mirror of
https://github.com/SickGear/SickGear.git
synced 2024-11-15 01:15:05 +00:00
Update CacheControl 0.12.11 (c05ef9e) → 0.13.1 (783a338).
This commit is contained in:
JackDandy
parent
674a2bd755
commit
6846c6cf8a
15 changed files with 395 additions and 326 deletions
|
|
@ -2,6 +2,7 @@
|
|||
|
||||
* Update Beautiful Soup 4.11.1 (r642) to 4.12.2
|
||||
* Update certifi 2023.05.07 to 2023.07.22
|
||||
* Update CacheControl 0.12.11 (c05ef9e) to 0.13.1 (783a338)
|
||||
* Update feedparser 6.0.10 (859ac57) to 6.0.10 (9865dec)
|
||||
* Update package resource API 67.5.1 (f51eccd) to 68.1.2 (1ef36f2)
|
||||
* Update soupsieve 2.3.2.post1 (792d566) to 2.4.1 (2e66beb)
|
||||
|
|
|
|||
|
|
@ -8,11 +8,21 @@ Make it easy to import from cachecontrol without long namespaces.
|
|||
"""
|
||||
__author__ = "Eric Larson"
|
||||
__email__ = "eric@ionrock.org"
|
||||
__version__ = "0.12.11"
|
||||
__version__ = "0.13.1"
|
||||
|
||||
from .wrapper import CacheControl
|
||||
from .adapter import CacheControlAdapter
|
||||
from .controller import CacheController
|
||||
from cachecontrol.adapter import CacheControlAdapter
|
||||
from cachecontrol.controller import CacheController
|
||||
from cachecontrol.wrapper import CacheControl
|
||||
|
||||
__all__ = [
|
||||
"__author__",
|
||||
"__email__",
|
||||
"__version__",
|
||||
"CacheControlAdapter",
|
||||
"CacheController",
|
||||
"CacheControl",
|
||||
]
|
||||
|
||||
import logging
|
||||
|
||||
logging.getLogger(__name__).addHandler(logging.NullHandler())
|
||||
|
|
|
|||
|
|
@ -1,8 +1,11 @@
|
|||
# SPDX-FileCopyrightText: 2015 Eric Larson
|
||||
#
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
from __future__ import annotations
|
||||
|
||||
import logging
|
||||
from argparse import ArgumentParser
|
||||
from typing import TYPE_CHECKING
|
||||
|
||||
import requests
|
||||
|
||||
|
|
@ -10,16 +13,19 @@ from cachecontrol.adapter import CacheControlAdapter
|
|||
from cachecontrol.cache import DictCache
|
||||
from cachecontrol.controller import logger
|
||||
|
||||
from argparse import ArgumentParser
|
||||
if TYPE_CHECKING:
|
||||
from argparse import Namespace
|
||||
|
||||
from cachecontrol.controller import CacheController
|
||||
|
||||
|
||||
def setup_logging():
|
||||
def setup_logging() -> None:
|
||||
logger.setLevel(logging.DEBUG)
|
||||
handler = logging.StreamHandler()
|
||||
logger.addHandler(handler)
|
||||
|
||||
|
||||
def get_session():
|
||||
def get_session() -> requests.Session:
|
||||
adapter = CacheControlAdapter(
|
||||
DictCache(), cache_etags=True, serializer=None, heuristic=None
|
||||
)
|
||||
|
|
@ -27,17 +33,17 @@ def get_session():
|
|||
sess.mount("http://", adapter)
|
||||
sess.mount("https://", adapter)
|
||||
|
||||
sess.cache_controller = adapter.controller
|
||||
sess.cache_controller = adapter.controller # type: ignore[attr-defined]
|
||||
return sess
|
||||
|
||||
|
||||
def get_args():
|
||||
def get_args() -> Namespace:
|
||||
parser = ArgumentParser()
|
||||
parser.add_argument("url", help="The URL to try and cache")
|
||||
return parser.parse_args()
|
||||
|
||||
|
||||
def main(args=None):
|
||||
def main() -> None:
|
||||
args = get_args()
|
||||
sess = get_session()
|
||||
|
||||
|
|
@ -48,10 +54,13 @@ def main(args=None):
|
|||
setup_logging()
|
||||
|
||||
# try setting the cache
|
||||
sess.cache_controller.cache_response(resp.request, resp.raw)
|
||||
cache_controller: CacheController = (
|
||||
sess.cache_controller # type: ignore[attr-defined]
|
||||
)
|
||||
cache_controller.cache_response(resp.request, resp.raw)
|
||||
|
||||
# Now try to get it
|
||||
if sess.cache_controller.cached_request(resp.request):
|
||||
if cache_controller.cached_request(resp.request):
|
||||
print("Cached!")
|
||||
else:
|
||||
print("Not cached :(")
|
||||
|
|
|
|||
|
|
@ -1,16 +1,26 @@
|
|||
# SPDX-FileCopyrightText: 2015 Eric Larson
|
||||
#
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
from __future__ import annotations
|
||||
|
||||
import types
|
||||
import functools
|
||||
import types
|
||||
import zlib
|
||||
from typing import TYPE_CHECKING, Any, Collection, Mapping
|
||||
|
||||
from requests.adapters import HTTPAdapter
|
||||
|
||||
from .controller import CacheController, PERMANENT_REDIRECT_STATUSES
|
||||
from .cache import DictCache
|
||||
from .filewrapper import CallbackFileWrapper
|
||||
from cachecontrol.cache import DictCache
|
||||
from cachecontrol.controller import PERMANENT_REDIRECT_STATUSES, CacheController
|
||||
from cachecontrol.filewrapper import CallbackFileWrapper
|
||||
|
||||
if TYPE_CHECKING:
|
||||
from requests import PreparedRequest, Response
|
||||
from urllib3 import HTTPResponse
|
||||
|
||||
from cachecontrol.cache import BaseCache
|
||||
from cachecontrol.heuristics import BaseHeuristic
|
||||
from cachecontrol.serialize import Serializer
|
||||
|
||||
|
||||
class CacheControlAdapter(HTTPAdapter):
|
||||
|
|
@ -18,16 +28,16 @@ class CacheControlAdapter(HTTPAdapter):
|
|||
|
||||
def __init__(
|
||||
self,
|
||||
cache=None,
|
||||
cache_etags=True,
|
||||
controller_class=None,
|
||||
serializer=None,
|
||||
heuristic=None,
|
||||
cacheable_methods=None,
|
||||
*args,
|
||||
**kw
|
||||
):
|
||||
super(CacheControlAdapter, self).__init__(*args, **kw)
|
||||
cache: BaseCache | None = None,
|
||||
cache_etags: bool = True,
|
||||
controller_class: type[CacheController] | None = None,
|
||||
serializer: Serializer | None = None,
|
||||
heuristic: BaseHeuristic | None = None,
|
||||
cacheable_methods: Collection[str] | None = None,
|
||||
*args: Any,
|
||||
**kw: Any,
|
||||
) -> None:
|
||||
super().__init__(*args, **kw)
|
||||
self.cache = DictCache() if cache is None else cache
|
||||
self.heuristic = heuristic
|
||||
self.cacheable_methods = cacheable_methods or ("GET",)
|
||||
|
|
@ -37,7 +47,16 @@ class CacheControlAdapter(HTTPAdapter):
|
|||
self.cache, cache_etags=cache_etags, serializer=serializer
|
||||
)
|
||||
|
||||
def send(self, request, cacheable_methods=None, **kw):
|
||||
def send(
|
||||
self,
|
||||
request: PreparedRequest,
|
||||
stream: bool = False,
|
||||
timeout: None | float | tuple[float, float] | tuple[float, None] = None,
|
||||
verify: bool | str = True,
|
||||
cert: (None | bytes | str | tuple[bytes | str, bytes | str]) = None,
|
||||
proxies: Mapping[str, str] | None = None,
|
||||
cacheable_methods: Collection[str] | None = None,
|
||||
) -> Response:
|
||||
"""
|
||||
Send a request. Use the request information to see if it
|
||||
exists in the cache and cache the response if we need to and can.
