v2.8.1

Allow non-native-string method names.

AUTHORS.rst

@@ -158,3 +158,5 @@ Patches and Suggestions
 - Ulrich Petri (`@ulope <https://github.com/ulope>`_)
 - Muhammad Yasoob Ullah Khalid <yasoob.khld@gmail.com> (`@yasoob <https://github.com/yasoob>`_)
 - Paul van der Linden (`@pvanderlinden <https://github.com/pvanderlinden>`_)
+- Colin Dickson (`@colindickson <https://github.com/colindickson>`_)
+- Smiley Barry (`@smiley <https://github.com/smiley>`_)

HISTORY.rst

@@ -3,6 +3,69 @@
 Release History
 ---------------
 
+2.8.1 (2015-10-13)
+++++++++++++++++++
+
+**Bugfixes**
+
+- Update certificate bundle to match ``certifi`` 2015.9.6.2's weak certificate
+  bundle.
+- Fix a bug in 2.8.0 where requests would raise ``ConnectTimeout`` instead of
+  ``ConnectionError``
+- When using the PreparedRequest flow, requests will now correctly respect the
+  ``json`` parameter. Broken in 2.8.0.
+- When using the PreparedRequest flow, requests will now correctly handle a
+  Unicode-string method name on Python 2. Broken in 2.8.0.
+
+2.8.0 (2015-10-05)
+++++++++++++++++++
+
+**Minor Improvements** (Backwards Compatible)
+
+- Requests now supports per-host proxies. This allows the ``proxies``
+  dictionary to have entries of the form
+  ``{'<scheme>://<hostname>': '<proxy>'}``. Host-specific proxies will be used
+  in preference to the previously-supported scheme-specific ones, but the
+  previous syntax will continue to work.
+- ``Response.raise_for_status`` now prints the URL that failed as part of the
+  exception message.
+- ``requests.utils.get_netrc_auth`` now takes an ``raise_errors`` kwarg,
+  defaulting to ``False``. When ``True``, errors parsing ``.netrc`` files cause
+  exceptions to be thrown.
+- Change to bundled projects import logic to make it easier to unbundle
+  requests downstream.
+- Changed the default User-Agent string to avoid leaking data on Linux: now
+  contains only the requests version.
+
+**Bugfixes**
+
+- The ``json`` parameter to ``post()`` and friends will now only be used if
+  neither ``data`` nor ``files`` are present, consistent with the
+  documentation.
+- We now ignore empty fields in the ``NO_PROXY`` enviroment variable.
+- Fixed problem where ``httplib.BadStatusLine`` would get raised if combining
+  ``stream=True`` with ``contextlib.closing``.
+- Prevented bugs where we would attempt to return the same connection back to
+  the connection pool twice when sending a Chunked body.
+- Miscellaneous minor internal changes.
+- Digest Auth support is now thread safe.
+
+**Updates**
+
+- Updated urllib3 to 1.12.
+
+2.7.0 (2015-05-03)
+++++++++++++++++++
+
+This is the first release that follows our new release process. For more, see
+`our documentation
+<http://docs.python-requests.org/en/latest/community/release-process/>`_.
+
+**Bugfixes**
+
+- Updated urllib3 to 1.10.4, resolving several bugs involving chunked transfer
+  encoding and response framing.
+
 2.6.2 (2015-04-23)
 ++++++++++++++++++
 
@@ -150,7 +213,7 @@ Release History
 **Bugfixes**
 
 - Only parse the URL once (#2353)
-- Allow Content-Length header to always be overriden (#2332)
+- Allow Content-Length header to always be overridden (#2332)
 - Properly handle files in HTTPDigestAuth (#2333)
 - Cap redirect_cache size to prevent memory abuse (#2299)
 - Fix HTTPDigestAuth handling of redirects after authenticating successfully
@@ -498,7 +561,7 @@ This is not a backwards compatible change.
 - Digest Authentication improvements.
 - Ensure proxy exclusions work properly.
 - Clearer UnicodeError exceptions.
-- Automatic casting of URLs to tsrings (fURL and such)
+- Automatic casting of URLs to strings (fURL and such)
 - Bugfixes.
 
 0.13.6 (2012-08-06)
@@ -550,7 +613,7 @@ This is not a backwards compatible change.
 
 - Removal of Requests.async in favor of `grequests <https://github.com/kennethreitz/grequests>`_
 - Allow disabling of cookie persistiance.
-- New implimentation of safe_mode
+- New implementation of safe_mode
 - cookies.get now supports default argument
 - Session cookies not saved when Session.request is called with return_response=False
 - Env: no_proxy support.
@@ -667,7 +730,7 @@ This is not a backwards compatible change.
 
 * ``Response.content`` is now bytes-only. (*Backwards Incompatible*)
 * New ``Response.text`` is unicode-only.
-* If no ``Response.encoding`` is specified and ``chardet`` is available, ``Respoonse.text`` will guess an encoding.
+* If no ``Response.encoding`` is specified and ``chardet`` is available, ``Response.text`` will guess an encoding.
 * Default to ISO-8859-1 (Western) encoding for "text" subtypes.
 * Removal of `decode_unicode`. (*Backwards Incompatible*)
 * New multiple-hooks system.
@@ -787,7 +850,7 @@ This is not a backwards compatible change.
 0.7.5 (2011-11-04)
 ++++++++++++++++++
 
-* Response.content = None if there was an invalid repsonse.
+* Response.content = None if there was an invalid response.
 * Redirection auth handling.
 
 0.7.4 (2011-10-26)
@@ -874,7 +937,7 @@ This is not a backwards compatible change.
 ++++++++++++++++++
 
 * New callback hook system
-* New persistient sessions object and context manager
+* New persistent sessions object and context manager
 * Transparent Dict-cookie handling
 * Status code reference object
 * Removed Response.cached
@@ -908,7 +971,7 @@ This is not a backwards compatible change.
 * Redirect Fixes
 * settings.verbose stream writing
 * Querystrings for all methods
-* URLErrors (Connection Refused, Timeout, Invalid URLs) are treated as explicity raised
+* URLErrors (Connection Refused, Timeout, Invalid URLs) are treated as explicitly raised
   ``r.requests.get('hwe://blah'); r.raise_for_status()``
 
 

README.rst

@@ -77,7 +77,6 @@ Contribute
 ----------
 
 #. Check for open issues or open a fresh issue to start a discussion around a feature idea or a bug. There is a `Contributor Friendly`_ tag for issues that should be ideal for people who are not very familiar with the codebase yet.
-#. If you feel uncomfortable or uncertain about an issue or your changes, feel free to email @sigmavirus24 and he will happily help you via email, Skype, remote pairing or whatever you are comfortable with.
 #. Fork `the repository`_ on GitHub to start making your changes to the **master** branch (or branch off of it).
 #. Write a test which shows that the bug was fixed or that the feature works as expected.
 #. Send a pull request and bug the maintainer until it gets merged and published. :) Make sure to add yourself to AUTHORS_.

docs/_templates/sidebarintro.html

@@ -34,13 +34,13 @@
 <li><a href="http://jp.python-requests.org/">Japanese</a></li>
 <li><a href="http://cn.python-requests.org/">Chinese</a></li>
 <li><a href="http://pt.python-requests.org/">Portuguese</a></li>
-
-
+<li><a href="http://it.python-requests.org/">Italian</a></li>
+<li><a href="http://es.python-requests.org/">Spanish</a></li>
 </ul>
 
 <h3>Useful Links</h3>
 <ul>
-  <li><a href="http://docs.python-requests.org/en/latest/community/recommended/">Recomended Packages and Extensions</a></li>
+  <li><a href="http://docs.python-requests.org/en/latest/community/recommended/">Recommended Packages and Extensions</a></li>
 
