_tls_proxy(self.host, sock)
tls_in_tls = True
elif self._tunnel_scheme == "http":
self.proxy_is_verified = False
# If we're tunneling it means we're connected to our proxy.
self._has_connected_to_proxy = True
self._tunnel() # type: ignore[attr-defined]
# Override the host with the one we're requesting data from.
server_hostname = self._tunnel_host
if self.server_hostname is not None:
server_hostname = self.server_hostname
is_time_off = datetime.date.today() < RECENT_DATE
if is_time_off:
warnings.warn(
(
f"System time is way off (before {RECENT_DATE}). This will probably "
"lead to SSL verification errors"
),
SystemTimeWarning,
)
# Remove trailing '.' from fqdn hostnames to allow certificate validation
server_hostname_rm_dot = server_hostname.rstrip(".")
sock_and_verified = _ssl_wrap_socket_and_match_hostname(
sock=sock,
cert_reqs=self.cert_reqs,
ssl_version=self.ssl_version,
ssl_minimum_version=self.ssl_minimum_version,
ssl_maximum_version=self.ssl_maximum_version,
ca_certs=self.ca_certs,
ca_cert_dir=self.ca_cert_dir,
ca_cert_data=self.ca_cert_data,
cert_file=self.cert_file,
key_file=self.key_file,
key_password=self.key_password,
server_hostname=server_hostname_rm_dot,
ssl_context=self.ssl_context,
tls_in_tls=tls_in_tls,
assert_hostname=self.assert_hostname,
assert_fingerprint=self.assert_fingerprint,
)
self.sock = sock_and_verified.socket
# Forwarding proxies can never have a verified target since
# the proxy is the one doing the verification. Should instead
# use a CONNECT tunnel in order to verify the target.
# See: https://github.com/urllib3/urllib3/issues/3267.
if self.proxy_is_forwarding:
self.is_verified = False
else:
self.is_verified = sock_and_verified.is_verified
# If there's a proxy to be connected to we are fully connected.
# This is set twice (once above and here) due to forwarding proxies
# not using tunnelling.
self._has_connected_to_proxy = bool(self.proxy)
# Set `self.proxy_is_verified` unless it's already set while
# establishing a tunnel.
if self._has_connected_to_proxy and self.proxy_is_verified is None:
self.proxy_is_verified = sock_and_verified.is_verified
def _connect_tls_proxy(self, hostname: str, sock: socket.socket) -> ssl.SSLSocket:
"""
Establish a TLS connection to the proxy using the provided SSL context.
"""
# `_connect_tls_proxy` is called when self._tunnel_host is truthy.
proxy_config = typing.cast(ProxyConfig, self.proxy_config)
ssl_context = proxy_config.ssl_context
sock_and_verified = _ssl_wrap_socket_and_match_hostname(
sock,
cert_reqs=self.cert_reqs,
ssl_version=self.ssl_version,
ssl_minimum_version=self.ssl_minimum_version,
ssl_maximum_version=self.ssl_maximum_version,
ca_certs=self.ca_certs,
ca_cert_dir=self.ca_cert_dir,
ca_cert_data=self.ca_cert_data,
server_hostname=hostname,
ssl_context=ssl_context,
assert_hostname=proxy_config.assert_hostname,
assert_fingerprint=proxy_config.assert_fingerprint,
# Features that aren't implemented for proxies yet:
cert_file=None,
key_file=None,
key_password=None,
tls_in_tls=False,
)
self.proxy_is_verified = sock_and_verified.is_verified
return sock_and_verified.socket # type: ignore[return-value]
class _WrappedAndVerifiedSocket(typing.NamedTuple):
"""
Wrapped socket and whether the connection is
verified after the TLS handshake
"""
socket: ssl.SSLSocket | SSLTransport
is_verified: bool
def _ssl_wrap_socket_and_match_hostname(
sock: socket.socket,
*,
cert_reqs: None | str | int,
ssl_version: None | str | int,
ssl_minimum_version: int | None,
ssl_maximum_version: int | None,
cert_file: str | None,
key_file: str | None,
key_password: str | None,
ca_certs: str | None,
ca_cert_dir: str | None,
ca_cert_data: None | str | bytes,
assert_hostname: None | str | Literal[False],
assert_fingerprint: str | None,
server_hostname: str | None,
ssl_context: ssl.SSLContext | None,
tls_in_tls: bool = False,
) -> _WrappedAndVerifiedSocket:
"""Logic for constructing an SSLContext from all TLS parameters, passing
that down into ssl_wrap_socket, and then doing certificate verification
either via hostname or fingerprint. This function exists to guarantee
that both proxies and targets have the same behavior when connecting via TLS.
"""
default_ssl_context = False
if ssl_context is None:
default_ssl_context = True
context = create_urllib3_context(
ssl_version=resolve_ssl_version(ssl_version),
ssl_minimum_version=ssl_minimum_version,
ssl_maximum_version=ssl_maximum_version,
cert_reqs=resolve_cert_reqs(cert_reqs),
)
else:
context = ssl_context
context.verify_mode = resolve_cert_reqs(cert_reqs)
# In some cases, we want to verify hostnames ourselves
if (
# `ssl` can't verify fingerprints or alternate hostnames
assert_fingerprint
or assert_hostname
# assert_hostname can be set to False to disable hostname checking
or assert_hostname is False
# We still support OpenSSL 1.0.2, which prevents us from verifying
# hostnames easily: https://github.com/pyca/pyopenssl/pull/933
or ssl_.IS_PYOPENSSL
or not ssl_.HAS_NEVER_CHECK_COMMON_NAME
):
context.check_hostname = False
# Try to load OS default certs if none are given. We need to do the hasattr() check
# for custom pyOpenSSL SSLContext objects because they don't support
# load_default_certs().
if (
not ca_certs
and not ca_cert_dir
and not ca_cert_data
and default_ssl_context
and hasattr(context, "load_default_certs")
):
context.load_default_certs()
# Ensure that IPv6 addresses are in the proper format and don't have a
# scope ID. Python's SSL module fails to recognize scoped IPv6 addresses
# and interprets them as DNS hostnames.
if server_hostname is not None:
normalized = server_hostname.strip("[]")
if "%" in normalized:
normalized = normalized[: normalized.rfind("%")]
if is_ipaddress(normalized):
server_hostname = normalized
ssl_sock = ssl_wrap_socket(
sock=sock,
keyfile=key_file,
certfile=cert_file,
key_password=key_password,
ca_certs=ca_certs,
ca_cert_dir=ca_cert_dir,
ca_cert_data=ca_cert_data,
server_hostname=server_hostname,
ssl_context=context,
tls_in_tls=tls_in_tls,
)
try:
if assert_fingerprint:
_assert_fingerprint(
ssl_sock.getpeercert(binary_form=True), assert_fingerprint
)
elif (
context.verify_mode != ssl.CERT_NONE
and not context.check_hostname
and assert_hostname is not False
):
cert: _TYPE_PEER_CERT_RET_DICT = ssl_sock.getpeercert() # type: ignore[assignment]
# Need to signal to our match_hostname whether to use 'commonName' or not.
# If we're using our own constructed SSLContext we explicitly set 'False'
# because PyPy hard-codes 'True' from SSLContext.hostname_checks_common_name.
if default_ssl_context:
hostname_checks_common_name = False
else:
hostname_checks_common_name = (
getattr(context, "hostname_checks_common_name", False) or False
)
_match_hostname(
cert,
assert_hostname or server_hostname, # type: ignore[arg-type]
hostname_checks_common_name,
)
return _WrappedAndVerifiedSocket(
socket=ssl_sock,
is_verified=context.verify_mode == ssl.CERT_REQUIRED
or bool(assert_fingerprint),
)
except BaseException:
ssl_sock.close()
raise
def _match_hostname(
cert: _TYPE_PEER_CERT_RET_DICT | None,
asserted_hostname: str,
hostname_checks_common_name: bool = False,
) -> None:
# Our upstream implementation of ssl.match_hostname()
# only applies this normalization to IP addresses so it doesn't
# match DNS SANs so we do the same thing!
stripped_hostname = asserted_hostname.strip("[]")
if is_ipaddress(stripped_hostname):
asserted_hostname = stripped_hostname
try:
match_hostname(cert, asserted_hostname, hostname_checks_common_name)
except CertificateError as e:
log.warning(
"Certificate did not match expected hostname: %s. Certificate: %s",
asserted_hostname,
cert,
)
# Add cert to exception and reraise so client code can inspect
# the cert when catching the exception, if they want to
e._peer_cert = cert # type: ignore[attr-defined]
raise
def _wrap_proxy_error(err: Exception, proxy_scheme: str | None) -> ProxyError:
# Look for the phrase 'wrong version number', if found
# then we should warn the user that we're very sure that
# this proxy is HTTP-only and they have a configuration issue.
error_normalized = " ".join(re.split("[^a-z]", str(err).lower()))
is_likely_http_proxy = (
"wrong version number" in error_normalized
or "unknown protocol" in error_normalized
or "record layer failure" in error_normalized
)
http_proxy_warning = (
". Your proxy appears to only use HTTP and not HTTPS, "
"try changing your proxy URL to be HTTP. See: "
"https://urllib3.readthedocs.io/en/latest/advanced-usage.html"
"#https-proxy-error-http-proxy"
)
new_err = ProxyError(
f"Unable to connect to proxy"
f"{http_proxy_warning if is_likely_http_proxy and proxy_scheme == 'https' else ''}",
err,
)
new_err.__cause__ = err
return new_err
def _get_default_user_agent() -> str:
return f"python-urllib3/{__version__}"
class DummyConnection:
"""Used to detect a failed ConnectionCls import."""
if not ssl:
HTTPSConnection = DummyConnection # type: ignore[misc, assignment] # noqa: F811
VerifiedHTTPSConnection = HTTPSConnection
def _url_from_connection(
conn: HTTPConnection | HTTPSConnection, path: str | None = None
) -> str:
"""Returns the URL from a given connection. This is mainly used for testing and logging."""
scheme = "https" if isinstance(conn, HTTPSConnection) else "http"
return Url(scheme=scheme, host=conn.host, port=conn.port, path=path).url

File diff suppressed because it is too large Load Diff

View File

@ -1,16 +0,0 @@
from __future__ import annotations
import urllib3.connection
from ...connectionpool import HTTPConnectionPool, HTTPSConnectionPool
from .connection import EmscriptenHTTPConnection, EmscriptenHTTPSConnection
def inject_into_urllib3() -> None:
# override connection classes to use emscripten specific classes
# n.b. mypy complains about the overriding of classes below
# if it isn't ignored
HTTPConnectionPool.ConnectionCls = EmscriptenHTTPConnection
HTTPSConnectionPool.ConnectionCls = EmscriptenHTTPSConnection
urllib3.connection.HTTPConnection = EmscriptenHTTPConnection # type: ignore[misc,assignment]
urllib3.connection.HTTPSConnection = EmscriptenHTTPSConnection # type: ignore[misc,assignment]

View File

@ -1,254 +0,0 @@
from __future__ import annotations
import os
import typing
# use http.client.HTTPException for consistency with non-emscripten
from http.client import HTTPException as HTTPException # noqa: F401
from http.client import ResponseNotReady
from ..._base_connection import _TYPE_BODY
from ...connection import HTTPConnection, ProxyConfig, port_by_scheme
from ...exceptions import TimeoutError
from ...response import BaseHTTPResponse
from ...util.connection import _TYPE_SOCKET_OPTIONS
from ...util.timeout import _DEFAULT_TIMEOUT, _TYPE_TIMEOUT
from ...util.url import Url
from .fetch import _RequestError, _TimeoutError, send_request, send_streaming_request
from .request import EmscriptenRequest
from .response import EmscriptenHttpResponseWrapper, EmscriptenResponse
if typing.TYPE_CHECKING:
from ..._base_connection import BaseHTTPConnection, BaseHTTPSConnection
class EmscriptenHTTPConnection:
default_port: typing.ClassVar[int] = port_by_scheme["http"]
default_socket_options: typing.ClassVar[_TYPE_SOCKET_OPTIONS]
timeout: None | (float)
host: str
port: int
blocksize: int
source_address: tuple[str, int] | None
socket_options: _TYPE_SOCKET_OPTIONS | None
proxy: Url | None
proxy_config: ProxyConfig | None
is_verified: bool = False
proxy_is_verified: bool | None = None
_response: EmscriptenResponse | None
def __init__(
self,
host: str,
port: int = 0,
*,
timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT,
source_address: tuple[str, int] | None = None,
blocksize: int = 8192,
socket_options: _TYPE_SOCKET_OPTIONS | None = None,
proxy: Url | None = None,
proxy_config: ProxyConfig | None = None,
) -> None:
self.host = host
self.port = port
self.timeout = timeout if isinstance(timeout, float) else 0.0
self.scheme = "http"
self._closed = True
self._response = None
# ignore these things because we don't
# have control over that stuff
self.proxy = None
self.proxy_config = None
self.blocksize = blocksize
self.source_address = None
self.socket_options = None
self.is_verified = False
def set_tunnel(
self,
host: str,
port: int | None = 0,
headers: typing.Mapping[str, str] | None = None,
scheme: str = "http",
) -> None:
pass
def connect(self) -> None:
pass
def request(
self,
method: str,
url: str,
body: _TYPE_BODY | None = None,
headers: typing.Mapping[str, str] | None = None,
# We know *at least* botocore is depending on the order of the
# first 3 parameters so to be safe we only mark the later ones
# as keyword-only to ensure we have space to extend.
*,
chunked: bool = False,
preload_content: bool = True,
decode_content: bool = True,
enforce_content_length: bool = True,
) -> None:
self._closed = False
if url.startswith("/"):
# no scheme / host / port included, make a full url
url = f"{self.scheme}://{self.host}:{self.port}" + url
request = EmscriptenRequest(
url=url,
method=method,
timeout=self.timeout if self.timeout else 0,
decode_content=decode_content,
)
request.set_body(body)
if headers:
for k, v in headers.items():
request.set_header(k, v)
self._response = None
try:
if not preload_content:
self._response = send_streaming_request(request)
if self._response is None:
self._response = send_request(request)
except _TimeoutError as e:
raise TimeoutError(e.message) from e
except _RequestError as e:
raise HTTPException(e.message) from e
def getresponse(self) -> BaseHTTPResponse:
if self._response is not None:
return EmscriptenHttpResponseWrapper(
internal_response=self._response,
url=self._response.request.url,
connection=self,
)
else:
raise ResponseNotReady()
def close(self) -> None:
self._closed = True
self._response = None
@property
def is_closed(self) -> bool:
"""Whether the connection either is brand new or has been previously closed.
If this property is True then both ``is_connected`` and ``has_connected_to_proxy``
properties must be False.
"""
return self._closed
@property
def is_connected(self) -> bool:
"""Whether the connection is actively connected to any origin (proxy or target)"""
return True
@property
def has_connected_to_proxy(self) -> bool:
"""Whether the connection has successfully connected to its proxy.
This returns False if no proxy is in use. Used to determine whether
errors are coming from the proxy layer or from tunnelling to the target origin.
"""
return False
class EmscriptenHTTPSConnection(EmscriptenHTTPConnection):
default_port = port_by_scheme["https"]
# all this is basically ignored, as browser handles https
cert_reqs: int | str | None = None
ca_certs: str | None = None
ca_cert_dir: str | None = None
ca_cert_data: None | str | bytes = None
cert_file: str | None
key_file: str | None
key_password: str | None
ssl_context: typing.Any | None
ssl_version: int | str | None = None
ssl_minimum_version: int | None = None
ssl_maximum_version: int | None = None
assert_hostname: None | str | typing.Literal[False]
assert_fingerprint: str | None = None
def __init__(
self,
host: str,
port: int = 0,
*,
timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT,
source_address: tuple[str, int] | None = None,
blocksize: int = 16384,
socket_options: None
| _TYPE_SOCKET_OPTIONS = HTTPConnection.default_socket_options,
proxy: Url | None = None,
proxy_config: ProxyConfig | None = None,
cert_reqs: int | str | None = None,
assert_hostname: None | str | typing.Literal[False] = None,
assert_fingerprint: str | None = None,
server_hostname: str | None = None,
ssl_context: typing.Any | None = None,
ca_certs: str | None = None,
ca_cert_dir: str | None = None,
ca_cert_data: None | str | bytes = None,
ssl_minimum_version: int | None = None,
ssl_maximum_version: int | None = None,
ssl_version: int | str | None = None, # Deprecated
cert_file: str | None = None,
key_file: str | None = None,
key_password: str | None = None,
) -> None:
super().__init__(
host,
port=port,
timeout=timeout,
source_address=source_address,
blocksize=blocksize,
socket_options=socket_options,
proxy=proxy,
proxy_config=proxy_config,
)
self.scheme = "https"
self.key_file = key_file
self.cert_file = cert_file
self.key_password = key_password
self.ssl_context = ssl_context
self.server_hostname = server_hostname
self.assert_hostname = assert_hostname
self.assert_fingerprint = assert_fingerprint
self.ssl_version = ssl_version
self.ssl_minimum_version = ssl_minimum_version
self.ssl_maximum_version = ssl_maximum_version
self.ca_certs = ca_certs and os.path.expanduser(ca_certs)
self.ca_cert_dir = ca_cert_dir and os.path.expanduser(ca_cert_dir)
self.ca_cert_data = ca_cert_data
self.cert_reqs = None
# The browser will automatically verify all requests.
# We have no control over that setting.
self.is_verified = True
def set_cert(
self,
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,
assert_hostname: None | str | typing.Literal[False] = None,
assert_fingerprint: str | None = None,
ca_cert_dir: str | None = None,
ca_cert_data: None | str | bytes = None,
) -> None:
pass
# verify that this class implements BaseHTTP(s) connection correctly
if typing.TYPE_CHECKING:
_supports_http_protocol: BaseHTTPConnection = EmscriptenHTTPConnection("", 0)
_supports_https_protocol: BaseHTTPSConnection = EmscriptenHTTPSConnection("", 0)

View File

@ -1,110 +0,0 @@
let Status = {
SUCCESS_HEADER: -1,
SUCCESS_EOF: -2,
ERROR_TIMEOUT: -3,
ERROR_EXCEPTION: -4,
};
let connections = {};
let nextConnectionID = 1;
const encoder = new TextEncoder();
self.addEventListener("message", async function (event) {
if (event.data.close) {
let connectionID = event.data.close;
delete connections[connectionID];
return;
} else if (event.data.getMore) {
let connectionID = event.data.getMore;
let { curOffset, value, reader, intBuffer, byteBuffer } =
connections[connectionID];
// if we still have some in buffer, then just send it back straight away
if (!value || curOffset >= value.length) {
// read another buffer if required
try {
let readResponse = await reader.read();
if (readResponse.done) {
// read everything - clear connection and return
delete connections[connectionID];
Atomics.store(intBuffer, 0, Status.SUCCESS_EOF);
Atomics.notify(intBuffer, 0);
// finished reading successfully
// return from event handler
return;
}
curOffset = 0;
connections[connectionID].value = readResponse.value;
value = readResponse.value;
} catch (error) {
console.log("Request exception:", error);
let errorBytes = encoder.encode(error.message);
let written = errorBytes.length;
byteBuffer.set(errorBytes);
intBuffer[1] = written;
Atomics.store(intBuffer, 0, Status.ERROR_EXCEPTION);
Atomics.notify(intBuffer, 0);
}
}
// send as much buffer as we can
let curLen = value.length - curOffset;
if (curLen > byteBuffer.length) {
curLen = byteBuffer.length;
}
byteBuffer.set(value.subarray(curOffset, curOffset + curLen), 0);
Atomics.store(intBuffer, 0, curLen); // store current length in bytes
Atomics.notify(intBuffer, 0);
curOffset += curLen;
connections[connectionID].curOffset = curOffset;
return;
} else {
// start fetch
let connectionID = nextConnectionID;
nextConnectionID += 1;
const intBuffer = new Int32Array(event.data.buffer);
const byteBuffer = new Uint8Array(event.data.buffer, 8);
try {
const response = await fetch(event.data.url, event.data.fetchParams);
// return the headers first via textencoder
var headers = [];
for (const pair of response.headers.entries()) {
headers.push([pair[0], pair[1]]);
}
let headerObj = {
headers: headers,
status: response.status,
connectionID,
};
const headerText = JSON.stringify(headerObj);
let headerBytes = encoder.encode(headerText);
let written = headerBytes.length;
byteBuffer.set(headerBytes);
intBuffer[1] = written;
// make a connection
connections[connectionID] = {
reader: response.body.getReader(),
intBuffer: intBuffer,
byteBuffer: byteBuffer,
value: undefined,
curOffset: 0,
};
// set header ready
Atomics.store(intBuffer, 0, Status.SUCCESS_HEADER);
Atomics.notify(intBuffer, 0);
// all fetching after this goes through a new postmessage call with getMore
// this allows for parallel requests
} catch (error) {
console.log("Request exception:", error);
let errorBytes = encoder.encode(error.message);
let written = errorBytes.length;
byteBuffer.set(errorBytes);
intBuffer[1] = written;
Atomics.store(intBuffer, 0, Status.ERROR_EXCEPTION);
Atomics.notify(intBuffer, 0);
}
}
});
self.postMessage({ inited: true });