|
||||
|
|
@ -54,13 +73,17 @@ class CacheControlAdapter(HTTPAdapter):
|
|||
# check for etags and add headers if appropriate
|
||||
request.headers.update(self.controller.conditional_headers(request))
|
||||
|
||||
resp = super(CacheControlAdapter, self).send(request, **kw)
|
||||
resp = super().send(request, stream, timeout, verify, cert, proxies)
|
||||
|
||||
return resp
|
||||
|
||||
def build_response(
|
||||
self, request, response, from_cache=False, cacheable_methods=None
|
||||
):
|
||||
self,
|
||||
request: PreparedRequest,
|
||||
response: HTTPResponse,
|
||||
from_cache: bool = False,
|
||||
cacheable_methods: Collection[str] | None = None,
|
||||
) -> Response:
|
||||
"""
|
||||
Build a response by making a request or using the cache.
|
||||
|
||||
|
|
@ -102,36 +125,37 @@ class CacheControlAdapter(HTTPAdapter):
|
|||
else:
|
||||
# Wrap the response file with a wrapper that will cache the
|
||||
# response when the stream has been consumed.
|
||||
response._fp = CallbackFileWrapper(
|
||||
response._fp,
|
||||
response._fp = CallbackFileWrapper( # type: ignore[attr-defined]
|
||||
response._fp, # type: ignore[attr-defined]
|
||||
functools.partial(
|
||||
self.controller.cache_response, request, response
|
||||
),
|
||||
)
|
||||
if response.chunked:
|
||||
super_update_chunk_length = response._update_chunk_length
|
||||
super_update_chunk_length = response._update_chunk_length # type: ignore[attr-defined]
|
||||
|
||||
def _update_chunk_length(self):
|
||||
def _update_chunk_length(self: HTTPResponse) -> None:
|
||||
super_update_chunk_length()
|
||||
if self.chunk_left == 0:
|
||||
self._fp._close()
|
||||
self._fp._close() # type: ignore[attr-defined]
|
||||
|
||||
response._update_chunk_length = types.MethodType(
|
||||
response._update_chunk_length = types.MethodType( # type: ignore[attr-defined]
|
||||
_update_chunk_length, response
|
||||
)
|
||||
|
||||
resp = super(CacheControlAdapter, self).build_response(request, response)
|
||||
resp: Response = super().build_response(request, response) # type: ignore[no-untyped-call]
|
||||
|
||||
# See if we should invalidate the cache.
|
||||
if request.method in self.invalidating_methods and resp.ok:
|
||||
assert request.url is not None
|
||||
cache_url = self.controller.cache_url(request.url)
|
||||
self.cache.delete(cache_url)
|
||||
|
||||
# Give the request a from_cache attr to let people use it
|
||||
resp.from_cache = from_cache
|
||||
resp.from_cache = from_cache # type: ignore[attr-defined]
|
||||
|
||||
return resp
|
||||
|
||||
def close(self):
|
||||
def close(self) -> None:
|
||||
self.cache.close()
|
||||
super(CacheControlAdapter, self).close()
|
||||
super().close() # type: ignore[no-untyped-call]
|
||||
|
|
|
|||
|
|
@ -6,38 +6,46 @@
|
|||
The cache object API for implementing caches. The default is a thread
|
||||
safe in-memory dictionary.
|
||||
"""
|
||||
from __future__ import annotations
|
||||
|
||||
from threading import Lock
|
||||
from typing import IO, TYPE_CHECKING, MutableMapping
|
||||
|
||||
if TYPE_CHECKING:
|
||||
from datetime import datetime
|
||||
|
||||
|
||||
class BaseCache(object):
|
||||
|
||||
def get(self, key):
|
||||
class BaseCache:
|
||||
def get(self, key: str) -> bytes | None:
|
||||
raise NotImplementedError()
|
||||
|
||||
def set(self, key, value, expires=None):
|
||||
def set(
|
||||
self, key: str, value: bytes, expires: int | datetime | None = None
|
||||
) -> None:
|
||||
raise NotImplementedError()
|
||||
|
||||
def delete(self, key):
|
||||
def delete(self, key: str) -> None:
|
||||
raise NotImplementedError()
|
||||
|
||||
def close(self):
|
||||
def close(self) -> None:
|
||||
pass
|
||||
|
||||
|
||||
class DictCache(BaseCache):
|
||||
|
||||
def __init__(self, init_dict=None):
|
||||
def __init__(self, init_dict: MutableMapping[str, bytes] | None = None) -> None:
|
||||
self.lock = Lock()
|
||||
self.data = init_dict or {}
|
||||
|
||||
def get(self, key):
|
||||
def get(self, key: str) -> bytes | None:
|
||||
return self.data.get(key, None)
|
||||
|
||||
def set(self, key, value, expires=None):
|
||||
def set(
|
||||
self, key: str, value: bytes, expires: int | datetime | None = None
|
||||
) -> None:
|
||||
with self.lock:
|
||||
self.data.update({key: value})
|
||||
|
||||
def delete(self, key):
|
||||
def delete(self, key: str) -> None:
|
||||
with self.lock:
|
||||
if key in self.data:
|
||||
self.data.pop(key)
|
||||
|
|
@ -55,10 +63,11 @@ class SeparateBodyBaseCache(BaseCache):
|
|||
|
||||
Similarly, the body should be loaded separately via ``get_body()``.
|
||||
"""
|
||||
def set_body(self, key, body):
|
||||
|
||||
def set_body(self, key: str, body: bytes) -> None:
|
||||
raise NotImplementedError()
|
||||
|
||||
def get_body(self, key):
|
||||
def get_body(self, key: str) -> IO[bytes] | None:
|
||||
"""
|
||||
Return the body as file-like object.