   <p></p>
 

docs/api.rst

@@ -81,6 +81,8 @@ Status Code Lookup
     >>> requests.codes['\o/']
     200
 
+.. _api-cookies:
+
 Cookies
 ~~~~~~~
 

docs/community/faq.rst

@@ -79,9 +79,9 @@ when servers are using `Virtual Hosting`_. When such servers are hosting
 more than one SSL site they need to be able to return the appropriate
 certificate based on the hostname the client is connecting to.
 
-Python3's SSL module includes native support for SNI. This support has not been
-back ported to Python2. For information on using SNI with Requests on Python2
-refer to this `Stack Overflow answer`_.
+Python3 and Python 2.7.9+ include native support for SNI in their SSL modules.
+For information on using SNI with Requests on Python < 2.7.9 refer to this
+`Stack Overflow answer`_.
 
 .. _`Server-Name-Indication`: https://en.wikipedia.org/wiki/Server_Name_Indication
 .. _`virtual hosting`: https://en.wikipedia.org/wiki/Virtual_hosting

docs/community/release-process.rst

@@ -0,0 +1,54 @@
+Release Process and Rules
+=========================
+
+.. versionadded:: v2.6.2
+
+Starting with the version to be released after ``v2.6.2``, the following rules
+will govern and describe how the Requests core team produces a new release.
+
+Major Releases
+--------------
+
+A major release will include breaking changes. When it is versioned, it will
+be versioned as ``vX.0.0``. For example, if the previous release was
+``v10.2.7`` the next version will be ``v11.0.0``.
+
+Breaking changes are changes that break backwards compatibility with prior
+versions. If the project were to change the ``text`` attribute on a
+``Response`` object to a method, that would only happen in a Major release.
+
+Major releases may also include miscellaneous bug fixes and upgrades to
+vendored packages. The core developers of Requests are committed to providing
+a good user experience. This means we're also committed to preserving
+backwards compatibility as much as possible. Major releases will be infrequent
+and will need strong justifications before they are considered.
+
+Minor Releases
+--------------
+
+A minor release will not include breaking changes but may include
+miscellaneous bug fixes and upgrades to vendored packages. If the previous
+version of Requests released was ``v10.2.7`` a minor release would be
+versioned as ``v10.3.0``.
+
+Minor releases will be backwards compatible with releases that have the same
+major version number. In other words, all versions that would start with
+``v10.`` should be compatible with each other.
+
+Hotfix Releases
+---------------
+
+A hotfix release will only include bug fixes that were missed when the project
+released the previous version. If the previous version of Requests released
+``v10.2.7`` the hotfix release would be versioned as ``v10.2.8``.
+
+Hotfixes will **not** include upgrades to vendored dependences after
+``v2.6.2``
+
+Reasoning
+---------
+
+In the 2.5 and 2.6 release series, the Requests core team upgraded vendored
+dependencies and caused a great deal of headaches for both users and the core
+team. To reduce this pain, we're forming a concrete set of procedures so
+expectations will be properly set.

docs/community/updates.rst

@@ -1,8 +1,6 @@
 .. _updates:
 
 
-
-
 Community Updates
 =================
 
@@ -19,23 +17,14 @@ The best way to track the development of Requests is through
 Twitter
 -------
 
-I often tweet about new features and releases of Requests.
+The author, Kenneth Reitz, often tweets about new features and releases of Requests.
 
 Follow `@kennethreitz <https://twitter.com/kennethreitz>`_ for updates.
 
 
-Mailing List
-------------
 
-There's a low-volume mailing list for Requests. To subscribe to the
-mailing list, send an email to
-`requests@librelist.org <mailto:requests@librelist.org>`_.
-
-
-
-
-Software Updates
-================
+Release and Version History
+===========================
 
 .. include:: ../../HISTORY.rst
 

docs/dev/contributing.rst

@@ -0,0 +1,145 @@
+.. _contributing:
+
+Contributor's Guide
+===================
+
+If you're reading this you're probably interested in contributing to
+Requests. First, We'd like to say: thank you! Open source projects
+live-and-die based on the support they receive from others, and the fact that
+you're even considering supporting Requests is very generous of
+you.
+
+This document lays out guidelines and advice for contributing to Requests.
+If you're thinking of contributing, start by reading this thoroughly and
+getting a feel for how contributing to the project works. If you have any
+questions, feel free to reach out to either `Ian Cordasco`_ or `Cory Benfield`_,
+the primary maintainers.
+
+The guide is split into sections based on the type of contribution you're
+thinking of making, with a section that covers general guidelines for all
+contributors.
+
+.. _Ian Cordasco: http://www.coglib.com/~icordasc/
+.. _Cory Benfield: https://lukasa.co.uk/about
+
+
+All Contributions
+-----------------
+
+Be Cordial
+~~~~~~~~~~
+
+**Be cordial or be on your way.**
+
+Requests has one very important rule governing all forms of contribution,
+including reporting bugs or requesting features. This golden rule is
+`be cordial or be on your way`_. **All contributions are welcome**, as long as
+everyone involved is treated with respect.
+
+.. _be cordial or be on your way: http://kennethreitz.org/be-cordial-or-be-on-your-way/
+
+.. _early-feedback:
+
+Get Early Feedback
+~~~~~~~~~~~~~~~~~~
+
+If you are contributing, do not feel the need to sit on your contribution until
+it is perfectly polished and complete. It helps everyone involved for you to
+seek feedback as early as you possibly can. Submitting an early, unfinished
+version of your contribution for feedback in no way prejudices your chances of
+getting that contribution accepted, and can save you from putting a lot of work
+into a contribution that is not suitable for the project.
+
+Contribution Suitability
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+The project maintainer has the last word on whether or not a contribution is
+suitable for Requests. All contributions will be considered, but from time
+to time contributions will be rejected because they do not suit the project.
+
+If your contribution is rejected, don't despair! So long as you followed these
+guidelines, you'll have a much better chance of getting your next contribution
+accepted.
+
+
+Code Contributions
+------------------
+
+Steps
+~~~~~
+
+When contributing code, you'll want to follow this checklist:
+
+1. Fork the repository on GitHub.
+2. Run the tests to confirm they all pass on your system. If they don't, you'll
+   need to investigate why they fail. If you're unable to diagnose this
+   yourself, raise it as a bug report by following the guidelines in this
+   document: :ref:`bug-reports`.
+3. Write tests that demonstrate your bug or feature. Ensure that they fail.
+4. Make your change.
+5. Run the entire test suite again, confirming that all tests pass *including
+   the ones you just added*.
+6. Send a GitHub Pull Request to the main repository's ``master`` branch.
+   GitHub Pull Requests are the expected method of code collaboration on this
+   project.
+
+The following sub-sections go into more detail on some of the points above.
+
+Code Review
+~~~~~~~~~~~
+
+Contributions will not be merged until they've been code reviewed. You should
+implement any code review feedback unless you strongly object to it. In the
+event that you object to the code review feedback, you should make your case
+clearly and calmly. If, after doing so, the feedback is judged to still apply,
+you must either apply the feedback or withdraw your contribution.
+
+New Contributors
+~~~~~~~~~~~~~~~~
+
+If you are new or relatively new to Open Source, welcome! Requests aims to
+be a gentle introduction to the world of Open Source. If you're concerned about
+how best to contribute, please consider mailing a maintainer (listed above) and
+asking for help.
+
+Please also check the :ref:`early-feedback` section.
+
+Documentation Contributions
+---------------------------
+
+Documentation improvements are always welcome! The documentation files live in
+the ``docs/`` directory of the codebase. They're written in
+`reStructuredText`_, and use `Sphinx`_ to generate the full suite of
+documentation.
+
+When contributing documentation, please attempt to follow the style of the
+documentation files. This means a soft-limit of 79 characters wide in your text
+files and a semi-formal prose style.
+
+.. _reStructuredText: http://docutils.sourceforge.net/rst.html
+.. _Sphinx: http://sphinx-doc.org/index.html
+
+
+.. _bug-reports:
+
+Bug Reports
+-----------
+
+Bug reports are hugely important! Before you raise one, though, please check
+through the `GitHub issues`_, **both open and closed**, to confirm that the bug
+hasn't been reported before. Duplicate bug reports are a huge drain on the time
+of other contributors, and should be avoided as much as possible.
+
+.. _GitHub issues: https://github.com/kennethreitz/requests/issues
+
+
+Feature Requests
+----------------
+
+Requests is in a perpetual feature freeze. The maintainers believe that
+requests contains every major feature currently required by the vast majority
+of users.
+
+If you believe there is a feature missing, feel free to raise a feature
+request, but please do be aware that the overwhelming likelihood is that your
+feature request will not be accepted.

docs/index.rst

@@ -48,7 +48,7 @@ Testimonials
 Her Majesty's Government, Amazon, Google, Twilio, Runscope, Mozilla, Heroku,
 PayPal, NPR, Obama for America, Transifex, Native Instruments, The Washington
 Post, Twitter, SoundCloud, Kippt, Readability, Sony, and Federal US Institutions that prefer to be unnamed
-use Requests internally. It has been downloaded over 40,000,000 times from PyPI.
+use Requests internally. It has been downloaded over 60,000,000 times from PyPI.
 
 **Armin Ronacher**
     Requests is the perfect example how beautiful an API can be with the
@@ -115,10 +115,11 @@ Requests ecosystem and community.
 
    community/faq
    community/recommended
-   community/out-there.rst
+   community/out-there
    community/support
    community/vulnerabilities
    community/updates
+   community/release-process
 
 API Documentation
 -----------------
@@ -141,6 +142,7 @@ you.
 .. toctree::
    :maxdepth: 1
 
+   dev/contributing
    dev/philosophy
    dev/todo
    dev/authors

docs/user/advanced.rst

@@ -12,7 +12,10 @@ Session Objects
 
 The Session object allows you to persist certain parameters across
 requests. It also persists cookies across all requests made from the
-Session instance.
+Session instance, and will use ``urllib3``'s `connection pooling`_. So if
+you're making several requests to the same host, the underlying TCP
+connection will be reused, which can result in a significant performance
+increase (see `HTTP persistent connection`_).
 
 A Session object has all the methods of the main Requests API.
 
@@ -42,6 +45,33 @@ Any dictionaries that you pass to a request method will be merged with the
 session-level values that are set. The method-level parameters override session
 parameters.
 
+Note, however, that method-level parameters will *not* be persisted across
+requests, even if using a session. This example will only send the cookies
+with the first request, but not the second::
+
+    s = requests.Session()
+    r = s.get('http://httpbin.org/cookies', cookies={'from-my': 'browser'})
+    print(r.text)
+    # '{"cookies": {"from-my": "browser"}}'
+
+    r = s.get('http://httpbin.org/cookies')
+    print(r.text)
+    # '{"cookies": {}}'
+
+
+If you want to manually add cookies to your session, use the
+:ref:`Cookie utility functions <api-cookies>` to manipulate
+:attr:`Session.cookies <requests.Session.cookies>`.
+
+Sessions can also be used as context managers::
+
+    with requests.Session() as s:
+        s.get('http://httpbin.org/cookies/set/sessioncookie/123456789')
+
+This will make sure the session is closed as soon as the ``with`` block is
+exited, even if unhandled exceptions occurred.
+
+
 .. admonition:: Remove a Value From a Dict Parameter
 
     Sometimes you'll want to omit session-level keys from a dict parameter. To
@@ -171,7 +201,11 @@ I don't have SSL setup on this domain, so it fails. Excellent. GitHub does thoug
     >>> requests.get('https://github.com', verify=True)
     <Response [200]>
 
-You can pass ``verify`` the path to a CA_BUNDLE file with certificates of trusted CAs. This list of trusted CAs can also be specified through the ``REQUESTS_CA_BUNDLE`` environment variable.
+You can pass ``verify`` the path to a CA_BUNDLE file with certificates of trusted CAs::
+
+    >>> requests.get('https://github.com', verify='/path/to/certfile')
+
+This list of trusted CAs can also be specified through the ``REQUESTS_CA_BUNDLE`` environment variable.
 
 Requests can also ignore verifying the SSL certificate if you set ``verify`` to False.
 
@@ -194,6 +228,28 @@ If you specify a wrong path or an invalid cert::
     >>> requests.get('https://kennethreitz.com', cert='/wrong_path/server.pem')
     SSLError: [Errno 336265225] _ssl.c:347: error:140B0009:SSL routines:SSL_CTX_use_PrivateKey_file:PEM lib
 
+.. _ca-certificates:
+
+CA Certificates
+---------------
+
+By default Requests bundles a set of root CAs that it trusts, sourced from the
+`Mozilla trust store`_. However, these are only updated once for each Requests
+version. This means that if you pin a Requests version your certificates can
+become extremely out of date.
+
+From Requests version 2.4.0 onwards, Requests will attempt to use certificates
+from `certifi`_ if it is present on the system. This allows for users to update
+their trusted certificates without having to change the code that runs on their
+system.
+
+For the sake of security we recommend upgrading certifi frequently!
+
+.. _HTTP persistent connection: https://en.wikipedia.org/wiki/HTTP_persistent_connection
+.. _connection pooling: https://urllib3.readthedocs.org/en/latest/pools.html
+.. _certifi: http://certifi.io/
+.. _Mozilla trust store: https://hg.mozilla.org/mozilla-central/raw-file/tip/security/nss/lib/ckfw/builtins/certdata.txt
+
 .. _body-content-workflow:
 
 Body Content Workflow
@@ -268,13 +324,19 @@ Requests also supports Chunked transfer encoding for outgoing and incoming reque
 To send a chunk-encoded request, simply provide a generator (or any iterator without
 a length) for your body::
 
-
     def gen():
         yield 'hi'
         yield 'there'
 
     requests.post('http://some.url/chunked', data=gen())
 
+For chunked encoded responses, it's best to iterate over the data using
+:meth:`Response.iter_content() <requests.models.Response.iter_content>`. In
+an ideal situation you'll have set ``stream=True`` on the request, in which
+case you can iterate chunk-by-chunk by calling ``iter_content`` with a chunk
+size parameter of ``None``. If you want to set a maximum size of the chunk,
+you can set a chunk size parameter to any integer.
+
 
 .. _multipart:
 
@@ -446,7 +508,19 @@ To use HTTP Basic Auth with your proxy, use the `http://user:password@host/` syn
         "http": "http://user:pass@10.10.1.10:3128/",
     }
 
+To give a proxy for a specific scheme and host, use the
+`scheme://hostname` form for the key.  This will match for
+any request to the given scheme and exact hostname.
+
+::
+
+    proxies = {
+      "http://10.20.1.128": "http://10.10.1.10:5323",
+    }
+
 Note that proxy URLs must include the scheme.
+    
+.. _compliance:
 
 Compliance
 ----------
@@ -474,6 +548,8 @@ specification in this case. If you require a different encoding, you can
 manually set the :attr:`Response.encoding <requests.Response.encoding>`
 property, or use the raw :attr:`Response.content <requests.Response.content>`.
 
+.. _http-verbs:
+
 HTTP Verbs
 ----------
 
@@ -647,6 +723,8 @@ headers.
 Excellent. Time to write a Python program that abuses the GitHub API in all
 kinds of exciting ways, 4995 more times.
 
+.. _link-headers:
+
 Link Headers
 ------------
 
@@ -669,6 +747,8 @@ Requests will automatically parse these link headers and make them easily consum
     >>> r.links["last"]
     {'url': 'https://api.github.com/users/kennethreitz/repos?page=7&per_page=10', 'rel': 'last'}
 
+.. _transport-adapters:
+
 Transport Adapters
 ------------------
 
@@ -737,6 +817,8 @@ SSLv3:
 .. _`described here`: http://www.kennethreitz.org/essays/the-future-of-python-http
 .. _`urllib3`: https://github.com/shazow/urllib3
 
+.. _blocking-or-nonblocking:
+
 Blocking Or Non-Blocking?
 -------------------------
 
@@ -754,6 +836,8 @@ Two excellent examples are `grequests`_ and `requests-futures`_.
 .. _`grequests`: https://github.com/kennethreitz/grequests
 .. _`requests-futures`: https://github.com/ross/requests-futures
 
+.. _timeouts:
+
 Timeouts
 --------
 
@@ -791,21 +875,3 @@ coffee.
     r = requests.get('https://github.com', timeout=None)
 
 .. _`connect()`: http://linux.die.net/man/2/connect
-
-CA Certificates
----------------
-
-By default Requests bundles a set of root CAs that it trusts, sourced from the
-`Mozilla trust store`_. However, these are only updated once for each Requests
-version. This means that if you pin a Requests version your certificates can
-become extremely out of date.
-
-From Requests version 2.4.0 onwards, Requests will attempt to use certificates
-from `certifi`_ if it is present on the system. This allows for users to update
-their trusted certificates without having to change the code that runs on their
-system.
-
-For the sake of security we recommend upgrading certifi frequently!
-
-.. _certifi: http://certifi.io/
-.. _Mozilla trust store: https://hg.mozilla.org/mozilla-central/raw-file/tip/security/nss/lib/ckfw/builtins/certdata.txt

docs/user/quickstart.rst

@@ -73,13 +73,12 @@ You can see that the URL has been correctly encoded by printing the URL::
 Note that any dictionary key whose value is ``None`` will not be added to the
 URL's query string.
 
-In order to pass a list of items as a value you must mark the key as
-referring to a list like string by appending ``[]`` to the key::
+You can also pass a list of items as a value::
 
-    >>> payload = {'key1': 'value1', 'key2[]': ['value2', 'value3']}
+    >>> payload = {'key1': 'value1', 'key2': ['value2', 'value3']}
     >>> r = requests.get("http://httpbin.org/get", params=payload)
     >>> print(r.url)
-    http://httpbin.org/get?key1=value1&key2%5B%5D=value2&key2%5B%5D=value3
+    http://httpbin.org/get?key1=value1&key2=value2&key2=value3
 
 Response Content
 ----------------
@@ -146,8 +145,14 @@ There's also a builtin JSON decoder, in case you're dealing with JSON data::
     [{u'repository': {u'open_issues': 0, u'url': 'https://github.com/...
 
 In case the JSON decoding fails, ``r.json`` raises an exception. For example, if
-the response gets a 401 (Unauthorized), attempting ``r.json`` raises ``ValueError:
-No JSON object could be decoded``
+the response gets a 204 (No Content), or if the response contains invalid JSON,
+attempting ``r.json`` raises ``ValueError: No JSON object could be decoded``.
+
+It should be noted that the success of the call to ``r.json`` does **not**
+indicate the success of the response. Some servers may return a JSON object in a
+failed response (e.g. error details with HTTP 500). Such JSON will be decoded
+and returned. To check that a request is successful, use
+``r.raise_for_status()`` or check ``r.status_code`` is what you expect.
 
 
 Raw Response Content
@@ -182,7 +187,7 @@ Custom Headers
 If you'd like to add HTTP headers to a request, simply pass in a ``dict`` to the
 ``headers`` parameter.
 
-For example, we didn't specify our content-type in the previous example::
+For example, we didn't specify our user-agent in the previous example::
 
     >>> import json
     >>> url = 'https://api.github.com/some/endpoint'
@@ -230,6 +235,15 @@ For example, the GitHub API v3 accepts JSON-Encoded POST/PATCH data::
 
     >>> r = requests.post(url, data=json.dumps(payload))
 
+Instead of encoding the ``dict`` yourself, you can also pass it directly using
+the ``json`` parameter (added in version 2.4.2) and it will be encoded automatically::
+
+    >>> import json
+    >>> url = 'https://api.github.com/some/endpoint'
+    >>> payload = {'some': 'data'}
+
+    >>> r = requests.post(url, json=payload)
+
 
 POST a Multipart-Encoded File
 -----------------------------
@@ -249,7 +263,7 @@ Requests makes it simple to upload Multipart-encoded files::
       ...
     }
 
-You can set the filename, content_type and headers explicitly:
+You can set the filename, content_type and headers explicitly::
 
     >>> url = 'http://httpbin.org/post'
     >>> files = {'file': ('report.xls', open('report.xls', 'rb'), 'application/vnd.ms-excel', {'Expires': '0'})}
@@ -360,10 +374,10 @@ times with different values, but requests combines them so they can be
 represented in the dictionary within a single mapping, as per
 `RFC 7230 <http://tools.ietf.org/html/rfc7230#section-3.2>`_:
 
-    > A recipient MAY combine multiple header fields with the same field name
-    > into one "field-name: field-value" pair, without changing the semantics
-    > of the message, by appending each subsequent field value to the combined
-    > field value in order, separated by a comma.
+    A recipient MAY combine multiple header fields with the same field name
+    into one "field-name: field-value" pair, without changing the semantics
+    of the message, by appending each subsequent field value to the combined
+    field value in order, separated by a comma.
 
 Cookies
 -------

requests/__init__.py

@@ -6,7 +6,7 @@
 #          /
 
 """
-requests HTTP library
+Requests HTTP library
 ~~~~~~~~~~~~~~~~~~~~~
 