View File

@ -1,418 +0,0 @@
"""
Support for streaming http requests in emscripten.
A few caveats -
Firstly, you can't do streaming http in the main UI thread, because atomics.wait isn't allowed.
Streaming only works if you're running pyodide in a web worker.
Secondly, this uses an extra web worker and SharedArrayBuffer to do the asynchronous fetch
operation, so it requires that you have crossOriginIsolation enabled, by serving over https
(or from localhost) with the two headers below set:
Cross-Origin-Opener-Policy: same-origin
Cross-Origin-Embedder-Policy: require-corp
You can tell if cross origin isolation is successfully enabled by looking at the global crossOriginIsolated variable in
javascript console. If it isn't, streaming requests will fallback to XMLHttpRequest, i.e. getting the whole
request into a buffer and then returning it. it shows a warning in the javascript console in this case.
Finally, the webworker which does the streaming fetch is created on initial import, but will only be started once
control is returned to javascript. Call `await wait_for_streaming_ready()` to wait for streaming fetch.
NB: in this code, there are a lot of javascript objects. They are named js_*
to make it clear what type of object they are.
"""
from __future__ import annotations
import io
import json
from email.parser import Parser
from importlib.resources import files
from typing import TYPE_CHECKING, Any
import js # type: ignore[import-not-found]
from pyodide.ffi import ( # type: ignore[import-not-found]
JsArray,
JsException,
JsProxy,
to_js,
)
if TYPE_CHECKING:
from typing_extensions import Buffer
from .request import EmscriptenRequest
from .response import EmscriptenResponse
"""
There are some headers that trigger unintended CORS preflight requests.
See also https://github.com/koenvo/pyodide-http/issues/22
"""
HEADERS_TO_IGNORE = ("user-agent",)
SUCCESS_HEADER = -1
SUCCESS_EOF = -2
ERROR_TIMEOUT = -3
ERROR_EXCEPTION = -4
_STREAMING_WORKER_CODE = (
files(__package__)
.joinpath("emscripten_fetch_worker.js")
.read_text(encoding="utf-8")
)
class _RequestError(Exception):
def __init__(
self,
message: str | None = None,
*,
request: EmscriptenRequest | None = None,
response: EmscriptenResponse | None = None,
):
self.request = request
self.response = response
self.message = message
super().__init__(self.message)
class _StreamingError(_RequestError):
pass
class _TimeoutError(_RequestError):
pass
def _obj_from_dict(dict_val: dict[str, Any]) -> JsProxy:
return to_js(dict_val, dict_converter=js.Object.fromEntries)
class _ReadStream(io.RawIOBase):
def __init__(
self,
int_buffer: JsArray,
byte_buffer: JsArray,
timeout: float,
worker: JsProxy,
connection_id: int,
request: EmscriptenRequest,
):
self.int_buffer = int_buffer
self.byte_buffer = byte_buffer
self.read_pos = 0
self.read_len = 0
self.connection_id = connection_id
self.worker = worker
self.timeout = int(1000 * timeout) if timeout > 0 else None
self.is_live = True
self._is_closed = False
self.request: EmscriptenRequest | None = request
def __del__(self) -> None:
self.close()
# this is compatible with _base_connection
def is_closed(self) -> bool:
return self._is_closed
# for compatibility with RawIOBase
@property
def closed(self) -> bool:
return self.is_closed()
def close(self) -> None:
if not self.is_closed():
self.read_len = 0
self.read_pos = 0
self.int_buffer = None
self.byte_buffer = None
self._is_closed = True
self.request = None
if self.is_live:
self.worker.postMessage(_obj_from_dict({"close": self.connection_id}))
self.is_live = False
super().close()
def readable(self) -> bool:
return True
def writable(self) -> bool:
return False
def seekable(self) -> bool:
return False
def readinto(self, byte_obj: Buffer) -> int:
if not self.int_buffer:
raise _StreamingError(
"No buffer for stream in _ReadStream.readinto",
request=self.request,
response=None,
)
if self.read_len == 0:
# wait for the worker to send something
js.Atomics.store(self.int_buffer, 0, ERROR_TIMEOUT)
self.worker.postMessage(_obj_from_dict({"getMore": self.connection_id}))
if (
js.Atomics.wait(self.int_buffer, 0, ERROR_TIMEOUT, self.timeout)
== "timed-out"
):
raise _TimeoutError
data_len = self.int_buffer[0]
if data_len > 0:
self.read_len = data_len
self.read_pos = 0
elif data_len == ERROR_EXCEPTION:
string_len = self.int_buffer[1]
# decode the error string
js_decoder = js.TextDecoder.new()
json_str = js_decoder.decode(self.byte_buffer.slice(0, string_len))
raise _StreamingError(
f"Exception thrown in fetch: {json_str}",
request=self.request,
response=None,
)
else:
# EOF, free the buffers and return zero
# and free the request
self.is_live = False
self.close()
return 0
# copy from int32array to python bytes
ret_length = min(self.read_len, len(memoryview(byte_obj)))
subarray = self.byte_buffer.subarray(
self.read_pos, self.read_pos + ret_length
).to_py()
memoryview(byte_obj)[0:ret_length] = subarray
self.read_len -= ret_length
self.read_pos += ret_length
return ret_length
class _StreamingFetcher:
def __init__(self) -> None:
# make web-worker and data buffer on startup
self.streaming_ready = False
js_data_blob = js.Blob.new(
[_STREAMING_WORKER_CODE], _obj_from_dict({"type": "application/javascript"})
)
def promise_resolver(js_resolve_fn: JsProxy, js_reject_fn: JsProxy) -> None:
def onMsg(e: JsProxy) -> None:
self.streaming_ready = True
js_resolve_fn(e)
def onErr(e: JsProxy) -> None:
js_reject_fn(e) # Defensive: never happens in ci
self.js_worker.onmessage = onMsg
self.js_worker.onerror = onErr
js_data_url = js.URL.createObjectURL(js_data_blob)
self.js_worker = js.globalThis.Worker.new(js_data_url)
self.js_worker_ready_promise = js.globalThis.Promise.new(promise_resolver)
def send(self, request: EmscriptenRequest) -> EmscriptenResponse:
headers = {
k: v for k, v in request.headers.items() if k not in HEADERS_TO_IGNORE
}
body = request.body
fetch_data = {"headers": headers, "body": to_js(body), "method": request.method}
# start the request off in the worker
timeout = int(1000 * request.timeout) if request.timeout > 0 else None
js_shared_buffer = js.SharedArrayBuffer.new(1048576)
js_int_buffer = js.Int32Array.new(js_shared_buffer)
js_byte_buffer = js.Uint8Array.new(js_shared_buffer, 8)
js.Atomics.store(js_int_buffer, 0, ERROR_TIMEOUT)
js.Atomics.notify(js_int_buffer, 0)
js_absolute_url = js.URL.new(request.url, js.location).href
self.js_worker.postMessage(
_obj_from_dict(
{
"buffer": js_shared_buffer,
"url": js_absolute_url,
"fetchParams": fetch_data,
}
)
)
# wait for the worker to send something
js.Atomics.wait(js_int_buffer, 0, ERROR_TIMEOUT, timeout)
if js_int_buffer[0] == ERROR_TIMEOUT:
raise _TimeoutError(
"Timeout connecting to streaming request",
request=request,
response=None,
)
elif js_int_buffer[0] == SUCCESS_HEADER:
# got response
# header length is in second int of intBuffer
string_len = js_int_buffer[1]
# decode the rest to a JSON string
js_decoder = js.TextDecoder.new()
# this does a copy (the slice) because decode can't work on shared array
# for some silly reason
json_str = js_decoder.decode(js_byte_buffer.slice(0, string_len))
# get it as an object
response_obj = json.loads(json_str)
return EmscriptenResponse(
request=request,
status_code=response_obj["status"],
headers=response_obj["headers"],
body=_ReadStream(
js_int_buffer,
js_byte_buffer,
request.timeout,
self.js_worker,
response_obj["connectionID"],
request,
),
)
elif js_int_buffer[0] == ERROR_EXCEPTION:
string_len = js_int_buffer[1]
# decode the error string
js_decoder = js.TextDecoder.new()
json_str = js_decoder.decode(js_byte_buffer.slice(0, string_len))
raise _StreamingError(
f"Exception thrown in fetch: {json_str}", request=request, response=None
)
else:
raise _StreamingError(
f"Unknown status from worker in fetch: {js_int_buffer[0]}",
request=request,
response=None,
)
# check if we are in a worker or not
def is_in_browser_main_thread() -> bool:
return hasattr(js, "window") and hasattr(js, "self") and js.self == js.window
def is_cross_origin_isolated() -> bool:
return hasattr(js, "crossOriginIsolated") and js.crossOriginIsolated
def is_in_node() -> bool:
return (
hasattr(js, "process")
and hasattr(js.process, "release")
and hasattr(js.process.release, "name")
and js.process.release.name == "node"
)
def is_worker_available() -> bool:
return hasattr(js, "Worker") and hasattr(js, "Blob")
_fetcher: _StreamingFetcher | None = None
if is_worker_available() and (
(is_cross_origin_isolated() and not is_in_browser_main_thread())
and (not is_in_node())
):
_fetcher = _StreamingFetcher()
else:
_fetcher = None
def send_streaming_request(request: EmscriptenRequest) -> EmscriptenResponse | None:
if _fetcher and streaming_ready():
return _fetcher.send(request)
else:
_show_streaming_warning()
return None
_SHOWN_TIMEOUT_WARNING = False
def _show_timeout_warning() -> None:
global _SHOWN_TIMEOUT_WARNING
if not _SHOWN_TIMEOUT_WARNING:
_SHOWN_TIMEOUT_WARNING = True
message = "Warning: Timeout is not available on main browser thread"
js.console.warn(message)
_SHOWN_STREAMING_WARNING = False
def _show_streaming_warning() -> None:
global _SHOWN_STREAMING_WARNING
if not _SHOWN_STREAMING_WARNING:
_SHOWN_STREAMING_WARNING = True
message = "Can't stream HTTP requests because: \n"
if not is_cross_origin_isolated():
message += " Page is not cross-origin isolated\n"
if is_in_browser_main_thread():
message += " Python is running in main browser thread\n"
if not is_worker_available():
message += " Worker or Blob classes are not available in this environment." # Defensive: this is always False in browsers that we test in
if streaming_ready() is False:
message += """ Streaming fetch worker isn't ready. If you want to be sure that streaming fetch
is working, you need to call: 'await urllib3.contrib.emscripten.fetch.wait_for_streaming_ready()`"""
from js import console
console.warn(message)
def send_request(request: EmscriptenRequest) -> EmscriptenResponse:
try:
js_xhr = js.XMLHttpRequest.new()
if not is_in_browser_main_thread():
js_xhr.responseType = "arraybuffer"
if request.timeout:
js_xhr.timeout = int(request.timeout * 1000)
else:
js_xhr.overrideMimeType("text/plain; charset=ISO-8859-15")
if request.timeout:
# timeout isn't available on the main thread - show a warning in console
# if it is set
_show_timeout_warning()
js_xhr.open(request.method, request.url, False)
for name, value in request.headers.items():
if name.lower() not in HEADERS_TO_IGNORE:
js_xhr.setRequestHeader(name, value)
js_xhr.send(to_js(request.body))
headers = dict(Parser().parsestr(js_xhr.getAllResponseHeaders()))
if not is_in_browser_main_thread():
body = js_xhr.response.to_py().tobytes()
else:
body = js_xhr.response.encode("ISO-8859-15")
return EmscriptenResponse(
status_code=js_xhr.status, headers=headers, body=body, request=request
)
except JsException as err:
if err.name == "TimeoutError":
raise _TimeoutError(err.message, request=request)
elif err.name == "NetworkError":
raise _RequestError(err.message, request=request)
else:
# general http error
raise _RequestError(err.message, request=request)
def streaming_ready() -> bool | None:
if _fetcher:
return _fetcher.streaming_ready
else:
return None # no fetcher, return None to signify that
async def wait_for_streaming_ready() -> bool:
if _fetcher:
await _fetcher.js_worker_ready_promise
return True
else:
return False

View File

@ -1,22 +0,0 @@
from __future__ import annotations
from dataclasses import dataclass, field
from ..._base_connection import _TYPE_BODY
@dataclass
class EmscriptenRequest:
method: str
url: str
params: dict[str, str] | None = None
body: _TYPE_BODY | None = None
headers: dict[str, str] = field(default_factory=dict)
timeout: float = 0
decode_content: bool = True
def set_header(self, name: str, value: str) -> None:
self.headers[name.capitalize()] = value
def set_body(self, body: _TYPE_BODY | None) -> None:
self.body = body

View File

@ -1,276 +0,0 @@
from __future__ import annotations
import json as _json
import logging
import typing
from contextlib import contextmanager
from dataclasses import dataclass
from http.client import HTTPException as HTTPException
from io import BytesIO, IOBase
from ...exceptions import InvalidHeader, TimeoutError
from ...response import BaseHTTPResponse
from ...util.retry import Retry
from .request import EmscriptenRequest
if typing.TYPE_CHECKING:
from ..._base_connection import BaseHTTPConnection, BaseHTTPSConnection
log = logging.getLogger(__name__)
@dataclass
class EmscriptenResponse:
status_code: int
headers: dict[str, str]
body: IOBase | bytes
request: EmscriptenRequest
class EmscriptenHttpResponseWrapper(BaseHTTPResponse):
def __init__(
self,
internal_response: EmscriptenResponse,
url: str | None = None,
connection: BaseHTTPConnection | BaseHTTPSConnection | None = None,
):
self._pool = None # set by pool class
self._body = None
self._response = internal_response
self._url = url
self._connection = connection
self._closed = False
super().__init__(
headers=internal_response.headers,
status=internal_response.status_code,
request_url=url,
version=0,
reason="",
decode_content=True,
)
self.length_remaining = self._init_length(self._response.request.method)
self.length_is_certain = False
@property
def url(self) -> str | None:
return self._url
@url.setter
def url(self, url: str | None) -> None:
self._url = url
@property
def connection(self) -> BaseHTTPConnection | BaseHTTPSConnection | None:
return self._connection
@property
def retries(self) -> Retry | None:
return self._retries
@retries.setter
def retries(self, retries: Retry | None) -> None:
# Override the request_url if retries has a redirect location.
self._retries = retries
def stream(
self, amt: int | None = 2**16, decode_content: bool | None = None
) -> typing.Generator[bytes, None, None]:
"""
A generator wrapper for the read() method. A call will block until
``amt`` bytes have been read from the connection or until the
connection is closed.
:param amt:
How much of the content to read. The generator will return up to
much data per iteration, but may return less. This is particularly
likely when using compressed data. However, the empty string will
never be returned.
:param decode_content:
If True, will attempt to decode the body based on the
'content-encoding' header.
"""
while True:
data = self.read(amt=amt, decode_content=decode_content)
if data:
yield data
else:
break
def _init_length(self, request_method: str | None) -> int | None:
length: int | None
content_length: str | None = self.headers.get("content-length")
if content_length is not None:
try:
# RFC 7230 section 3.3.2 specifies multiple content lengths can
# be sent in a single Content-Length header
# (e.g. Content-Length: 42, 42). This line ensures the values
# are all valid ints and that as long as the `set` length is 1,
# all values are the same. Otherwise, the header is invalid.
lengths = {int(val) for val in content_length.split(",")}
if len(lengths) > 1:
raise InvalidHeader(
"Content-Length contained multiple "
"unmatching values (%s)" % content_length
)
length = lengths.pop()
except ValueError:
length = None
else:
if length < 0:
length = None
else: # if content_length is None
length = None
# Check for responses that shouldn't include a body
if (
self.status in (204, 304)
or 100 <= self.status < 200
or request_method == "HEAD"
):
length = 0
return length
def read(
self,
amt: int | None = None,
decode_content: bool | None = None, # ignored because browser decodes always
cache_content: bool = False,
) -> bytes:
if (
self._closed
or self._response is None
or (isinstance(self._response.body, IOBase) and self._response.body.closed)
):
return b""
with self._error_catcher():
# body has been preloaded as a string by XmlHttpRequest
if not isinstance(self._response.body, IOBase):
self.length_remaining = len(self._response.body)
self.length_is_certain = True
# wrap body in IOStream
self._response.body = BytesIO(self._response.body)
if amt is not None:
# don't cache partial content
cache_content = False
data = self._response.body.read(amt)
if self.length_remaining is not None:
self.length_remaining = max(self.length_remaining - len(data), 0)
if (self.length_is_certain and self.length_remaining == 0) or len(
data
) < amt:
# definitely finished reading, close response stream
self._response.body.close()
return typing.cast(bytes, data)
else: # read all we can (and cache it)
data = self._response.body.read()
if cache_content:
self._body = data
if self.length_remaining is not None:
self.length_remaining = max(self.length_remaining - len(data), 0)
if len(data) == 0 or (
self.length_is_certain and self.length_remaining == 0
):
# definitely finished reading, close response stream
self._response.body.close()
return typing.cast(bytes, data)
def read_chunked(
self,
amt: int | None = None,
decode_content: bool | None = None,
) -> typing.Generator[bytes, None, None]:
# chunked is handled by browser
while True:
bytes = self.read(amt, decode_content)
if not bytes:
break
yield bytes
def release_conn(self) -> None:
if not self._pool or not self._connection:
return None
self._pool._put_conn(self._connection)
self._connection = None
def drain_conn(self) -> None:
self.close()
@property
def data(self) -> bytes:
if self._body:
return self._body
else:
return self.read(cache_content=True)
def json(self) -> typing.Any:
"""
Parses the body of the HTTP response as JSON.
To use a custom JSON decoder pass the result of :attr:`HTTPResponse.data` to the decoder.
This method can raise either `UnicodeDecodeError` or `json.JSONDecodeError`.
Read more :ref:`here <json>`.
"""
data = self.data.decode("utf-8")
return _json.loads(data)
def close(self) -> None:
if not self._closed:
if isinstance(self._response.body, IOBase):
self._response.body.close()
if self._connection:
self._connection.close()
self._connection = None
self._closed = True
@contextmanager
def _error_catcher(self) -> typing.Generator[None, None, None]:
"""
Catch Emscripten specific exceptions thrown by fetch.py,
instead re-raising urllib3 variants, so that low-level exceptions
are not leaked in the high-level api.
On exit, release the connection back to the pool.
"""
from .fetch import _RequestError, _TimeoutError # avoid circular import
clean_exit = False
try:
yield
# If no exception is thrown, we should avoid cleaning up
# unnecessarily.
clean_exit = True
except _TimeoutError as e:
raise TimeoutError(str(e))
except _RequestError as e:
raise HTTPException(str(e))
finally:
# If we didn't terminate cleanly, we need to throw away our
# connection.
if not clean_exit:
# The response may not be closed but we're not going to use it
# anymore so close it now
if (
isinstance(self._response.body, IOBase)
and not self._response.body.closed
):
self._response.body.close()
# release the connection back to the pool
self.release_conn()
else:
# If we have read everything from the response stream,
# return the connection back to the pool.
if (
isinstance(self._response.body, IOBase)
and self._response.body.closed
):
self.release_conn()

