Update cachecontrol library 0.12.5 (007e8ca) → 0.12.6 (167a605).

CHANGES.md

@@ -45,6 +45,7 @@
 * Update included fallback timezone info file to 2020d
 * Update attr 20.1.0.dev0 (4bd6827) to 20.2.0 (4f74fba)
 * Update Beautiful Soup 4.8.2 (r559) to 4.9.3 (r593)
+* Update cachecontrol library 0.12.5 (007e8ca) to 0.12.6 (167a605)
 * Update dateutil 2.8.1 (43b7838) to 2.8.1 (c496b4f)
 * Change add diskcache_py3 5.0.1 (9670fbb)
 * Change add diskcache_py2 4.1.0 (b0451e0)

lib/cachecontrol/__init__.py

@@ -1,11 +1,18 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 """CacheControl import Interface.
 
 Make it easy to import from cachecontrol without long namespaces.
 """
 __author__ = "Eric Larson"
 __email__ = "eric@ionrock.org"
-__version__ = "0.12.5"
+__version__ = "0.12.6"
 
 from .wrapper import CacheControl
 from .adapter import CacheControlAdapter
 from .controller import CacheController
+
+import logging
+logging.getLogger(__name__).addHandler(logging.NullHandler())

lib/cachecontrol/_cmd.py

@@ -1,3 +1,7 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 import logging
 
 import requests

lib/cachecontrol/adapter.py

@@ -1,16 +1,20 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 import types
 import functools
 import zlib
 
 from requests.adapters import HTTPAdapter
 
-from .controller import CacheController
+from .controller import CacheController, PERMANENT_REDIRECT_STATUSES
 from .cache import DictCache
 from .filewrapper import CallbackFileWrapper
 
 
 class CacheControlAdapter(HTTPAdapter):
-    invalidating_methods = {"PUT", "DELETE"}
+    invalidating_methods = {"PUT", "PATCH", "DELETE"}
 
     def __init__(
         self,
@@ -93,7 +97,7 @@ class CacheControlAdapter(HTTPAdapter):
                 response = cached_response
 
             # We always cache the 301 responses
-            elif response.status == 301:
+            elif int(response.status) in PERMANENT_REDIRECT_STATUSES:
                 self.controller.cache_response(request, response)
             else:
                 # Wrap the response file with a wrapper that will cache the

lib/cachecontrol/cache.py

@@ -1,3 +1,7 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 """
 The cache object API for implementing caches. The default is a thread
 safe in-memory dictionary.
@@ -10,7 +14,7 @@ class BaseCache(object):
     def get(self, key):
         raise NotImplementedError()
 
-    def set(self, key, value):
+    def set(self, key, value, expires=None):
         raise NotImplementedError()
 
     def delete(self, key):
@@ -29,7 +33,7 @@ class DictCache(BaseCache):
     def get(self, key):
         return self.data.get(key, None)
 
-    def set(self, key, value):
+    def set(self, key, value, expires=None):
         with self.lock:
             self.data.update({key: value})
 

lib/cachecontrol/caches/__init__.py

@@ -1,2 +1,6 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 from .file_cache import FileCache  # noqa
 from .redis_cache import RedisCache  # noqa

lib/cachecontrol/caches/file_cache.py

@@ -1,3 +1,7 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 import hashlib
 import os
 from textwrap import dedent
@@ -114,7 +118,7 @@ class FileCache(BaseCache):
         except FileNotFoundError:
             return None
 
-    def set(self, key, value):
+    def set(self, key, value, expires=None):
         name = self._fn(key)
 
         # Make sure the directory exists

lib/cachecontrol/caches/redis_cache.py

@@ -1,3 +1,7 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 from __future__ import division
 
 from datetime import datetime

lib/cachecontrol/compat.py

@@ -1,3 +1,7 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 try:
     from urllib.parse import urljoin
 except ImportError:
@@ -9,7 +13,6 @@ try:
 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:

lib/cachecontrol/controller.py

@@ -1,3 +1,7 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 """
 The httplib2 algorithms ported for use with requests.
 """
@@ -17,6 +21,8 @@ logger = logging.getLogger(__name__)
 
 URI = re.compile(r"^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\?([^#]*))?(#(.*))?")
 
+PERMANENT_REDIRECT_STATUSES = (301, 308)
+
 
 def parse_uri(uri):
     """Parses a URI using the regex given in Appendix B of RFC 3986.
@@ -37,7 +43,7 @@ class CacheController(object):
         self.cache = DictCache() if cache is None else cache
         self.cache_etags = cache_etags
         self.serializer = serializer or Serializer()
-        self.cacheable_status_codes = status_codes or (200, 203, 300, 301)
+        self.cacheable_status_codes = status_codes or (200, 203, 300, 301, 308)
 
     @classmethod
     def _urlnorm(cls, uri):
@@ -147,17 +153,18 @@ class CacheController(object):
             logger.warning("Cache entry deserialization failed, entry ignored")
             return False
 