 Requests is an HTTP library, written in Python, for human beings. Basic GET
@@ -42,8 +42,8 @@ is at <http://python-requests.org>.
 """
 
 __title__ = 'requests'
-__version__ = '2.6.2'
-__build__ = 0x020602
+__version__ = '2.8.1'
+__build__ = 0x020801
 __author__ = 'Kenneth Reitz'
 __license__ = 'Apache 2.0'
 __copyright__ = 'Copyright 2015 Kenneth Reitz'

requests/adapters.py

@@ -17,11 +17,14 @@ from .packages.urllib3.util import Timeout as TimeoutSauce
 from .packages.urllib3.util.retry import Retry
 from .compat import urlparse, basestring
 from .utils import (DEFAULT_CA_BUNDLE_PATH, get_encoding_from_headers,
-                    prepend_scheme_if_needed, get_auth_from_url, urldefragauth)
+                    prepend_scheme_if_needed, get_auth_from_url, urldefragauth,
+                    select_proxy)
 from .structures import CaseInsensitiveDict
+from .packages.urllib3.exceptions import ClosedPoolError
 from .packages.urllib3.exceptions import ConnectTimeoutError
 from .packages.urllib3.exceptions import HTTPError as _HTTPError
 from .packages.urllib3.exceptions import MaxRetryError
+from .packages.urllib3.exceptions import NewConnectionError
 from .packages.urllib3.exceptions import ProxyError as _ProxyError
 from .packages.urllib3.exceptions import ProtocolError
 from .packages.urllib3.exceptions import ReadTimeoutError
@@ -35,6 +38,7 @@ from .auth import _basic_auth_str
 DEFAULT_POOLBLOCK = False
 DEFAULT_POOLSIZE = 10
 DEFAULT_RETRIES = 0
+DEFAULT_POOL_TIMEOUT = None
 
 
 class BaseAdapter(object):
@@ -237,8 +241,7 @@ class HTTPAdapter(BaseAdapter):
         :param url: The URL to connect to.
         :param proxies: (optional) A Requests-style dictionary of proxies used on this request.
         """
-        proxies = proxies or {}
-        proxy = proxies.get(urlparse(url.lower()).scheme)
+        proxy = select_proxy(url, proxies)
 
         if proxy:
             proxy = prepend_scheme_if_needed(proxy, 'http')
@@ -271,12 +274,10 @@ class HTTPAdapter(BaseAdapter):
         :class:`HTTPAdapter <requests.adapters.HTTPAdapter>`.
 
         :param request: The :class:`PreparedRequest <PreparedRequest>` being sent.
-        :param proxies: A dictionary of schemes to proxy URLs.
+        :param proxies: A dictionary of schemes or schemes and hosts to proxy URLs.
         """
-        proxies = proxies or {}
+        proxy = select_proxy(request.url, proxies)
         scheme = urlparse(request.url).scheme
-        proxy = proxies.get(scheme)
-
         if proxy and scheme != 'https':
             url = urldefragauth(request.url)
         else:
@@ -309,7 +310,6 @@ class HTTPAdapter(BaseAdapter):
         :class:`HTTPAdapter <requests.adapters.HTTPAdapter>`.
 
         :param proxies: The url of the proxy being used for this request.
-        :param kwargs: Optional additional keyword arguments.
         """
         headers = {}
         username, password = get_auth_from_url(proxy)
@@ -326,8 +326,8 @@ class HTTPAdapter(BaseAdapter):
         :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 (`connect timeout, read
-            timeout <user/advanced.html#timeouts>`_) tuple.
+            data before giving up, as a float, or a :ref:`(connect timeout,
+            read timeout) <timeouts>` tuple.
         :type timeout: float or tuple
         :param verify: (optional) Whether to verify SSL certificates.
         :param cert: (optional) Any user-provided SSL certificate to be trusted.
@@ -375,7 +375,7 @@ class HTTPAdapter(BaseAdapter):
                 if hasattr(conn, 'proxy_pool'):
                     conn = conn.proxy_pool
 
-                low_conn = conn._get_conn(timeout=timeout)
+                low_conn = conn._get_conn(timeout=DEFAULT_POOL_TIMEOUT)
 
                 try:
                     low_conn.putrequest(request.method,
@@ -407,22 +407,24 @@ class HTTPAdapter(BaseAdapter):
                     # Then, reraise so that we can handle the actual exception.
                     low_conn.close()
                     raise
-                else:
-                    # All is well, return the connection to the pool.
-                    conn._put_conn(low_conn)
 
         except (ProtocolError, socket.error) as err:
             raise ConnectionError(err, request=request)
 
         except MaxRetryError as e:
             if isinstance(e.reason, ConnectTimeoutError):
-                raise ConnectTimeout(e, request=request)
+                # 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)
 
             raise ConnectionError(e, request=request)
 
+        except ClosedPoolError as e:
+            raise ConnectionError(e, request=request)
+
         except _ProxyError as e:
             raise ProxyError(e)
 

requests/api.py

@@ -27,8 +27,8 @@ def request(method, url, **kwargs):
     :param files: (optional) Dictionary of ``'name': file-like-objects`` (or ``{'name': ('filename', fileobj)}``) for multipart encoding upload.
     :param auth: (optional) Auth tuple 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 (`connect timeout, read timeout
-        <user/advanced.html#timeouts>`_) tuple.
+        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. Set to True if POST/PUT/DELETE redirect following is allowed.
     :type allow_redirects: bool
@@ -55,17 +55,18 @@ def request(method, url, **kwargs):
     return response
 
 
-def get(url, **kwargs):
+def get(url, params=None, **kwargs):
     """Sends a GET request.
 
     :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 \*\*kwargs: Optional arguments that ``request`` takes.
     :return: :class:`Response <Response>` object
     :rtype: requests.Response
     """
 
     kwargs.setdefault('allow_redirects', True)
-    return request('get', url, **kwargs)
+    return request('get', url, params=params, **kwargs)
 
 
 def options(url, **kwargs):

requests/auth.py

@@ -11,6 +11,7 @@ import os
 import re
 import time
 import hashlib
+import threading
 
 from base64 import b64encode
 
@@ -63,19 +64,26 @@ class HTTPDigestAuth(AuthBase):
     def __init__(self, username, password):
         self.username = username
         self.password = password
-        self.last_nonce = ''
-        self.nonce_count = 0
-        self.chal = {}
-        self.pos = None
-        self.num_401_calls = 1
+        # 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):
 
-        realm = self.chal['realm']
-        nonce = self.chal['nonce']
-        qop = self.chal.get('qop')
-        algorithm = self.chal.get('algorithm')
-        opaque = self.chal.get('opaque')
+        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')
 
         if algorithm is None:
             _algorithm = 'MD5'
@@ -114,12 +122,12 @@ class HTTPDigestAuth(AuthBase):
         HA1 = hash_utf8(A1)
         HA2 = hash_utf8(A2)
 
-        if nonce == self.last_nonce:
-            self.nonce_count += 1
+        if nonce == self._thread_local.last_nonce:
+            self._thread_local.nonce_count += 1
         else:
-            self.nonce_count = 1
-        ncvalue = '%08x' % self.nonce_count
-        s = str(self.nonce_count).encode('utf-8')
+            self._thread_local.nonce_count = 1
+        ncvalue = '%08x' % self._thread_local.nonce_count
+        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)
@@ -139,7 +147,7 @@ class HTTPDigestAuth(AuthBase):
             # XXX handle auth-int.
             return None
 
-        self.last_nonce = nonce
+        self._thread_local.last_nonce = nonce
 
         # XXX should the partial digests be encoded too?
         base = 'username="%s", realm="%s", nonce="%s", uri="%s", ' \
@@ -158,28 +166,27 @@ class HTTPDigestAuth(AuthBase):
     def handle_redirect(self, r, **kwargs):
         """Reset num_401_calls counter on redirects."""
         if r.is_redirect:
-            self.num_401_calls = 1
+            self._thread_local.num_401_calls = 1
 
     def handle_401(self, r, **kwargs):
         """Takes the given response and tries digest-auth, if needed."""
 
-        if self.pos is not None:
+        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.pos)
-        num_401_calls = getattr(self, 'num_401_calls', 1)
+            r.request.body.seek(self._thread_local.pos)
         s_auth = r.headers.get('www-authenticate', '')
 
-        if 'digest' in s_auth.lower() and num_401_calls < 2:
+        if 'digest' in s_auth.lower() and self._thread_local.num_401_calls < 2:
 
-            self.num_401_calls += 1
+            self._thread_local.num_401_calls += 1
             pat = re.compile(r'digest ', flags=re.IGNORECASE)
-            self.chal = parse_dict_header(pat.sub('', s_auth, count=1))
+            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.raw.release_conn()
+            r.close()
             prep = r.request.copy()
             extract_cookies_to_jar(prep._cookies, r.request, r.raw)
             prep.prepare_cookies(prep._cookies)
@@ -192,21 +199,25 @@ class HTTPDigestAuth(AuthBase):
 
             return _r
 
-        self.num_401_calls = 1
+        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.last_nonce:
+        if self._thread_local.last_nonce:
             r.headers['Authorization'] = self.build_digest_header(r.method, r.url)
         try:
-            self.pos = r.body.tell()
+            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.pos = 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

requests/cookies.py

@@ -415,11 +415,14 @@ def morsel_to_cookie(morsel):
 
     expires = None
     if morsel['max-age']:
-        expires = time.time() + morsel['max-age']
+        try:
+            expires = int(time.time() + int(morsel['max-age']))
+        except ValueError:
+            raise TypeError('max-age: %s must be integer' % morsel['max-age'])
     elif morsel['expires']:
         time_template = '%a, %d-%b-%Y %H:%M:%S GMT'
-        expires = time.mktime(
-            time.strptime(morsel['expires'], time_template)) - time.timezone
+        expires = int(time.mktime(
+            time.strptime(morsel['expires'], time_template)) - time.timezone)
     return create_cookie(
         comment=morsel['comment'],
         comment_url=bool(morsel['comment']),

requests/hooks.py

@@ -12,34 +12,23 @@ Available hooks:
     The response generated from a Request.
 