View File

@ -1,548 +0,0 @@
"""
Module for using pyOpenSSL as a TLS backend. This module was relevant before
the standard library ``ssl`` module supported SNI, but now that we've dropped
support for Python 2.7 all relevant Python versions support SNI so
**this module is no longer recommended**.
This needs the following packages installed:
* `pyOpenSSL`_ (tested with 16.0.0)
* `cryptography`_ (minimum 1.3.4, from pyopenssl)
* `idna`_ (minimum 2.0)
However, pyOpenSSL depends on cryptography, so while we use all three directly here we
end up having relatively few packages required.
You can install them with the following command:
.. code-block:: bash
$ python -m pip install pyopenssl cryptography idna
To activate certificate checking, call
:func:`~urllib3.contrib.pyopenssl.inject_into_urllib3` from your Python code
before you begin making HTTP requests. This can be done in a ``sitecustomize``
module, or at any other time before your application begins using ``urllib3``,
like this:
.. code-block:: python
try:
import urllib3.contrib.pyopenssl
urllib3.contrib.pyopenssl.inject_into_urllib3()
except ImportError:
pass
.. _pyopenssl: https://www.pyopenssl.org
.. _cryptography: https://cryptography.io
.. _idna: https://github.com/kjd/idna
"""
from __future__ import annotations
import OpenSSL.SSL # type: ignore[import-untyped]
from cryptography import x509
try:
from cryptography.x509 import UnsupportedExtension # type: ignore[attr-defined]
except ImportError:
# UnsupportedExtension is gone in cryptography >= 2.1.0
class UnsupportedExtension(Exception): # type: ignore[no-redef]
pass
import logging
import ssl
import typing
from io import BytesIO
from socket import socket as socket_cls
from socket import timeout
from .. import util
if typing.TYPE_CHECKING:
from OpenSSL.crypto import X509 # type: ignore[import-untyped]
__all__ = ["inject_into_urllib3", "extract_from_urllib3"]
# Map from urllib3 to PyOpenSSL compatible parameter-values.
_openssl_versions: dict[int, int] = {
util.ssl_.PROTOCOL_TLS: OpenSSL.SSL.SSLv23_METHOD, # type: ignore[attr-defined]
util.ssl_.PROTOCOL_TLS_CLIENT: OpenSSL.SSL.SSLv23_METHOD, # type: ignore[attr-defined]
ssl.PROTOCOL_TLSv1: OpenSSL.SSL.TLSv1_METHOD,
}
if hasattr(ssl, "PROTOCOL_TLSv1_1") and hasattr(OpenSSL.SSL, "TLSv1_1_METHOD"):
_openssl_versions[ssl.PROTOCOL_TLSv1_1] = OpenSSL.SSL.TLSv1_1_METHOD
if hasattr(ssl, "PROTOCOL_TLSv1_2") and hasattr(OpenSSL.SSL, "TLSv1_2_METHOD"):
_openssl_versions[ssl.PROTOCOL_TLSv1_2] = OpenSSL.SSL.TLSv1_2_METHOD
_stdlib_to_openssl_verify = {
ssl.CERT_NONE: OpenSSL.SSL.VERIFY_NONE,
ssl.CERT_OPTIONAL: OpenSSL.SSL.VERIFY_PEER,
ssl.CERT_REQUIRED: OpenSSL.SSL.VERIFY_PEER
+ OpenSSL.SSL.VERIFY_FAIL_IF_NO_PEER_CERT,
}
_openssl_to_stdlib_verify = {v: k for k, v in _stdlib_to_openssl_verify.items()}
# The SSLvX values are the most likely to be missing in the future
# but we check them all just to be sure.
_OP_NO_SSLv2_OR_SSLv3: int = getattr(OpenSSL.SSL, "OP_NO_SSLv2", 0) | getattr(
OpenSSL.SSL, "OP_NO_SSLv3", 0
)
_OP_NO_TLSv1: int = getattr(OpenSSL.SSL, "OP_NO_TLSv1", 0)
_OP_NO_TLSv1_1: int = getattr(OpenSSL.SSL, "OP_NO_TLSv1_1", 0)
_OP_NO_TLSv1_2: int = getattr(OpenSSL.SSL, "OP_NO_TLSv1_2", 0)
_OP_NO_TLSv1_3: int = getattr(OpenSSL.SSL, "OP_NO_TLSv1_3", 0)
_openssl_to_ssl_minimum_version: dict[int, int] = {
ssl.TLSVersion.MINIMUM_SUPPORTED: _OP_NO_SSLv2_OR_SSLv3,
ssl.TLSVersion.TLSv1: _OP_NO_SSLv2_OR_SSLv3,
ssl.TLSVersion.TLSv1_1: _OP_NO_SSLv2_OR_SSLv3 | _OP_NO_TLSv1,
ssl.TLSVersion.TLSv1_2: _OP_NO_SSLv2_OR_SSLv3 | _OP_NO_TLSv1 | _OP_NO_TLSv1_1,
ssl.TLSVersion.TLSv1_3: (
_OP_NO_SSLv2_OR_SSLv3 | _OP_NO_TLSv1 | _OP_NO_TLSv1_1 | _OP_NO_TLSv1_2
),
ssl.TLSVersion.MAXIMUM_SUPPORTED: (
_OP_NO_SSLv2_OR_SSLv3 | _OP_NO_TLSv1 | _OP_NO_TLSv1_1 | _OP_NO_TLSv1_2
),
}
_openssl_to_ssl_maximum_version: dict[int, int] = {
ssl.TLSVersion.MINIMUM_SUPPORTED: (
_OP_NO_SSLv2_OR_SSLv3
| _OP_NO_TLSv1
| _OP_NO_TLSv1_1
| _OP_NO_TLSv1_2
| _OP_NO_TLSv1_3
),
ssl.TLSVersion.TLSv1: (
_OP_NO_SSLv2_OR_SSLv3 | _OP_NO_TLSv1_1 | _OP_NO_TLSv1_2 | _OP_NO_TLSv1_3
),
ssl.TLSVersion.TLSv1_1: _OP_NO_SSLv2_OR_SSLv3 | _OP_NO_TLSv1_2 | _OP_NO_TLSv1_3,
ssl.TLSVersion.TLSv1_2: _OP_NO_SSLv2_OR_SSLv3 | _OP_NO_TLSv1_3,
ssl.TLSVersion.TLSv1_3: _OP_NO_SSLv2_OR_SSLv3,
ssl.TLSVersion.MAXIMUM_SUPPORTED: _OP_NO_SSLv2_OR_SSLv3,
}
# OpenSSL will only write 16K at a time
SSL_WRITE_BLOCKSIZE = 16384
orig_util_SSLContext = util.ssl_.SSLContext
log = logging.getLogger(__name__)
def inject_into_urllib3() -> None:
"Monkey-patch urllib3 with PyOpenSSL-backed SSL-support."
_validate_dependencies_met()
util.SSLContext = PyOpenSSLContext # type: ignore[assignment]
util.ssl_.SSLContext = PyOpenSSLContext # type: ignore[assignment]
util.IS_PYOPENSSL = True
util.ssl_.IS_PYOPENSSL = True
def extract_from_urllib3() -> None:
"Undo monkey-patching by :func:`inject_into_urllib3`."
util.SSLContext = orig_util_SSLContext
util.ssl_.SSLContext = orig_util_SSLContext
util.IS_PYOPENSSL = False
util.ssl_.IS_PYOPENSSL = False
def _validate_dependencies_met() -> None:
"""
Verifies that PyOpenSSL's package-level dependencies have been met.
Throws `ImportError` if they are not met.
"""
# Method added in `cryptography==1.1`; not available in older versions
from cryptography.x509.extensions import Extensions
if getattr(Extensions, "get_extension_for_class", None) is None:
raise ImportError(
"'cryptography' module missing required functionality. "
"Try upgrading to v1.3.4 or newer."
)
# pyOpenSSL 0.14 and above use cryptography for OpenSSL bindings. The _x509
# attribute is only present on those versions.
from OpenSSL.crypto import X509
x509 = X509()
if getattr(x509, "_x509", None) is None:
raise ImportError(
"'pyOpenSSL' module missing required functionality. "
"Try upgrading to v0.14 or newer."
)
def _dnsname_to_stdlib(name: str) -> str | None:
"""
Converts a dNSName SubjectAlternativeName field to the form used by the
standard library on the given Python version.
Cryptography produces a dNSName as a unicode string that was idna-decoded
from ASCII bytes. We need to idna-encode that string to get it back, and
then on Python 3 we also need to convert to unicode via UTF-8 (the stdlib
uses PyUnicode_FromStringAndSize on it, which decodes via UTF-8).
If the name cannot be idna-encoded then we return None signalling that
the name given should be skipped.
"""
def idna_encode(name: str) -> bytes | None:
"""
Borrowed wholesale from the Python Cryptography Project. It turns out
that we can't just safely call `idna.encode`: it can explode for
wildcard names. This avoids that problem.
"""
import idna
try:
for prefix in ["*.", "."]:
if name.startswith(prefix):
name = name[len(prefix) :]
return prefix.encode("ascii") + idna.encode(name)
return idna.encode(name)
except idna.core.IDNAError:
return None
# Don't send IPv6 addresses through the IDNA encoder.
if ":" in name:
return name
encoded_name = idna_encode(name)
if encoded_name is None:
return None
return encoded_name.decode("utf-8")
def get_subj_alt_name(peer_cert: X509) -> list[tuple[str, str]]:
"""
Given an PyOpenSSL certificate, provides all the subject alternative names.
"""
cert = peer_cert.to_cryptography()
# We want to find the SAN extension. Ask Cryptography to locate it (it's
# faster than looping in Python)
try:
ext = cert.extensions.get_extension_for_class(x509.SubjectAlternativeName).value
except x509.ExtensionNotFound:
# No such extension, return the empty list.
return []
except (
x509.DuplicateExtension,
UnsupportedExtension,
x509.UnsupportedGeneralNameType,
UnicodeError,
) as e:
# A problem has been found with the quality of the certificate. Assume
# no SAN field is present.
log.warning(
"A problem was encountered with the certificate that prevented "
"urllib3 from finding the SubjectAlternativeName field. This can "
"affect certificate validation. The error was %s",
e,
)
return []
# We want to return dNSName and iPAddress fields. We need to cast the IPs
# back to strings because the match_hostname function wants them as
# strings.
# Sadly the DNS names need to be idna encoded and then, on Python 3, UTF-8
# decoded. This is pretty frustrating, but that's what the standard library
# does with certificates, and so we need to attempt to do the same.
# We also want to skip over names which cannot be idna encoded.
names = [
("DNS", name)
for name in map(_dnsname_to_stdlib, ext.get_values_for_type(x509.DNSName))
if name is not None
]
names.extend(
("IP Address", str(name)) for name in ext.get_values_for_type(x509.IPAddress)
)
return names
class WrappedSocket:
"""API-compatibility wrapper for Python OpenSSL's Connection-class."""
def __init__(
self,
connection: OpenSSL.SSL.Connection,
socket: socket_cls,
suppress_ragged_eofs: bool = True,
) -> None:
self.connection = connection
self.socket = socket
self.suppress_ragged_eofs = suppress_ragged_eofs
self._io_refs = 0
self._closed = False
def fileno(self) -> int:
return self.socket.fileno()
# Copy-pasted from Python 3.5 source code
def _decref_socketios(self) -> None:
if self._io_refs > 0:
self._io_refs -= 1
if self._closed:
self.close()
def recv(self, *args: typing.Any, **kwargs: typing.Any) -> bytes:
try:
data = self.connection.recv(*args, **kwargs)
except OpenSSL.SSL.SysCallError as e:
if self.suppress_ragged_eofs and e.args == (-1, "Unexpected EOF"):
return b""
else:
raise OSError(e.args[0], str(e)) from e
except OpenSSL.SSL.ZeroReturnError:
if self.connection.get_shutdown() == OpenSSL.SSL.RECEIVED_SHUTDOWN:
return b""
else:
raise
except OpenSSL.SSL.WantReadError as e:
if not util.wait_for_read(self.socket, self.socket.gettimeout()):
raise timeout("The read operation timed out") from e
else:
return self.recv(*args, **kwargs)
# TLS 1.3 post-handshake authentication
except OpenSSL.SSL.Error as e:
raise ssl.SSLError(f"read error: {e!r}") from e
else:
return data # type: ignore[no-any-return]
def recv_into(self, *args: typing.Any, **kwargs: typing.Any) -> int:
try:
return self.connection.recv_into(*args, **kwargs) # type: ignore[no-any-return]
except OpenSSL.SSL.SysCallError as e:
if self.suppress_ragged_eofs and e.args == (-1, "Unexpected EOF"):
return 0
else:
raise OSError(e.args[0], str(e)) from e
except OpenSSL.SSL.ZeroReturnError:
if self.connection.get_shutdown() == OpenSSL.SSL.RECEIVED_SHUTDOWN:
return 0
else:
raise
except OpenSSL.SSL.WantReadError as e:
if not util.wait_for_read(self.socket, self.socket.gettimeout()):
raise timeout("The read operation timed out") from e
else:
return self.recv_into(*args, **kwargs)
# TLS 1.3 post-handshake authentication
except OpenSSL.SSL.Error as e:
raise ssl.SSLError(f"read error: {e!r}") from e
def settimeout(self, timeout: float) -> None:
return self.socket.settimeout(timeout)
def _send_until_done(self, data: bytes) -> int:
while True:
try:
return self.connection.send(data) # type: ignore[no-any-return]
except OpenSSL.SSL.WantWriteError as e:
if not util.wait_for_write(self.socket, self.socket.gettimeout()):
raise timeout() from e
continue
except OpenSSL.SSL.SysCallError as e:
raise OSError(e.args[0], str(e)) from e
def sendall(self, data: bytes) -> None:
total_sent = 0
while total_sent < len(data):
sent = self._send_until_done(
data[total_sent : total_sent + SSL_WRITE_BLOCKSIZE]
)
total_sent += sent
def shutdown(self) -> None:
# FIXME rethrow compatible exceptions should we ever use this
self.connection.shutdown()
def close(self) -> None:
self._closed = True
if self._io_refs <= 0:
self._real_close()
def _real_close(self) -> None:
try:
return self.connection.close() # type: ignore[no-any-return]
except OpenSSL.SSL.Error:
return
def getpeercert(
self, binary_form: bool = False
) -> dict[str, list[typing.Any]] | None:
x509 = self.connection.get_peer_certificate()
if not x509:
return x509 # type: ignore[no-any-return]
if binary_form:
return OpenSSL.crypto.dump_certificate(OpenSSL.crypto.FILETYPE_ASN1, x509) # type: ignore[no-any-return]
return {
"subject": ((("commonName", x509.get_subject().CN),),), # type: ignore[dict-item]
"subjectAltName": get_subj_alt_name(x509),
}
def version(self) -> str:
return self.connection.get_protocol_version_name() # type: ignore[no-any-return]
WrappedSocket.makefile = socket_cls.makefile # type: ignore[attr-defined]
class PyOpenSSLContext:
"""
I am a wrapper class for the PyOpenSSL ``Context`` object. I am responsible
for translating the interface of the standard library ``SSLContext`` object
to calls into PyOpenSSL.
"""
def __init__(self, protocol: int) -> None:
self.protocol = _openssl_versions[protocol]
self._ctx = OpenSSL.SSL.Context(self.protocol)
self._options = 0
self.check_hostname = False
self._minimum_version: int = ssl.TLSVersion.MINIMUM_SUPPORTED
self._maximum_version: int = ssl.TLSVersion.MAXIMUM_SUPPORTED
@property
def options(self) -> int:
return self._options
@options.setter
def options(self, value: int) -> None:
self._options = value
self._set_ctx_options()
@property
def verify_mode(self) -> int:
return _openssl_to_stdlib_verify[self._ctx.get_verify_mode()]
@verify_mode.setter
def verify_mode(self, value: ssl.VerifyMode) -> None:
self._ctx.set_verify(_stdlib_to_openssl_verify[value], _verify_callback)
def set_default_verify_paths(self) -> None:
self._ctx.set_default_verify_paths()
def set_ciphers(self, ciphers: bytes | str) -> None:
if isinstance(ciphers, str):
ciphers = ciphers.encode("utf-8")
self._ctx.set_cipher_list(ciphers)
def load_verify_locations(
self,
cafile: str | None = None,
capath: str | None = None,
cadata: bytes | None = None,
) -> None:
if cafile is not None:
cafile = cafile.encode("utf-8") # type: ignore[assignment]
if capath is not None:
capath = capath.encode("utf-8") # type: ignore[assignment]
try:
self._ctx.load_verify_locations(cafile, capath)
if cadata is not None:
self._ctx.load_verify_locations(BytesIO(cadata))
except OpenSSL.SSL.Error as e:
raise ssl.SSLError(f"unable to load trusted certificates: {e!r}") from e
def load_cert_chain(
self,
certfile: str,
keyfile: str | None = None,
password: str | None = None,
) -> None:
try:
self._ctx.use_certificate_chain_file(certfile)
if password is not None:
if not isinstance(password, bytes):
password = password.encode("utf-8") # type: ignore[assignment]
self._ctx.set_passwd_cb(lambda *_: password)
self._ctx.use_privatekey_file(keyfile or certfile)
except OpenSSL.SSL.Error as e:
raise ssl.SSLError(f"Unable to load certificate chain: {e!r}") from e
def set_alpn_protocols(self, protocols: list[bytes | str]) -> None:
protocols = [util.util.to_bytes(p, "ascii") for p in protocols]
return self._ctx.set_alpn_protos(protocols) # type: ignore[no-any-return]
def wrap_socket(
self,
sock: socket_cls,
server_side: bool = False,
do_handshake_on_connect: bool = True,
suppress_ragged_eofs: bool = True,
server_hostname: bytes | str | None = None,
) -> WrappedSocket:
cnx = OpenSSL.SSL.Connection(self._ctx, sock)
# If server_hostname is an IP, don't use it for SNI, per RFC6066 Section 3
if server_hostname and not util.ssl_.is_ipaddress(server_hostname):
if isinstance(server_hostname, str):
server_hostname = server_hostname.encode("utf-8")
cnx.set_tlsext_host_name(server_hostname)
cnx.set_connect_state()
while True:
try:
cnx.do_handshake()
except OpenSSL.SSL.WantReadError as e:
if not util.wait_for_read(sock, sock.gettimeout()):
raise timeout("select timed out") from e
continue
except OpenSSL.SSL.Error as e:
raise ssl.SSLError(f"bad handshake: {e!r}") from e
break
return WrappedSocket(cnx, sock)
def _set_ctx_options(self) -> None:
self._ctx.set_options(
self._options
| _openssl_to_ssl_minimum_version[self._minimum_version]
| _openssl_to_ssl_maximum_version[self._maximum_version]
)
@property
def minimum_version(self) -> int:
return self._minimum_version
@minimum_version.setter
def minimum_version(self, minimum_version: int) -> None:
self._minimum_version = minimum_version
self._set_ctx_options()
@property
def maximum_version(self) -> int:
return self._maximum_version
@maximum_version.setter
def maximum_version(self, maximum_version: int) -> None:
self._maximum_version = maximum_version
self._set_ctx_options()
def _verify_callback(
cnx: OpenSSL.SSL.Connection,
x509: X509,
err_no: int,
err_depth: int,
return_code: int,
) -> bool:
return err_no == 0

View File

@ -1,230 +0,0 @@
"""
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

View File

@ -1,321 +0,0 @@
from __future__ import annotations
import socket
import typing
import warnings
from email.errors import MessageDefect
from http.client import IncompleteRead as httplib_IncompleteRead
if typing.TYPE_CHECKING:
from .connection import HTTPConnection
from .connectionpool import ConnectionPool
from .response import HTTPResponse
from .util.retry import Retry
# Base Exceptions
class HTTPError(Exception):
"""Base exception used by this module."""
class HTTPWarning(Warning):
"""Base warning used by this module."""
_TYPE_REDUCE_RESULT = typing.Tuple[
typing.Callable[..., object], typing.Tuple[object, ...]
]
class PoolError(HTTPError):
"""Base exception for errors caused within a pool."""
def __init__(self, pool: ConnectionPool, message: str) -> None:
self.pool = pool
super().__init__(f"{pool}: {message}")
def __reduce__(self) -> _TYPE_REDUCE_RESULT:
# For pickling purposes.
return self.__class__, (None, None)
class RequestError(PoolError):
"""Base exception for PoolErrors that have associated URLs."""
def __init__(self, pool: ConnectionPool, url: str, message: str) -> None:
self.url = url
super().__init__(pool, message)
def __reduce__(self) -> _TYPE_REDUCE_RESULT:
# For pickling purposes.
return self.__class__, (None, self.url, None)
class SSLError(HTTPError):
"""Raised when SSL certificate fails in an HTTPS connection."""
class ProxyError(HTTPError):
"""Raised when the connection to a proxy fails."""
# The original error is also available as __cause__.
original_error: Exception
def __init__(self, message: str, error: Exception) -> None:
super().__init__(message, error)
self.original_error = error
class DecodeError(HTTPError):
"""Raised when automatic decoding based on Content-Type fails."""
class ProtocolError(HTTPError):
"""Raised when something unexpected happens mid-request/response."""
#: 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 str url: The requested Url
:param reason: The underlying error
:type reason: :class:`Exception`
"""
def __init__(
self, pool: ConnectionPool, url: str, reason: Exception | None = None
) -> None:
self.reason = reason
message = f"Max retries exceeded with url: {url} (Caused by {reason!r})"
super().__init__(pool, url, message)
class HostChangedError(RequestError):
"""Raised when an existing pool gets a request for a foreign host."""
def __init__(
self, pool: ConnectionPool, url: str, retries: Retry | int = 3
) -> None:
message = f"Tried to open a foreign host with url: {url}"
super().__init__(pool, url, message)
self.retries = retries
class TimeoutStateError(HTTPError):
"""Raised when passing an invalid state to a timeout"""
class TimeoutError(HTTPError):
"""Raised when a socket timeout error occurs.
Catching this error will catch both :exc:`ReadTimeoutErrors
<ReadTimeoutError>` and :exc:`ConnectTimeoutErrors <ConnectTimeoutError>`.
"""
class ReadTimeoutError(TimeoutError, RequestError):
"""Raised when a socket timeout occurs while receiving data from a server"""
# 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"""
class NewConnectionError(ConnectTimeoutError, HTTPError):
"""Raised when we fail to establish a new connection. Usually ECONNREFUSED."""
def __init__(self, conn: HTTPConnection, message: str) -> None:
self.conn = conn
super().__init__(f"{conn}: {message}")
@property
def pool(self) -> HTTPConnection:
warnings.warn(
"The 'pool' property is deprecated and will be removed "
"in urllib3 v2.1.0. Use 'conn' instead.",
DeprecationWarning,
stacklevel=2,
)
return self.conn
class NameResolutionError(NewConnectionError):
"""Raised when host name resolution fails."""
def __init__(self, host: str, conn: HTTPConnection, reason: socket.gaierror):
message = f"Failed to resolve '{host}' ({reason})"
super().__init__(conn, message)
class EmptyPoolError(PoolError):
"""Raised when a pool runs out of connections and no more are allowed."""
class FullPoolError(PoolError):
"""Raised when we try to add a connection to a full pool in blocking mode."""
class ClosedPoolError(PoolError):
"""Raised when a request enters a pool after the pool has been closed."""
class LocationValueError(ValueError, HTTPError):
"""Raised when there is something wrong with a given URL input."""
class LocationParseError(LocationValueError):
"""Raised when get_host or similar fails to parse the URL input."""
def __init__(self, location: str) -> None:
message = f"Failed to parse: {location}"
super().__init__(message)
self.location = location
class URLSchemeUnknown(LocationValueError):
"""Raised when a URL input has an unsupported scheme."""
def __init__(self, scheme: str):
message = f"Not supported URL scheme {scheme}"
super().__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"""
class InsecureRequestWarning(SecurityWarning):
"""Warned when making an unverified HTTPS request."""
class NotOpenSSLWarning(SecurityWarning):
"""Warned when using unsupported SSL library"""
class SystemTimeWarning(SecurityWarning):
"""Warned when system time is suspected to be wrong"""
class InsecurePlatformWarning(SecurityWarning):
"""Warned when certain TLS/SSL configuration is not available on a platform."""
class DependencyWarning(HTTPWarning):
"""
Warned when an attempt is made to import a module with missing optional
dependencies.
"""
class ResponseNotChunked(ProtocolError, ValueError):
"""Response needs to be chunked in order to read it as chunks."""
class BodyNotHttplibCompatible(HTTPError):
"""
Body should be :class:`http.client.HTTPResponse` like
(have an fp attribute which returns raw chunks) for read_chunked().
"""
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.
"""
partial: int # type: ignore[assignment]
expected: int
def __init__(self, partial: int, expected: int) -> None:
self.partial = partial
self.expected = expected
def __repr__(self) -> str:
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: HTTPResponse, length: bytes) -> None:
self.partial: int = response.tell() # type: ignore[assignment]
self.expected: int | None = response.length_remaining
self.response = response
self.length = length
def __repr__(self) -> str:
return "InvalidChunkLength(got length %r, %i bytes read)" % (
self.length,
self.partial,
)
class InvalidHeader(HTTPError):
"""The header provided was somehow invalid."""
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: str | None) -> None:
# '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 = f"Proxy URL had unsupported scheme {scheme}, should use http:// or https://"
super().__init__(message)
class ProxySchemeUnsupported(ValueError):
"""Fetching HTTPS resources through HTTPS proxies is unsupported"""
class HeaderParsingError(HTTPError):
"""Raised by assert_header_parsing, but we convert it to a log.warning statement."""
def __init__(
self, defects: list[MessageDefect], unparsed_data: bytes | str | None
) -> None:
message = f"{defects or 'Unknown'}, unparsed data: {unparsed_data!r}"
super().__init__(message)
class UnrewindableBodyError(HTTPError):
"""urllib3 encountered an error when trying to rewind a body"""

View File

@ -1,341 +0,0 @@
from __future__ import annotations
import email.utils
import mimetypes
import typing
_TYPE_FIELD_VALUE = typing.Union[str, bytes]
_TYPE_FIELD_VALUE_TUPLE = typing.Union[
_TYPE_FIELD_VALUE,
typing.Tuple[str, _TYPE_FIELD_VALUE],
typing.Tuple[str, _TYPE_FIELD_VALUE, str],
]
def guess_content_type(
filename: str | None, default: str = "application/octet-stream"
) -> str:
"""
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: str, value: _TYPE_FIELD_VALUE) -> str:
"""
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``.
:returns:
An RFC-2231-formatted unicode string.
.. deprecated:: 2.0.0
Will be removed in urllib3 v2.1.0. This is not valid for
``multipart/form-data`` header parameters.
"""
import warnings
warnings.warn(
"'format_header_param_rfc2231' is deprecated and will be "
"removed in urllib3 v2.1.0. This is not valid for "
"multipart/form-data header parameters.",
DeprecationWarning,
stacklevel=2,
)
if isinstance(value, bytes):
value = value.decode("utf-8")
if not any(ch in value for ch in '"\\\r\n'):
result = f'{name}="{value}"'
try:
result.encode("ascii")
except (UnicodeEncodeError, UnicodeDecodeError):
pass
else:
return result
value = email.utils.encode_rfc2231(value, "utf-8")
value = f"{name}*={value}"
return value
def format_multipart_header_param(name: str, value: _TYPE_FIELD_VALUE) -> str:
"""
Format and quote a single multipart header parameter.
This follows the `WHATWG HTML Standard`_ as of 2021/06/10, matching
the behavior of current browser and curl versions. Values are
assumed to be UTF-8. The ``\\n``, ``\\r``, and ``"`` characters are
percent encoded.
.. _WHATWG HTML Standard:
https://html.spec.whatwg.org/multipage/
form-control-infrastructure.html#multipart-form-data
:param name:
The name of the parameter, an ASCII-only ``str``.
:param value:
The value of the parameter, a ``str`` or UTF-8 encoded
``bytes``.
:returns:
A string ``name="value"`` with the escaped value.
.. versionchanged:: 2.0.0
Matches the WHATWG HTML Standard as of 2021/06/10. Control
characters are no longer percent encoded.
.. versionchanged:: 2.0.0
Renamed from ``format_header_param_html5`` and
``format_header_param``. The old names will be removed in
urllib3 v2.1.0.
"""
if isinstance(value, bytes):
value = value.decode("utf-8")
# percent encode \n \r "
value = value.translate({10: "%0A", 13: "%0D", 34: "%22"})
return f'{name}="{value}"'
def format_header_param_html5(name: str, value: _TYPE_FIELD_VALUE) -> str:
"""
.. deprecated:: 2.0.0
Renamed to :func:`format_multipart_header_param`. Will be
removed in urllib3 v2.1.0.
"""
import warnings
warnings.warn(
"'format_header_param_html5' has been renamed to "
"'format_multipart_header_param'. The old name will be "
"removed in urllib3 v2.1.0.",
DeprecationWarning,
stacklevel=2,
)
return format_multipart_header_param(name, value)
def format_header_param(name: str, value: _TYPE_FIELD_VALUE) -> str:
"""
.. deprecated:: 2.0.0
Renamed to :func:`format_multipart_header_param`. Will be
removed in urllib3 v2.1.0.
"""
import warnings
warnings.warn(
"'format_header_param' has been renamed to "
"'format_multipart_header_param'. The old name will be "
"removed in urllib3 v2.1.0.",
DeprecationWarning,
stacklevel=2,
)
return format_multipart_header_param(name, value)
class RequestField:
"""
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.
.. versionchanged:: 2.0.0
The ``header_formatter`` parameter is deprecated and will
be removed in urllib3 v2.1.0.
"""
def __init__(
self,
name: str,
data: _TYPE_FIELD_VALUE,
filename: str | None = None,
headers: typing.Mapping[str, str] | None = None,
header_formatter: typing.Callable[[str, _TYPE_FIELD_VALUE], str] | None = None,
):
self._name = name
self._filename = filename
self.data = data
self.headers: dict[str, str | None] = {}
if headers:
self.headers = dict(headers)
if header_formatter is not None:
import warnings
warnings.warn(
"The 'header_formatter' parameter is deprecated and "
"will be removed in urllib3 v2.1.0.",
DeprecationWarning,
stacklevel=2,
)
self.header_formatter = header_formatter
else:
self.header_formatter = format_multipart_header_param
@classmethod
def from_tuples(
cls,
fieldname: str,
value: _TYPE_FIELD_VALUE_TUPLE,
header_formatter: typing.Callable[[str, _TYPE_FIELD_VALUE], str] | None = None,
) -> RequestField:
"""
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.
"""
filename: str | None
content_type: str | None
data: _TYPE_FIELD_VALUE
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: str, value: _TYPE_FIELD_VALUE) -> str:
"""
Override this method to change how each multipart header
parameter is formatted. By default, this calls
:func:`format_multipart_header_param`.
:param name:
The name of the parameter, an ASCII-only ``str``.
:param value:
The value of the parameter, a ``str`` or UTF-8 encoded
``bytes``.
:meta public:
"""
return self.header_formatter(name, value)
def _render_parts(
self,
header_parts: (
dict[str, _TYPE_FIELD_VALUE | None]
| typing.Sequence[tuple[str, _TYPE_FIELD_VALUE | None]]
),
) -> str:
"""
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"; ...`.
"""
iterable: typing.Iterable[tuple[str, _TYPE_FIELD_VALUE | None]]
parts = []
if isinstance(header_parts, dict):
iterable = header_parts.items()
else:
iterable = header_parts
for name, value in iterable:
if value is not None:
parts.append(self._render_part(name, value))
return "; ".join(parts)
def render_headers(self) -> str:
"""
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(f"{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(f"{header_name}: {header_value}")
lines.append("\r\n")
return "\r\n".join(lines)
def make_multipart(
self,
content_disposition: str | None = None,
content_type: str | None = None,
content_location: str | None = None,
) -> 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_disposition:
The 'Content-Disposition' of the request body. Defaults to 'form-data'
:param content_type:
The 'Content-Type' of the request body.
:param content_location:
The 'Content-Location' of the request body.
"""
content_disposition = (content_disposition or "form-data") + "; ".join(
[
"",
self._render_parts(
(("name", self._name), ("filename", self._filename))
),
]
)
self.headers["Content-Disposition"] = content_disposition
self.headers["Content-Type"] = content_type
self.headers["Content-Location"] = content_location