|
||||
"""
|
||||
|
|
|
|||
|
|
@ -2,8 +2,7 @@
|
|||
#
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
|
||||
from .file_cache import FileCache, SeparateBodyFileCache
|
||||
from .redis_cache import RedisCache
|
||||
|
||||
from cachecontrol.caches.file_cache import FileCache, SeparateBodyFileCache
|
||||
from cachecontrol.caches.redis_cache import RedisCache
|
||||
|
||||
__all__ = ["FileCache", "SeparateBodyFileCache", "RedisCache"]
|
||||
|
|
|
|||
|
|
@ -1,22 +1,25 @@
|
|||
# SPDX-FileCopyrightText: 2015 Eric Larson
|
||||
#
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
from __future__ import annotations
|
||||
|
||||
import gc
|
||||
import hashlib
|
||||
import os
|
||||
from textwrap import dedent
|
||||
from typing import IO, TYPE_CHECKING, Union
|
||||
from pathlib import Path
|
||||
|
||||
from ..cache import BaseCache, SeparateBodyBaseCache
|
||||
from ..controller import CacheController
|
||||
from cachecontrol.cache import BaseCache, SeparateBodyBaseCache
|
||||
from cachecontrol.controller import CacheController
|
||||
|
||||
try:
|
||||
FileNotFoundError
|
||||
except NameError:
|
||||
# py2.X
|
||||
FileNotFoundError = (IOError, OSError)
|
||||
if TYPE_CHECKING:
|
||||
from datetime import datetime
|
||||
|
||||
from filelock import BaseFileLock
|
||||
|
||||
|
||||
def _secure_open_write(filename, fmode):
|
||||
def _secure_open_write(filename: str, fmode: int) -> IO[bytes]:
|
||||
# We only want to write to this file, so open it in write only mode
|
||||
flags = os.O_WRONLY
|
||||
|
||||
|
|
@ -40,7 +43,7 @@ def _secure_open_write(filename, fmode):
|
|||
try:
|
||||
os.remove(filename)
|
||||
gc.collect(2)
|
||||
except (IOError, OSError):
|
||||
except OSError:
|
||||
# The file must not exist already, so we can just skip ahead to opening
|
||||
pass
|
||||
|
||||
|
|
@ -63,23 +66,23 @@ class _FileCacheMixin:
|
|||
|
||||
def __init__(
|
||||
self,
|
||||
directory,
|
||||
forever=False,
|
||||
filemode=0o0600,
|
||||
dirmode=0o0700,
|
||||
lock_class=None,
|
||||
):
|
||||
|
||||
directory: Union[str, Path],
|
||||
forever: bool = False,
|
||||
filemode: int = 0o0600,
|
||||
dirmode: int = 0o0700,
|
||||
lock_class: type[BaseFileLock] | None = None,
|
||||
) -> None:
|
||||
try:
|
||||
if lock_class is None:
|
||||
from filelock import FileLock
|
||||
|
||||
lock_class = FileLock
|
||||
except ImportError:
|
||||
notice = dedent(
|
||||
"""
|
||||
NOTE: In order to use the FileCache you must have
|
||||
filelock installed. You can install it via pip:
|
||||
pip install filelock
|
||||
pip install cachecontrol[filecache]
|
||||
"""
|
||||
)
|
||||
raise ImportError(notice)
|
||||
|
|
@ -91,17 +94,17 @@ class _FileCacheMixin:
|
|||
self.lock_class = lock_class
|
||||
|
||||
@staticmethod
|
||||
def encode(x):
|
||||
def encode(x: str) -> str:
|
||||
return hashlib.sha224(x.encode()).hexdigest()
|
||||
|
||||
def _fn(self, name):
|
||||
def _fn(self, name: str) -> str:
|
||||
# NOTE: This method should not change as some may depend on it.
|
||||
# See: https://github.com/ionrock/cachecontrol/issues/63
|
||||
hashed = self.encode(name)
|
||||
parts = list(hashed[:5]) + [hashed]
|
||||
return os.path.join(self.directory, *parts)
|
||||
|
||||
def get(self, key):
|
||||
def get(self, key: str) -> bytes | None:
|
||||
name = self._fn(key)
|
||||
try:
|
||||
with open(name, "rb") as fh:
|
||||
|
|
@ -110,18 +113,20 @@ class _FileCacheMixin:
|
|||
except FileNotFoundError:
|
||||
return None
|
||||
|
||||
def set(self, key, value, expires=None):
|
||||
def set(
|
||||
self, key: str, value: bytes, expires: int | datetime | None = None
|
||||
) -> None:
|
||||
name = self._fn(key)
|
||||
self._write(name, value)
|
||||
|
||||
def _write(self, path, data: bytes):
|
||||
def _write(self, path: str, data: bytes) -> None:
|
||||
"""
|
||||
Safely write the data to the given path.
|
||||
"""
|
||||
# Make sure the directory exists
|
||||
try:
|
||||
os.makedirs(os.path.dirname(path), self.dirmode)
|
||||
except (IOError, OSError):
|
||||
except OSError:
|
||||
pass
|
||||
|
||||
with self.lock_class(path + ".lock"):
|
||||
|
|
@ -129,7 +134,7 @@ class _FileCacheMixin:
|
|||
with _secure_open_write(path, self.filemode) as fh:
|
||||
fh.write(data)
|
||||
|
||||
def _delete(self, key, suffix):
|
||||
def _delete(self, key: str, suffix: str) -> None:
|
||||
name = self._fn(key) + suffix
|
||||
if not self.forever:
|
||||
try:
|
||||
|
|
@ -144,7 +149,7 @@ class FileCache(_FileCacheMixin, BaseCache):
|
|||
downloads.
|
||||
"""
|
||||
|
||||
def delete(self, key):
|
||||
def delete(self, key: str) -> None:
|
||||
self._delete(key, "")
|
||||
|
||||
|
||||
|
|
@ -154,23 +159,23 @@ class SeparateBodyFileCache(_FileCacheMixin, SeparateBodyBaseCache):
|
|||
peak memory usage.
|
||||
"""
|
||||
|
||||
def get_body(self, key):
|
||||
def get_body(self, key: str) -> IO[bytes] | None:
|
||||
name = self._fn(key) + ".body"
|
||||
try:
|
||||
return open(name, "rb")
|
||||
except FileNotFoundError:
|
||||
return None
|
||||
|
||||
def set_body(self, key, body):
|
||||
def set_body(self, key: str, body: bytes) -> None:
|
||||
name = self._fn(key) + ".body"
|
||||
self._write(name, body)
|
||||
|
||||
def delete(self, key):
|
||||
def delete(self, key: str) -> None:
|
||||
self._delete(key, "")
|
||||
self._delete(key, ".body")
|
||||
|
||||
|
||||
def url_to_file_path(url, filecache):
|
||||
def url_to_file_path(url: str, filecache: FileCache) -> str:
|
||||
"""Return the file cache path based on the URL.
|
||||
|
||||
This does not ensure the file exists!
|
||||
|
|
|
|||
|
|
@ -1,39 +1,48 @@
|
|||
# SPDX-FileCopyrightText: 2015 Eric Larson
|
||||
#
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
from __future__ import annotations
|
||||
|
||||
from __future__ import division
|
||||
|
||||
from datetime import datetime
|
||||
from datetime import datetime, timezone
|
||||
from typing import TYPE_CHECKING
|
||||
|
||||
from cachecontrol.cache import BaseCache
|
||||
|
||||
if TYPE_CHECKING:
|
||||
from redis import Redis
|
||||
|
||||
|
||||
class RedisCache(BaseCache):
|
||||
|
||||
def __init__(self, conn):
|
||||
def __init__(self, conn: Redis[bytes]) -> None:
|
||||
self.conn = conn
|
||||
|
||||
def get(self, key):
|
||||
def get(self, key: str) -> bytes | None:
|
||||
return self.conn.get(key)
|
||||
|
||||
def set(self, key, value, expires=None):
|
||||
def set(
|
||||
self, key: str, value: bytes, expires: int | datetime | None = None
|
||||
) -> None:
|
||||
if not expires:
|
||||
self.conn.set(key, value)
|
||||
elif isinstance(expires, datetime):
|
||||
expires = expires - datetime.utcnow()
|
||||
self.conn.setex(key, int(expires.total_seconds()), value)
|
||||
now_utc = datetime.now(timezone.utc)
|
||||
if expires.tzinfo is None:
|
||||
now_utc = now_utc.replace(tzinfo=None)
|
||||
delta = expires - now_utc
|
||||
self.conn.setex(key, int(delta.total_seconds()), value)
|
||||
else:
|
||||
self.conn.setex(key, expires, value)
|
||||
|
||||
def delete(self, key):
|
||||
def delete(self, key: str) -> None:
|
||||
self.conn.delete(key)
|
||||
|
||||
def clear(self):
|
||||
def clear(self) -> None:
|
||||
"""Helper for clearing all the keys in a database. Use with
|
||||
caution!"""