 """
-
-
 HOOKS = ['response']
 
-
 def default_hooks():
-    hooks = {}
-    for event in HOOKS:
-        hooks[event] = []
-    return hooks
+    return dict((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 dict()
-
-    if key in hooks:
-        hooks = hooks.get(key)
-
+    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

requests/models.py

@@ -30,7 +30,8 @@ from .utils import (
     iter_slices, guess_json_utf, super_len, to_native_string)
 from .compat import (
     cookielib, urlunparse, urlsplit, urlencode, str, bytes, StringIO,
-    is_py2, chardet, json, builtin_str, basestring)
+    is_py2, chardet, builtin_str, basestring)
+from .compat import json as complexjson
 from .status_codes import codes
 
 #: The set of HTTP status codes that indicate an automatically
@@ -42,12 +43,11 @@ REDIRECT_STATI = (
     codes.temporary_redirect, # 307
     codes.permanent_redirect, # 308
 )
+
 DEFAULT_REDIRECT_LIMIT = 30
 CONTENT_CHUNK_SIZE = 10 * 1024
 ITER_CHUNK_SIZE = 512
 
-json_dumps = json.dumps
-
 
 class RequestEncodingMixin(object):
     @property
@@ -149,8 +149,7 @@ class RequestEncodingMixin(object):
             else:
                 fdata = fp.read()
 
-            rf = RequestField(name=k, data=fdata,
-                              filename=fn, headers=fh)
+            rf = RequestField(name=k, data=fdata, filename=fn, headers=fh)
             rf.make_multipart(content_type=ft)
             new_fields.append(rf)
 
@@ -193,7 +192,7 @@ class Request(RequestHooksMixin):
     :param headers: dictionary of headers to send.
     :param files: dictionary of {filename: fileobject} files to multipart upload.
     :param data: the body to attach to the request. If a dictionary is provided, form-encoding will take place.
-    :param json: json for the body to attach to the request (if data is not specified).
+    :param json: json for the body to attach to the request (if files or data is not specified).
     :param params: dictionary of URL parameters to append to the URL.
     :param auth: Auth handler or (user, pass) tuple.
     :param cookies: dictionary or CookieJar of cookies to attach to this request.
@@ -207,17 +206,8 @@ class Request(RequestHooksMixin):
       <PreparedRequest [GET]>
 
     """
-    def __init__(self,
-        method=None,
-        url=None,
-        headers=None,
-        files=None,
-        data=None,
-        params=None,
-        auth=None,
-        cookies=None,
-        hooks=None,
-        json=None):
+    def __init__(self, method=None, url=None, headers=None, files=None,
+        data=None, params=None, auth=None, cookies=None, hooks=None, json=None):
 
         # Default empty dicts for dict params.
         data = [] if data is None else data
@@ -296,8 +286,7 @@ class PreparedRequest(RequestEncodingMixin, RequestHooksMixin):
         self.hooks = default_hooks()
 
     def prepare(self, method=None, url=None, headers=None, files=None,
-                data=None, params=None, auth=None, cookies=None, hooks=None,
-                json=None):
+        data=None, params=None, auth=None, cookies=None, hooks=None, json=None):
         """Prepares the entire request with the given parameters."""
 
         self.prepare_method(method)
@@ -306,6 +295,7 @@ class PreparedRequest(RequestEncodingMixin, RequestHooksMixin):
         self.prepare_cookies(cookies)
         self.prepare_body(data, files, json)
         self.prepare_auth(auth, url)
+
         # Note that prepare_auth must be last to enable authentication schemes
         # such as OAuth to work on a fully prepared request.
 
@@ -329,7 +319,7 @@ class PreparedRequest(RequestEncodingMixin, RequestHooksMixin):
         """Prepares the given HTTP method."""
         self.method = method
         if self.method is not None:
-            self.method = self.method.upper()
+            self.method = to_native_string(self.method.upper())
 
     def prepare_url(self, url, params):
         """Prepares the given HTTP URL."""
@@ -357,9 +347,10 @@ class PreparedRequest(RequestEncodingMixin, RequestHooksMixin):
             raise InvalidURL(*e.args)
 
         if not scheme:
-            raise MissingSchema("Invalid URL {0!r}: No schema supplied. "
-                                "Perhaps you meant http://{0}?".format(
-                                    to_native_string(url, 'utf8')))
+            error = ("Invalid URL {0!r}: No schema supplied. Perhaps you meant http://{0}?")
+            error = error.format(to_native_string(url, 'utf8'))
+
+            raise MissingSchema(error)
 
         if not host:
             raise InvalidURL("Invalid URL %r: No host supplied" % url)
@@ -423,9 +414,9 @@ class PreparedRequest(RequestEncodingMixin, RequestHooksMixin):
         content_type = None
         length = None
 
-        if json is not None:
+        if not data and json is not None:
             content_type = 'application/json'
-            body = json_dumps(json)
+            body = complexjson.dumps(json)
 
         is_stream = all([
             hasattr(data, '__iter__'),
@@ -452,7 +443,7 @@ class PreparedRequest(RequestEncodingMixin, RequestHooksMixin):
             if files:
                 (body, content_type) = self._encode_files(files, data)
             else:
-                if data and json is None:
+                if data:
                     body = self._encode_params(data)
                     if isinstance(data, basestring) or hasattr(data, 'read'):
                         content_type = None
@@ -537,16 +528,8 @@ class Response(object):
     """
 
     __attrs__ = [
-        '_content',
-        'status_code',
-        'headers',
-        'url',
-        'history',
-        'encoding',
-        'reason',
-        'cookies',
-        'elapsed',
-        'request',
+        '_content', 'status_code', 'headers', 'url', 'history',
+        'encoding', 'reason', 'cookies', 'elapsed', 'request'
     ]
 
     def __init__(self):
@@ -666,9 +649,10 @@ class Response(object):
         If decode_unicode is True, content will be decoded using the best
         available encoding based on the response.
         """
+
         def generate():
-            try:
-                # Special case for urllib3.
+            # Special case for urllib3.
+            if hasattr(self.raw, 'stream'):
                 try:
                     for chunk in self.raw.stream(chunk_size, decode_content=True):
                         yield chunk
@@ -678,7 +662,7 @@ class Response(object):
                     raise ContentDecodingError(e)
                 except ReadTimeoutError as e:
                     raise ConnectionError(e)
-            except AttributeError:
+            else:
                 # Standard file-like object.
                 while True:
                     chunk = self.raw.read(chunk_size)
@@ -809,14 +793,16 @@ class Response(object):
             encoding = guess_json_utf(self.content)
             if encoding is not None:
                 try:
-                    return json.loads(self.content.decode(encoding), **kwargs)
+                    return complexjson.loads(
+                        self.content.decode(encoding), **kwargs
+                    )
                 except UnicodeDecodeError:
                     # Wrong UTF codec detected; usually because it's not UTF-8
                     # but some other 8-bit codec.  This is an RFC violation,
                     # and the server didn't bother to tell us what codec *was*
                     # used.
                     pass
-        return json.loads(self.text, **kwargs)
+        return complexjson.loads(self.text, **kwargs)
 
     @property
     def links(self):
@@ -842,10 +828,10 @@ class Response(object):
         http_error_msg = ''
 
         if 400 <= self.status_code < 500:
-            http_error_msg = '%s Client Error: %s' % (self.status_code, self.reason)
+            http_error_msg = '%s Client Error: %s for url: %s' % (self.status_code, self.reason, self.url)
 
         elif 500 <= self.status_code < 600:
-            http_error_msg = '%s Server Error: %s' % (self.status_code, self.reason)
+            http_error_msg = '%s Server Error: %s for url: %s' % (self.status_code, self.reason, self.url)
 
         if http_error_msg:
             raise HTTPError(http_error_msg, response=self)
@@ -856,4 +842,7 @@ class Response(object):
 
         *Note: Should not normally need to be called explicitly.*
         """
+        if not self._content_consumed:
+            return self.raw.close()
+
         return self.raw.release_conn()

requests/packages/README.rst

@@ -1,8 +1,11 @@
 If you are planning to submit a pull request to requests with any changes in 
-this library do not go any further. These are independent libraries which we 
-vendor into requests. Any changes necessary to these libraries must be made in 
+this library do not go any further. These are independent libraries which we
+vendor into requests. Any changes necessary to these libraries must be made in
 them and submitted as separate pull requests to those libraries.
 
 urllib3 pull requests go here: https://github.com/shazow/urllib3
 
 chardet pull requests go here: https://github.com/chardet/chardet
+
+See https://github.com/kennethreitz/requests/pull/1812#issuecomment-30854316
+for the reasoning behind this.

requests/packages/__init__.py

@@ -1,3 +1,36 @@
-from __future__ import absolute_import
+'''
+Debian and other distributions "unbundle" requests' vendored dependencies, and
+rewrite all imports to use the global versions of ``urllib3`` and ``chardet``.
+The problem with this is that not only requests itself imports those
+dependencies, but third-party code outside of the distros' control too.
 
-from . import urllib3
+In reaction to these problems, the distro maintainers replaced
+``requests.packages`` with a magical "stub module" that imports the correct
+modules. The implementations were varying in quality and all had severe
+problems. For example, a symlink (or hardlink) that links the correct modules
+into place introduces problems regarding object identity, since you now have
+two modules in `sys.modules` with the same API, but different identities::
+
+    requests.packages.urllib3 is not urllib3
+
+With version ``2.5.2``, requests started to maintain its own stub, so that
+distro-specific breakage would be reduced to a minimum, even though the whole
+issue is not requests' fault in the first place. See
+https://github.com/kennethreitz/requests/pull/2375 for the corresponding pull
+request.
+'''
+
+from __future__ import absolute_import
+import sys
+
+try:
+    from . import urllib3
+except ImportError:
+    import urllib3
+    sys.modules['%s.urllib3' % __name__] = urllib3
+
+try:
+    from . import chardet
+except ImportError:
+    import chardet
+    sys.modules['%s.chardet' % __name__] = chardet

requests/packages/urllib3/__init__.py

@@ -4,7 +4,7 @@ urllib3 - Thread-safe connection pooling and re-using.
 