View File

@ -1,89 +0,0 @@
from __future__ import annotations
import binascii
import codecs
import os
import typing
from io import BytesIO
from .fields import _TYPE_FIELD_VALUE_TUPLE, RequestField
writer = codecs.lookup("utf-8")[3]
_TYPE_FIELDS_SEQUENCE = typing.Sequence[
typing.Union[typing.Tuple[str, _TYPE_FIELD_VALUE_TUPLE], RequestField]
]
_TYPE_FIELDS = typing.Union[
_TYPE_FIELDS_SEQUENCE,
typing.Mapping[str, _TYPE_FIELD_VALUE_TUPLE],
]
def choose_boundary() -> str:
"""
Our embarrassingly-simple replacement for mimetools.choose_boundary.
"""
return binascii.hexlify(os.urandom(16)).decode()
def iter_field_objects(fields: _TYPE_FIELDS) -> typing.Iterable[RequestField]:
"""
Iterate over fields.
Supports list of (k, v) tuples and dicts, and lists of
:class:`~urllib3.fields.RequestField`.
"""
iterable: typing.Iterable[RequestField | tuple[str, _TYPE_FIELD_VALUE_TUPLE]]
if isinstance(fields, typing.Mapping):
iterable = fields.items()
else:
iterable = fields
for field in iterable:
if isinstance(field, RequestField):
yield field
else:
yield RequestField.from_tuples(*field)
def encode_multipart_formdata(
fields: _TYPE_FIELDS, boundary: str | None = None
) -> tuple[bytes, str]:
"""
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`).
Values are processed by :func:`urllib3.fields.RequestField.from_tuples`.
: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(f"--{boundary}\r\n".encode("latin-1"))
writer(body).write(field.render_headers())
data = field.data
if isinstance(data, int):
data = str(data) # Backwards compatibility
if isinstance(data, str):
writer(body).write(data)
else:
body.write(data)
body.write(b"\r\n")
body.write(f"--{boundary}--\r\n".encode("latin-1"))
content_type = f"multipart/form-data; boundary={boundary}"
return body.getvalue(), content_type

View File

@ -1,229 +0,0 @@
from __future__ import annotations
import threading
import types
import typing
import h2.config # type: ignore[import-untyped]
import h2.connection # type: ignore[import-untyped]
import h2.events # type: ignore[import-untyped]
import urllib3.connection
import urllib3.util.ssl_
from urllib3.response import BaseHTTPResponse
from ._collections import HTTPHeaderDict
from .connection import HTTPSConnection
from .connectionpool import HTTPSConnectionPool
orig_HTTPSConnection = HTTPSConnection
T = typing.TypeVar("T")
class _LockedObject(typing.Generic[T]):
"""
A wrapper class that hides a specific object behind a lock.
The goal here is to provide a simple way to protect access to an object
that cannot safely be simultaneously accessed from multiple threads. The
intended use of this class is simple: take hold of it with a context
manager, which returns the protected object.
"""
def __init__(self, obj: T):
self.lock = threading.RLock()
self._obj = obj
def __enter__(self) -> T:
self.lock.acquire()
return self._obj
def __exit__(
self,
exc_type: type[BaseException] | None,
exc_val: BaseException | None,
exc_tb: types.TracebackType | None,
) -> None:
self.lock.release()
class HTTP2Connection(HTTPSConnection):
def __init__(
self, host: str, port: int | None = None, **kwargs: typing.Any
) -> None:
self._h2_conn = self._new_h2_conn()
self._h2_stream: int | None = None
self._h2_headers: list[tuple[bytes, bytes]] = []
if "proxy" in kwargs or "proxy_config" in kwargs: # Defensive:
raise NotImplementedError("Proxies aren't supported with HTTP/2")
super().__init__(host, port, **kwargs)
def _new_h2_conn(self) -> _LockedObject[h2.connection.H2Connection]:
config = h2.config.H2Configuration(client_side=True)
return _LockedObject(h2.connection.H2Connection(config=config))
def connect(self) -> None:
super().connect()
with self._h2_conn as h2_conn:
h2_conn.initiate_connection()
self.sock.sendall(h2_conn.data_to_send())
def putrequest(
self,
method: str,
url: str,
skip_host: bool = False,
skip_accept_encoding: bool = False,
) -> None:
with self._h2_conn as h2_conn:
self._request_url = url
self._h2_stream = h2_conn.get_next_available_stream_id()
if ":" in self.host:
authority = f"[{self.host}]:{self.port or 443}"
else:
authority = f"{self.host}:{self.port or 443}"
self._h2_headers.extend(
(
(b":scheme", b"https"),
(b":method", method.encode()),
(b":authority", authority.encode()),
(b":path", url.encode()),
)
)
def putheader(self, header: str, *values: str) -> None: # type: ignore[override]
for value in values:
self._h2_headers.append(
(header.encode("utf-8").lower(), value.encode("utf-8"))
)
def endheaders(self) -> None: # type: ignore[override]
with self._h2_conn as h2_conn:
h2_conn.send_headers(
stream_id=self._h2_stream,
headers=self._h2_headers,
end_stream=True,
)
if data_to_send := h2_conn.data_to_send():
self.sock.sendall(data_to_send)
def send(self, data: bytes) -> None: # type: ignore[override] # Defensive:
if not data:
return
raise NotImplementedError("Sending data isn't supported yet")
def getresponse( # type: ignore[override]
self,
) -> HTTP2Response:
status = None
data = bytearray()
with self._h2_conn as h2_conn:
end_stream = False
while not end_stream:
# TODO: Arbitrary read value.
if received_data := self.sock.recv(65535):
events = h2_conn.receive_data(received_data)
for event in events:
if isinstance(event, h2.events.ResponseReceived):
headers = HTTPHeaderDict()
for header, value in event.headers:
if header == b":status":
status = int(value.decode())
else:
headers.add(
header.decode("ascii"), value.decode("ascii")
)
elif isinstance(event, h2.events.DataReceived):
data += event.data
h2_conn.acknowledge_received_data(
event.flow_controlled_length, event.stream_id
)
elif isinstance(event, h2.events.StreamEnded):
end_stream = True
if data_to_send := h2_conn.data_to_send():
self.sock.sendall(data_to_send)
# We always close to not have to handle connection management.
self.close()
assert status is not None
return HTTP2Response(
status=status,
headers=headers,
request_url=self._request_url,
data=bytes(data),
)
def close(self) -> None:
with self._h2_conn as h2_conn:
try:
h2_conn.close_connection()
if data := h2_conn.data_to_send():
self.sock.sendall(data)
except Exception:
pass
# Reset all our HTTP/2 connection state.
self._h2_conn = self._new_h2_conn()
self._h2_stream = None
self._h2_headers = []
super().close()
class HTTP2Response(BaseHTTPResponse):
# TODO: This is a woefully incomplete response object, but works for non-streaming.
def __init__(
self,
status: int,
headers: HTTPHeaderDict,
request_url: str,
data: bytes,
decode_content: bool = False, # TODO: support decoding
) -> None:
super().__init__(
status=status,
headers=headers,
# Following CPython, we map HTTP versions to major * 10 + minor integers
version=20,
# No reason phrase in HTTP/2
reason=None,
decode_content=decode_content,
request_url=request_url,
)
self._data = data
self.length_remaining = 0
@property
def data(self) -> bytes:
return self._data
def get_redirect_location(self) -> None:
return None
def close(self) -> None:
pass
def inject_into_urllib3() -> None:
HTTPSConnectionPool.ConnectionCls = HTTP2Connection
urllib3.connection.HTTPSConnection = HTTP2Connection # type: ignore[misc]
# TODO: Offer 'http/1.1' as well, but for testing purposes this is handy.
urllib3.util.ssl_.ALPN_PROTOCOLS = ["h2"]
def extract_from_urllib3() -> None:
HTTPSConnectionPool.ConnectionCls = orig_HTTPSConnection
urllib3.connection.HTTPSConnection = orig_HTTPSConnection # type: ignore[misc]
urllib3.util.ssl_.ALPN_PROTOCOLS = ["http/1.1"]

View File

@ -1,638 +0,0 @@
from __future__ import annotations
import functools
import logging
import typing
import warnings
from types import TracebackType
from urllib.parse import urljoin
from ._collections import HTTPHeaderDict, RecentlyUsedContainer
from ._request_methods import RequestMethods
from .connection import ProxyConfig
from .connectionpool import HTTPConnectionPool, HTTPSConnectionPool, port_by_scheme
from .exceptions import (
LocationValueError,
MaxRetryError,
ProxySchemeUnknown,
URLSchemeUnknown,
)
from .response import BaseHTTPResponse
from .util.connection import _TYPE_SOCKET_OPTIONS
from .util.proxy import connection_requires_http_tunnel
from .util.retry import Retry
from .util.timeout import Timeout
from .util.url import Url, parse_url
if typing.TYPE_CHECKING:
import ssl
from typing import Literal
__all__ = ["PoolManager", "ProxyManager", "proxy_from_url"]
log = logging.getLogger(__name__)
SSL_KEYWORDS = (
"key_file",
"cert_file",
"cert_reqs",
"ca_certs",
"ca_cert_data",
"ssl_version",
"ssl_minimum_version",
"ssl_maximum_version",
"ca_cert_dir",
"ssl_context",
"key_password",
"server_hostname",
)
# Default value for `blocksize` - a new parameter introduced to
# http.client.HTTPConnection & http.client.HTTPSConnection in Python 3.7
_DEFAULT_BLOCKSIZE = 16384
_SelfT = typing.TypeVar("_SelfT")
class PoolKey(typing.NamedTuple):
"""
All known keyword arguments that could be provided to the pool manager, its
pools, or the underlying connections.
All custom key schemes should include the fields in this key at a minimum.
"""
key_scheme: str
key_host: str
key_port: int | None
key_timeout: Timeout | float | int | None
key_retries: Retry | bool | int | None
key_block: bool | None
key_source_address: tuple[str, int] | None
key_key_file: str | None
key_key_password: str | None
key_cert_file: str | None
key_cert_reqs: str | None
key_ca_certs: str | None
key_ca_cert_data: str | bytes | None
key_ssl_version: int | str | None
key_ssl_minimum_version: ssl.TLSVersion | None
key_ssl_maximum_version: ssl.TLSVersion | None
key_ca_cert_dir: str | None
key_ssl_context: ssl.SSLContext | None
key_maxsize: int | None
key_headers: frozenset[tuple[str, str]] | None
key__proxy: Url | None
key__proxy_headers: frozenset[tuple[str, str]] | None
key__proxy_config: ProxyConfig | None
key_socket_options: _TYPE_SOCKET_OPTIONS | None
key__socks_options: frozenset[tuple[str, str]] | None
key_assert_hostname: bool | str | None
key_assert_fingerprint: str | None
key_server_hostname: str | None
key_blocksize: int | None
def _default_key_normalizer(
key_class: type[PoolKey], request_context: dict[str, typing.Any]
) -> PoolKey:
"""
Create a pool key out of a request context dictionary.
According to RFC 3986, both the scheme and host are case-insensitive.
Therefore, this function normalizes both before constructing the pool
key for an HTTPS request. If you wish to change this behaviour, provide
alternate callables to ``key_fn_by_scheme``.
:param key_class:
The class to use when constructing the key. This should be a namedtuple
with the ``scheme`` and ``host`` keys at a minimum.
:type key_class: namedtuple
:param request_context:
A dictionary-like object that contain the context for a request.
:type request_context: dict
:return: A namedtuple that can be used as a connection pool key.
:rtype: PoolKey
"""
# Since we mutate the dictionary, make a copy first
context = request_context.copy()
context["scheme"] = context["scheme"].lower()
context["host"] = context["host"].lower()
# These are both dictionaries and need to be transformed into frozensets
for key in ("headers", "_proxy_headers", "_socks_options"):
if key in context and context[key] is not None:
context[key] = frozenset(context[key].items())
# The socket_options key may be a list and needs to be transformed into a
# tuple.
socket_opts = context.get("socket_options")
if socket_opts is not None:
context["socket_options"] = tuple(socket_opts)
# Map the kwargs to the names in the namedtuple - this is necessary since
# namedtuples can't have fields starting with '_'.
for key in list(context.keys()):
context["key_" + key] = context.pop(key)
# Default to ``None`` for keys missing from the context
for field in key_class._fields:
if field not in context:
context[field] = None
# Default key_blocksize to _DEFAULT_BLOCKSIZE if missing from the context
if context.get("key_blocksize") is None:
context["key_blocksize"] = _DEFAULT_BLOCKSIZE
return key_class(**context)
#: A dictionary that maps a scheme to a callable that creates a pool key.
#: This can be used to alter the way pool keys are constructed, if desired.
#: Each PoolManager makes a copy of this dictionary so they can be configured
#: globally here, or individually on the instance.
key_fn_by_scheme = {
"http": functools.partial(_default_key_normalizer, PoolKey),
"https": functools.partial(_default_key_normalizer, PoolKey),
}
pool_classes_by_scheme = {"http": HTTPConnectionPool, "https": HTTPSConnectionPool}
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 headers:
Headers to include with all requests, unless other headers are given
explicitly.
:param \\**connection_pool_kw:
Additional parameters are used to create fresh
:class:`urllib3.connectionpool.ConnectionPool` instances.
Example:
.. code-block:: python
import urllib3
http = urllib3.PoolManager(num_pools=2)
resp1 = http.request("GET", "https://google.com/")
resp2 = http.request("GET", "https://google.com/mail")
resp3 = http.request("GET", "https://yahoo.com/")
print(len(http.pools))
# 2
"""
proxy: Url | None = None
proxy_config: ProxyConfig | None = None
def __init__(
self,
num_pools: int = 10,
headers: typing.Mapping[str, str] | None = None,
**connection_pool_kw: typing.Any,
) -> None:
super().__init__(headers)
self.connection_pool_kw = connection_pool_kw
self.pools: RecentlyUsedContainer[PoolKey, HTTPConnectionPool]
self.pools = RecentlyUsedContainer(num_pools)
# Locally set the pool classes and keys so other PoolManagers can
# override them.
self.pool_classes_by_scheme = pool_classes_by_scheme
self.key_fn_by_scheme = key_fn_by_scheme.copy()
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.clear()
# Return False to re-raise any potential exceptions
return False
def _new_pool(
self,
scheme: str,
host: str,
port: int,
request_context: dict[str, typing.Any] | None = None,
) -> HTTPConnectionPool:
"""
Create a new :class:`urllib3.connectionpool.ConnectionPool` based on host, port, scheme, and
any additional pool keyword arguments.
If ``request_context`` is provided, it is provided as keyword arguments
to the pool class used. This method is used to actually create the
connection pools handed out by :meth:`connection_from_url` and
companion methods. It is intended to be overridden for customization.
"""
pool_cls: type[HTTPConnectionPool] = self.pool_classes_by_scheme[scheme]
if request_context is None:
request_context = self.connection_pool_kw.copy()
# Default blocksize to _DEFAULT_BLOCKSIZE if missing or explicitly
# set to 'None' in the request_context.
if request_context.get("blocksize") is None:
request_context["blocksize"] = _DEFAULT_BLOCKSIZE
# Although the context has everything necessary to create the pool,
# this function has historically only used the scheme, host, and port
# in the positional args. When an API change is acceptable these can
# be removed.
for key in ("scheme", "host", "port"):
request_context.pop(key, None)
if scheme == "http":
for kw in SSL_KEYWORDS:
request_context.pop(kw, None)
return pool_cls(host, port, **request_context)
def clear(self) -> None:
"""
Empty our store of pools and direct them all to close.
This will not affect in-flight connections, but they will not be
re-used after completion.
"""
self.pools.clear()
def connection_from_host(
self,
host: str | None,
port: int | None = None,
scheme: str | None = "http",
pool_kwargs: dict[str, typing.Any] | None = None,
) -> HTTPConnectionPool:
"""
Get a :class:`urllib3.connectionpool.ConnectionPool` based on the host, port, and scheme.
If ``port`` isn't given, it will be derived from the ``scheme`` using
``urllib3.connectionpool.port_by_scheme``. If ``pool_kwargs`` is
provided, it is merged with the instance's ``connection_pool_kw``
variable and used to create the new connection pool, if one is
needed.
"""
if not host:
raise LocationValueError("No host specified.")
request_context = self._merge_pool_kwargs(pool_kwargs)
request_context["scheme"] = scheme or "http"
if not port:
port = port_by_scheme.get(request_context["scheme"].lower(), 80)
request_context["port"] = port
request_context["host"] = host
return self.connection_from_context(request_context)
def connection_from_context(
self, request_context: dict[str, typing.Any]
) -> HTTPConnectionPool:
"""
Get a :class:`urllib3.connectionpool.ConnectionPool` based on the request context.
``request_context`` must at least contain the ``scheme`` key and its
value must be a key in ``key_fn_by_scheme`` instance variable.
"""
if "strict" in request_context:
warnings.warn(
"The 'strict' parameter is no longer needed on Python 3+. "
"This will raise an error in urllib3 v2.1.0.",
DeprecationWarning,
)
request_context.pop("strict")
scheme = request_context["scheme"].lower()
pool_key_constructor = self.key_fn_by_scheme.get(scheme)
if not pool_key_constructor:
raise URLSchemeUnknown(scheme)
pool_key = pool_key_constructor(request_context)
return self.connection_from_pool_key(pool_key, request_context=request_context)
def connection_from_pool_key(
self, pool_key: PoolKey, request_context: dict[str, typing.Any]
) -> HTTPConnectionPool:
"""
Get a :class:`urllib3.connectionpool.ConnectionPool` based on the provided pool key.
``pool_key`` should be a namedtuple that only contains immutable
objects. At a minimum it must have the ``scheme``, ``host``, and
``port`` fields.
"""
with self.pools.lock:
# 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
scheme = request_context["scheme"]
host = request_context["host"]
port = request_context["port"]
pool = self._new_pool(scheme, host, port, request_context=request_context)
self.pools[pool_key] = pool
return pool
def connection_from_url(
self, url: str, pool_kwargs: dict[str, typing.Any] | None = None
) -> HTTPConnectionPool:
"""
Similar to :func:`urllib3.connectionpool.connection_from_url`.
If ``pool_kwargs`` is not provided and a new pool needs to be
constructed, ``self.connection_pool_kw`` is used to initialize
the :class:`urllib3.connectionpool.ConnectionPool`. If ``pool_kwargs``
is provided, it is used instead. Note that if a new pool does not
need to be created for the request, the provided ``pool_kwargs`` are
not used.
"""
u = parse_url(url)
return self.connection_from_host(
u.host, port=u.port, scheme=u.scheme, pool_kwargs=pool_kwargs
)
def _merge_pool_kwargs(
self, override: dict[str, typing.Any] | None
) -> dict[str, typing.Any]:
"""
Merge a dictionary of override values for self.connection_pool_kw.
This does not modify self.connection_pool_kw and returns a new dict.
Any keys in the override dictionary with a value of ``None`` are
removed from the merged dictionary.
"""
base_pool_kwargs = self.connection_pool_kw.copy()
if override:
for key, value in override.items():
if value is None:
try:
del base_pool_kwargs[key]
except KeyError:
pass
else:
base_pool_kwargs[key] = value
return base_pool_kwargs
def _proxy_requires_url_absolute_form(self, parsed_url: Url) -> bool:
"""
Indicates if the proxy requires the complete destination URL in the
request. Normally this is only needed when not using an HTTP CONNECT
tunnel.
"""
if self.proxy is None:
return False
return not connection_requires_http_tunnel(
self.proxy, self.proxy_config, parsed_url.scheme
)
def urlopen( # type: ignore[override]
self, method: str, url: str, redirect: bool = True, **kw: typing.Any
) -> BaseHTTPResponse:
"""
Same as :meth:`urllib3.HTTPConnectionPool.urlopen`
with custom cross-host redirect logic and only sends the request-uri
portion of the ``url``.
The given ``url`` parameter must be absolute, such that an appropriate
:class:`urllib3.connectionpool.ConnectionPool` can be chosen for it.
"""
u = parse_url(url)
if u.scheme is None:
warnings.warn(
"URLs without a scheme (ie 'https://') are deprecated and will raise an error "
"in a future version of urllib3. To avoid this DeprecationWarning ensure all URLs "
"start with 'https://' or 'http://'. Read more in this issue: "
"https://github.com/urllib3/urllib3/issues/2920",
category=DeprecationWarning,
stacklevel=2,
)
conn = self.connection_from_host(u.host, port=u.port, scheme=u.scheme)
kw["assert_same_host"] = False
kw["redirect"] = False
if "headers" not in kw:
kw["headers"] = self.headers
if self._proxy_requires_url_absolute_form(u):
response = conn.urlopen(method, url, **kw)
else:
response = conn.urlopen(method, u.request_uri, **kw)
redirect_location = redirect and response.get_redirect_location()
if not redirect_location:
return response
# Support relative URLs for redirecting.
redirect_location = urljoin(url, 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.
kw["body"] = None
kw["headers"] = HTTPHeaderDict(kw["headers"])._prepare_for_method_change()
retries = kw.get("retries")
if not isinstance(retries, Retry):
retries = Retry.from_int(retries, redirect=redirect)
# Strip headers marked as unsafe to forward to the redirected location.
# Check remove_headers_on_redirect to avoid a potential network call within
# conn.is_same_host() which may use socket.gethostbyname() in the future.
if retries.remove_headers_on_redirect and not conn.is_same_host(
redirect_location
):
new_headers = kw["headers"].copy()
for header in kw["headers"]:
if header.lower() in retries.remove_headers_on_redirect:
new_headers.pop(header, None)
kw["headers"] = new_headers
try:
retries = retries.increment(method, url, response=response, _pool=conn)
except MaxRetryError:
if retries.raise_on_redirect:
response.drain_conn()
raise
return response
kw["retries"] = retries
kw["redirect"] = redirect
log.info("Redirecting %s -> %s", url, redirect_location)
response.drain_conn()
return self.urlopen(method, redirect_location, **kw)
class ProxyManager(PoolManager):
"""
Behaves just like :class:`PoolManager`, but sends all requests through
the defined proxy, using the CONNECT method for HTTPS URLs.
:param proxy_url:
The URL of the proxy to be used.
:param proxy_headers:
A dictionary containing headers that will be sent to the proxy. In case
of HTTP they are being sent with each request, while in the
HTTPS/CONNECT case they are sent only once. Could be used for proxy
authentication.
:param proxy_ssl_context:
The proxy SSL context is used to establish the TLS connection to the
proxy when using HTTPS proxies.
:param use_forwarding_for_https:
(Defaults to False) If set to True will forward requests to the HTTPS
proxy to be made on behalf of the client instead of creating a TLS
tunnel via the CONNECT method. **Enabling this flag means that request
and response headers and content will be visible from the HTTPS proxy**
whereas tunneling keeps request and response headers and content
private. IP address, target hostname, SNI, and port are always visible
to an HTTPS proxy even when this flag is disabled.
:param proxy_assert_hostname:
The hostname of the certificate to verify against.
:param proxy_assert_fingerprint:
The fingerprint of the certificate to verify against.
Example:
.. code-block:: python
import urllib3
proxy = urllib3.ProxyManager("https://localhost:3128/")
resp1 = proxy.request("GET", "https://google.com/")
resp2 = proxy.request("GET", "https://httpbin.org/")
print(len(proxy.pools))
# 1
resp3 = proxy.request("GET", "https://httpbin.org/")
resp4 = proxy.request("GET", "https://twitter.com/")
print(len(proxy.pools))
# 3
"""
def __init__(
self,
proxy_url: str,
num_pools: int = 10,
headers: typing.Mapping[str, str] | None = None,
proxy_headers: typing.Mapping[str, str] | None = None,
proxy_ssl_context: ssl.SSLContext | None = None,
use_forwarding_for_https: bool = False,
proxy_assert_hostname: None | str | Literal[False] = None,
proxy_assert_fingerprint: str | None = None,
**connection_pool_kw: typing.Any,
) -> None:
if isinstance(proxy_url, HTTPConnectionPool):
str_proxy_url = f"{proxy_url.scheme}://{proxy_url.host}:{proxy_url.port}"
else:
str_proxy_url = proxy_url
proxy = parse_url(str_proxy_url)
if proxy.scheme not in ("http", "https"):
raise ProxySchemeUnknown(proxy.scheme)
if not proxy.port:
port = port_by_scheme.get(proxy.scheme, 80)
proxy = proxy._replace(port=port)
self.proxy = proxy
self.proxy_headers = proxy_headers or {}
self.proxy_ssl_context = proxy_ssl_context
self.proxy_config = ProxyConfig(
proxy_ssl_context,
use_forwarding_for_https,
proxy_assert_hostname,
proxy_assert_fingerprint,
)
connection_pool_kw["_proxy"] = self.proxy
connection_pool_kw["_proxy_headers"] = self.proxy_headers
connection_pool_kw["_proxy_config"] = self.proxy_config
super().__init__(num_pools, headers, **connection_pool_kw)
def connection_from_host(
self,
host: str | None,
port: int | None = None,
scheme: str | None = "http",
pool_kwargs: dict[str, typing.Any] | None = None,
) -> HTTPConnectionPool:
if scheme == "https":
return super().connection_from_host(
host, port, scheme, pool_kwargs=pool_kwargs
)
return super().connection_from_host(
self.proxy.host, self.proxy.port, self.proxy.scheme, pool_kwargs=pool_kwargs # type: ignore[union-attr]
)
def _set_proxy_headers(
self, url: str, headers: typing.Mapping[str, str] | None = None
) -> typing.Mapping[str, str]:
"""
Sets headers needed by proxies: specifically, the Accept and Host
headers. Only sets headers not provided by the user.
"""
headers_ = {"Accept": "*/*"}
netloc = parse_url(url).netloc
if netloc:
headers_["Host"] = netloc
if headers:
headers_.update(headers)
return headers_
def urlopen( # type: ignore[override]
self, method: str, url: str, redirect: bool = True, **kw: typing.Any
) -> BaseHTTPResponse:
"Same as HTTP(S)ConnectionPool.urlopen, ``url`` must be absolute."
u = parse_url(url)
if not connection_requires_http_tunnel(self.proxy, self.proxy_config, u.scheme):
# For connections using HTTP CONNECT, httplib sets the necessary
# headers on the CONNECT to the proxy. If we're not using CONNECT,
# we'll definitely need to set 'Host' at the very least.
headers = kw.get("headers", self.headers)
kw["headers"] = self._set_proxy_headers(url, headers)
return super().urlopen(method, url, redirect=redirect, **kw)
def proxy_from_url(url: str, **kw: typing.Any) -> ProxyManager:
return ProxyManager(proxy_url=url, **kw)