|
||||
for key in self.conn.keys():
|
||||
self.conn.delete(key)
|
||||
|
||||
def close(self):
|
||||
def close(self) -> None:
|
||||
"""Redis uses connection pooling, no need to close the connection."""
|
||||
pass
|
||||
|
|
|
|||
|
|
@ -1,32 +0,0 @@
|
|||
# SPDX-FileCopyrightText: 2015 Eric Larson
|
||||
#
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
|
||||
try:
|
||||
from urllib.parse import urljoin
|
||||
except ImportError:
|
||||
from urlparse import urljoin
|
||||
|
||||
|
||||
try:
|
||||
import cPickle as pickle
|
||||
except ImportError:
|
||||
import pickle
|
||||
|
||||
# Handle the case where the requests module has been patched to not have
|
||||
# urllib3 bundled as part of its source.
|
||||
try:
|
||||
from requests.packages.urllib3.response import HTTPResponse
|
||||
except ImportError:
|
||||
from urllib3.response import HTTPResponse
|
||||
|
||||
try:
|
||||
from requests.packages.urllib3.util import is_fp_closed
|
||||
except ImportError:
|
||||
from urllib3.util import is_fp_closed
|
||||
|
||||
# Replicate some six behaviour
|
||||
try:
|
||||
text_type = unicode
|
||||
except NameError:
|
||||
text_type = str
|
||||
|
|
@ -5,17 +5,27 @@
|
|||
"""
|
||||
The httplib2 algorithms ported for use with requests.
|
||||
"""
|
||||
from __future__ import annotations
|
||||
|
||||
import calendar
|
||||
import logging
|
||||
import re
|
||||
import calendar
|
||||
import time
|
||||
from email.utils import parsedate_tz
|
||||
from typing import TYPE_CHECKING, Collection, Mapping
|
||||
|
||||
from requests.structures import CaseInsensitiveDict
|
||||
|
||||
from .cache import DictCache, SeparateBodyBaseCache
|
||||
from .serialize import Serializer
|
||||
from cachecontrol.cache import DictCache, SeparateBodyBaseCache
|
||||
from cachecontrol.serialize import Serializer
|
||||
|
||||
if TYPE_CHECKING:
|
||||
from typing import Literal
|
||||
|
||||
from requests import PreparedRequest
|
||||
from urllib3 import HTTPResponse
|
||||
|
||||
from cachecontrol.cache import BaseCache
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
|
@ -24,20 +34,26 @@ URI = re.compile(r"^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\?([^#]*))?(#(.*))?")
|
|||
PERMANENT_REDIRECT_STATUSES = (301, 308)
|
||||
|
||||
|
||||
def parse_uri(uri):
|
||||
def parse_uri(uri: str) -> tuple[str, str, str, str, str]:
|
||||
"""Parses a URI using the regex given in Appendix B of RFC 3986.
|
||||
|
||||
(scheme, authority, path, query, fragment) = parse_uri(uri)
|
||||
"""
|
||||
groups = URI.match(uri).groups()
|
||||
match = URI.match(uri)
|
||||
assert match is not None
|
||||
groups = match.groups()
|
||||
return (groups[1], groups[3], groups[4], groups[6], groups[8])
|
||||
|
||||
|
||||
class CacheController(object):
|
||||
class CacheController:
|
||||
"""An interface to see if request should cached or not."""
|
||||
|
||||
def __init__(
|
||||
self, cache=None, cache_etags=True, serializer=None, status_codes=None
|
||||
self,
|
||||
cache: BaseCache | None = None,
|
||||
cache_etags: bool = True,
|
||||
serializer: Serializer | None = None,
|
||||
status_codes: Collection[int] | None = None,
|
||||
):
|
||||
self.cache = DictCache() if cache is None else cache
|
||||
self.cache_etags = cache_etags
|
||||
|
|
@ -45,7 +61,7 @@ class CacheController(object):
|
|||
self.cacheable_status_codes = status_codes or (200, 203, 300, 301, 308)
|
||||
|
||||
@classmethod
|
||||
def _urlnorm(cls, uri):
|
||||
def _urlnorm(cls, uri: str) -> str:
|
||||
"""Normalize the URL to create a safe key for the cache"""
|
||||
(scheme, authority, path, query, fragment) = parse_uri(uri)
|
||||
if not scheme or not authority:
|
||||
|
|
@ -65,10 +81,10 @@ class CacheController(object):
|
|||
return defrag_uri
|
||||
|
||||
@classmethod
|
||||
def cache_url(cls, uri):
|
||||
def cache_url(cls, uri: str) -> str:
|
||||
return cls._urlnorm(uri)
|
||||
|
||||
def parse_cache_control(self, headers):
|
||||
def parse_cache_control(self, headers: Mapping[str, str]) -> dict[str, int | None]:
|
||||
known_directives = {
|
||||
# https://tools.ietf.org/html/rfc7234#section-5.2
|
||||
"max-age": (int, True),
|
||||
|
|
@ -87,7 +103,7 @@ class CacheController(object):
|
|||
|
||||
cc_headers = headers.get("cache-control", headers.get("Cache-Control", ""))
|
||||
|
||||
retval = {}
|
||||
retval: dict[str, int | None] = {}
|
||||
|
||||
for cc_directive in cc_headers.split(","):
|
||||
if not cc_directive.strip():
|
||||
|
|
@ -122,11 +138,12 @@ class CacheController(object):
|
|||
|
||||
return retval
|
||||
|
||||
def _load_from_cache(self, request):
|
||||
def _load_from_cache(self, request: PreparedRequest) -> HTTPResponse | None:
|
||||
"""
|
||||
Load a cached response, or return None if it's not available.
|
||||
"""
|
||||
cache_url = request.url
|
||||
assert cache_url is not None
|
||||
cache_data = self.cache.get(cache_url)
|
||||
if cache_data is None:
|
||||
logger.debug("No cache entry available")
|
||||
|
|
@ -142,11 +159,12 @@ class CacheController(object):
|
|||
logger.warning("Cache entry deserialization failed, entry ignored")
|
||||
return result
|
||||
|
||||
def cached_request(self, request):
|
||||
def cached_request(self, request: PreparedRequest) -> HTTPResponse | Literal[False]:
|
||||
"""
|
||||
Return a cached response if it exists in the cache, otherwise
|
||||
return False.
|
||||
"""
|
||||
assert request.url is not None
|
||||
cache_url = self.cache_url(request.url)
|
||||
logger.debug('Looking up "%s" in the cache', cache_url)
|
||||
cc = self.parse_cache_control(request.headers)
|
||||
|
|
@ -182,7 +200,7 @@ class CacheController(object):
|
|||
logger.debug(msg)
|
||||
return resp
|
||||
|
||||
headers = CaseInsensitiveDict(resp.headers)
|
||||
headers: CaseInsensitiveDict[str] = CaseInsensitiveDict(resp.headers)
|
||||
if not headers or "date" not in headers:
|
||||
if "etag" not in headers:
|
||||
# Without date or etag, the cached response can never be used
|
||||
|
|
@ -193,7 +211,9 @@ class CacheController(object):
|
|||
return False
|
||||
|
||||
now = time.time()
|
||||
date = calendar.timegm(parsedate_tz(headers["date"]))
|
||||
time_tuple = parsedate_tz(headers["date"])
|
||||
assert time_tuple is not None
|
||||
date = calendar.timegm(time_tuple[:6])
|
||||
current_age = max(0, now - date)
|
||||
logger.debug("Current age based on date: %i", current_age)
|
||||
|
||||
|
|
@ -207,28 +227,30 @@ class CacheController(object):
|
|||
freshness_lifetime = 0
|
||||
|
||||
# Check the max-age pragma in the cache control header
|
||||
if "max-age" in resp_cc:
|
||||
freshness_lifetime = resp_cc["max-age"]
|
||||
max_age = resp_cc.get("max-age")
|
||||
if max_age is not None:
|
||||
freshness_lifetime = max_age
|
||||
logger.debug("Freshness lifetime from max-age: %i", freshness_lifetime)
|
||||
|
||||
# If there isn't a max-age, check for an expires header
|
||||
elif "expires" in headers:
|
||||
expires = parsedate_tz(headers["expires"])
|
||||
if expires is not None:
|
||||
expire_time = calendar.timegm(expires) - date
|
||||
expire_time = calendar.timegm(expires[:6]) - date
|
||||
freshness_lifetime = max(0, expire_time)
|
||||
logger.debug("Freshness lifetime from expires: %i", freshness_lifetime)