 __author__ = 'Andrey Petrov (andrey.petrov@shazow.net)'
 __license__ = 'MIT'
-__version__ = '1.10.3'
+__version__ = '1.12'
 
 
 from .connectionpool import (
@@ -57,9 +57,12 @@ del NullHandler
 
 import warnings
 # SecurityWarning's always go off by default.
-warnings.simplefilter('always', exceptions.SecurityWarning)
+warnings.simplefilter('always', exceptions.SecurityWarning, append=True)
+# SubjectAltNameWarning's should go off once per host
+warnings.simplefilter('default', exceptions.SubjectAltNameWarning)
 # InsecurePlatformWarning's don't vary between requests, so we keep it default.
-warnings.simplefilter('default', exceptions.InsecurePlatformWarning)
+warnings.simplefilter('default', exceptions.InsecurePlatformWarning,
+                      append=True)
 
 def disable_warnings(category=exceptions.HTTPWarning):
     """

requests/packages/urllib3/_collections.py

@@ -97,14 +97,7 @@ class RecentlyUsedContainer(MutableMapping):
             return list(iterkeys(self._container))
 
 
-_dict_setitem = dict.__setitem__
-_dict_getitem = dict.__getitem__
-_dict_delitem = dict.__delitem__
-_dict_contains = dict.__contains__
-_dict_setdefault = dict.setdefault
-
-
-class HTTPHeaderDict(dict):
+class HTTPHeaderDict(MutableMapping):
     """
     :param headers:
         An iterable of field-value pairs. Must not contain multiple field names
@@ -139,7 +132,8 @@ class HTTPHeaderDict(dict):
     """
 
     def __init__(self, headers=None, **kwargs):
-        dict.__init__(self)
+        super(HTTPHeaderDict, self).__init__()
+        self._container = {}
         if headers is not None:
             if isinstance(headers, HTTPHeaderDict):
                 self._copy_from(headers)
@@ -149,38 +143,44 @@ class HTTPHeaderDict(dict):
             self.extend(kwargs)
 
     def __setitem__(self, key, val):
-        return _dict_setitem(self, key.lower(), (key, val))
+        self._container[key.lower()] = (key, val)
+        return self._container[key.lower()]
 
     def __getitem__(self, key):
-        val = _dict_getitem(self, key.lower())
+        val = self._container[key.lower()]
         return ', '.join(val[1:])
 
     def __delitem__(self, key):
-        return _dict_delitem(self, key.lower())
+        del self._container[key.lower()]
 
     def __contains__(self, key):
-        return _dict_contains(self, key.lower())
+        return key.lower() in self._container
 
     def __eq__(self, other):
         if not isinstance(other, Mapping) and not hasattr(other, 'keys'):
             return False
         if not isinstance(other, type(self)):
             other = type(self)(other)
-        return dict((k1, self[k1]) for k1 in self) == dict((k2, other[k2]) for k2 in other)
+        return (dict((k.lower(), v) for k, v in self.itermerged()) ==
+                dict((k.lower(), v) for k, v in other.itermerged()))
 
     def __ne__(self, other):
         return not self.__eq__(other)
 
-    values = MutableMapping.values
-    get = MutableMapping.get
-    update = MutableMapping.update
-    
     if not PY3: # Python 2
         iterkeys = MutableMapping.iterkeys
         itervalues = MutableMapping.itervalues
 
     __marker = object()
 
+    def __len__(self):
+        return len(self._container)
+
+    def __iter__(self):
+        # Only provide the originally cased names
+        for vals in self._container.values():
+            yield vals[0]
+
     def pop(self, key, default=__marker):
         '''D.pop(k[,d]) -> v, remove specified key and return the corresponding value.
           If key is not found, d is returned if given, otherwise KeyError is raised.
@@ -216,7 +216,7 @@ class HTTPHeaderDict(dict):
         key_lower = key.lower()
         new_vals = key, val
         # Keep the common case aka no item present as fast as possible
-        vals = _dict_setdefault(self, key_lower, new_vals)
+        vals = self._container.setdefault(key_lower, new_vals)
         if new_vals is not vals:
             # new_vals was not inserted, as there was a previous one
             if isinstance(vals, list):
@@ -225,7 +225,7 @@ class HTTPHeaderDict(dict):
             else:
                 # vals should be a tuple then, i.e. only one item so far
                 # Need to convert the tuple to list for further extension
-                _dict_setitem(self, key_lower, [vals[0], vals[1], val])
+                self._container[key_lower] = [vals[0], vals[1], val]
 
     def extend(self, *args, **kwargs):
         """Generic import function for any type of header-like object.
@@ -236,7 +236,7 @@ class HTTPHeaderDict(dict):
             raise TypeError("extend() takes at most 1 positional "
                             "arguments ({} given)".format(len(args)))
         other = args[0] if len(args) >= 1 else ()
-        
+
         if isinstance(other, HTTPHeaderDict):
             for key, val in other.iteritems():
                 self.add(key, val)
@@ -257,7 +257,7 @@ class HTTPHeaderDict(dict):
         """Returns a list of all the values for the named field. Returns an
         empty list if the key doesn't exist."""
         try:
-            vals = _dict_getitem(self, key.lower())
+            vals = self._container[key.lower()]
         except KeyError:
             return []
         else:
@@ -276,11 +276,11 @@ class HTTPHeaderDict(dict):
 
     def _copy_from(self, other):
         for key in other:
-            val = _dict_getitem(other, key)
+            val = other.getlist(key)
             if isinstance(val, list):
                 # Don't need to convert tuples
                 val = list(val)
-            _dict_setitem(self, key, val)
+            self._container[key.lower()] = [key] + val
 
     def copy(self):
         clone = type(self)()
@@ -290,14 +290,14 @@ class HTTPHeaderDict(dict):
     def iteritems(self):
         """Iterate over all header lines, including duplicate ones."""
         for key in self:
-            vals = _dict_getitem(self, key)
+            vals = self._container[key.lower()]
             for val in vals[1:]:
                 yield vals[0], val
 
     def itermerged(self):
         """Iterate over all headers, merging duplicate ones together."""
         for key in self:
-            val = _dict_getitem(self, key)
+            val = self._container[key.lower()]
             yield val[0], ', '.join(val[1:])
 
     def items(self):
@@ -307,16 +307,16 @@ class HTTPHeaderDict(dict):
     def from_httplib(cls, message): # Python 2
         """Read headers from a Python 2 httplib message object."""
         # python2.7 does not expose a proper API for exporting multiheaders
-        # efficiently. This function re-reads raw lines from the message 
+        # efficiently. This function re-reads raw lines from the message
         # object and extracts the multiheaders properly.
         headers = []
-         
+
         for line in message.headers:
             if line.startswith((' ', '\t')):
                 key, value = headers[-1]
                 headers[-1] = (key, value + '\r\n' + line.rstrip())
                 continue
-    
+
             key, value = line.split(':', 1)
             headers.append((key, value.strip()))
 

requests/packages/urllib3/connection.py

@@ -1,7 +1,7 @@
 import datetime
 import sys
 import socket
-from socket import timeout as SocketTimeout
+from socket import error as SocketError, timeout as SocketTimeout
 import warnings
 from .packages import six
 
@@ -36,9 +36,10 @@ except NameError:  # Python 2:
 
 
 from .exceptions import (
+    NewConnectionError,
     ConnectTimeoutError,
+    SubjectAltNameWarning,
     SystemTimeWarning,
-    SecurityWarning,
 )
 from .packages.ssl_match_hostname import match_hostname
 
@@ -133,11 +134,15 @@ class HTTPConnection(_HTTPConnection, object):
             conn = connection.create_connection(
                 (self.host, self.port), self.timeout, **extra_kw)
 
-        except SocketTimeout:
+        except SocketTimeout as e:
             raise ConnectTimeoutError(
                 self, "Connection to %s timed out. (connect timeout=%s)" %
                 (self.host, self.timeout))
 
+        except SocketError as e:
+            raise NewConnectionError(
+                self, "Failed to establish a new connection: %s" % e)
+
         return conn
 
     def _prepare_conn(self, conn):
@@ -185,17 +190,23 @@ class VerifiedHTTPSConnection(HTTPSConnection):
     """
     cert_reqs = None
     ca_certs = None
+    ca_cert_dir = None
     ssl_version = None
     assert_fingerprint = None
 
     def set_cert(self, key_file=None, cert_file=None,
                  cert_reqs=None, ca_certs=None,
-                 assert_hostname=None, assert_fingerprint=None):
+                 assert_hostname=None, assert_fingerprint=None,
+                 ca_cert_dir=None):
+
+        if (ca_certs or ca_cert_dir) and cert_reqs is None:
+            cert_reqs = 'CERT_REQUIRED'
 
         self.key_file = key_file
         self.cert_file = cert_file
         self.cert_reqs = cert_reqs
         self.ca_certs = ca_certs
+        self.ca_cert_dir = ca_cert_dir
         self.assert_hostname = assert_hostname
         self.assert_fingerprint = assert_fingerprint
 
@@ -234,6 +245,7 @@ class VerifiedHTTPSConnection(HTTPSConnection):
         self.sock = ssl_wrap_socket(conn, self.key_file, self.cert_file,
                                     cert_reqs=resolved_cert_reqs,
                                     ca_certs=self.ca_certs,
+                                    ca_cert_dir=self.ca_cert_dir,
                                     server_hostname=hostname,
                                     ssl_version=resolved_ssl_version)
 
@@ -245,10 +257,11 @@ class VerifiedHTTPSConnection(HTTPSConnection):
             cert = self.sock.getpeercert()
             if not cert.get('subjectAltName', ()):
                 warnings.warn((
-                    'Certificate has no `subjectAltName`, falling back to check for a `commonName` for now. '
-                    'This feature is being removed by major browsers and deprecated by RFC 2818. '
-                    '(See https://github.com/shazow/urllib3/issues/497 for details.)'),
-                    SecurityWarning
+                    'Certificate for {0} has no `subjectAltName`, falling back to check for a '
+                    '`commonName` for now. This feature is being removed by major browsers and '
+                    'deprecated by RFC 2818. (See https://github.com/shazow/urllib3/issues/497 '
+                    'for details.)'.format(hostname)),
+                    SubjectAltNameWarning
                 )
             match_hostname(cert, self.assert_hostname or hostname)
 

requests/packages/urllib3/connectionpool.py

@@ -17,14 +17,17 @@ from .exceptions import (
     ClosedPoolError,
     ProtocolError,
     EmptyPoolError,
+    HeaderParsingError,
     HostChangedError,
     LocationValueError,
     MaxRetryError,
     ProxyError,
+    ConnectTimeoutError,
     ReadTimeoutError,
     SSLError,
     TimeoutError,
     InsecureRequestWarning,
+    NewConnectionError,
 )
 from .packages.ssl_match_hostname import CertificateError
 from .packages import six
@@ -38,9 +41,10 @@ from .request import RequestMethods
 from .response import HTTPResponse
 
 from .util.connection import is_connection_dropped
+from .util.response import assert_header_parsing
 from .util.retry import Retry
 from .util.timeout import Timeout
-from .util.url import get_host
+from .util.url import get_host, Url
 
 
 xrange = six.moves.xrange
@@ -120,7 +124,7 @@ class HTTPConnectionPool(ConnectionPool, RequestMethods):
 
     :param maxsize:
         Number of connections to save that can be reused. More than 1 is useful
-        in multithreaded situations. If ``block`` is set to false, more
+        in multithreaded situations. If ``block`` is set to False, more
         connections will be created but they will not be saved once they've
         been used.
 
@@ -381,8 +385,19 @@ class HTTPConnectionPool(ConnectionPool, RequestMethods):
         log.debug("\"%s %s %s\" %s %s" % (method, url, http_version,
                                           httplib_response.status,
                                           httplib_response.length))
+
+        try:
+            assert_header_parsing(httplib_response.msg)
+        except HeaderParsingError as hpe:  # Platform-specific: Python 3
+            log.warning(
+                'Failed to parse headers (url=%s): %s',
+                self._absolute_url(url), hpe, exc_info=True)
+
         return httplib_response
 
+    def _absolute_url(self, path):
+        return Url(scheme=self.scheme, host=self.host, port=self.port, path=path).url
+
     def close(self):
         """
         Close all pooled connections and disable the pool.
@@ -568,27 +583,24 @@ class HTTPConnectionPool(ConnectionPool, RequestMethods):
             # Close the connection. If a connection is reused on which there
             # was a Certificate error, the next request will certainly raise
             # another Certificate error.
-            if conn:
-                conn.close()
-                conn = None
+            conn = conn and conn.close()
+            release_conn = True
             raise SSLError(e)
 
         except SSLError:
             # Treat SSLError separately from BaseSSLError to preserve
             # traceback.
-            if conn:
-                conn.close()
-                conn = None
+            conn = conn and conn.close()
+            release_conn = True
             raise
 
-        except (TimeoutError, HTTPException, SocketError, ConnectionError) as e:
-            if conn:
-                # Discard the connection for these exceptions. It will be
-                # be replaced during the next _get_conn() call.
-                conn.close()
-                conn = None
+        except (TimeoutError, HTTPException, SocketError, ProtocolError) as e:
+            # Discard the connection for these exceptions. It will be
+            # be replaced during the next _get_conn() call.
+            conn = conn and conn.close()
+            release_conn = True
 
-            if isinstance(e, SocketError) and self.proxy:
+            if isinstance(e, (SocketError, NewConnectionError)) and self.proxy:
                 e = ProxyError('Cannot connect to proxy.', e)
             elif isinstance(e, (SocketError, HTTPException)):
                 e = ProtocolError('Connection aborted.', e)
@@ -626,6 +638,9 @@ class HTTPConnectionPool(ConnectionPool, RequestMethods):
                 retries = retries.increment(method, url, response=response, _pool=self)
             except MaxRetryError:
                 if retries.raise_on_redirect:
+                    # Release the connection for this response, since we're not
+                    # returning it to be released manually.
+                    response.release_conn()
                     raise
                 return response
 
@@ -662,10 +677,10 @@ class HTTPSConnectionPool(HTTPConnectionPool):
     ``assert_hostname`` and ``host`` in this order to verify connections.
     If ``assert_hostname`` is False, no verification is done.
 