View File

@ -1,2 +0,0 @@
# Instruct type checkers to look for inline type annotations in this package.
# See PEP 561.

File diff suppressed because it is too large Load Diff

View File

@ -1,42 +0,0 @@
# For backwards compatibility, provide imports that used to be here.
from __future__ import annotations
from .connection import is_connection_dropped
from .request import SKIP_HEADER, SKIPPABLE_HEADERS, make_headers
from .response import is_fp_closed
from .retry import Retry
from .ssl_ import (
ALPN_PROTOCOLS,
IS_PYOPENSSL,
SSLContext,
assert_fingerprint,
create_urllib3_context,
resolve_cert_reqs,
resolve_ssl_version,
ssl_wrap_socket,
)
from .timeout import Timeout
from .url import Url, parse_url
from .wait import wait_for_read, wait_for_write
__all__ = (
"IS_PYOPENSSL",
"SSLContext",
"ALPN_PROTOCOLS",
"Retry",
"Timeout",
"Url",
"assert_fingerprint",
"create_urllib3_context",
"is_connection_dropped",
"is_fp_closed",
"parse_url",
"make_headers",
"resolve_cert_reqs",
"resolve_ssl_version",
"ssl_wrap_socket",
"wait_for_read",
"wait_for_write",
"SKIP_HEADER",
"SKIPPABLE_HEADERS",
)

View File

@ -1,137 +0,0 @@
from __future__ import annotations
import socket
import typing
from ..exceptions import LocationParseError
from .timeout import _DEFAULT_TIMEOUT, _TYPE_TIMEOUT
_TYPE_SOCKET_OPTIONS = typing.Sequence[typing.Tuple[int, int, typing.Union[int, bytes]]]
if typing.TYPE_CHECKING:
from .._base_connection import BaseHTTPConnection
def is_connection_dropped(conn: BaseHTTPConnection) -> bool: # Platform-specific
"""
Returns True if the connection is dropped and should be closed.
:param conn: :class:`urllib3.connection.HTTPConnection` object.
"""
return not conn.is_connected
# This function is copied from socket.py in the Python 2.7 standard
# library test suite. Added to its signature is only `socket_options`.
# One additional modification is that we avoid binding to IPv6 servers
# discovered in DNS if the system doesn't have IPv6 functionality.
def create_connection(
address: tuple[str, int],
timeout: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT,
source_address: tuple[str, int] | None = None,
socket_options: _TYPE_SOCKET_OPTIONS | None = None,
) -> socket.socket:
"""Connect to *address* and return the socket object.
Convenience function. Connect to *address* (a 2-tuple ``(host,
port)``) and return the socket object. Passing the optional
*timeout* parameter will set the timeout on the socket instance
before attempting to connect. If no *timeout* is supplied, the
global default timeout setting returned by :func:`socket.getdefaulttimeout`
is used. If *source_address* is set it must be a tuple of (host, port)
for the socket to bind as a source address before making the connection.
An host of '' or port 0 tells the OS to use the default.
"""
host, port = address
if host.startswith("["):
host = host.strip("[]")
err = None
# Using the value from allowed_gai_family() in the context of getaddrinfo lets
# us select whether to work with IPv4 DNS records, IPv6 records, or both.
# The original create_connection function always returns all records.
family = allowed_gai_family()
try:
host.encode("idna")
except UnicodeError:
raise LocationParseError(f"'{host}', label empty or too long") from None
for res in socket.getaddrinfo(host, port, family, socket.SOCK_STREAM):
af, socktype, proto, canonname, sa = res
sock = None
try:
sock = socket.socket(af, socktype, proto)
# If provided, set socket level options before connecting.
_set_socket_options(sock, socket_options)
if timeout is not _DEFAULT_TIMEOUT:
sock.settimeout(timeout)
if source_address:
sock.bind(source_address)
sock.connect(sa)
# Break explicitly a reference cycle
err = None
return sock
except OSError as _:
err = _
if sock is not None:
sock.close()
if err is not None:
try:
raise err
finally:
# Break explicitly a reference cycle
err = None
else:
raise OSError("getaddrinfo returns an empty list")
def _set_socket_options(
sock: socket.socket, options: _TYPE_SOCKET_OPTIONS | None
) -> None:
if options is None:
return
for opt in options:
sock.setsockopt(*opt)
def allowed_gai_family() -> socket.AddressFamily:
"""This function is designed to work in the context of
getaddrinfo, where family=socket.AF_UNSPEC is the default and
will perform a DNS search for both IPv6 and IPv4 records."""
family = socket.AF_INET
if HAS_IPV6:
family = socket.AF_UNSPEC
return family
def _has_ipv6(host: str) -> bool:
"""Returns True if the system can bind an IPv6 address."""
sock = None
has_ipv6 = False
if socket.has_ipv6:
# has_ipv6 returns true if cPython was compiled with IPv6 support.
# It does not tell us if the system has IPv6 support enabled. To
# determine that we must bind to an IPv6 address.
# https://github.com/urllib3/urllib3/pull/611
# https://bugs.python.org/issue658327
try:
sock = socket.socket(socket.AF_INET6)
sock.bind((host, 0))
has_ipv6 = True
except Exception:
pass
if sock:
sock.close()
return has_ipv6
HAS_IPV6 = _has_ipv6("::1")

View File

@ -1,43 +0,0 @@
from __future__ import annotations
import typing
from .url import Url
if typing.TYPE_CHECKING:
from ..connection import ProxyConfig
def connection_requires_http_tunnel(
proxy_url: Url | None = None,
proxy_config: ProxyConfig | None = None,
destination_scheme: str | None = None,
) -> bool:
"""
Returns True if the connection requires an HTTP CONNECT through the proxy.
:param URL proxy_url:
URL of the proxy.
:param ProxyConfig proxy_config:
Proxy configuration from poolmanager.py
:param str destination_scheme:
The scheme of the destination. (i.e https, http, etc)
"""
# If we're not using a proxy, no way to use a tunnel.
if proxy_url is None:
return False
# HTTP destinations never require tunneling, we always forward.
if destination_scheme == "http":
return False
# Support for forwarding with HTTPS proxies and HTTPS destinations.
if (
proxy_url.scheme == "https"
and proxy_config
and proxy_config.use_forwarding_for_https
):
return False
# Otherwise always use a tunnel.
return True

View File

@ -1,256 +0,0 @@
from __future__ import annotations
import io
import typing
from base64 import b64encode
from enum import Enum
from ..exceptions import UnrewindableBodyError
from .util import to_bytes
if typing.TYPE_CHECKING:
from typing import Final
# Pass as a value within ``headers`` to skip
# emitting some HTTP headers that are added automatically.
# The only headers that are supported are ``Accept-Encoding``,
# ``Host``, and ``User-Agent``.
SKIP_HEADER = "@@@SKIP_HEADER@@@"
SKIPPABLE_HEADERS = frozenset(["accept-encoding", "host", "user-agent"])
ACCEPT_ENCODING = "gzip,deflate"
try:
try:
import brotlicffi as _unused_module_brotli # type: ignore[import-not-found] # noqa: F401
except ImportError:
import brotli as _unused_module_brotli # type: ignore[import-not-found] # noqa: F401
except ImportError:
pass
else:
ACCEPT_ENCODING += ",br"
try:
import zstandard as _unused_module_zstd # type: ignore[import-not-found] # noqa: F401
except ImportError:
pass
else:
ACCEPT_ENCODING += ",zstd"
class _TYPE_FAILEDTELL(Enum):
token = 0
_FAILEDTELL: Final[_TYPE_FAILEDTELL] = _TYPE_FAILEDTELL.token
_TYPE_BODY_POSITION = typing.Union[int, _TYPE_FAILEDTELL]
# When sending a request with these methods we aren't expecting
# a body so don't need to set an explicit 'Content-Length: 0'
# The reason we do this in the negative instead of tracking methods
# which 'should' have a body is because unknown methods should be
# treated as if they were 'POST' which *does* expect a body.
_METHODS_NOT_EXPECTING_BODY = {"GET", "HEAD", "DELETE", "TRACE", "OPTIONS", "CONNECT"}
def make_headers(
keep_alive: bool | None = None,
accept_encoding: bool | list[str] | str | None = None,
user_agent: str | None = None,
basic_auth: str | None = None,
proxy_basic_auth: str | None = None,
disable_cache: bool | None = None,
) -> dict[str, str]:
"""
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'. If either the ``brotli`` or
``brotlicffi`` package is installed 'gzip,deflate,br' is used instead.
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.
:param proxy_basic_auth:
Colon-separated username:password string for 'proxy-authorization: basic ...'
auth header.
:param disable_cache:
If ``True``, adds 'cache-control: no-cache' header.
Example:
.. code-block:: python
import urllib3
print(urllib3.util.make_headers(keep_alive=True, user_agent="Batman/1.0"))
# {'connection': 'keep-alive', 'user-agent': 'Batman/1.0'}
print(urllib3.util.make_headers(accept_encoding=True))
# {'accept-encoding': 'gzip,deflate'}
"""
headers: dict[str, str] = {}
if accept_encoding:
if isinstance(accept_encoding, str):
pass
elif isinstance(accept_encoding, list):
accept_encoding = ",".join(accept_encoding)
else:
accept_encoding = ACCEPT_ENCODING
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"
] = f"Basic {b64encode(basic_auth.encode('latin-1')).decode()}"
if proxy_basic_auth:
headers[
"proxy-authorization"
] = f"Basic {b64encode(proxy_basic_auth.encode('latin-1')).decode()}"
if disable_cache:
headers["cache-control"] = "no-cache"
return headers
def set_file_position(
body: typing.Any, pos: _TYPE_BODY_POSITION | None
) -> _TYPE_BODY_POSITION | None:
"""
If a position is provided, move file to that point.
Otherwise, we'll attempt to record a position for future use.
"""
if pos is not None:
rewind_body(body, pos)
elif getattr(body, "tell", None) is not None:
try:
pos = body.tell()
except OSError:
# This differentiates from None, allowing us to catch
# a failed `tell()` later when trying to rewind the body.
pos = _FAILEDTELL
return pos
def rewind_body(body: typing.IO[typing.AnyStr], body_pos: _TYPE_BODY_POSITION) -> None:
"""
Attempt to rewind body to a certain position.
Primarily used for request redirects and retries.
:param body:
File-like object that supports seek.
:param int pos:
Position to seek to in file.
"""
body_seek = getattr(body, "seek", None)
if body_seek is not None and isinstance(body_pos, int):
try:
body_seek(body_pos)
except OSError as e:
raise UnrewindableBodyError(
"An error occurred when rewinding request body for redirect/retry."
) from e
elif body_pos is _FAILEDTELL:
raise UnrewindableBodyError(
"Unable to record file position for rewinding "
"request body during a redirect/retry."
)
else:
raise ValueError(
f"body_pos must be of type integer, instead it was {type(body_pos)}."
)
class ChunksAndContentLength(typing.NamedTuple):
chunks: typing.Iterable[bytes] | None
content_length: int | None
def body_to_chunks(
body: typing.Any | None, method: str, blocksize: int
) -> ChunksAndContentLength:
"""Takes the HTTP request method, body, and blocksize and
transforms them into an iterable of chunks to pass to
socket.sendall() and an optional 'Content-Length' header.
A 'Content-Length' of 'None' indicates the length of the body
can't be determined so should use 'Transfer-Encoding: chunked'
for framing instead.
"""
chunks: typing.Iterable[bytes] | None
content_length: int | None
# No body, we need to make a recommendation on 'Content-Length'
# based on whether that request method is expected to have
# a body or not.
if body is None:
chunks = None
if method.upper() not in _METHODS_NOT_EXPECTING_BODY:
content_length = 0
else:
content_length = None
# Bytes or strings become bytes
elif isinstance(body, (str, bytes)):
chunks = (to_bytes(body),)
content_length = len(chunks[0])
# File-like object, TODO: use seek() and tell() for length?
elif hasattr(body, "read"):
def chunk_readable() -> typing.Iterable[bytes]:
nonlocal body, blocksize
encode = isinstance(body, io.TextIOBase)
while True:
datablock = body.read(blocksize)
if not datablock:
break
if encode:
datablock = datablock.encode("iso-8859-1")
yield datablock
chunks = chunk_readable()
content_length = None
# Otherwise we need to start checking via duck-typing.
else:
try:
# Check if the body implements the buffer API.
mv = memoryview(body)
except TypeError:
try:
# Check if the body is an iterable
chunks = iter(body)
content_length = None
except TypeError:
raise TypeError(
f"'body' must be a bytes-like object, file-like "
f"object, or iterable. Instead was {body!r}"
) from None
else:
# Since it implements the buffer API can be passed directly to socket.sendall()
chunks = (body,)
content_length = mv.nbytes
return ChunksAndContentLength(chunks=chunks, content_length=content_length)

View File

@ -1,101 +0,0 @@
from __future__ import annotations
import http.client as httplib
from email.errors import MultipartInvariantViolationDefect, StartBoundaryNotFoundDefect
from ..exceptions import HeaderParsingError
def is_fp_closed(obj: object) -> bool:
"""
Checks whether a given file-like object is closed.
:param obj:
The file-like object to check.
"""
try:
# Check `isclosed()` first, in case Python3 doesn't set `closed`.
# GH Issue #928
return obj.isclosed() # type: ignore[no-any-return, attr-defined]
except AttributeError:
pass
try:
# Check via the official file-like-object way.
return obj.closed # type: ignore[no-any-return, attr-defined]
except AttributeError:
pass
try:
# Check if the object is a container for another file-like object that
# gets released on exhaustion (e.g. HTTPResponse).
return obj.fp is None # type: ignore[attr-defined]
except AttributeError:
pass
raise ValueError("Unable to determine whether fp is closed.")
def assert_header_parsing(headers: httplib.HTTPMessage) -> None:
"""
Asserts whether all headers have been successfully parsed.
Extracts encountered errors from the result of parsing headers.
Only works on Python 3.
:param http.client.HTTPMessage headers: Headers to verify.
:raises urllib3.exceptions.HeaderParsingError:
If parsing errors are found.
"""
# This will fail silently if we pass in the wrong kind of parameter.
# To make debugging easier add an explicit check.
if not isinstance(headers, httplib.HTTPMessage):
raise TypeError(f"expected httplib.Message, got {type(headers)}.")
unparsed_data = None
# get_payload is actually email.message.Message.get_payload;
# we're only interested in the result if it's not a multipart message
if not headers.is_multipart():
payload = headers.get_payload()
if isinstance(payload, (bytes, str)):
unparsed_data = payload
# httplib is assuming a response body is available
# when parsing headers even when httplib only sends
# header data to parse_headers() This results in
# defects on multipart responses in particular.
# See: https://github.com/urllib3/urllib3/issues/800
# So we ignore the following defects:
# - StartBoundaryNotFoundDefect:
# The claimed start boundary was never found.
# - MultipartInvariantViolationDefect:
# A message claimed to be a multipart but no subparts were found.
defects = [
defect
for defect in headers.defects
if not isinstance(
defect, (StartBoundaryNotFoundDefect, MultipartInvariantViolationDefect)
)
]
if defects or unparsed_data:
raise HeaderParsingError(defects=defects, unparsed_data=unparsed_data)
def is_response_to_head(response: httplib.HTTPResponse) -> bool:
"""
Checks whether the request of a response has been a HEAD-request.
:param http.client.HTTPResponse response:
Response to check if the originating request
used 'HEAD' as a method.
"""
# FIXME: Can we do this somehow without accessing private httplib _method?
method_str = response._method # type: str # type: ignore[attr-defined]
return method_str.upper() == "HEAD"