|
||||
|
||||
# Determine if we are setting freshness limit in the
|
||||
# request. Note, this overrides what was in the response.
|
||||
if "max-age" in cc:
|
||||
freshness_lifetime = cc["max-age"]
|
||||
max_age = cc.get("max-age")
|
||||
if max_age is not None:
|
||||
freshness_lifetime = max_age
|
||||
logger.debug(
|
||||
"Freshness lifetime from request max-age: %i", freshness_lifetime
|
||||
)
|
||||
|
||||
if "min-fresh" in cc:
|
||||
min_fresh = cc["min-fresh"]
|
||||
min_fresh = cc.get("min-fresh")
|
||||
if min_fresh is not None:
|
||||
# adjust our current age by our min fresh
|
||||
current_age += min_fresh
|
||||
logger.debug("Adjusted current age from min-fresh: %i", current_age)
|
||||
|
|
@ -247,12 +269,12 @@ class CacheController(object):
|
|||
# return the original handler
|
||||
return False
|
||||
|
||||
def conditional_headers(self, request):
|
||||
def conditional_headers(self, request: PreparedRequest) -> dict[str, str]:
|
||||
resp = self._load_from_cache(request)
|
||||
new_headers = {}
|
||||
|
||||
if resp:
|
||||
headers = CaseInsensitiveDict(resp.headers)
|
||||
headers: CaseInsensitiveDict[str] = CaseInsensitiveDict(resp.headers)
|
||||
|
||||
if "etag" in headers:
|
||||
new_headers["If-None-Match"] = headers["ETag"]
|
||||
|
|
@ -262,7 +284,14 @@ class CacheController(object):
|
|||
|
||||
return new_headers
|
||||
|
||||
def _cache_set(self, cache_url, request, response, body=None, expires_time=None):
|
||||
def _cache_set(
|
||||
self,
|
||||
cache_url: str,
|
||||
request: PreparedRequest,
|
||||
response: HTTPResponse,
|
||||
body: bytes | None = None,
|
||||
expires_time: int | None = None,
|
||||
) -> None:
|
||||
"""
|
||||
Store the data in the cache.
|
||||
"""
|
||||
|
|
@ -285,7 +314,13 @@ class CacheController(object):
|
|||
expires=expires_time,
|
||||
)
|
||||
|
||||
def cache_response(self, request, response, body=None, status_codes=None):
|
||||
def cache_response(
|
||||
self,
|
||||
request: PreparedRequest,
|
||||
response: HTTPResponse,
|
||||
body: bytes | None = None,
|
||||
status_codes: Collection[int] | None = None,
|
||||
) -> None:
|
||||
"""
|
||||
Algorithm for caching requests.
|
||||
|
||||
|
|
@ -300,10 +335,14 @@ class CacheController(object):
|
|||
)
|
||||
return
|
||||
|
||||
response_headers = CaseInsensitiveDict(response.headers)
|
||||
response_headers: CaseInsensitiveDict[str] = CaseInsensitiveDict(
|
||||
response.headers
|
||||
)
|
||||
|
||||
if "date" in response_headers:
|
||||
date = calendar.timegm(parsedate_tz(response_headers["date"]))
|
||||
time_tuple = parsedate_tz(response_headers["date"])
|
||||
assert time_tuple is not None
|
||||
date = calendar.timegm(time_tuple[:6])
|
||||
else:
|
||||
date = 0
|
||||
|
||||
|
|
@ -322,6 +361,7 @@ class CacheController(object):
|
|||
cc_req = self.parse_cache_control(request.headers)
|
||||
cc = self.parse_cache_control(response_headers)
|
||||
|
||||
assert request.url is not None
|
||||
cache_url = self.cache_url(request.url)
|
||||
logger.debug('Updating cache with response from "%s"', cache_url)
|
||||
|
||||
|
|
@ -354,11 +394,11 @@ class CacheController(object):
|
|||
if response_headers.get("expires"):
|
||||
expires = parsedate_tz(response_headers["expires"])
|
||||
if expires is not None:
|
||||
expires_time = calendar.timegm(expires) - date
|
||||
expires_time = calendar.timegm(expires[:6]) - date
|
||||
|
||||
expires_time = max(expires_time, 14 * 86400)
|
||||
|
||||
logger.debug("etag object cached for {0} seconds".format(expires_time))
|
||||
logger.debug(f"etag object cached for {expires_time} seconds")
|
||||
logger.debug("Caching due to etag")
|
||||
self._cache_set(cache_url, request, response, body, expires_time)
|
||||
|
||||
|
|
@ -372,11 +412,14 @@ class CacheController(object):
|
|||
# is no date header then we can't do anything about expiring
|
||||
# the cache.
|
||||
elif "date" in response_headers:
|
||||
date = calendar.timegm(parsedate_tz(response_headers["date"]))
|
||||
time_tuple = parsedate_tz(response_headers["date"])
|
||||
assert time_tuple is not None
|
||||
date = calendar.timegm(time_tuple[:6])
|
||||
# cache when there is a max-age > 0
|
||||
if "max-age" in cc and cc["max-age"] > 0:
|
||||
max_age = cc.get("max-age")
|
||||
if max_age is not None and max_age > 0:
|
||||
logger.debug("Caching b/c date exists and max-age > 0")
|
||||
expires_time = cc["max-age"]
|
||||
expires_time = max_age
|
||||
self._cache_set(
|
||||
cache_url,
|
||||
request,
|
||||
|
|
@ -391,12 +434,12 @@ class CacheController(object):
|
|||
if response_headers["expires"]:
|
||||
expires = parsedate_tz(response_headers["expires"])
|
||||
if expires is not None:
|
||||
expires_time = calendar.timegm(expires) - date
|
||||
expires_time = calendar.timegm(expires[:6]) - date
|
||||
else:
|
||||
expires_time = None
|
||||
|
||||
logger.debug(
|
||||
"Caching b/c of expires header. expires in {0} seconds".format(
|
||||
"Caching b/c of expires header. expires in {} seconds".format(
|
||||
expires_time
|
||||
)
|
||||
)
|
||||
|
|
@ -408,13 +451,16 @@ class CacheController(object):
|
|||
expires_time,
|
||||
)
|
||||
|
||||
def update_cached_response(self, request, response):
|
||||
def update_cached_response(
|
||||
self, request: PreparedRequest, response: HTTPResponse
|
||||
) -> HTTPResponse:
|
||||
"""On a 304 we will get a new set of headers that we want to
|
||||
update our cached value with, assuming we have one.
|
||||
|
||||
This should only ever be called when we've sent an ETag and
|
||||
gotten a 304 as the response.