-    The ``key_file``, ``cert_file``, ``cert_reqs``, ``ca_certs`` and
-    ``ssl_version`` are only used if :mod:`ssl` is available and are fed into
-    :meth:`urllib3.util.ssl_wrap_socket` to upgrade the connection socket
-    into an SSL socket.
+    The ``key_file``, ``cert_file``, ``cert_reqs``, ``ca_certs``,
+    ``ca_cert_dir``, and ``ssl_version`` are only used if :mod:`ssl` is
+    available and are fed into :meth:`urllib3.util.ssl_wrap_socket` to upgrade
+    the connection socket into an SSL socket.
     """
 
     scheme = 'https'
@@ -678,15 +693,20 @@ class HTTPSConnectionPool(HTTPConnectionPool):
                  key_file=None, cert_file=None, cert_reqs=None,
                  ca_certs=None, ssl_version=None,
                  assert_hostname=None, assert_fingerprint=None,
-                 **conn_kw):
+                 ca_cert_dir=None, **conn_kw):
 
         HTTPConnectionPool.__init__(self, host, port, strict, timeout, maxsize,
                                     block, headers, retries, _proxy, _proxy_headers,
                                     **conn_kw)
+
+        if ca_certs and cert_reqs is None:
+            cert_reqs = 'CERT_REQUIRED'
+
         self.key_file = key_file
         self.cert_file = cert_file
         self.cert_reqs = cert_reqs
         self.ca_certs = ca_certs
+        self.ca_cert_dir = ca_cert_dir
         self.ssl_version = ssl_version
         self.assert_hostname = assert_hostname
         self.assert_fingerprint = assert_fingerprint
@@ -702,6 +722,7 @@ class HTTPSConnectionPool(HTTPConnectionPool):
                           cert_file=self.cert_file,
                           cert_reqs=self.cert_reqs,
                           ca_certs=self.ca_certs,
+                          ca_cert_dir=self.ca_cert_dir,
                           assert_hostname=self.assert_hostname,
                           assert_fingerprint=self.assert_fingerprint)
             conn.ssl_version = self.ssl_version

requests/packages/urllib3/contrib/appengine.py

@@ -0,0 +1,222 @@
+import logging
+import os
+import warnings
+
+from ..exceptions import (
+    HTTPError,
+    HTTPWarning,
+    MaxRetryError,
+    ProtocolError,
+    TimeoutError,
+    SSLError
+)
+
+from ..packages.six import BytesIO
+from ..request import RequestMethods
+from ..response import HTTPResponse
+from ..util.timeout import Timeout
+from ..util.retry import Retry
+
+try:
+    from google.appengine.api import urlfetch
+except ImportError:
+    urlfetch = None
+
+
+log = logging.getLogger(__name__)
+
+
+class AppEnginePlatformWarning(HTTPWarning):
+    pass
+
+
+class AppEnginePlatformError(HTTPError):
+    pass
+
+
+class AppEngineManager(RequestMethods):
+    """
+    Connection manager for Google App Engine sandbox applications.
+
+    This manager uses the URLFetch service directly instead of using the
+    emulated httplib, and is subject to URLFetch limitations as described in
+    the App Engine documentation here:
+
+        https://cloud.google.com/appengine/docs/python/urlfetch
+
+    Notably it will raise an AppEnginePlatformError if:
+        * URLFetch is not available.
+        * If you attempt to use this on GAEv2 (Managed VMs), as full socket
+          support is available.
+        * If a request size is more than 10 megabytes.
+        * If a response size is more than 32 megabtyes.
+        * If you use an unsupported request method such as OPTIONS.
+
+    Beyond those cases, it will raise normal urllib3 errors.
+    """
+
+    def __init__(self, headers=None, retries=None, validate_certificate=True):
+        if not urlfetch:
+            raise AppEnginePlatformError(
+                "URLFetch is not available in this environment.")
+
+        if is_prod_appengine_v2():
+            raise AppEnginePlatformError(
+                "Use normal urllib3.PoolManager instead of AppEngineManager"
+                "on Managed VMs, as using URLFetch is not necessary in "
+                "this environment.")
+
+        warnings.warn(
+            "urllib3 is using URLFetch on Google App Engine sandbox instead "
+            "of sockets. To use sockets directly instead of URLFetch see "
+            "https://urllib3.readthedocs.org/en/latest/contrib.html.",
+            AppEnginePlatformWarning)
+
+        RequestMethods.__init__(self, headers)
+        self.validate_certificate = validate_certificate
+
+        self.retries = retries or Retry.DEFAULT
+
+    def __enter__(self):
+        return self
+
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        # Return False to re-raise any potential exceptions
+        return False
+
+    def urlopen(self, method, url, body=None, headers=None,
+                retries=None, redirect=True, timeout=Timeout.DEFAULT_TIMEOUT,
+                **response_kw):
+
+        retries = self._get_retries(retries, redirect)
+
+        try:
+            response = urlfetch.fetch(
+                url,
+                payload=body,
+                method=method,
+                headers=headers or {},
+                allow_truncated=False,
+                follow_redirects=(
+                    redirect and
+                    retries.redirect != 0 and
+                    retries.total),
+                deadline=self._get_absolute_timeout(timeout),
+                validate_certificate=self.validate_certificate,
+            )
+        except urlfetch.DeadlineExceededError as e:
+            raise TimeoutError(self, e)
+
+        except urlfetch.InvalidURLError as e:
+            if 'too large' in e.message:
+                raise AppEnginePlatformError(
+                    "URLFetch request too large, URLFetch only "
+                    "supports requests up to 10mb in size.", e)
+            raise ProtocolError(e)
+
+        except urlfetch.DownloadError as e:
+            if 'Too many redirects' in e.message:
+                raise MaxRetryError(self, url, reason=e)
+            raise ProtocolError(e)
+
+        except urlfetch.ResponseTooLargeError as e:
+            raise AppEnginePlatformError(
+                "URLFetch response too large, URLFetch only supports"
+                "responses up to 32mb in size.", e)
+
+        except urlfetch.SSLCertificateError as e:
+            raise SSLError(e)
+
+        except urlfetch.InvalidMethodError as e:
+            raise AppEnginePlatformError(
+                "URLFetch does not support method: %s" % method, e)
+
+        http_response = self._urlfetch_response_to_http_response(
+            response, **response_kw)
+
+        # Check for redirect response
+        if (http_response.get_redirect_location() and
+                retries.raise_on_redirect and redirect):
+            raise MaxRetryError(self, url, "too many redirects")
+
+        # Check if we should retry the HTTP response.
+        if retries.is_forced_retry(method, status_code=http_response.status):
+            retries = retries.increment(
+                method, url, response=http_response, _pool=self)
+            log.info("Forced retry: %s" % url)
+            retries.sleep()
+            return self.urlopen(
+                method, url,
+                body=body, headers=headers,
+                retries=retries, redirect=redirect,
+                timeout=timeout, **response_kw)
+
+        return http_response
+
+    def _urlfetch_response_to_http_response(self, urlfetch_resp, **response_kw):
+
+        if is_prod_appengine_v1():
+            # Production GAE handles deflate encoding automatically, but does
+            # not remove the encoding header.
+            content_encoding = urlfetch_resp.headers.get('content-encoding')
+
+            if content_encoding == 'deflate':
+                del urlfetch_resp.headers['content-encoding']
+
+        return HTTPResponse(
+            # In order for decoding to work, we must present the content as
+            # a file-like object.
+            body=BytesIO(urlfetch_resp.content),
+            headers=urlfetch_resp.headers,
+            status=urlfetch_resp.status_code,
+            **response_kw
+        )
+
+    def _get_absolute_timeout(self, timeout):
+        if timeout is Timeout.DEFAULT_TIMEOUT:
+            return 5  # 5s is the default timeout for URLFetch.
+        if isinstance(timeout, Timeout):
+            if not timeout.read is timeout.connect:
+                warnings.warn(
+                    "URLFetch does not support granular timeout settings, "
+                    "reverting to total timeout.", AppEnginePlatformWarning)
+            return timeout.total
+        return timeout
+
+    def _get_retries(self, retries, redirect):
+        if not isinstance(retries, Retry):
+            retries = Retry.from_int(
+                retries, redirect=redirect, default=self.retries)
+
+        if retries.connect or retries.read or retries.redirect:
+            warnings.warn(
+                "URLFetch only supports total retries and does not "
+                "recognize connect, read, or redirect retry parameters.",
+                AppEnginePlatformWarning)
+
+        return retries
+
+
+def is_appengine():
+    return (is_local_appengine() or
+            is_prod_appengine_v1() or
+            is_prod_appengine_v2())
+
+
+def is_appengine_sandbox():
+    return is_appengine() and not is_prod_appengine_v2()
+
+
+def is_local_appengine():
+    return ('APPENGINE_RUNTIME' in os.environ and
+            'Development/' in os.environ['SERVER_SOFTWARE'])
+
+
+def is_prod_appengine_v1():
+    return ('APPENGINE_RUNTIME' in os.environ and
+            'Google App Engine/' in os.environ['SERVER_SOFTWARE'] and
+            not is_prod_appengine_v2())
+
+
+def is_prod_appengine_v2():
+    return os.environ.get('GAE_VM', False) == 'true'

requests/packages/urllib3/contrib/pyopenssl.py

@@ -85,6 +85,14 @@ _openssl_verify = {
 
 DEFAULT_SSL_CIPHER_LIST = util.ssl_.DEFAULT_CIPHERS
 
+# OpenSSL will only write 16K at a time
+SSL_WRITE_BLOCKSIZE = 16384
+
+try:
+    _ = memoryview
+    has_memoryview = True
+except NameError:
+    has_memoryview = False
 
 orig_util_HAS_SNI = util.HAS_SNI
 orig_connection_ssl_wrap_socket = connection.ssl_wrap_socket
@@ -204,13 +212,21 @@ class WrappedSocket(object):
                 continue
 
     def sendall(self, data):
-        while len(data):
-            sent = self._send_until_done(data)
-            data = data[sent:]
+        if has_memoryview and not isinstance(data, memoryview):
+            data = memoryview(data)
+
+        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):
+        # FIXME rethrow compatible exceptions should we ever use this
+        self.connection.shutdown()
 
     def close(self):
         if self._makefile_refs < 1:
-            return self.connection.shutdown()
+            return self.connection.close()
         else:
             self._makefile_refs -= 1
 
@@ -251,7 +267,7 @@ def _verify_callback(cnx, x509, err_no, err_depth, return_code):
 
 def ssl_wrap_socket(sock, keyfile=None, certfile=None, cert_reqs=None,
                     ca_certs=None, server_hostname=None,
-                    ssl_version=None):
+                    ssl_version=None, ca_cert_dir=None):
     ctx = OpenSSL.SSL.Context(_openssl_versions[ssl_version])
     if certfile:
         keyfile = keyfile or certfile  # Match behaviour of the normal python ssl library
@@ -260,9 +276,9 @@ def ssl_wrap_socket(sock, keyfile=None, certfile=None, cert_reqs=None,
         ctx.use_privatekey_file(keyfile)
     if cert_reqs != ssl.CERT_NONE:
         ctx.set_verify(_openssl_verify[cert_reqs], _verify_callback)
-    if ca_certs:
+    if ca_certs or ca_cert_dir:
         try:
-            ctx.load_verify_locations(ca_certs, None)
+            ctx.load_verify_locations(ca_certs, ca_cert_dir)
         except OpenSSL.SSL.Error as e:
             raise ssl.SSLError('bad ca_certs: %r' % ca_certs, e)
     else:
@@ -287,7 +303,7 @@ def ssl_wrap_socket(sock, keyfile=None, certfile=None, cert_reqs=None,
                 raise timeout('select timed out')
             continue
         except OpenSSL.SSL.Error as e:
-            raise ssl.SSLError('bad handshake', e)
+            raise ssl.SSLError('bad handshake: %r' % e)
         break
 
     return WrappedSocket(cnx, sock)

requests/packages/urllib3/exceptions.py

@@ -112,6 +112,9 @@ class ConnectTimeoutError(TimeoutError):
     "Raised when a socket timeout occurs while connecting to a server"
     pass
 
+class NewConnectionError(ConnectTimeoutError, PoolError):
+    "Raised when we fail to establish a new connection. Usually ECONNREFUSED."
+    pass
 
 class EmptyPoolError(PoolError):
     "Raised when a pool runs out of connections and no more are allowed."
@@ -149,6 +152,11 @@ class SecurityWarning(HTTPWarning):
     pass
 