View File

@ -1,529 +0,0 @@
from __future__ import annotations
import email
import logging
import random
import re
import time
import typing
from itertools import takewhile
from types import TracebackType
from ..exceptions import (
ConnectTimeoutError,
InvalidHeader,
MaxRetryError,
ProtocolError,
ProxyError,
ReadTimeoutError,
ResponseError,
)
from .util import reraise
if typing.TYPE_CHECKING:
from ..connectionpool import ConnectionPool
from ..response import BaseHTTPResponse
log = logging.getLogger(__name__)
# Data structure for representing the metadata of requests that result in a retry.
class RequestHistory(typing.NamedTuple):
method: str | None
url: str | None
error: Exception | None
status: int | None
redirect_location: str | None
class Retry:
"""Retry configuration.
Each retry attempt will create a new Retry object with updated values, so
they can be safely reused.
Retries can be defined as a default for a pool:
.. code-block:: python
retries = Retry(connect=5, read=2, redirect=5)
http = PoolManager(retries=retries)
response = http.request("GET", "https://example.com/")
Or per-request (which overrides the default for the pool):
.. code-block:: python
response = http.request("GET", "https://example.com/", retries=Retry(10))
Retries can be disabled by passing ``False``:
.. code-block:: python
response = http.request("GET", "https://example.com/", retries=False)
Errors will be wrapped in :class:`~urllib3.exceptions.MaxRetryError` unless
retries are disabled, in which case the causing exception will be raised.
:param int total:
Total number of retries to allow. Takes precedence over other counts.
Set to ``None`` to remove this constraint and fall back on other
counts.
Set to ``0`` to fail on the first retry.
Set to ``False`` to disable and imply ``raise_on_redirect=False``.
:param int connect:
How many connection-related errors to retry on.
These are errors raised before the request is sent to the remote server,
which we assume has not triggered the server to process the request.
Set to ``0`` to fail on the first retry of this type.
:param int read:
How many times to retry on read errors.
These errors are raised after the request was sent to the server, so the
request may have side-effects.
Set to ``0`` to fail on the first retry of this type.
:param int redirect:
How many redirects to perform. Limit this to avoid infinite redirect
loops.
A redirect is a HTTP response with a status code 301, 302, 303, 307 or
308.
Set to ``0`` to fail on the first retry of this type.
Set to ``False`` to disable and imply ``raise_on_redirect=False``.
:param int status:
How many times to retry on bad status codes.
These are retries made on responses, where status code matches
``status_forcelist``.
Set to ``0`` to fail on the first retry of this type.
:param int other:
How many times to retry on other errors.
Other errors are errors that are not connect, read, redirect or status errors.
These errors might be raised after the request was sent to the server, so the
request might have side-effects.
Set to ``0`` to fail on the first retry of this type.
If ``total`` is not set, it's a good idea to set this to 0 to account
for unexpected edge cases and avoid infinite retry loops.
:param Collection allowed_methods:
Set of uppercased HTTP method verbs that we should retry on.
By default, we only retry on methods which are considered to be
idempotent (multiple requests with the same parameters end with the
same state). See :attr:`Retry.DEFAULT_ALLOWED_METHODS`.
Set to a ``None`` value to retry on any verb.
:param Collection status_forcelist:
A set of integer HTTP status codes that we should force a retry on.
A retry is initiated if the request method is in ``allowed_methods``
and the response status code is in ``status_forcelist``.
By default, this is disabled with ``None``.
:param float backoff_factor:
A backoff factor to apply between attempts after the second try
(most errors are resolved immediately by a second try without a
delay). urllib3 will sleep for::
{backoff factor} * (2 ** ({number of previous retries}))
seconds. If `backoff_jitter` is non-zero, this sleep is extended by::
random.uniform(0, {backoff jitter})
seconds. For example, if the backoff_factor is 0.1, then :func:`Retry.sleep` will
sleep for [0.0s, 0.2s, 0.4s, 0.8s, ...] between retries. No backoff will ever
be longer than `backoff_max`.
By default, backoff is disabled (factor set to 0).
:param bool raise_on_redirect: Whether, if the number of redirects is
exhausted, to raise a MaxRetryError, or to return a response with a
response code in the 3xx range.
:param bool raise_on_status: Similar meaning to ``raise_on_redirect``:
whether we should raise an exception, or return a response,
if status falls in ``status_forcelist`` range and retries have
been exhausted.
:param tuple history: The history of the request encountered during
each call to :meth:`~Retry.increment`. The list is in the order
the requests occurred. Each list item is of class :class:`RequestHistory`.
:param bool respect_retry_after_header:
Whether to respect Retry-After header on status codes defined as
:attr:`Retry.RETRY_AFTER_STATUS_CODES` or not.
:param Collection remove_headers_on_redirect:
Sequence of headers to remove from the request when a response
indicating a redirect is returned before firing off the redirected
request.
"""
#: Default methods to be used for ``allowed_methods``
DEFAULT_ALLOWED_METHODS = frozenset(
["HEAD", "GET", "PUT", "DELETE", "OPTIONS", "TRACE"]
)
#: Default status codes to be used for ``status_forcelist``
RETRY_AFTER_STATUS_CODES = frozenset([413, 429, 503])
#: Default headers to be used for ``remove_headers_on_redirect``
DEFAULT_REMOVE_HEADERS_ON_REDIRECT = frozenset(["Cookie", "Authorization"])
#: Default maximum backoff time.
DEFAULT_BACKOFF_MAX = 120
# Backward compatibility; assigned outside of the class.
DEFAULT: typing.ClassVar[Retry]
def __init__(
self,
total: bool | int | None = 10,
connect: int | None = None,
read: int | None = None,
redirect: bool | int | None = None,
status: int | None = None,
other: int | None = None,
allowed_methods: typing.Collection[str] | None = DEFAULT_ALLOWED_METHODS,
status_forcelist: typing.Collection[int] | None = None,
backoff_factor: float = 0,
backoff_max: float = DEFAULT_BACKOFF_MAX,
raise_on_redirect: bool = True,
raise_on_status: bool = True,
history: tuple[RequestHistory, ...] | None = None,
respect_retry_after_header: bool = True,
remove_headers_on_redirect: typing.Collection[
str
] = DEFAULT_REMOVE_HEADERS_ON_REDIRECT,
backoff_jitter: float = 0.0,
) -> None:
self.total = total
self.connect = connect
self.read = read
self.status = status
self.other = other
if redirect is False or total is False:
redirect = 0
raise_on_redirect = False
self.redirect = redirect
self.status_forcelist = status_forcelist or set()
self.allowed_methods = allowed_methods
self.backoff_factor = backoff_factor
self.backoff_max = backoff_max
self.raise_on_redirect = raise_on_redirect
self.raise_on_status = raise_on_status
self.history = history or ()
self.respect_retry_after_header = respect_retry_after_header
self.remove_headers_on_redirect = frozenset(
h.lower() for h in remove_headers_on_redirect
)
self.backoff_jitter = backoff_jitter
def new(self, **kw: typing.Any) -> Retry:
params = dict(
total=self.total,
connect=self.connect,
read=self.read,
redirect=self.redirect,
status=self.status,
other=self.other,
allowed_methods=self.allowed_methods,
status_forcelist=self.status_forcelist,
backoff_factor=self.backoff_factor,
backoff_max=self.backoff_max,
raise_on_redirect=self.raise_on_redirect,
raise_on_status=self.raise_on_status,
history=self.history,
remove_headers_on_redirect=self.remove_headers_on_redirect,
respect_retry_after_header=self.respect_retry_after_header,
backoff_jitter=self.backoff_jitter,
)
params.update(kw)
return type(self)(**params) # type: ignore[arg-type]
@classmethod
def from_int(
cls,
retries: Retry | bool | int | None,
redirect: bool | int | None = True,
default: Retry | bool | int | None = None,
) -> Retry:
"""Backwards-compatibility for the old retries format."""
if retries is None:
retries = default if default is not None else cls.DEFAULT
if isinstance(retries, Retry):
return retries
redirect = bool(redirect) and None
new_retries = cls(retries, redirect=redirect)
log.debug("Converted retries value: %r -> %r", retries, new_retries)
return new_retries
def get_backoff_time(self) -> float:
"""Formula for computing the current backoff
:rtype: float
"""
# We want to consider only the last consecutive errors sequence (Ignore redirects).
consecutive_errors_len = len(
list(
takewhile(lambda x: x.redirect_location is None, reversed(self.history))
)
)
if consecutive_errors_len <= 1:
return 0
backoff_value = self.backoff_factor * (2 ** (consecutive_errors_len - 1))
if self.backoff_jitter != 0.0:
backoff_value += random.random() * self.backoff_jitter
return float(max(0, min(self.backoff_max, backoff_value)))
def parse_retry_after(self, retry_after: str) -> float:
seconds: float
# Whitespace: https://tools.ietf.org/html/rfc7230#section-3.2.4
if re.match(r"^\s*[0-9]+\s*$", retry_after):
seconds = int(retry_after)
else:
retry_date_tuple = email.utils.parsedate_tz(retry_after)
if retry_date_tuple is None:
raise InvalidHeader(f"Invalid Retry-After header: {retry_after}")
retry_date = email.utils.mktime_tz(retry_date_tuple)
seconds = retry_date - time.time()
seconds = max(seconds, 0)
return seconds
def get_retry_after(self, response: BaseHTTPResponse) -> float | None:
"""Get the value of Retry-After in seconds."""
retry_after = response.headers.get("Retry-After")
if retry_after is None:
return None
return self.parse_retry_after(retry_after)
def sleep_for_retry(self, response: BaseHTTPResponse) -> bool:
retry_after = self.get_retry_after(response)
if retry_after:
time.sleep(retry_after)
return True
return False
def _sleep_backoff(self) -> None:
backoff = self.get_backoff_time()
if backoff <= 0:
return
time.sleep(backoff)
def sleep(self, response: BaseHTTPResponse | None = None) -> None:
"""Sleep between retry attempts.
This method will respect a server's ``Retry-After`` response header
and sleep the duration of the time requested. If that is not present, it
will use an exponential backoff. By default, the backoff factor is 0 and
this method will return immediately.
"""
if self.respect_retry_after_header and response:
slept = self.sleep_for_retry(response)
if slept:
return
self._sleep_backoff()
def _is_connection_error(self, err: Exception) -> bool:
"""Errors when we're fairly sure that the server did not receive the
request, so it should be safe to retry.
"""
if isinstance(err, ProxyError):
err = err.original_error
return isinstance(err, ConnectTimeoutError)
def _is_read_error(self, err: Exception) -> bool:
"""Errors that occur after the request has been started, so we should
assume that the server began processing it.
"""
return isinstance(err, (ReadTimeoutError, ProtocolError))
def _is_method_retryable(self, method: str) -> bool:
"""Checks if a given HTTP method should be retried upon, depending if
it is included in the allowed_methods
"""
if self.allowed_methods and method.upper() not in self.allowed_methods:
return False
return True
def is_retry(
self, method: str, status_code: int, has_retry_after: bool = False
) -> bool:
"""Is this method/status code retryable? (Based on allowlists and control
variables such as the number of total retries to allow, whether to
respect the Retry-After header, whether this header is present, and
whether the returned status code is on the list of status codes to
be retried upon on the presence of the aforementioned header)
"""
if not self._is_method_retryable(method):
return False
if self.status_forcelist and status_code in self.status_forcelist:
return True
return bool(
self.total
and self.respect_retry_after_header
and has_retry_after
and (status_code in self.RETRY_AFTER_STATUS_CODES)
)
def is_exhausted(self) -> bool:
"""Are we out of retries?"""
retry_counts = [
x
for x in (
self.total,
self.connect,
self.read,
self.redirect,
self.status,
self.other,
)
if x
]
if not retry_counts:
return False
return min(retry_counts) < 0
def increment(
self,
method: str | None = None,
url: str | None = None,
response: BaseHTTPResponse | None = None,
error: Exception | None = None,
_pool: ConnectionPool | None = None,
_stacktrace: TracebackType | None = None,
) -> Retry:
"""Return a new Retry object with incremented retry counters.
:param response: A response object, or None, if the server did not
return a response.
:type response: :class:`~urllib3.response.BaseHTTPResponse`
:param Exception error: An error encountered during the request, or
None if the response was received successfully.
:return: A new ``Retry`` object.
"""
if self.total is False and error:
# Disabled, indicate to re-raise the error.
raise reraise(type(error), error, _stacktrace)
total = self.total
if total is not None:
total -= 1
connect = self.connect
read = self.read
redirect = self.redirect
status_count = self.status
other = self.other
cause = "unknown"
status = None
redirect_location = None
if error and self._is_connection_error(error):
# Connect retry?
if connect is False:
raise reraise(type(error), error, _stacktrace)
elif connect is not None:
connect -= 1
elif error and self._is_read_error(error):
# Read retry?
if read is False or method is None or not self._is_method_retryable(method):
raise reraise(type(error), error, _stacktrace)
elif read is not None:
read -= 1
elif error:
# Other retry?
if other is not None:
other -= 1
elif response and response.get_redirect_location():
# Redirect retry?
if redirect is not None:
redirect -= 1
cause = "too many redirects"
response_redirect_location = response.get_redirect_location()
if response_redirect_location:
redirect_location = response_redirect_location
status = response.status
else:
# Incrementing because of a server error like a 500 in
# status_forcelist and the given method is in the allowed_methods
cause = ResponseError.GENERIC_ERROR
if response and response.status:
if status_count is not None:
status_count -= 1
cause = ResponseError.SPECIFIC_ERROR.format(status_code=response.status)
status = response.status
history = self.history + (
RequestHistory(method, url, error, status, redirect_location),
)
new_retry = self.new(
total=total,
connect=connect,
read=read,
redirect=redirect,
status=status_count,
other=other,
history=history,
)
if new_retry.is_exhausted():
reason = error or ResponseError(cause)
raise MaxRetryError(_pool, url, reason) from reason # type: ignore[arg-type]
log.debug("Incremented Retry for (url='%s'): %r", url, new_retry)
return new_retry
def __repr__(self) -> str:
return (
f"{type(self).__name__}(total={self.total}, connect={self.connect}, "
f"read={self.read}, redirect={self.redirect}, status={self.status})"
)
# For backwards compatibility (equivalent to pre-v1.9):
Retry.DEFAULT = Retry(3)

View File

@ -1,509 +0,0 @@
from __future__ import annotations
import hmac
import os
import socket
import sys
import typing
import warnings
from binascii import unhexlify
from hashlib import md5, sha1, sha256
from ..exceptions import ProxySchemeUnsupported, SSLError
from .url import _BRACELESS_IPV6_ADDRZ_RE, _IPV4_RE
SSLContext = None
SSLTransport = None
HAS_NEVER_CHECK_COMMON_NAME = False
IS_PYOPENSSL = False
ALPN_PROTOCOLS = ["http/1.1"]
_TYPE_VERSION_INFO = typing.Tuple[int, int, int, str, int]
# Maps the length of a digest to a possible hash function producing this digest
HASHFUNC_MAP = {32: md5, 40: sha1, 64: sha256}
def _is_bpo_43522_fixed(
implementation_name: str,
version_info: _TYPE_VERSION_INFO,
pypy_version_info: _TYPE_VERSION_INFO | None,
) -> bool:
"""Return True for CPython 3.8.9+, 3.9.3+ or 3.10+ and PyPy 7.3.8+ where
setting SSLContext.hostname_checks_common_name to False works.
Outside of CPython and PyPy we don't know which implementations work
or not so we conservatively use our hostname matching as we know that works
on all implementations.
https://github.com/urllib3/urllib3/issues/2192#issuecomment-821832963
https://foss.heptapod.net/pypy/pypy/-/issues/3539
"""
if implementation_name == "pypy":
# https://foss.heptapod.net/pypy/pypy/-/issues/3129
return pypy_version_info >= (7, 3, 8) # type: ignore[operator]
elif implementation_name == "cpython":
major_minor = version_info[:2]
micro = version_info[2]
return (
(major_minor == (3, 8) and micro >= 9)
or (major_minor == (3, 9) and micro >= 3)
or major_minor >= (3, 10)
)
else: # Defensive:
return False
def _is_has_never_check_common_name_reliable(
openssl_version: str,
openssl_version_number: int,
implementation_name: str,
version_info: _TYPE_VERSION_INFO,
pypy_version_info: _TYPE_VERSION_INFO | None,
) -> bool:
# As of May 2023, all released versions of LibreSSL fail to reject certificates with
# only common names, see https://github.com/urllib3/urllib3/pull/3024
is_openssl = openssl_version.startswith("OpenSSL ")
# Before fixing OpenSSL issue #14579, the SSL_new() API was not copying hostflags
# like X509_CHECK_FLAG_NEVER_CHECK_SUBJECT, which tripped up CPython.
# https://github.com/openssl/openssl/issues/14579
# This was released in OpenSSL 1.1.1l+ (>=0x101010cf)
is_openssl_issue_14579_fixed = openssl_version_number >= 0x101010CF
return is_openssl and (
is_openssl_issue_14579_fixed
or _is_bpo_43522_fixed(implementation_name, version_info, pypy_version_info)
)
if typing.TYPE_CHECKING:
from ssl import VerifyMode
from typing import Literal, TypedDict
from .ssltransport import SSLTransport as SSLTransportType
class _TYPE_PEER_CERT_RET_DICT(TypedDict, total=False):
subjectAltName: tuple[tuple[str, str], ...]
subject: tuple[tuple[tuple[str, str], ...], ...]
serialNumber: str
# Mapping from 'ssl.PROTOCOL_TLSX' to 'TLSVersion.X'
_SSL_VERSION_TO_TLS_VERSION: dict[int, int] = {}
try: # Do we have ssl at all?
import ssl
from ssl import ( # type: ignore[assignment]
CERT_REQUIRED,
HAS_NEVER_CHECK_COMMON_NAME,
OP_NO_COMPRESSION,
OP_NO_TICKET,
OPENSSL_VERSION,
OPENSSL_VERSION_NUMBER,
PROTOCOL_TLS,
PROTOCOL_TLS_CLIENT,
OP_NO_SSLv2,
OP_NO_SSLv3,
SSLContext,
TLSVersion,
)
PROTOCOL_SSLv23 = PROTOCOL_TLS
# Setting SSLContext.hostname_checks_common_name = False didn't work before CPython
# 3.8.9, 3.9.3, and 3.10 (but OK on PyPy) or OpenSSL 1.1.1l+
if HAS_NEVER_CHECK_COMMON_NAME and not _is_has_never_check_common_name_reliable(
OPENSSL_VERSION,
OPENSSL_VERSION_NUMBER,
sys.implementation.name,
sys.version_info,
sys.pypy_version_info if sys.implementation.name == "pypy" else None, # type: ignore[attr-defined]
):
HAS_NEVER_CHECK_COMMON_NAME = False
# Need to be careful here in case old TLS versions get
# removed in future 'ssl' module implementations.
for attr in ("TLSv1", "TLSv1_1", "TLSv1_2"):
try:
_SSL_VERSION_TO_TLS_VERSION[getattr(ssl, f"PROTOCOL_{attr}")] = getattr(
TLSVersion, attr
)
except AttributeError: # Defensive:
continue
from .ssltransport import SSLTransport # type: ignore[assignment]
except ImportError:
OP_NO_COMPRESSION = 0x20000 # type: ignore[assignment]
OP_NO_TICKET = 0x4000 # type: ignore[assignment]
OP_NO_SSLv2 = 0x1000000 # type: ignore[assignment]
OP_NO_SSLv3 = 0x2000000 # type: ignore[assignment]
PROTOCOL_SSLv23 = PROTOCOL_TLS = 2 # type: ignore[assignment]
PROTOCOL_TLS_CLIENT = 16 # type: ignore[assignment]
_TYPE_PEER_CERT_RET = typing.Union["_TYPE_PEER_CERT_RET_DICT", bytes, None]
def assert_fingerprint(cert: bytes | None, fingerprint: str) -> None:
"""
Checks if given fingerprint matches the supplied certificate.
:param cert:
Certificate as bytes object.
:param fingerprint:
Fingerprint as string of hexdigits, can be interspersed by colons.
"""
if cert is None:
raise SSLError("No certificate for the peer.")
fingerprint = fingerprint.replace(":", "").lower()
digest_length = len(fingerprint)
hashfunc = HASHFUNC_MAP.get(digest_length)
if not hashfunc:
raise SSLError(f"Fingerprint of invalid length: {fingerprint}")
# We need encode() here for py32; works on py2 and p33.
fingerprint_bytes = unhexlify(fingerprint.encode())
cert_digest = hashfunc(cert).digest()
if not hmac.compare_digest(cert_digest, fingerprint_bytes):
raise SSLError(
f'Fingerprints did not match. Expected "{fingerprint}", got "{cert_digest.hex()}"'
)
def resolve_cert_reqs(candidate: None | int | str) -> VerifyMode:
"""
Resolves the argument to a numeric constant, which can be passed to
the wrap_socket function/method from the ssl module.
Defaults to :data:`ssl.CERT_REQUIRED`.
If given a string it is assumed to be the name of the constant in the
:mod:`ssl` module or its abbreviation.
(So you can specify `REQUIRED` instead of `CERT_REQUIRED`.
If it's neither `None` nor a string we assume it is already the numeric
constant which can directly be passed to wrap_socket.
"""
if candidate is None:
return CERT_REQUIRED
if isinstance(candidate, str):
res = getattr(ssl, candidate, None)
if res is None:
res = getattr(ssl, "CERT_" + candidate)
return res # type: ignore[no-any-return]
return candidate # type: ignore[return-value]
def resolve_ssl_version(candidate: None | int | str) -> int:
"""
like resolve_cert_reqs
"""
if candidate is None:
return PROTOCOL_TLS
if isinstance(candidate, str):
res = getattr(ssl, candidate, None)
if res is None:
res = getattr(ssl, "PROTOCOL_" + candidate)
return typing.cast(int, res)
return candidate
def create_urllib3_context(
ssl_version: int | None = None,
cert_reqs: int | None = None,
options: int | None = None,
ciphers: str | None = None,
ssl_minimum_version: int | None = None,
ssl_maximum_version: int | None = None,
) -> ssl.SSLContext:
"""Creates and configures an :class:`ssl.SSLContext` instance for use with urllib3.
:param ssl_version:
The desired protocol version to use. This will default to
PROTOCOL_SSLv23 which will negotiate the highest protocol that both
the server and your installation of OpenSSL support.
This parameter is deprecated instead use 'ssl_minimum_version'.
:param ssl_minimum_version:
The minimum version of TLS to be used. Use the 'ssl.TLSVersion' enum for specifying the value.
:param ssl_maximum_version:
The maximum version of TLS to be used. Use the 'ssl.TLSVersion' enum for specifying the value.
Not recommended to set to anything other than 'ssl.TLSVersion.MAXIMUM_SUPPORTED' which is the
default value.
:param cert_reqs:
Whether to require the certificate verification. This defaults to
``ssl.CERT_REQUIRED``.
:param options:
Specific OpenSSL options. These default to ``ssl.OP_NO_SSLv2``,
``ssl.OP_NO_SSLv3``, ``ssl.OP_NO_COMPRESSION``, and ``ssl.OP_NO_TICKET``.
:param ciphers:
Which cipher suites to allow the server to select. Defaults to either system configured
ciphers if OpenSSL 1.1.1+, otherwise uses a secure default set of ciphers.
:returns:
Constructed SSLContext object with specified options
:rtype: SSLContext
"""
if SSLContext is None:
raise TypeError("Can't create an SSLContext object without an ssl module")
# This means 'ssl_version' was specified as an exact value.
if ssl_version not in (None, PROTOCOL_TLS, PROTOCOL_TLS_CLIENT):
# Disallow setting 'ssl_version' and 'ssl_minimum|maximum_version'
# to avoid conflicts.
if ssl_minimum_version is not None or ssl_maximum_version is not None:
raise ValueError(
"Can't specify both 'ssl_version' and either "
"'ssl_minimum_version' or 'ssl_maximum_version'"
)
# 'ssl_version' is deprecated and will be removed in the future.
else:
# Use 'ssl_minimum_version' and 'ssl_maximum_version' instead.
ssl_minimum_version = _SSL_VERSION_TO_TLS_VERSION.get(
ssl_version, TLSVersion.MINIMUM_SUPPORTED
)
ssl_maximum_version = _SSL_VERSION_TO_TLS_VERSION.get(
ssl_version, TLSVersion.MAXIMUM_SUPPORTED
)
# This warning message is pushing users to use 'ssl_minimum_version'
# instead of both min/max. Best practice is to only set the minimum version and
# keep the maximum version to be it's default value: 'TLSVersion.MAXIMUM_SUPPORTED'
warnings.warn(
"'ssl_version' option is deprecated and will be "
"removed in urllib3 v2.1.0. Instead use 'ssl_minimum_version'",
category=DeprecationWarning,
stacklevel=2,
)
# PROTOCOL_TLS is deprecated in Python 3.10 so we always use PROTOCOL_TLS_CLIENT
context = SSLContext(PROTOCOL_TLS_CLIENT)
if ssl_minimum_version is not None:
context.minimum_version = ssl_minimum_version
else: # Python <3.10 defaults to 'MINIMUM_SUPPORTED' so explicitly set TLSv1.2 here
context.minimum_version = TLSVersion.TLSv1_2
if ssl_maximum_version is not None:
context.maximum_version = ssl_maximum_version
# Unless we're given ciphers defer to either system ciphers in
# the case of OpenSSL 1.1.1+ or use our own secure default ciphers.
if ciphers:
context.set_ciphers(ciphers)
# Setting the default here, as we may have no ssl module on import
cert_reqs = ssl.CERT_REQUIRED if cert_reqs is None else cert_reqs
if options is None:
options = 0
# SSLv2 is easily broken and is considered harmful and dangerous
options |= OP_NO_SSLv2
# SSLv3 has several problems and is now dangerous
options |= OP_NO_SSLv3
# Disable compression to prevent CRIME attacks for OpenSSL 1.0+
# (issue #309)
options |= OP_NO_COMPRESSION
# TLSv1.2 only. Unless set explicitly, do not request tickets.
# This may save some bandwidth on wire, and although the ticket is encrypted,
# there is a risk associated with it being on wire,
# if the server is not rotating its ticketing keys properly.
options |= OP_NO_TICKET
context.options |= options
# Enable post-handshake authentication for TLS 1.3, see GH #1634. PHA is
# necessary for conditional client cert authentication with TLS 1.3.
# The attribute is None for OpenSSL <= 1.1.0 or does not exist when using
# an SSLContext created by pyOpenSSL.
if getattr(context, "post_handshake_auth", None) is not None:
context.post_handshake_auth = True
# The order of the below lines setting verify_mode and check_hostname
# matter due to safe-guards SSLContext has to prevent an SSLContext with
# check_hostname=True, verify_mode=NONE/OPTIONAL.
# We always set 'check_hostname=False' for pyOpenSSL so we rely on our own
# 'ssl.match_hostname()' implementation.
if cert_reqs == ssl.CERT_REQUIRED and not IS_PYOPENSSL:
context.verify_mode = cert_reqs
context.check_hostname = True
else:
context.check_hostname = False
context.verify_mode = cert_reqs
try:
context.hostname_checks_common_name = False
except AttributeError: # Defensive: for CPython < 3.8.9 and 3.9.3; for PyPy < 7.3.8
pass
# Enable logging of TLS session keys via defacto standard environment variable
# 'SSLKEYLOGFILE', if the feature is available (Python 3.8+). Skip empty values.
if hasattr(context, "keylog_filename"):
sslkeylogfile = os.environ.get("SSLKEYLOGFILE")
if sslkeylogfile:
context.keylog_filename = sslkeylogfile
return context
@typing.overload
def ssl_wrap_socket(
sock: socket.socket,
keyfile: str | None = ...,
certfile: str | None = ...,
cert_reqs: int | None = ...,
ca_certs: str | None = ...,
server_hostname: str | None = ...,
ssl_version: int | None = ...,
ciphers: str | None = ...,
ssl_context: ssl.SSLContext | None = ...,
ca_cert_dir: str | None = ...,
key_password: str | None = ...,
ca_cert_data: None | str | bytes = ...,
tls_in_tls: Literal[False] = ...,
) -> ssl.SSLSocket:
...
@typing.overload
def ssl_wrap_socket(
sock: socket.socket,
keyfile: str | None = ...,
certfile: str | None = ...,
cert_reqs: int | None = ...,
ca_certs: str | None = ...,
server_hostname: str | None = ...,
ssl_version: int | None = ...,
ciphers: str | None = ...,
ssl_context: ssl.SSLContext | None = ...,
ca_cert_dir: str | None = ...,
key_password: str | None = ...,
ca_cert_data: None | str | bytes = ...,
tls_in_tls: bool = ...,
) -> ssl.SSLSocket | SSLTransportType:
...
def ssl_wrap_socket(
sock: socket.socket,
keyfile: str | None = None,
certfile: str | None = None,
cert_reqs: int | None = None,
ca_certs: str | None = None,
server_hostname: str | None = None,
ssl_version: int | None = None,
ciphers: str | None = None,
ssl_context: ssl.SSLContext | None = None,
ca_cert_dir: str | None = None,
key_password: str | None = None,
ca_cert_data: None | str | bytes = None,
tls_in_tls: bool = False,
) -> ssl.SSLSocket | SSLTransportType:
"""
All arguments except for server_hostname, ssl_context, tls_in_tls, ca_cert_data and
ca_cert_dir have the same meaning as they do when using
:func:`ssl.create_default_context`, :meth:`ssl.SSLContext.load_cert_chain`,
:meth:`ssl.SSLContext.set_ciphers` and :meth:`ssl.SSLContext.wrap_socket`.
:param server_hostname:
When SNI is supported, the expected hostname of the certificate
:param ssl_context:
A pre-made :class:`SSLContext` object. If none is provided, one will
be created using :func:`create_urllib3_context`.
:param ciphers:
A string of ciphers we wish the client to support.
:param ca_cert_dir:
A directory containing CA certificates in multiple separate files, as
supported by OpenSSL's -CApath flag or the capath argument to
SSLContext.load_verify_locations().
:param key_password:
Optional password if the keyfile is encrypted.
:param ca_cert_data:
Optional string containing CA certificates in PEM format suitable for
passing as the cadata parameter to SSLContext.load_verify_locations()
:param tls_in_tls:
Use SSLTransport to wrap the existing socket.
"""
context = ssl_context
if context is None:
# Note: This branch of code and all the variables in it are only used in tests.
# We should consider deprecating and removing this code.
context = create_urllib3_context(ssl_version, cert_reqs, ciphers=ciphers)
if ca_certs or ca_cert_dir or ca_cert_data:
try:
context.load_verify_locations(ca_certs, ca_cert_dir, ca_cert_data)
except OSError as e:
raise SSLError(e) from e
elif ssl_context is None and hasattr(context, "load_default_certs"):
# try to load OS default certs; works well on Windows.
context.load_default_certs()
# Attempt to detect if we get the goofy behavior of the
# keyfile being encrypted and OpenSSL asking for the
# passphrase via the terminal and instead error out.
if keyfile and key_password is None and _is_key_file_encrypted(keyfile):
raise SSLError("Client private key is encrypted, password is required")
if certfile:
if key_password is None:
context.load_cert_chain(certfile, keyfile)
else:
context.load_cert_chain(certfile, keyfile, key_password)
try:
context.set_alpn_protocols(ALPN_PROTOCOLS)
except NotImplementedError: # Defensive: in CI, we always have set_alpn_protocols
pass
ssl_sock = _ssl_wrap_socket_impl(sock, context, tls_in_tls, server_hostname)
return ssl_sock
def is_ipaddress(hostname: str | bytes) -> bool:
"""Detects whether the hostname given is an IPv4 or IPv6 address.
Also detects IPv6 addresses with Zone IDs.
:param str hostname: Hostname to examine.
:return: True if the hostname is an IP address, False otherwise.
"""
if isinstance(hostname, bytes):
# IDN A-label bytes are ASCII compatible.
hostname = hostname.decode("ascii")
return bool(_IPV4_RE.match(hostname) or _BRACELESS_IPV6_ADDRZ_RE.match(hostname))
def _is_key_file_encrypted(key_file: str) -> bool:
"""Detects if a key file is encrypted or not."""
with open(key_file) as f:
for line in f:
# Look for Proc-Type: 4,ENCRYPTED
if "ENCRYPTED" in line:
return True
return False
def _ssl_wrap_socket_impl(
sock: socket.socket,
ssl_context: ssl.SSLContext,
tls_in_tls: bool,
server_hostname: str | None = None,
) -> ssl.SSLSocket | SSLTransportType:
if tls_in_tls:
if not SSLTransport:
# Import error, ssl is not available.
raise ProxySchemeUnsupported(
"TLS in TLS requires support for the 'ssl' module"
)
SSLTransport._validate_ssl_context_for_tls_in_tls(ssl_context)
return SSLTransport(sock, ssl_context, server_hostname)
return ssl_context.wrap_socket(sock, server_hostname=server_hostname)