-        # If we have a cached 301, return it immediately. We don't
-        # need to test our response for other headers b/c it is
+        # If we have a cached permanent redirect, return it immediately. We
+        # don't need to test our response for other headers b/c it is
         # intrinsically "cacheable" as it is Permanent.
+        #
         # See:
         #   https://tools.ietf.org/html/rfc7231#section-6.4.2
         #
         # Client can try to refresh the value by repeating the request
         # with cache busting headers as usual (ie no-cache).
-        if resp.status == 301:
+        if int(resp.status) in PERMANENT_REDIRECT_STATUSES:
             msg = (
-                'Returning cached "301 Moved Permanently" response '
+                'Returning cached permanent redirect response '
                 "(ignoring date and etag information)"
             )
             logger.debug(msg)
@@ -261,6 +268,13 @@ class CacheController(object):
 
         response_headers = CaseInsensitiveDict(response.headers)
 
+        if 'date' in response_headers:
+            date = calendar.timegm(
+                parsedate_tz(response_headers['date'])
+            )
+        else:
+            date = 0
+
         # If we've been given a body, our response has a Content-Length, that
         # Content-Length is valid then we can check to see if the body we've
         # been given matches the expected size, and if it doesn't we'll just
@@ -304,35 +318,60 @@ class CacheController(object):
 
         # If we've been given an etag, then keep the response
         if self.cache_etags and "etag" in response_headers:
+            expires_time = 0
+            if response_headers.get('expires'):
+                expires = parsedate_tz(response_headers['expires'])
+                if expires is not None:
+                    expires_time = calendar.timegm(expires) - date
+
+            expires_time = max(expires_time, 14 * 86400)
+
+            logger.debug('etag object cached for {0} seconds'.format(expires_time))
             logger.debug("Caching due to etag")
             self.cache.set(
-                cache_url, self.serializer.dumps(request, response, body=body)
+                cache_url,
+                self.serializer.dumps(request, response, body),
+                expires=expires_time
             )
 
-        # Add to the cache any 301s. We do this before looking that
-        # the Date headers.
-        elif response.status == 301:
-            logger.debug("Caching permanant redirect")
-            self.cache.set(cache_url, self.serializer.dumps(request, response))
+        # Add to the cache any permanent redirects. We do this before looking
+        # that the Date headers.
+        elif int(response.status) in PERMANENT_REDIRECT_STATUSES:
+            logger.debug("Caching permanent redirect")
+            self.cache.set(cache_url, self.serializer.dumps(request, response, b''))
 
         # Add to the cache if the response headers demand it. If there
         # 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'])
+            )
             # cache when there is a max-age > 0
             if "max-age" in cc and cc["max-age"] > 0:
                 logger.debug("Caching b/c date exists and max-age > 0")
+                expires_time = cc['max-age']
                 self.cache.set(
-                    cache_url, self.serializer.dumps(request, response, body=body)
+                    cache_url,
+                    self.serializer.dumps(request, response, body),
+                    expires=expires_time
                 )
 
             # If the request can expire, it means we should cache it
             # in the meantime.
             elif "expires" in response_headers:
                 if response_headers["expires"]:
-                    logger.debug("Caching b/c of expires header")
+                    expires = parsedate_tz(response_headers['expires'])
+                    if expires is not None:
+                        expires_time = calendar.timegm(expires) - date
+                    else:
+                        expires_time = None
+
+                    logger.debug('Caching b/c of expires header. expires in {0} seconds'.format(expires_time))
                     self.cache.set(
-                        cache_url, self.serializer.dumps(request, response, body=body)
+                        cache_url,
+                        self.serializer.dumps(request, response, body=body),
+                        expires=expires_time,
                     )
 
     def update_cached_response(self, request, response):
@@ -371,6 +410,7 @@ class CacheController(object):
         cached_response.status = 200
 
         # update our cache
-        self.cache.set(cache_url, self.serializer.dumps(request, cached_response))
+        body = cached_response.read(decode_content=False)
+        self.cache.set(cache_url, self.serializer.dumps(request, cached_response, body))
 
         return cached_response

lib/cachecontrol/filewrapper.py

@@ -1,3 +1,7 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 from io import BytesIO
 
 

lib/cachecontrol/heuristics.py

@@ -1,3 +1,7 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 import calendar
 import time
 

lib/cachecontrol/serialize.py

@@ -1,3 +1,7 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 import base64
 import io
 import json
@@ -17,26 +21,14 @@ 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):
+    def dumps(self, request, response, body):
         response_headers = CaseInsensitiveDict(response.headers)
 
-        if body is None:
-            body = response.read(decode_content=False)
-
-            # NOTE: 99% sure this is dead code. I'm only leaving it
-            #       here b/c I don't have a test yet to prove
-            #       it. Basically, before using
-            #       `cachecontrol.filewrapper.CallbackFileWrapper`,
-            #       this made an effort to reset the file handle. The
-            #       `CallbackFileWrapper` short circuits this code by
-            #       setting the body as the content is consumed, the
-            #       result being a `body` argument is *always* passed
-            #       into cache_response, and in turn,
-            #       `Serializer.dump`.
-            response._fp = io.BytesIO(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

lib/cachecontrol/wrapper.py

@@ -1,3 +1,7 @@
+# SPDX-FileCopyrightText: 2015 Eric Larson
+#
+# SPDX-License-Identifier: Apache-2.0
+
 from .adapter import CacheControlAdapter
 from .cache import DictCache