 
+class SubjectAltNameWarning(SecurityWarning):
+    "Warned when connecting to a host with a certificate missing a SAN."
+    pass
+
+
 class InsecureRequestWarning(SecurityWarning):
     "Warned when making an unverified HTTPS request."
     pass
@@ -167,3 +175,19 @@ class InsecurePlatformWarning(SecurityWarning):
 class ResponseNotChunked(ProtocolError, ValueError):
     "Response needs to be chunked in order to read it as chunks."
     pass
+
+
+class ProxySchemeUnknown(AssertionError, ValueError):
+    "ProxyManager does not support the supplied scheme"
+    # TODO(t-8ch): Stop inheriting from AssertionError in v2.0.
+
+    def __init__(self, scheme):
+        message = "Not supported proxy scheme %s" % scheme
+        super(ProxySchemeUnknown, self).__init__(message)
+
+
+class HeaderParsingError(HTTPError):
+    "Raised by assert_header_parsing, but we convert it to a log.warning statement."
+    def __init__(self, defects, unparsed_data):
+        message = '%s, unparsed data: %r' % (defects or 'Unknown', unparsed_data)
+        super(HeaderParsingError, self).__init__(message)

requests/packages/urllib3/poolmanager.py

@@ -8,7 +8,7 @@ except ImportError:
 from ._collections import RecentlyUsedContainer
 from .connectionpool import HTTPConnectionPool, HTTPSConnectionPool
 from .connectionpool import port_by_scheme
-from .exceptions import LocationValueError, MaxRetryError
+from .exceptions import LocationValueError, MaxRetryError, ProxySchemeUnknown
 from .request import RequestMethods
 from .util.url import parse_url
 from .util.retry import Retry
@@ -227,8 +227,8 @@ class ProxyManager(PoolManager):
             port = port_by_scheme.get(proxy.scheme, 80)
             proxy = proxy._replace(port=port)
 
-        assert proxy.scheme in ("http", "https"), \
-            'Not supported proxy scheme %s' % proxy.scheme
+        if proxy.scheme not in ("http", "https"):
+            raise ProxySchemeUnknown(proxy.scheme)
 
         self.proxy = proxy
         self.proxy_headers = proxy_headers or {}

requests/packages/urllib3/request.py

@@ -71,14 +71,22 @@ class RequestMethods(object):
                                             headers=headers,
                                             **urlopen_kw)
 
-    def request_encode_url(self, method, url, fields=None, **urlopen_kw):
+    def request_encode_url(self, method, url, fields=None, headers=None,
+                           **urlopen_kw):
         """
         Make a request using :meth:`urlopen` with the ``fields`` encoded in
         the url. This is useful for request methods like GET, HEAD, DELETE, etc.
         """
+        if headers is None:
+            headers = self.headers
+
+        extra_kw = {'headers': headers}
+        extra_kw.update(urlopen_kw)
+
         if fields:
             url += '?' + urlencode(fields)
-        return self.urlopen(method, url, **urlopen_kw)
+
+        return self.urlopen(method, url, **extra_kw)
 
     def request_encode_body(self, method, url, fields=None, headers=None,
                             encode_multipart=True, multipart_boundary=None,

requests/packages/urllib3/response.py

@@ -1,7 +1,4 @@
-try:
-    import http.client as httplib
-except ImportError:
-    import httplib
+from contextlib import contextmanager
 import zlib
 import io
 from socket import timeout as SocketTimeout
@@ -11,8 +8,9 @@ from .exceptions import (
     ProtocolError, DecodeError, ReadTimeoutError, ResponseNotChunked
 )
 from .packages.six import string_types as basestring, binary_type, PY3
+from .packages.six.moves import http_client as httplib
 from .connection import HTTPException, BaseSSLError
-from .util.response import is_fp_closed
+from .util.response import is_fp_closed, is_response_to_head
 
 
 class DeflateDecoder(object):
@@ -126,14 +124,15 @@ class HTTPResponse(io.IOBase):
         # Are we using the chunked-style of transfer encoding?
         self.chunked = False
         self.chunk_left = None
-        tr_enc = self.headers.get('transfer-encoding', '')
-        if tr_enc.lower() == "chunked":
+        tr_enc = self.headers.get('transfer-encoding', '').lower()
+        # Don't incur the penalty of creating a list and then discarding it
+        encodings = (enc.strip() for enc in tr_enc.split(","))
+        if "chunked" in encodings:
             self.chunked = True
 
         # We certainly don't want to preload content when the response is chunked.
-        if not self.chunked:
-            if preload_content and not self._body:
-                self._body = self.read(decode_content=decode_content)
+        if not self.chunked and preload_content and not self._body:
+            self._body = self.read(decode_content=decode_content)
 
     def get_redirect_location(self):
         """
@@ -179,9 +178,8 @@ class HTTPResponse(io.IOBase):
         # Note: content-encoding value should be case-insensitive, per RFC 7230
         # Section 3.2
         content_encoding = self.headers.get('content-encoding', '').lower()
-        if self._decoder is None:
-            if content_encoding in self.CONTENT_DECODERS:
-                self._decoder = _get_decoder(content_encoding)
+        if self._decoder is None and content_encoding in self.CONTENT_DECODERS:
+            self._decoder = _get_decoder(content_encoding)
 
     def _decode(self, data, decode_content, flush_decoder):
         """
@@ -202,6 +200,47 @@ class HTTPResponse(io.IOBase):
 
         return data
 
+    @contextmanager
+    def _error_catcher(self):
+        """
+        Catch low-level python exceptions, 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.
+        """
+        try:
+            try:
+                yield
+
+            except SocketTimeout:
+                # FIXME: Ideally we'd like to include the url in the ReadTimeoutError but
+                # there is yet no clean way to get at it from this context.
+                raise ReadTimeoutError(self._pool, None, 'Read timed out.')
+
+            except BaseSSLError as e:
+                # FIXME: Is there a better way to differentiate between SSLErrors?
+                if 'read operation timed out' not in str(e):  # Defensive:
+                    # This shouldn't happen but just in case we're missing an edge
+                    # case, let's avoid swallowing SSL errors.
+                    raise
+
+                raise ReadTimeoutError(self._pool, None, 'Read timed out.')
+
+            except HTTPException as e:
+                # This includes IncompleteRead.
+                raise ProtocolError('Connection broken: %r' % e, e)
+        except Exception:
+            # The response may not be closed but we're not going to use it anymore
+            # so close it now to ensure that the connection is released back to the pool.
+            if self._original_response and not self._original_response.isclosed():
+                self._original_response.close()
+
+            raise
+        finally:
+            if self._original_response and self._original_response.isclosed():
+                self.release_conn()
+
     def read(self, amt=None, decode_content=None, cache_content=False):
         """
         Similar to :meth:`httplib.HTTPResponse.read`, but with two additional
@@ -231,45 +270,28 @@ class HTTPResponse(io.IOBase):
             return
 
         flush_decoder = False
+        data = None
 
-        try:
-            try:
-                if amt is None:
-                    # cStringIO doesn't like amt=None
-                    data = self._fp.read()
+        with self._error_catcher():
+            if amt is None:
+                # cStringIO doesn't like amt=None
+                data = self._fp.read()
+                flush_decoder = True
+            else:
+                cache_content = False
+                data = self._fp.read(amt)
+                if amt != 0 and not data:  # Platform-specific: Buggy versions of Python.
+                    # Close the connection when no data is returned
+                    #
+                    # This is redundant to what httplib/http.client _should_
+                    # already do.  However, versions of python released before
+                    # December 15, 2012 (http://bugs.python.org/issue16298) do
+                    # not properly close the connection in all cases. There is
+                    # no harm in redundantly calling close.
+                    self._fp.close()
                     flush_decoder = True
-                else:
-                    cache_content = False
-                    data = self._fp.read(amt)
-                    if amt != 0 and not data:  # Platform-specific: Buggy versions of Python.
-                        # Close the connection when no data is returned
-                        #
-                        # This is redundant to what httplib/http.client _should_
-                        # already do.  However, versions of python released before
-                        # December 15, 2012 (http://bugs.python.org/issue16298) do
-                        # not properly close the connection in all cases. There is
-                        # no harm in redundantly calling close.
-                        self._fp.close()
-                        flush_decoder = True
-
-            except SocketTimeout:
-                # FIXME: Ideally we'd like to include the url in the ReadTimeoutError but
-                # there is yet no clean way to get at it from this context.
-                raise ReadTimeoutError(self._pool, None, 'Read timed out.')
-
-            except BaseSSLError as e:
-                # FIXME: Is there a better way to differentiate between SSLErrors?
-                if 'read operation timed out' not in str(e):  # Defensive:
-                    # This shouldn't happen but just in case we're missing an edge
-                    # case, let's avoid swallowing SSL errors.
-                    raise
-
-                raise ReadTimeoutError(self._pool, None, 'Read timed out.')
-
-            except HTTPException as e:
-                # This includes IncompleteRead.
-                raise ProtocolError('Connection broken: %r' % e, e)
 
+        if data:
             self._fp_bytes_read += len(data)
 
             data = self._decode(data, decode_content, flush_decoder)
@@ -277,11 +299,8 @@ class HTTPResponse(io.IOBase):
             if cache_content:
                 self._body = data
 
-            return data
+        return data
 
-        finally:
-            if self._original_response and self._original_response.isclosed():
-                self.release_conn()
 
     def stream(self, amt=2**16, decode_content=None):
         """
@@ -299,10 +318,9 @@ class HTTPResponse(io.IOBase):
             If True, will attempt to decode the body based on the
             'content-encoding' header.
         """
-        self._init_decoder()
         if self.chunked:
-            for line in self.read_chunked(amt):
-                yield self._decode(line, decode_content, True)
+            for line in self.read_chunked(amt, decode_content=decode_content):
+                yield line
         else:
             while not is_fp_closed(self._fp):
                 data = self.read(amt=amt, decode_content=decode_content)
@@ -320,6 +338,7 @@ class HTTPResponse(io.IOBase):
         with ``original_response=r``.
         """
         headers = r.msg
+
         if not isinstance(headers, HTTPHeaderDict):
             if PY3: # Python 3
                 headers = HTTPHeaderDict(headers.items())
@@ -387,58 +406,80 @@ class HTTPResponse(io.IOBase):
             b[:len(temp)] = temp
             return len(temp)
 
-    def read_chunked(self, amt=None):
-        # FIXME: Rewrite this method and make it a class with
-        #        a better structured logic.
+    def _update_chunk_length(self):
+        # First, we'll figure out length of a chunk and then
+        # we'll try to read it from socket.
+        if self.chunk_left is not None:
+            return
+        line = self._fp.fp.readline()
+        line = line.split(b';', 1)[0]
+        try:
+            self.chunk_left = int(line, 16)
+        except ValueError:
+            # Invalid chunked protocol response, abort.
+            self.close()
+            raise httplib.IncompleteRead(line)
+
+    def _handle_chunk(self, amt):
+        returned_chunk = None
+        if amt is None:
+            chunk = self._fp._safe_read(self.chunk_left)
+            returned_chunk = chunk
+            self._fp._safe_read(2)  # Toss the CRLF at the end of the chunk.
+            self.chunk_left = None
+        elif amt < self.chunk_left:
+            value = self._fp._safe_read(amt)
+            self.chunk_left = self.chunk_left - amt
+            returned_chunk = value
+        elif amt == self.chunk_left:
+            value = self._fp._safe_read(amt)
+            self._fp._safe_read(2)  # Toss the CRLF at the end of the chunk.
+            self.chunk_left = None
+            returned_chunk = value
+        else:  # amt > self.chunk_left
+            returned_chunk = self._fp._safe_read(self.chunk_left)
+            self._fp._safe_read(2)  # Toss the CRLF at the end of the chunk.
+            self.chunk_left = None
+        return returned_chunk
+
+    def read_chunked(self, amt=None, decode_content=None):
+        """
+        Similar to :meth:`HTTPResponse.read`, but with an additional
+        parameter: ``decode_content``.
+
+        :param decode_content:
+            If True, will attempt to decode the body based on the
+            'content-encoding' header.
+        """
+        self._init_decoder()
+        # FIXME: Rewrite this method and make it a class with a better structured logic.
         if not self.chunked:
             raise ResponseNotChunked("Response is not chunked. "
                 "Header 'transfer-encoding: chunked' is missing.")
-        while True:
-            # First, we'll figure out length of a chunk and then
-            # we'll try to read it from socket.
-            if self.chunk_left is None:
-                line = self._fp.fp.readline()
-                line = line.decode()
-                # See RFC 7230: Chunked Transfer Coding.
-                i = line.find(';')
-                if i >= 0:
-                    line = line[:i]  # Strip chunk-extensions.
-                try:
-                    self.chunk_left = int(line, 16)
-                except ValueError:
-                    # Invalid chunked protocol response, abort.
-                    self.close()
-                    raise httplib.IncompleteRead(''.join(line))
+
+        # Don't bother reading the body of a HEAD request.
+        if self._original_response and is_response_to_head(self._original_response):
+            self._original_response.close()
+            return
+
+        with self._error_catcher():
+            while True:
+                self._update_chunk_length()
                 if self.chunk_left == 0:
                     break
-            if amt is None:
-                chunk = self._fp._safe_read(self.chunk_left)
-                yield chunk
-                self._fp._safe_read(2)  # Toss the CRLF at the end of the chunk.
-                self.chunk_left = None
-            elif amt < self.chunk_left:
-                value = self._fp._safe_read(amt)
-                self.chunk_left = self.chunk_left - amt
-                yield value
-            elif amt == self.chunk_left:
-                value = self._fp._safe_read(amt)
-                self._fp._safe_read(2)  # Toss the CRLF at the end of the chunk.
-                self.chunk_left = None
-                yield value
-            else:  # amt > self.chunk_left
-                yield self._fp._safe_read(self.chunk_left)
-                self._fp._safe_read(2)  # Toss the CRLF at the end of the chunk.
-                self.chunk_left = None
+                chunk = self._handle_chunk(amt)
+                yield self._decode(chunk, decode_content=decode_content,
+                                   flush_decoder=True)
 