View File

@ -1,159 +0,0 @@
"""The match_hostname() function from Python 3.5, essential when using SSL."""
# Note: This file is under the PSF license as the code comes from the python
# stdlib. http://docs.python.org/3/license.html
# It is modified to remove commonName support.
from __future__ import annotations
import ipaddress
import re
import typing
from ipaddress import IPv4Address, IPv6Address
if typing.TYPE_CHECKING:
from .ssl_ import _TYPE_PEER_CERT_RET_DICT
__version__ = "3.5.0.1"
class CertificateError(ValueError):
pass
def _dnsname_match(
dn: typing.Any, hostname: str, max_wildcards: int = 1
) -> typing.Match[str] | None | bool:
"""Matching according to RFC 6125, section 6.4.3
http://tools.ietf.org/html/rfc6125#section-6.4.3
"""
pats = []
if not dn:
return False
# Ported from python3-syntax:
# leftmost, *remainder = dn.split(r'.')
parts = dn.split(r".")
leftmost = parts[0]
remainder = parts[1:]
wildcards = leftmost.count("*")
if wildcards > max_wildcards:
# Issue #17980: avoid denials of service by refusing more
# than one wildcard per fragment. A survey of established
# policy among SSL implementations showed it to be a
# reasonable choice.
raise CertificateError(
"too many wildcards in certificate DNS name: " + repr(dn)
)
# speed up common case w/o wildcards
if not wildcards:
return bool(dn.lower() == hostname.lower())
# RFC 6125, section 6.4.3, subitem 1.
# The client SHOULD NOT attempt to match a presented identifier in which
# the wildcard character comprises a label other than the left-most label.
if leftmost == "*":
# When '*' is a fragment by itself, it matches a non-empty dotless
# fragment.
pats.append("[^.]+")
elif leftmost.startswith("xn--") or hostname.startswith("xn--"):
# RFC 6125, section 6.4.3, subitem 3.
# The client SHOULD NOT attempt to match a presented identifier
# where the wildcard character is embedded within an A-label or
# U-label of an internationalized domain name.
pats.append(re.escape(leftmost))
else:
# Otherwise, '*' matches any dotless string, e.g. www*
pats.append(re.escape(leftmost).replace(r"\*", "[^.]*"))
# add the remaining fragments, ignore any wildcards
for frag in remainder:
pats.append(re.escape(frag))
pat = re.compile(r"\A" + r"\.".join(pats) + r"\Z", re.IGNORECASE)
return pat.match(hostname)
def _ipaddress_match(ipname: str, host_ip: IPv4Address | IPv6Address) -> bool:
"""Exact matching of IP addresses.
RFC 9110 section 4.3.5: "A reference identity of IP-ID contains the decoded
bytes of the IP address. An IP version 4 address is 4 octets, and an IP
version 6 address is 16 octets. [...] A reference identity of type IP-ID
matches if the address is identical to an iPAddress value of the
subjectAltName extension of the certificate."
"""
# OpenSSL may add a trailing newline to a subjectAltName's IP address
# Divergence from upstream: ipaddress can't handle byte str
ip = ipaddress.ip_address(ipname.rstrip())
return bool(ip.packed == host_ip.packed)
def match_hostname(
cert: _TYPE_PEER_CERT_RET_DICT | None,
hostname: str,
hostname_checks_common_name: bool = False,
) -> None:
"""Verify that *cert* (in decoded format as returned by
SSLSocket.getpeercert()) matches the *hostname*. RFC 2818 and RFC 6125
rules are followed, but IP addresses are not accepted for *hostname*.
CertificateError is raised on failure. On success, the function
returns nothing.
"""
if not cert:
raise ValueError(
"empty or no certificate, match_hostname needs a "
"SSL socket or SSL context with either "
"CERT_OPTIONAL or CERT_REQUIRED"
)
try:
# Divergence from upstream: ipaddress can't handle byte str
#
# The ipaddress module shipped with Python < 3.9 does not support
# scoped IPv6 addresses so we unconditionally strip the Zone IDs for
# now. Once we drop support for Python 3.9 we can remove this branch.
if "%" in hostname:
host_ip = ipaddress.ip_address(hostname[: hostname.rfind("%")])
else:
host_ip = ipaddress.ip_address(hostname)
except ValueError:
# Not an IP address (common case)
host_ip = None
dnsnames = []
san: tuple[tuple[str, str], ...] = cert.get("subjectAltName", ())
key: str
value: str
for key, value in san:
if key == "DNS":
if host_ip is None and _dnsname_match(value, hostname):
return
dnsnames.append(value)
elif key == "IP Address":
if host_ip is not None and _ipaddress_match(value, host_ip):
return
dnsnames.append(value)
# We only check 'commonName' if it's enabled and we're not verifying
# an IP address. IP addresses aren't valid within 'commonName'.
if hostname_checks_common_name and host_ip is None and not dnsnames:
for sub in cert.get("subject", ()):
for key, value in sub:
if key == "commonName":
if _dnsname_match(value, hostname):
return
dnsnames.append(value)
if len(dnsnames) > 1:
raise CertificateError(
"hostname %r "
"doesn't match either of %s" % (hostname, ", ".join(map(repr, dnsnames)))
)
elif len(dnsnames) == 1:
raise CertificateError(f"hostname {hostname!r} doesn't match {dnsnames[0]!r}")
else:
raise CertificateError("no appropriate subjectAltName fields were found")

View File

@ -1,280 +0,0 @@
from __future__ import annotations
import io
import socket
import ssl
import typing
from ..exceptions import ProxySchemeUnsupported
if typing.TYPE_CHECKING:
from typing import Literal
from .ssl_ import _TYPE_PEER_CERT_RET, _TYPE_PEER_CERT_RET_DICT
_SelfT = typing.TypeVar("_SelfT", bound="SSLTransport")
_WriteBuffer = typing.Union[bytearray, memoryview]
_ReturnValue = typing.TypeVar("_ReturnValue")
SSL_BLOCKSIZE = 16384
class SSLTransport:
"""
The SSLTransport wraps an existing socket and establishes an SSL connection.
Contrary to Python's implementation of SSLSocket, it allows you to chain
multiple TLS connections together. It's particularly useful if you need to
implement TLS within TLS.
The class supports most of the socket API operations.
"""
@staticmethod
def _validate_ssl_context_for_tls_in_tls(ssl_context: ssl.SSLContext) -> None:
"""
Raises a ProxySchemeUnsupported if the provided ssl_context can't be used
for TLS in TLS.
The only requirement is that the ssl_context provides the 'wrap_bio'
methods.
"""
if not hasattr(ssl_context, "wrap_bio"):
raise ProxySchemeUnsupported(
"TLS in TLS requires SSLContext.wrap_bio() which isn't "
"available on non-native SSLContext"
)
def __init__(
self,
socket: socket.socket,
ssl_context: ssl.SSLContext,
server_hostname: str | None = None,
suppress_ragged_eofs: bool = True,
) -> None:
"""
Create an SSLTransport around socket using the provided ssl_context.
"""
self.incoming = ssl.MemoryBIO()
self.outgoing = ssl.MemoryBIO()
self.suppress_ragged_eofs = suppress_ragged_eofs
self.socket = socket
self.sslobj = ssl_context.wrap_bio(
self.incoming, self.outgoing, server_hostname=server_hostname
)
# Perform initial handshake.
self._ssl_io_loop(self.sslobj.do_handshake)
def __enter__(self: _SelfT) -> _SelfT:
return self
def __exit__(self, *_: typing.Any) -> None:
self.close()
def fileno(self) -> int:
return self.socket.fileno()
def read(self, len: int = 1024, buffer: typing.Any | None = None) -> int | bytes:
return self._wrap_ssl_read(len, buffer)
def recv(self, buflen: int = 1024, flags: int = 0) -> int | bytes:
if flags != 0:
raise ValueError("non-zero flags not allowed in calls to recv")
return self._wrap_ssl_read(buflen)
def recv_into(
self,
buffer: _WriteBuffer,
nbytes: int | None = None,
flags: int = 0,
) -> None | int | bytes:
if flags != 0:
raise ValueError("non-zero flags not allowed in calls to recv_into")
if nbytes is None:
nbytes = len(buffer)
return self.read(nbytes, buffer)
def sendall(self, data: bytes, flags: int = 0) -> None:
if flags != 0:
raise ValueError("non-zero flags not allowed in calls to sendall")
count = 0
with memoryview(data) as view, view.cast("B") as byte_view:
amount = len(byte_view)
while count < amount:
v = self.send(byte_view[count:])
count += v
def send(self, data: bytes, flags: int = 0) -> int:
if flags != 0:
raise ValueError("non-zero flags not allowed in calls to send")
return self._ssl_io_loop(self.sslobj.write, data)
def makefile(
self,
mode: str,
buffering: int | None = None,
*,
encoding: str | None = None,
errors: str | None = None,
newline: str | None = None,
) -> typing.BinaryIO | typing.TextIO | socket.SocketIO:
"""
Python's httpclient uses makefile and buffered io when reading HTTP
messages and we need to support it.
This is unfortunately a copy and paste of socket.py makefile with small
changes to point to the socket directly.
"""
if not set(mode) <= {"r", "w", "b"}:
raise ValueError(f"invalid mode {mode!r} (only r, w, b allowed)")
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 = socket.SocketIO(self, rawmode) # type: ignore[arg-type]
self.socket._io_refs += 1 # type: ignore[attr-defined]
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
buffer: typing.BinaryIO
if reading and writing:
buffer = io.BufferedRWPair(raw, raw, buffering) # type: ignore[assignment]
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 # type: ignore[misc]
return text
def unwrap(self) -> None:
self._ssl_io_loop(self.sslobj.unwrap)
def close(self) -> None:
self.socket.close()
@typing.overload
def getpeercert(
self, binary_form: Literal[False] = ...
) -> _TYPE_PEER_CERT_RET_DICT | None:
...
@typing.overload
def getpeercert(self, binary_form: Literal[True]) -> bytes | None:
...
def getpeercert(self, binary_form: bool = False) -> _TYPE_PEER_CERT_RET:
return self.sslobj.getpeercert(binary_form) # type: ignore[return-value]
def version(self) -> str | None:
return self.sslobj.version()
def cipher(self) -> tuple[str, str, int] | None:
return self.sslobj.cipher()
def selected_alpn_protocol(self) -> str | None:
return self.sslobj.selected_alpn_protocol()
def selected_npn_protocol(self) -> str | None:
return self.sslobj.selected_npn_protocol()
def shared_ciphers(self) -> list[tuple[str, str, int]] | None:
return self.sslobj.shared_ciphers()
def compression(self) -> str | None:
return self.sslobj.compression()
def settimeout(self, value: float | None) -> None:
self.socket.settimeout(value)
def gettimeout(self) -> float | None:
return self.socket.gettimeout()
def _decref_socketios(self) -> None:
self.socket._decref_socketios() # type: ignore[attr-defined]
def _wrap_ssl_read(self, len: int, buffer: bytearray | None = None) -> int | bytes:
try:
return self._ssl_io_loop(self.sslobj.read, len, buffer)
except ssl.SSLError as e:
if e.errno == ssl.SSL_ERROR_EOF and self.suppress_ragged_eofs:
return 0 # eof, return 0.
else:
raise
# func is sslobj.do_handshake or sslobj.unwrap
@typing.overload
def _ssl_io_loop(self, func: typing.Callable[[], None]) -> None:
...
# func is sslobj.write, arg1 is data
@typing.overload
def _ssl_io_loop(self, func: typing.Callable[[bytes], int], arg1: bytes) -> int:
...
# func is sslobj.read, arg1 is len, arg2 is buffer
@typing.overload
def _ssl_io_loop(
self,
func: typing.Callable[[int, bytearray | None], bytes],
arg1: int,
arg2: bytearray | None,
) -> bytes:
...
def _ssl_io_loop(
self,
func: typing.Callable[..., _ReturnValue],
arg1: None | bytes | int = None,
arg2: bytearray | None = None,
) -> _ReturnValue:
"""Performs an I/O loop between incoming/outgoing and the socket."""
should_loop = True
ret = None
while should_loop:
errno = None
try:
if arg1 is None and arg2 is None:
ret = func()
elif arg2 is None:
ret = func(arg1)
else:
ret = func(arg1, arg2)
except ssl.SSLError as e:
if e.errno not in (ssl.SSL_ERROR_WANT_READ, ssl.SSL_ERROR_WANT_WRITE):
# WANT_READ, and WANT_WRITE are expected, others are not.
raise e
errno = e.errno
buf = self.outgoing.read()
self.socket.sendall(buf)
if errno is None:
should_loop = False
elif errno == ssl.SSL_ERROR_WANT_READ:
buf = self.socket.recv(SSL_BLOCKSIZE)
if buf:
self.incoming.write(buf)
else:
self.incoming.write_eof()
return typing.cast(_ReturnValue, ret)