|
||||
"""
|
||||
assert request.url is not None
|
||||
cache_url = self.cache_url(request.url)
|
||||
cached_response = self._load_from_cache(request)
|
||||
|
||||
|
|
@ -432,11 +478,11 @@ class CacheController(object):
|
|||
excluded_headers = ["content-length"]
|
||||
|
||||
cached_response.headers.update(
|
||||
dict(
|
||||
(k, v)
|
||||
for k, v in response.headers.items()
|
||||
{
|
||||
k: v
|
||||
for k, v in response.headers.items() # type: ignore[no-untyped-call]
|
||||
if k.lower() not in excluded_headers
|
||||
)
|
||||
}
|
||||
)
|
||||
|
||||
# we want a 200 b/c we have content via the cache
|
||||
|
|
|
|||
|
|
@ -1,12 +1,17 @@
|
|||
# SPDX-FileCopyrightText: 2015 Eric Larson
|
||||
#
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
from __future__ import annotations
|
||||
|
||||
from tempfile import NamedTemporaryFile
|
||||
import mmap
|
||||
from tempfile import NamedTemporaryFile
|
||||
from typing import TYPE_CHECKING, Any, Callable
|
||||
|
||||
if TYPE_CHECKING:
|
||||
from http.client import HTTPResponse
|
||||
|
||||
|
||||
class CallbackFileWrapper(object):
|
||||
class CallbackFileWrapper:
|
||||
"""
|
||||
Small wrapper around a fp object which will tee everything read into a
|
||||
buffer, and when that file is closed it will execute a callback with the
|
||||
|
|
@ -25,12 +30,14 @@ class CallbackFileWrapper(object):
|
|||
performance impact.
|
||||
"""
|
||||
|
||||
def __init__(self, fp, callback):
|
||||
def __init__(
|
||||
self, fp: HTTPResponse, callback: Callable[[bytes], None] | None
|
||||
) -> None:
|
||||
self.__buf = NamedTemporaryFile("rb+", delete=True)
|
||||
self.__fp = fp
|
||||
self.__callback = callback
|
||||
|
||||
def __getattr__(self, name):
|
||||
def __getattr__(self, name: str) -> Any:
|
||||
# The vaguaries of garbage collection means that self.__fp is
|
||||
# not always set. By using __getattribute__ and the private
|
||||
# name[0] allows looking up the attribute value and raising an
|
||||
|
|
@ -42,7 +49,7 @@ class CallbackFileWrapper(object):
|
|||
fp = self.__getattribute__("_CallbackFileWrapper__fp")
|
||||
return getattr(fp, name)
|
||||
|
||||
def __is_fp_closed(self):
|
||||
def __is_fp_closed(self) -> bool:
|
||||
try:
|
||||
return self.__fp.fp is None
|
||||
|
||||
|
|
@ -50,7 +57,8 @@ class CallbackFileWrapper(object):
|
|||
pass
|
||||
|
||||
try:
|
||||
return self.__fp.closed
|
||||
closed: bool = self.__fp.closed
|
||||
return closed
|
||||
|
||||
except AttributeError:
|
||||
pass
|
||||
|
|
@ -59,7 +67,7 @@ class CallbackFileWrapper(object):
|
|||
# TODO: Add some logging here...
|
||||
return False
|
||||
|
||||
def _close(self):
|
||||
def _close(self) -> None:
|
||||
if self.__callback:
|
||||
if self.__buf.tell() == 0:
|
||||
# Empty file:
|
||||
|
|
@ -86,8 +94,8 @@ class CallbackFileWrapper(object):
|
|||
# Important when caching big files.
|
||||
self.__buf.close()
|
||||
|
||||
def read(self, amt=None):
|
||||
data = self.__fp.read(amt)
|
||||
def read(self, amt: int | None = None) -> bytes:
|
||||
data: bytes = self.__fp.read(amt)
|
||||
if data:
|
||||
# We may be dealing with b'', a sign that things are over:
|
||||
# it's passed e.g. after we've already closed self.__buf.
|
||||
|
|
@ -97,8 +105,8 @@ class CallbackFileWrapper(object):
|
|||
|
||||
return data
|
||||
|
||||
def _safe_read(self, amt):
|
||||
data = self.__fp._safe_read(amt)
|
||||
def _safe_read(self, amt: int) -> bytes:
|
||||
data: bytes = self.__fp._safe_read(amt) # type: ignore[attr-defined]
|
||||
if amt == 2 and data == b"\r\n":
|
||||
# urllib executes this read to toss the CRLF at the end
|
||||
# of the chunk.
|
||||
|
|
|
|||
|
|
@ -1,29 +1,31 @@
|
|||
# SPDX-FileCopyrightText: 2015 Eric Larson
|
||||
#
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
from __future__ import annotations
|
||||
|
||||
import calendar
|
||||
import time
|
||||
|
||||
from datetime import datetime, timedelta, timezone
|
||||
from email.utils import formatdate, parsedate, parsedate_tz
|
||||
from typing import TYPE_CHECKING, Any, Mapping
|
||||
|
||||
from datetime import datetime, timedelta
|
||||
if TYPE_CHECKING:
|
||||
from urllib3 import HTTPResponse
|
||||
|
||||
TIME_FMT = "%a, %d %b %Y %H:%M:%S GMT"
|
||||
|
||||
|
||||
def expire_after(delta, date=None):
|
||||
date = date or datetime.utcnow()
|
||||
def expire_after(delta: timedelta, date: datetime | None = None) -> datetime:
|
||||
date = date or datetime.now(timezone.utc)
|
||||
return date + delta
|
||||
|
||||
|
||||
def datetime_to_header(dt):
|
||||
def datetime_to_header(dt: datetime) -> str:
|
||||
return formatdate(calendar.timegm(dt.timetuple()))
|
||||
|
||||
|
||||
class BaseHeuristic(object):
|
||||
|
||||
def warning(self, response):
|
||||
class BaseHeuristic:
|
||||
def warning(self, response: HTTPResponse) -> str | None:
|
||||
"""
|
||||
Return a valid 1xx warning header value describing the cache
|
||||
adjustments.
|
||||
|
|
@ -34,7 +36,7 @@ class BaseHeuristic(object):
|
|||
"""
|
||||
return '110 - "Response is Stale"'
|
||||
|
||||
def update_headers(self, response):
|
||||
def update_headers(self, response: HTTPResponse) -> dict[str, str]:
|
||||
"""Update the response headers with any new headers.
|
||||
|
||||
NOTE: This SHOULD always include some Warning header to
|
||||
|
|
@ -43,7 +45,7 @@ class BaseHeuristic(object):
|
|||
"""
|
||||
return {}
|
||||
|
||||
def apply(self, response):
|
||||
def apply(self, response: HTTPResponse) -> HTTPResponse:
|
||||
updated_headers = self.update_headers(response)
|
||||
|
||||
if updated_headers:
|
||||
|
|
@ -61,12 +63,12 @@ class OneDayCache(BaseHeuristic):
|
|||
future.
|
||||
"""
|
||||
|
||||
def update_headers(self, response):
|
||||
def update_headers(self, response: HTTPResponse) -> dict[str, str]:
|
||||
headers = {}
|
||||
|
||||
if "expires" not in response.headers:
|
||||
date = parsedate(response.headers["date"])
|
||||
expires = expire_after(timedelta(days=1), date=datetime(*date[:6]))
|
||||
expires = expire_after(timedelta(days=1), date=datetime(*date[:6], tzinfo=timezone.utc)) # type: ignore[misc]
|
||||
headers["expires"] = datetime_to_header(expires)
|
||||
headers["cache-control"] = "public"
|
||||
return headers
|
||||
|
|
@ -77,14 +79,14 @@ class ExpiresAfter(BaseHeuristic):
|
|||
Cache **all** requests for a defined time period.