View File

@ -1,275 +0,0 @@
from __future__ import annotations
import time
import typing
from enum import Enum
from socket import getdefaulttimeout
from ..exceptions import TimeoutStateError
if typing.TYPE_CHECKING:
from typing import Final
class _TYPE_DEFAULT(Enum):
# This value should never be passed to socket.settimeout() so for safety we use a -1.
# socket.settimout() raises a ValueError for negative values.
token = -1
_DEFAULT_TIMEOUT: Final[_TYPE_DEFAULT] = _TYPE_DEFAULT.token
_TYPE_TIMEOUT = typing.Optional[typing.Union[float, _TYPE_DEFAULT]]
class Timeout:
"""Timeout configuration.
Timeouts can be defined as a default for a pool:
.. code-block:: python
import urllib3
timeout = urllib3.util.Timeout(connect=2.0, read=7.0)
http = urllib3.PoolManager(timeout=timeout)
resp = http.request("GET", "https://example.com/")
print(resp.status)
Or per-request (which overrides the default for the pool):
.. code-block:: python
response = http.request("GET", "https://example.com/", timeout=Timeout(10))
Timeouts can be disabled by setting all the parameters to ``None``:
.. code-block:: python
no_timeout = Timeout(connect=None, read=None)
response = http.request("GET", "https://example.com/", timeout=no_timeout)
:param total:
This combines the connect and read timeouts into one; the read timeout
will be set to the time leftover from the connect attempt. In the
event that both a connect timeout and a total are specified, or a read
timeout and a total are specified, the shorter timeout will be applied.
Defaults to None.
:type total: int, float, or None
:param connect:
The maximum amount of time (in seconds) to wait for a connection
attempt to a server to succeed. Omitting the parameter will default the
connect timeout to the system default, probably `the global default
timeout in socket.py
<http://hg.python.org/cpython/file/603b4d593758/Lib/socket.py#l535>`_.
None will set an infinite timeout for connection attempts.
:type connect: int, float, or None
:param read:
The maximum amount of time (in seconds) to wait between consecutive
read operations for a response from the server. Omitting the parameter
will default the read timeout to the system default, probably `the
global default timeout in socket.py
<http://hg.python.org/cpython/file/603b4d593758/Lib/socket.py#l535>`_.
None will set an infinite timeout.
:type read: int, float, or None
.. note::
Many factors can affect the total amount of time for urllib3 to return
an HTTP response.
For example, Python's DNS resolver does not obey the timeout specified
on the socket. Other factors that can affect total request time include
high CPU load, high swap, the program running at a low priority level,
or other behaviors.
In addition, the read and total timeouts only measure the time between
read operations on the socket connecting the client and the server,
not the total amount of time for the request to return a complete
response. For most requests, the timeout is raised because the server
has not sent the first byte in the specified time. This is not always
the case; if a server streams one byte every fifteen seconds, a timeout
of 20 seconds will not trigger, even though the request will take
several minutes to complete.
"""
#: A sentinel object representing the default timeout value
DEFAULT_TIMEOUT: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT
def __init__(
self,
total: _TYPE_TIMEOUT = None,
connect: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT,
read: _TYPE_TIMEOUT = _DEFAULT_TIMEOUT,
) -> None:
self._connect = self._validate_timeout(connect, "connect")
self._read = self._validate_timeout(read, "read")
self.total = self._validate_timeout(total, "total")
self._start_connect: float | None = None
def __repr__(self) -> str:
return f"{type(self).__name__}(connect={self._connect!r}, read={self._read!r}, total={self.total!r})"
# __str__ provided for backwards compatibility
__str__ = __repr__
@staticmethod
def resolve_default_timeout(timeout: _TYPE_TIMEOUT) -> float | None:
return getdefaulttimeout() if timeout is _DEFAULT_TIMEOUT else timeout
@classmethod
def _validate_timeout(cls, value: _TYPE_TIMEOUT, name: str) -> _TYPE_TIMEOUT:
"""Check that a timeout attribute is valid.
:param value: The timeout value to validate
:param name: The name of the timeout attribute to validate. This is
used to specify in error messages.
:return: The validated and casted version of the given value.
:raises ValueError: If it is a numeric value less than or equal to
zero, or the type is not an integer, float, or None.
"""
if value is None or value is _DEFAULT_TIMEOUT:
return value
if isinstance(value, bool):
raise ValueError(
"Timeout cannot be a boolean value. It must "
"be an int, float or None."
)
try:
float(value)
except (TypeError, ValueError):
raise ValueError(
"Timeout value %s was %s, but it must be an "
"int, float or None." % (name, value)
) from None
try:
if value <= 0:
raise ValueError(
"Attempted to set %s timeout to %s, but the "
"timeout cannot be set to a value less "
"than or equal to 0." % (name, value)
)
except TypeError:
raise ValueError(
"Timeout value %s was %s, but it must be an "
"int, float or None." % (name, value)
) from None
return value
@classmethod
def from_float(cls, timeout: _TYPE_TIMEOUT) -> Timeout:
"""Create a new Timeout from a legacy timeout value.
The timeout value used by httplib.py sets the same timeout on the
connect(), and recv() socket requests. This creates a :class:`Timeout`
object that sets the individual timeouts to the ``timeout`` value
passed to this function.
:param timeout: The legacy timeout value.
:type timeout: integer, float, :attr:`urllib3.util.Timeout.DEFAULT_TIMEOUT`, or None
:return: Timeout object
:rtype: :class:`Timeout`
"""
return Timeout(read=timeout, connect=timeout)
def clone(self) -> Timeout:
"""Create a copy of the timeout object
Timeout properties are stored per-pool but each request needs a fresh
Timeout object to ensure each one has its own start/stop configured.
:return: a copy of the timeout object
:rtype: :class:`Timeout`
"""
# We can't use copy.deepcopy because that will also create a new object
# for _GLOBAL_DEFAULT_TIMEOUT, which socket.py uses as a sentinel to
# detect the user default.
return Timeout(connect=self._connect, read=self._read, total=self.total)
def start_connect(self) -> float:
"""Start the timeout clock, used during a connect() attempt
:raises urllib3.exceptions.TimeoutStateError: if you attempt
to start a timer that has been started already.
"""
if self._start_connect is not None:
raise TimeoutStateError("Timeout timer has already been started.")
self._start_connect = time.monotonic()
return self._start_connect
def get_connect_duration(self) -> float:
"""Gets the time elapsed since the call to :meth:`start_connect`.
:return: Elapsed time in seconds.
:rtype: float
:raises urllib3.exceptions.TimeoutStateError: if you attempt
to get duration for a timer that hasn't been started.
"""
if self._start_connect is None:
raise TimeoutStateError(
"Can't get connect duration for timer that has not started."
)
return time.monotonic() - self._start_connect
@property
def connect_timeout(self) -> _TYPE_TIMEOUT:
"""Get the value to use when setting a connection timeout.
This will be a positive float or integer, the value None
(never timeout), or the default system timeout.
:return: Connect timeout.
:rtype: int, float, :attr:`Timeout.DEFAULT_TIMEOUT` or None
"""
if self.total is None:
return self._connect
if self._connect is None or self._connect is _DEFAULT_TIMEOUT:
return self.total
return min(self._connect, self.total) # type: ignore[type-var]
@property
def read_timeout(self) -> float | None:
"""Get the value for the read timeout.
This assumes some time has elapsed in the connection timeout and
computes the read timeout appropriately.
If self.total is set, the read timeout is dependent on the amount of
time taken by the connect timeout. If the connection time has not been
established, a :exc:`~urllib3.exceptions.TimeoutStateError` will be
raised.
:return: Value to use for the read timeout.
:rtype: int, float or None
:raises urllib3.exceptions.TimeoutStateError: If :meth:`start_connect`
has not yet been called on this object.
"""
if (
self.total is not None
and self.total is not _DEFAULT_TIMEOUT
and self._read is not None
and self._read is not _DEFAULT_TIMEOUT
):
# In case the connect timeout has not yet been established.
if self._start_connect is None:
return self._read
return max(0, min(self.total - self.get_connect_duration(), self._read))
elif self.total is not None and self.total is not _DEFAULT_TIMEOUT:
return max(0, self.total - self.get_connect_duration())
else:
return self.resolve_default_timeout(self._read)

View File

@ -1,471 +0,0 @@
from __future__ import annotations
import re
import typing
from ..exceptions import LocationParseError
from .util import to_str
# We only want to normalize urls with an HTTP(S) scheme.
# urllib3 infers URLs without a scheme (None) to be http.
_NORMALIZABLE_SCHEMES = ("http", "https", None)
# Almost all of these patterns were derived from the
# 'rfc3986' module: https://github.com/python-hyper/rfc3986
_PERCENT_RE = re.compile(r"%[a-fA-F0-9]{2}")
_SCHEME_RE = re.compile(r"^(?:[a-zA-Z][a-zA-Z0-9+-]*:|/)")
_URI_RE = re.compile(
r"^(?:([a-zA-Z][a-zA-Z0-9+.-]*):)?"
r"(?://([^\\/?#]*))?"
r"([^?#]*)"
r"(?:\?([^#]*))?"
r"(?:#(.*))?$",
re.UNICODE | re.DOTALL,
)
_IPV4_PAT = r"(?:[0-9]{1,3}\.){3}[0-9]{1,3}"
_HEX_PAT = "[0-9A-Fa-f]{1,4}"
_LS32_PAT = "(?:{hex}:{hex}|{ipv4})".format(hex=_HEX_PAT, ipv4=_IPV4_PAT)
_subs = {"hex": _HEX_PAT, "ls32": _LS32_PAT}
_variations = [
# 6( h16 ":" ) ls32
"(?:%(hex)s:){6}%(ls32)s",
# "::" 5( h16 ":" ) ls32
"::(?:%(hex)s:){5}%(ls32)s",
# [ h16 ] "::" 4( h16 ":" ) ls32
"(?:%(hex)s)?::(?:%(hex)s:){4}%(ls32)s",
# [ *1( h16 ":" ) h16 ] "::" 3( h16 ":" ) ls32
"(?:(?:%(hex)s:)?%(hex)s)?::(?:%(hex)s:){3}%(ls32)s",
# [ *2( h16 ":" ) h16 ] "::" 2( h16 ":" ) ls32
"(?:(?:%(hex)s:){0,2}%(hex)s)?::(?:%(hex)s:){2}%(ls32)s",
# [ *3( h16 ":" ) h16 ] "::" h16 ":" ls32
"(?:(?:%(hex)s:){0,3}%(hex)s)?::%(hex)s:%(ls32)s",
# [ *4( h16 ":" ) h16 ] "::" ls32
"(?:(?:%(hex)s:){0,4}%(hex)s)?::%(ls32)s",
# [ *5( h16 ":" ) h16 ] "::" h16
"(?:(?:%(hex)s:){0,5}%(hex)s)?::%(hex)s",
# [ *6( h16 ":" ) h16 ] "::"
"(?:(?:%(hex)s:){0,6}%(hex)s)?::",
]
_UNRESERVED_PAT = r"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789._\-~"
_IPV6_PAT = "(?:" + "|".join([x % _subs for x in _variations]) + ")"
_ZONE_ID_PAT = "(?:%25|%)(?:[" + _UNRESERVED_PAT + "]|%[a-fA-F0-9]{2})+"
_IPV6_ADDRZ_PAT = r"\[" + _IPV6_PAT + r"(?:" + _ZONE_ID_PAT + r")?\]"
_REG_NAME_PAT = r"(?:[^\[\]%:/?#]|%[a-fA-F0-9]{2})*"
_TARGET_RE = re.compile(r"^(/[^?#]*)(?:\?([^#]*))?(?:#.*)?$")
_IPV4_RE = re.compile("^" + _IPV4_PAT + "$")
_IPV6_RE = re.compile("^" + _IPV6_PAT + "$")
_IPV6_ADDRZ_RE = re.compile("^" + _IPV6_ADDRZ_PAT + "$")
_BRACELESS_IPV6_ADDRZ_RE = re.compile("^" + _IPV6_ADDRZ_PAT[2:-2] + "$")
_ZONE_ID_RE = re.compile("(" + _ZONE_ID_PAT + r")\]$")
_HOST_PORT_PAT = ("^(%s|%s|%s)(?::0*?(|0|[1-9][0-9]{0,4}))?$") % (
_REG_NAME_PAT,
_IPV4_PAT,
_IPV6_ADDRZ_PAT,
)
_HOST_PORT_RE = re.compile(_HOST_PORT_PAT, re.UNICODE | re.DOTALL)
_UNRESERVED_CHARS = set(
"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789._-~"
)
_SUB_DELIM_CHARS = set("!$&'()*+,;=")
_USERINFO_CHARS = _UNRESERVED_CHARS | _SUB_DELIM_CHARS | {":"}
_PATH_CHARS = _USERINFO_CHARS | {"@", "/"}
_QUERY_CHARS = _FRAGMENT_CHARS = _PATH_CHARS | {"?"}
class Url(
typing.NamedTuple(
"Url",
[
("scheme", typing.Optional[str]),
("auth", typing.Optional[str]),
("host", typing.Optional[str]),
("port", typing.Optional[int]),
("path", typing.Optional[str]),
("query", typing.Optional[str]),
("fragment", typing.Optional[str]),
],
)
):
"""
Data structure for representing an HTTP URL. Used as a return value for
:func:`parse_url`. Both the scheme and host are normalized as they are
both case-insensitive according to RFC 3986.
"""
def __new__( # type: ignore[no-untyped-def]
cls,
scheme: str | None = None,
auth: str | None = None,
host: str | None = None,
port: int | None = None,
path: str | None = None,
query: str | None = None,
fragment: str | None = None,
):
if path and not path.startswith("/"):
path = "/" + path
if scheme is not None:
scheme = scheme.lower()
return super().__new__(cls, scheme, auth, host, port, path, query, fragment)
@property
def hostname(self) -> str | None:
"""For backwards-compatibility with urlparse. We're nice like that."""
return self.host
@property
def request_uri(self) -> str:
"""Absolute path including the query string."""
uri = self.path or "/"
if self.query is not None:
uri += "?" + self.query
return uri
@property
def authority(self) -> str | None:
"""
Authority component as defined in RFC 3986 3.2.
This includes userinfo (auth), host and port.
i.e.
userinfo@host:port
"""
userinfo = self.auth
netloc = self.netloc
if netloc is None or userinfo is None:
return netloc
else:
return f"{userinfo}@{netloc}"
@property
def netloc(self) -> str | None:
"""
Network location including host and port.
If you need the equivalent of urllib.parse's ``netloc``,
use the ``authority`` property instead.
"""
if self.host is None:
return None
if self.port:
return f"{self.host}:{self.port}"
return self.host
@property
def url(self) -> str:
"""
Convert self into a url
This function should more or less round-trip with :func:`.parse_url`. The
returned url may not be exactly the same as the url inputted to
:func:`.parse_url`, but it should be equivalent by the RFC (e.g., urls
with a blank port will have : removed).
Example:
.. code-block:: python
import urllib3
U = urllib3.util.parse_url("https://google.com/mail/")
print(U.url)
# "https://google.com/mail/"
print( urllib3.util.Url("https", "username:password",
"host.com", 80, "/path", "query", "fragment"
).url
)
# "https://username:password@host.com:80/path?query#fragment"
"""
scheme, auth, host, port, path, query, fragment = self
url = ""
# We use "is not None" we want things to happen with empty strings (or 0 port)
if scheme is not None:
url += scheme + "://"
if auth is not None:
url += auth + "@"
if host is not None:
url += host
if port is not None:
url += ":" + str(port)
if path is not None:
url += path
if query is not None:
url += "?" + query
if fragment is not None:
url += "#" + fragment
return url
def __str__(self) -> str:
return self.url
@typing.overload
def _encode_invalid_chars(
component: str, allowed_chars: typing.Container[str]
) -> str: # Abstract
...
@typing.overload
def _encode_invalid_chars(
component: None, allowed_chars: typing.Container[str]
) -> None: # Abstract
...
def _encode_invalid_chars(
component: str | None, allowed_chars: typing.Container[str]
) -> str | None:
"""Percent-encodes a URI component without reapplying
onto an already percent-encoded component.
"""
if component is None:
return component
component = to_str(component)
# Normalize existing percent-encoded bytes.
# Try to see if the component we're encoding is already percent-encoded
# so we can skip all '%' characters but still encode all others.
component, percent_encodings = _PERCENT_RE.subn(
lambda match: match.group(0).upper(), component
)
uri_bytes = component.encode("utf-8", "surrogatepass")
is_percent_encoded = percent_encodings == uri_bytes.count(b"%")
encoded_component = bytearray()
for i in range(0, len(uri_bytes)):
# Will return a single character bytestring
byte = uri_bytes[i : i + 1]
byte_ord = ord(byte)
if (is_percent_encoded and byte == b"%") or (
byte_ord < 128 and byte.decode() in allowed_chars
):
encoded_component += byte
continue
encoded_component.extend(b"%" + (hex(byte_ord)[2:].encode().zfill(2).upper()))
return encoded_component.decode()
def _remove_path_dot_segments(path: str) -> str:
# See http://tools.ietf.org/html/rfc3986#section-5.2.4 for pseudo-code
segments = path.split("/") # Turn the path into a list of segments
output = [] # Initialize the variable to use to store output
for segment in segments:
# '.' is the current directory, so ignore it, it is superfluous
if segment == ".":
continue
# Anything other than '..', should be appended to the output
if segment != "..":
output.append(segment)
# In this case segment == '..', if we can, we should pop the last
# element
elif output:
output.pop()
# If the path starts with '/' and the output is empty or the first string
# is non-empty
if path.startswith("/") and (not output or output[0]):
output.insert(0, "")
# If the path starts with '/.' or '/..' ensure we add one more empty
# string to add a trailing '/'
if path.endswith(("/.", "/..")):
output.append("")
return "/".join(output)
@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:
if host:
if scheme in _NORMALIZABLE_SCHEMES:
is_ipv6 = _IPV6_ADDRZ_RE.match(host)
if is_ipv6:
# IPv6 hosts of the form 'a::b%zone' are encoded in a URL as
# such per RFC 6874: 'a::b%25zone'. Unquote the ZoneID
# separator as necessary to return a valid RFC 4007 scoped IP.
match = _ZONE_ID_RE.search(host)
if match:
start, end = match.span(1)
zone_id = host[start:end]
if zone_id.startswith("%25") and zone_id != "%25":
zone_id = zone_id[3:]
else:
zone_id = zone_id[1:]
zone_id = _encode_invalid_chars(zone_id, _UNRESERVED_CHARS)
return f"{host[:start].lower()}%{zone_id}{host[end:]}"
else:
return host.lower()
elif not _IPV4_RE.match(host):
return to_str(
b".".join([_idna_encode(label) for label in host.split(".")]),
"ascii",
)
return host
def _idna_encode(name: str) -> bytes:
if not name.isascii():
try:
import idna
except ImportError:
raise LocationParseError(
"Unable to parse URL without the 'idna' module"
) from None
try:
return idna.encode(name.lower(), strict=True, std3_rules=True)
except idna.IDNAError:
raise LocationParseError(
f"Name '{name}' is not a valid IDNA label"
) from None
return name.lower().encode("ascii")
def _encode_target(target: str) -> str:
"""Percent-encodes a request target so that there are no invalid characters
Pre-condition for this function is that 'target' must start with '/'.
If that is the case then _TARGET_RE will always produce a match.
"""
match = _TARGET_RE.match(target)
if not match: # Defensive:
raise LocationParseError(f"{target!r} is not a valid request URI")
path, query = match.groups()
encoded_target = _encode_invalid_chars(path, _PATH_CHARS)
if query is not None:
query = _encode_invalid_chars(query, _QUERY_CHARS)
encoded_target += "?" + query
return encoded_target
def parse_url(url: str) -> Url:
"""
Given a url, return a parsed :class:`.Url` namedtuple. Best-effort is
performed to parse incomplete urls. Fields not provided will be None.
This parser is RFC 3986 and RFC 6874 compliant.
The parser logic and helper functions are based heavily on
work done in the ``rfc3986`` module.
:param str url: URL to parse into a :class:`.Url` namedtuple.
Partly backwards-compatible with :mod:`urllib.parse`.
Example:
.. code-block:: python
import urllib3
print( urllib3.util.parse_url('http://google.com/mail/'))
# Url(scheme='http', host='google.com', port=None, path='/mail/', ...)
print( urllib3.util.parse_url('google.com:80'))
# Url(scheme=None, host='google.com', port=80, path=None, ...)
print( urllib3.util.parse_url('/foo?bar'))
# Url(scheme=None, host=None, port=None, path='/foo', query='bar', ...)
"""
if not url:
# Empty
return Url()
source_url = url
if not _SCHEME_RE.search(url):
url = "//" + url
scheme: str | None
authority: str | None
auth: str | None
host: str | None
port: str | None
port_int: int | None
path: str | None
query: str | None
fragment: str | None
try:
scheme, authority, path, query, fragment = _URI_RE.match(url).groups() # type: ignore[union-attr]
normalize_uri = scheme is None or scheme.lower() in _NORMALIZABLE_SCHEMES
if scheme:
scheme = scheme.lower()
if authority:
auth, _, host_port = authority.rpartition("@")
auth = auth or None
host, port = _HOST_PORT_RE.match(host_port).groups() # type: ignore[union-attr]
if auth and normalize_uri:
auth = _encode_invalid_chars(auth, _USERINFO_CHARS)
if port == "":
port = None
else:
auth, host, port = None, None, None
if port is not None:
port_int = int(port)
if not (0 <= port_int <= 65535):
raise LocationParseError(url)
else:
port_int = None
host = _normalize_host(host, scheme)
if normalize_uri and path:
path = _remove_path_dot_segments(path)
path = _encode_invalid_chars(path, _PATH_CHARS)
if normalize_uri and query:
query = _encode_invalid_chars(query, _QUERY_CHARS)
if normalize_uri and fragment:
fragment = _encode_invalid_chars(fragment, _FRAGMENT_CHARS)
except (ValueError, AttributeError) as e:
raise LocationParseError(source_url) from e
# For the sake of backwards compatibility we put empty
# string values for path if there are any defined values
# beyond the path in the URL.
# TODO: Remove this when we break backwards compatibility.
if not path:
if query is not None or fragment is not None:
path = ""
else:
path = None
return Url(
scheme=scheme,
auth=auth,
host=host,
port=port_int,
path=path,
query=query,
fragment=fragment,
)

View File

@ -1,42 +0,0 @@
from __future__ import annotations
import typing
from types import TracebackType
def to_bytes(
x: str | bytes, encoding: str | None = None, errors: str | None = None
) -> bytes:
if isinstance(x, bytes):
return x
elif not isinstance(x, str):
raise TypeError(f"not expecting type {type(x).__name__}")
if encoding or errors:
return x.encode(encoding or "utf-8", errors=errors or "strict")
return x.encode()
def to_str(
x: str | bytes, encoding: str | None = None, errors: str | None = None
) -> str:
if isinstance(x, str):
return x
elif not isinstance(x, bytes):
raise TypeError(f"not expecting type {type(x).__name__}")
if encoding or errors:
return x.decode(encoding or "utf-8", errors=errors or "strict")
return x.decode()
def reraise(
tp: type[BaseException] | None,
value: BaseException,
tb: TracebackType | None = None,
) -> typing.NoReturn:
try:
if value.__traceback__ is not tb:
raise value.with_traceback(tb)
raise value
finally:
value = None # type: ignore[assignment]
tb = None

View File

@ -1,124 +0,0 @@
from __future__ import annotations
import select
import socket
from functools import partial
__all__ = ["wait_for_read", "wait_for_write"]
# How should we wait on sockets?
#
# There are two types of APIs you can use for waiting on sockets: the fancy
# modern stateful APIs like epoll/kqueue, and the older stateless APIs like
# select/poll. The stateful APIs are more efficient when you have a lots of
# sockets to keep track of, because you can set them up once and then use them
# lots of times. But we only ever want to wait on a single socket at a time
# and don't want to keep track of state, so the stateless APIs are actually
# more efficient. So we want to use select() or poll().
#
# Now, how do we choose between select() and poll()? On traditional Unixes,
# select() has a strange calling convention that makes it slow, or fail
# altogether, for high-numbered file descriptors. The point of poll() is to fix
# that, so on Unixes, we prefer poll().
#
# On Windows, there is no poll() (or at least Python doesn't provide a wrapper
# for it), but that's OK, because on Windows, select() doesn't have this
# strange calling convention; plain select() works fine.
#
# So: on Windows we use select(), and everywhere else we use poll(). We also
# fall back to select() in case poll() is somehow broken or missing.
def select_wait_for_socket(
sock: socket.socket,
read: bool = False,
write: bool = False,
timeout: float | None = None,
) -> bool:
if not read and not write:
raise RuntimeError("must specify at least one of read=True, write=True")
rcheck = []
wcheck = []
if read:
rcheck.append(sock)
if write:
wcheck.append(sock)
# When doing a non-blocking connect, most systems signal success by
# marking the socket writable. Windows, though, signals success by marked
# it as "exceptional". We paper over the difference by checking the write
# sockets for both conditions. (The stdlib selectors module does the same
# thing.)
fn = partial(select.select, rcheck, wcheck, wcheck)
rready, wready, xready = fn(timeout)
return bool(rready or wready or xready)
def poll_wait_for_socket(
sock: socket.socket,
read: bool = False,
write: bool = False,
timeout: float | None = None,
) -> bool:
if not read and not write:
raise RuntimeError("must specify at least one of read=True, write=True")
mask = 0
if read:
mask |= select.POLLIN
if write:
mask |= select.POLLOUT
poll_obj = select.poll()
poll_obj.register(sock, mask)
# For some reason, poll() takes timeout in milliseconds
def do_poll(t: float | None) -> list[tuple[int, int]]:
if t is not None:
t *= 1000
return poll_obj.poll(t)
return bool(do_poll(timeout))
def _have_working_poll() -> bool:
# Apparently some systems have a select.poll that fails as soon as you try
# to use it, either due to strange configuration or broken monkeypatching
# from libraries like eventlet/greenlet.
try:
poll_obj = select.poll()
poll_obj.poll(0)
except (AttributeError, OSError):
return False
else:
return True
def wait_for_socket(
sock: socket.socket,
read: bool = False,
write: bool = False,
timeout: float | None = None,
) -> bool:
# We delay choosing which implementation to use until the first time we're
# called. We could do it at import time, but then we might make the wrong
# decision if someone goes wild with monkeypatching select.poll after
# we're imported.
global wait_for_socket
if _have_working_poll():
wait_for_socket = poll_wait_for_socket
elif hasattr(select, "select"):
wait_for_socket = select_wait_for_socket
return wait_for_socket(sock, read, write, timeout)
def wait_for_read(sock: socket.socket, timeout: float | None = None) -> bool:
"""Waits for reading to be available on a given socket.
Returns True if the socket is readable, or False if the timeout expired.
"""
return wait_for_socket(sock, read=True, timeout=timeout)
def wait_for_write(sock: socket.socket, timeout: float | None = None) -> bool:
"""Waits for writing to be available on a given socket.
Returns True if the socket is readable, or False if the timeout expired.
"""
return wait_for_socket(sock, write=True, timeout=timeout)