|
||||
"""
|
||||
|
||||
def __init__(self, **kw):
|
||||
def __init__(self, **kw: Any) -> None:
|
||||
self.delta = timedelta(**kw)
|
||||
|
||||
def update_headers(self, response):
|
||||
def update_headers(self, response: HTTPResponse) -> dict[str, str]:
|
||||
expires = expire_after(self.delta)
|
||||
return {"expires": datetime_to_header(expires), "cache-control": "public"}
|
||||
|
||||
def warning(self, response):
|
||||
def warning(self, response: HTTPResponse) -> str | None:
|
||||
tmpl = "110 - Automatically cached for %s. Response might be stale"
|
||||
return tmpl % self.delta
|
||||
|
||||
|
|
@ -101,12 +103,23 @@ class LastModified(BaseHeuristic):
|
|||
http://lxr.mozilla.org/mozilla-release/source/netwerk/protocol/http/nsHttpResponseHead.cpp#397
|
||||
Unlike mozilla we limit this to 24-hr.
|
||||
"""
|
||||
|
||||
cacheable_by_default_statuses = {
|
||||
200, 203, 204, 206, 300, 301, 404, 405, 410, 414, 501
|
||||
200,
|
||||
203,
|
||||
204,
|
||||
206,
|
||||
300,
|
||||
301,
|
||||
404,
|
||||
405,
|
||||
410,
|
||||
414,
|
||||
501,
|
||||
}
|
||||
|
||||
def update_headers(self, resp):
|
||||
headers = resp.headers
|
||||
def update_headers(self, resp: HTTPResponse) -> dict[str, str]:
|
||||
headers: Mapping[str, str] = resp.headers
|
||||
|
||||
if "expires" in headers:
|
||||
return {}
|
||||
|
|
@ -120,9 +133,11 @@ class LastModified(BaseHeuristic):
|
|||
if "date" not in headers or "last-modified" not in headers:
|
||||
return {}
|
||||
|
||||
date = calendar.timegm(parsedate_tz(headers["date"]))
|
||||
time_tuple = parsedate_tz(headers["date"])
|
||||
assert time_tuple is not None
|
||||
date = calendar.timegm(time_tuple[:6])
|
||||
last_modified = parsedate(headers["last-modified"])
|
||||
if date is None or last_modified is None:
|
||||
if last_modified is None:
|
||||
return {}
|
||||
|
||||
now = time.time()
|
||||
|
|
@ -135,5 +150,5 @@ class LastModified(BaseHeuristic):
|
|||
expires = date + freshness_lifetime
|
||||
return {"expires": time.strftime(TIME_FMT, time.gmtime(expires))}
|
||||
|
||||
def warning(self, resp):
|
||||
def warning(self, resp: HTTPResponse) -> str | None:
|
||||
return None
|
||||
|
|
|
|||
0
lib/cachecontrol/py.typed
Normal file
0
lib/cachecontrol/py.typed
Normal file
|
|
@ -1,105 +1,91 @@
|
|||
# SPDX-FileCopyrightText: 2015 Eric Larson
|
||||
#
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
from __future__ import annotations
|
||||
|
||||
import base64
|
||||
import io
|
||||
import json
|
||||
import zlib
|
||||
from typing import IO, TYPE_CHECKING, Any, Mapping, cast
|
||||
|
||||
import msgpack
|
||||
from requests.structures import CaseInsensitiveDict
|
||||
from urllib3 import HTTPResponse
|
||||
|
||||
from .compat import HTTPResponse, pickle, text_type
|
||||
if TYPE_CHECKING:
|
||||
from requests import PreparedRequest
|
||||
|
||||
|
||||
def _b64_decode_bytes(b):
|
||||
return base64.b64decode(b.encode("ascii"))
|
||||
class Serializer:
|
||||
serde_version = "4"
|
||||
|
||||
|
||||
def _b64_decode_str(s):
|
||||
return _b64_decode_bytes(s).decode("utf8")
|
||||
|
||||
|
||||
_default_body_read = object()
|
||||
|
||||
|
||||
class Serializer(object):
|
||||
def dumps(self, request, response, body=None):
|
||||
response_headers = CaseInsensitiveDict(response.headers)
|
||||
def dumps(
|
||||
self,
|
||||
request: PreparedRequest,
|
||||
response: HTTPResponse,
|
||||
body: bytes | None = None,
|
||||
) -> bytes:
|
||||
response_headers: CaseInsensitiveDict[str] = CaseInsensitiveDict(
|
||||
response.headers
|
||||
)
|
||||
|
||||
if body is None:
|
||||
# When a body isn't passed in, we'll read the response. We
|
||||
# also update the response with a new file handler to be
|
||||
# sure it acts as though it was never read.
|
||||
body = response.read(decode_content=False)
|
||||
response._fp = io.BytesIO(body)
|
||||
response._fp = io.BytesIO(body) # type: ignore[attr-defined]
|
||||
response.length_remaining = len(body)
|
||||
|
||||
# NOTE: This is all a bit weird, but it's really important that on
|
||||
# Python 2.x these objects are unicode and not str, even when
|
||||
# they contain only ascii. The problem here is that msgpack
|
||||
# understands the difference between unicode and bytes and we
|
||||
# have it set to differentiate between them, however Python 2
|
||||
# doesn't know the difference. Forcing these to unicode will be
|
||||
# enough to have msgpack know the difference.
|
||||
data = {
|
||||
u"response": {
|
||||
u"body": body, # Empty bytestring if body is stored separately
|
||||
u"headers": dict(
|
||||
(text_type(k), text_type(v)) for k, v in response.headers.items()
|
||||
),
|
||||
u"status": response.status,
|
||||
u"version": response.version,
|
||||
u"reason": text_type(response.reason),
|
||||
u"strict": response.strict,
|
||||
u"decode_content": response.decode_content,
|
||||
"response": {
|
||||
"body": body, # Empty bytestring if body is stored separately
|
||||
"headers": {str(k): str(v) for k, v in response.headers.items()}, # type: ignore[no-untyped-call]
|
||||
"status": response.status,
|
||||
"version": response.version,
|
||||
"reason": str(response.reason),
|
||||
"decode_content": response.decode_content,
|
||||
}
|
||||
}
|
||||
|
||||
# Construct our vary headers
|
||||
data[u"vary"] = {}
|
||||
if u"vary" in response_headers:
|
||||
varied_headers = response_headers[u"vary"].split(",")
|
||||
data["vary"] = {}
|
||||
if "vary" in response_headers:
|
||||
varied_headers = response_headers["vary"].split(",")
|
||||
for header in varied_headers:
|
||||
header = text_type(header).strip()
|
||||
header = str(header).strip()
|
||||
header_value = request.headers.get(header, None)
|
||||
if header_value is not None:
|
||||
header_value = text_type(header_value)
|
||||
data[u"vary"][header] = header_value
|
||||
header_value = str(header_value)
|
||||
data["vary"][header] = header_value
|
||||
|
||||
return b",".join([b"cc=4", msgpack.dumps(data, use_bin_type=True)])
|
||||
return b",".join([f"cc={self.serde_version}".encode(), self.serialize(data)])
|
||||
|
||||
def loads(self, request, data, body_file=None):
|
||||
def serialize(self, data: dict[str, Any]) -> bytes:
|
||||
return cast(bytes, msgpack.dumps(data, use_bin_type=True))
|
||||
|
||||
def loads(
|
||||
self,
|
||||
request: PreparedRequest,
|
||||
data: bytes,
|
||||
body_file: IO[bytes] | None = None,
|
||||
) -> HTTPResponse | None:
|
||||
# Short circuit if we've been given an empty set of data
|
||||
if not data:
|
||||
return
|
||||
return None
|
||||
|
||||
# Determine what version of the serializer the data was serialized
|
||||
# with
|
||||
try:
|
||||
ver, data = data.split(b",", 1)
|
||||
except ValueError:
|
||||
ver = b"cc=0"
|
||||
# Previous versions of this library supported other serialization
|
||||
# formats, but these have all been removed.
|
||||
if not data.startswith(f"cc={self.serde_version},".encode()):
|
||||
return None
|
||||
|
||||
# Make sure that our "ver" is actually a version and isn't a false
|
||||
# positive from a , being in the data stream.
|
||||
if ver[:3] != b"cc=":
|
||||
data = ver + data
|
||||
ver = b"cc=0"
|
||||
data = data[5:]
|
||||
return self._loads_v4(request, data, body_file)
|
||||
|
||||
# Get the version number out of the cc=N
|
||||
ver = ver.split(b"=", 1)[-1].decode("ascii")
|
||||
|
||||
# Dispatch to the actual load method for the given version
|
||||
try:
|
||||
return getattr(self, "_loads_v{}".format(ver))(request, data, body_file)
|
||||
|
||||
except AttributeError:
|
||||
# This is a version we don't have a loads function for, so we'll
|
||||
# just treat it as a miss and return None
|
||||
return
|
||||
|
||||
def prepare_response(self, request, cached, body_file=None):
|
||||
def prepare_response(
|
||||
self,
|
||||
request: PreparedRequest,
|
||||
cached: Mapping[str, Any],
|
||||
body_file: IO[bytes] | None = None,
|
||||
) -> HTTPResponse | None:
|
||||
"""Verify our vary headers match and construct a real urllib3
|
||||
HTTPResponse object.
|
||||
"""
|
||||
|
|
@ -108,23 +94,26 @@ class Serializer(object):
|
|||
# This case is also handled in the controller code when creating
|
||||
# a cache entry, but is left here for backwards compatibility.
|
||||
if "*" in cached.get("vary", {}):
|
||||
return
|
||||
return None
|
||||
|
||||
# Ensure that the Vary headers for the cached response match our
|
||||
# request
|
||||
for header, value in cached.get("vary", {}).items():
|
||||
if request.headers.get(header, None) != value:
|
||||
return
|
||||
return None
|
||||
|
||||
body_raw = cached["response"].pop("body")
|
||||
|
||||
headers = CaseInsensitiveDict(data=cached["response"]["headers"])
|
||||
headers: CaseInsensitiveDict[str] = CaseInsensitiveDict(
|
||||
data=cached["response"]["headers"]
|
||||
)
|
||||
if headers.get("transfer-encoding", "") == "chunked":
|
||||
headers.pop("transfer-encoding")
|
||||
|
||||
cached["response"]["headers"] = headers
|
||||
|
||||
try:
|
||||
body: IO[bytes]
|
||||
if body_file is None:
|
||||
body = io.BytesIO(body_raw)
|
||||
else:
|
||||
|
|
@ -138,53 +127,20 @@ class Serializer(object):
|
|||
# TypeError: 'str' does not support the buffer interface
|
||||
body = io.BytesIO(body_raw.encode("utf8"))
|
||||
|
||||
# Discard any `strict` parameter serialized by older version of cachecontrol.
|
||||
cached["response"].pop("strict", None)
|
||||
|
||||
return HTTPResponse(body=body, preload_content=False, **cached["response"])
|
||||
|
||||
def _loads_v0(self, request, data, body_file=None):
|
||||
# The original legacy cache data. This doesn't contain enough
|
||||
# information to construct everything we need, so we'll treat this as
|
||||
# a miss.
|
||||
return
|
||||
|
||||
def _loads_v1(self, request, data, body_file=None):
|
||||
try:
|
||||
cached = pickle.loads(data)
|
||||
except ValueError:
|
||||
return
|
||||
|
||||
return self.prepare_response(request, cached, body_file)
|
||||
|
||||
def _loads_v2(self, request, data, body_file=None):
|
||||
assert body_file is None
|
||||
try:
|
||||
cached = json.loads(zlib.decompress(data).decode("utf8"))
|
||||
except (ValueError, zlib.error):
|
||||
return
|
||||
|
||||
# We need to decode the items that we've base64 encoded
|
||||
cached["response"]["body"] = _b64_decode_bytes(cached["response"]["body"])
|
||||
cached["response"]["headers"] = dict(
|
||||
(_b64_decode_str(k), _b64_decode_str(v))
|
||||
for k, v in cached["response"]["headers"].items()
|
||||
)
|
||||
cached["response"]["reason"] = _b64_decode_str(cached["response"]["reason"])
|
||||
cached["vary"] = dict(
|
||||
(_b64_decode_str(k), _b64_decode_str(v) if v is not None else v)
|
||||
for k, v in cached["vary"].items()
|
||||
)
|
||||
|
||||
return self.prepare_response(request, cached, body_file)
|
||||
|
||||
def _loads_v3(self, request, data, body_file):
|
||||
# Due to Python 2 encoding issues, it's impossible to know for sure
|
||||
# exactly how to load v3 entries, thus we'll treat these as a miss so
|
||||
# that they get rewritten out as v4 entries.
|
||||
return
|
||||
|
||||
def _loads_v4(self, request, data, body_file=None):
|
||||
def _loads_v4(
|
||||
self,
|
||||
request: PreparedRequest,
|
||||
data: bytes,
|
||||
body_file: IO[bytes] | None = None,
|
||||
) -> HTTPResponse | None:
|
||||
try:
|
||||
cached = msgpack.loads(data, raw=False)
|
||||
except ValueError:
|
||||
return
|
||||
return None
|
||||
|
||||
return self.prepare_response(request, cached, body_file)
|
||||
|
|
|
|||
|
|
@ -1,22 +1,32 @@
|
|||
# SPDX-FileCopyrightText: 2015 Eric Larson
|
||||
#
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
from __future__ import annotations
|
||||
|
||||
from .adapter import CacheControlAdapter
|
||||
from .cache import DictCache
|
||||
from typing import TYPE_CHECKING, Collection
|
||||
|
||||
from cachecontrol.adapter import CacheControlAdapter
|
||||
from cachecontrol.cache import DictCache
|
||||
|
||||
if TYPE_CHECKING:
|
||||
import requests
|
||||
|
||||
from cachecontrol.cache import BaseCache
|
||||
from cachecontrol.controller import CacheController
|
||||
from cachecontrol.heuristics import BaseHeuristic
|
||||
from cachecontrol.serialize import Serializer
|
||||
|
||||
|
||||
def CacheControl(
|
||||
sess,
|
||||
cache=None,
|
||||
cache_etags=True,
|
||||
serializer=None,
|
||||
heuristic=None,
|
||||
controller_class=None,
|
||||
adapter_class=None,
|
||||
cacheable_methods=None,
|
||||
):
|
||||
|
||||
sess: requests.Session,
|
||||
cache: BaseCache | None = None,
|
||||
cache_etags: bool = True,
|
||||
serializer: Serializer | None = None,
|
||||
heuristic: BaseHeuristic | None = None,
|
||||
controller_class: type[CacheController] | None = None,
|
||||
adapter_class: type[CacheControlAdapter] | None = None,
|
||||
cacheable_methods: Collection[str] | None = None,
|
||||
) -> requests.Session:
|
||||
cache = DictCache() if cache is None else cache
|
||||
adapter_class = adapter_class or CacheControlAdapter
|
||||
adapter = adapter_class(
|
||||
|
|
|
|||
Loading…
Reference in a new issue