-        # Chunk content ends with \r\n: discard it.
-        while True:
-            line = self._fp.fp.readline()
-            if not line:
-                # Some sites may not end with '\r\n'.
-                break
-            if line == b'\r\n':
-                break
-
-        # We read everything; close the "file".
-        self.release_conn()
+            # Chunk content ends with \r\n: discard it.
+            while True:
+                line = self._fp.fp.readline()
+                if not line:
+                    # Some sites may not end with '\r\n'.
+                    break
+                if line == b'\r\n':
+                    break
 
+            # We read everything; close the "file".
+            if self._original_response:
+                self._original_response.close()

requests/packages/urllib3/util/connection.py

@@ -60,6 +60,8 @@ def create_connection(address, timeout=socket._GLOBAL_DEFAULT_TIMEOUT,
     """
 
     host, port = address
+    if host.startswith('['):
+        host = host.strip('[]')
     err = None
     for res in socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM):
         af, socktype, proto, canonname, sa = res
@@ -78,16 +80,16 @@ def create_connection(address, timeout=socket._GLOBAL_DEFAULT_TIMEOUT,
             sock.connect(sa)
             return sock
 
-        except socket.error as _:
-            err = _
+        except socket.error as e:
+            err = e
             if sock is not None:
                 sock.close()
                 sock = None
 
     if err is not None:
         raise err
-    else:
-        raise socket.error("getaddrinfo returns an empty list")
+
+    raise socket.error("getaddrinfo returns an empty list")
 
 
 def _set_socket_options(sock, options):

requests/packages/urllib3/util/response.py

@@ -1,3 +1,8 @@
+from ..packages.six.moves import http_client as httplib
+
+from ..exceptions import HeaderParsingError
+
+
 def is_fp_closed(obj):
     """
     Checks whether a given file-like object is closed.
@@ -20,3 +25,49 @@ def is_fp_closed(obj):
         pass
 
     raise ValueError("Unable to determine whether fp is closed.")
+
+
+def assert_header_parsing(headers):
+    """
+    Asserts whether all headers have been successfully parsed.
+    Extracts encountered errors from the result of parsing headers.
+
+    Only works on Python 3.
+
+    :param headers: Headers to verify.
+    :type headers: `httplib.HTTPMessage`.
+
+    :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('expected httplib.Message, got {}.'.format(
+            type(headers)))
+
+    defects = getattr(headers, 'defects', None)
+    get_payload = getattr(headers, 'get_payload', None)
+
+    unparsed_data = None
+    if get_payload:  # Platform-specific: Python 3.
+        unparsed_data = get_payload()
+
+    if defects or unparsed_data:
+        raise HeaderParsingError(defects=defects, unparsed_data=unparsed_data)
+
+
+def is_response_to_head(response):
+    """
+    Checks, wether a the request of a response has been a HEAD-request.
+    Handles the quirks of AppEngine.
+
+    :param conn:
+    :type conn: :class:`httplib.HTTPResponse`
+    """
+    # FIXME: Can we do this somehow without accessing private httplib _method?
+    method = response._method
+    if isinstance(method, int):  # Platform-specific: Appengine
+        return method == 3
+    return method.upper() == 'HEAD'

requests/packages/urllib3/util/retry.py

@@ -94,7 +94,7 @@ class Retry(object):
 
         seconds. If the backoff_factor is 0.1, then :func:`.sleep` will sleep
         for [0.1s, 0.2s, 0.4s, ...] between retries. It will never be longer
-        than :attr:`Retry.MAX_BACKOFF`.
+        than :attr:`Retry.BACKOFF_MAX`.
 
         By default, backoff is disabled (set to 0).
 

requests/packages/urllib3/util/ssl_.py

@@ -8,6 +8,13 @@ SSLContext = None
 HAS_SNI = False
 create_default_context = None
 
+# Maps the length of a digest to a possible hash function producing this digest
+HASHFUNC_MAP = {
+    32: md5,
+    40: sha1,
+    64: sha256,
+}
+
 import errno
 import warnings
 
@@ -68,8 +75,11 @@ except ImportError:
             self.certfile = certfile
             self.keyfile = keyfile
 
-        def load_verify_locations(self, location):
-            self.ca_certs = location
+        def load_verify_locations(self, cafile=None, capath=None):
+            self.ca_certs = cafile
+
+            if capath is not None:
+                raise SSLError("CA directories not supported in older Pythons")
 
         def set_ciphers(self, cipher_suite):
             if not self.supports_set_ciphers:
@@ -112,31 +122,21 @@ def assert_fingerprint(cert, fingerprint):
         Fingerprint as string of hexdigits, can be interspersed by colons.
     """
 
-    # Maps the length of a digest to a possible hash function producing
-    # this digest.
-    hashfunc_map = {
-        16: md5,
-        20: sha1,
-        32: sha256,
-    }
-
     fingerprint = fingerprint.replace(':', '').lower()
-    digest_length, odd = divmod(len(fingerprint), 2)
-
-    if odd or digest_length not in hashfunc_map:
-        raise SSLError('Fingerprint is of invalid length.')
+    digest_length = len(fingerprint)
+    hashfunc = HASHFUNC_MAP.get(digest_length)
+    if not hashfunc:
+        raise SSLError(
+            'Fingerprint of invalid length: {0}'.format(fingerprint))
 
     # We need encode() here for py32; works on py2 and p33.
     fingerprint_bytes = unhexlify(fingerprint.encode())
 
-    hashfunc = hashfunc_map[digest_length]
-
     cert_digest = hashfunc(cert).digest()
 
-    if not cert_digest == fingerprint_bytes:
+    if cert_digest != fingerprint_bytes:
         raise SSLError('Fingerprints did not match. Expected "{0}", got "{1}".'
-                       .format(hexlify(fingerprint_bytes),
-                               hexlify(cert_digest)))
+                       .format(fingerprint, hexlify(cert_digest)))
 
 
 def resolve_cert_reqs(candidate):
@@ -243,10 +243,11 @@ def create_urllib3_context(ssl_version=None, cert_reqs=None,
 
 def ssl_wrap_socket(sock, keyfile=None, certfile=None, cert_reqs=None,
                     ca_certs=None, server_hostname=None,
-                    ssl_version=None, ciphers=None, ssl_context=None):
+                    ssl_version=None, ciphers=None, ssl_context=None,
+                    ca_cert_dir=None):
     """
-    All arguments except for server_hostname and ssl_context have the same
-    meaning as they do when using :func:`ssl.wrap_socket`.
+    All arguments except for server_hostname, ssl_context, and ca_cert_dir have
+    the same meaning as they do when using :func:`ssl.wrap_socket`.
 
     :param server_hostname:
         When SNI is supported, the expected hostname of the certificate
@@ -256,15 +257,19 @@ def ssl_wrap_socket(sock, keyfile=None, certfile=None, cert_reqs=None,
     :param ciphers:
         A string of ciphers we wish the client to support. This is not
         supported on Python 2.6 as the ssl module does not support it.
+    :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().
     """
     context = ssl_context
     if context is None:
         context = create_urllib3_context(ssl_version, cert_reqs,
                                          ciphers=ciphers)
 
-    if ca_certs:
+    if ca_certs or ca_cert_dir:
         try:
-            context.load_verify_locations(ca_certs)
+            context.load_verify_locations(ca_certs, ca_cert_dir)
         except IOError as e:  # Platform-specific: Python 2.6, 2.7, 3.2
             raise SSLError(e)
         # Py33 raises FileNotFoundError which subclasses OSError
@@ -273,6 +278,7 @@ def ssl_wrap_socket(sock, keyfile=None, certfile=None, cert_reqs=None,
             if e.errno == errno.ENOENT:
                 raise SSLError(e)
             raise
+
     if certfile:
         context.load_cert_chain(certfile, keyfile)
     if HAS_SNI:  # Platform-specific: OpenSSL with enabled SNI

requests/packages/urllib3/util/url.py

@@ -15,6 +15,8 @@ class Url(namedtuple('Url', url_attrs)):
 
     def __new__(cls, scheme=None, auth=None, host=None, port=None, path=None,
                 query=None, fragment=None):
+        if path and not path.startswith('/'):
+            path = '/' + path
         return super(Url, cls).__new__(cls, scheme, auth, host, port, path,
                                        query, fragment)
 

requests/sessions.py

@@ -62,12 +62,11 @@ def merge_setting(request_setting, session_setting, dict_class=OrderedDict):
     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.
-    for (k, v) in request_setting.items():
-        if v is None:
-            del merged_setting[k]
-
-    merged_setting = dict((k, v) for (k, v) in merged_setting.items() if v is not None)
+    # 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
 
@@ -275,6 +274,12 @@ class Session(SessionRedirectMixin):
       >>> s = requests.Session()
       >>> s.get('http://httpbin.org/get')
       200
+
+    Or as a context manager::
+
+      >>> with requests.Session() as s:
+      >>>     s.get('http://httpbin.org/get')
+      200
     """
 
     __attrs__ = [
@@ -294,9 +299,9 @@ class Session(SessionRedirectMixin):
         #: :class:`Request <Request>`.
         self.auth = None
 
-        #: Dictionary mapping protocol to the URL of the proxy (e.g.
-        #: {'http': 'foo.bar:3128'}) to be used on each
-        #: :class:`Request <Request>`.
+        #: 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.
@@ -320,7 +325,8 @@ class Session(SessionRedirectMixin):
         #: limit, a :class:`TooManyRedirects` exception is raised.
         self.max_redirects = DEFAULT_REDIRECT_LIMIT
 
-        #: Should we trust the environment?
+        #: Trust environement settings for proxy configuration, default
+        #: authentication and similar.
         self.trust_env = True
 
         #: A CookieJar containing all currently outstanding cookies set on this
@@ -405,8 +411,8 @@ class Session(SessionRedirectMixin):
         :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 or bytes to send in the body of the
-            :class:`Request`.
+        :param data: (optional) Dictionary, 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
@@ -418,13 +424,13 @@ class Session(SessionRedirectMixin):
         :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 (`connect timeout, read
-            timeout <user/advanced.html#timeouts>`_) tuple.
+            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 to the URL of
-            the proxy.
+        :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) if ``True``, the SSL cert will be verified.
@@ -432,9 +438,6 @@ class Session(SessionRedirectMixin):
         :param cert: (optional) if String, path to ssl client cert file (.pem).
             If Tuple, ('cert', 'key') pair.
         """
-
-        method = to_native_string(method)
-
         # Create the Request.
         req = Request(
             method = method.upper(),

requests/status_codes.py

@@ -82,7 +82,7 @@ _codes = {
 
 codes = LookupDict(name='status_codes')
 
-for (code, titles) in list(_codes.items()):
+for code, titles in _codes.items():
     for title in titles:
         setattr(codes, title, code)
         if not title.startswith('\\'):

requests/utils.py

@@ -67,7 +67,7 @@ def super_len(o):
         return len(o.getvalue())
 
 
-def get_netrc_auth(url):
+def get_netrc_auth(url, raise_errors=False):
     """Returns the Requests tuple auth for a given url from netrc."""
 
     try:
@@ -105,8 +105,9 @@ def get_netrc_auth(url):
                 return (_netrc[login_i], _netrc[2])
         except (NetrcParseError, IOError):
             # If there was a parsing error or a permissions issue reading the file,
-            # we'll just skip netrc auth
-            pass
+            # we'll just skip netrc auth unless explicitly asked to raise errors.
+            if raise_errors:
+                raise
 
     # AppEngine hackiness.
     except (ImportError, AttributeError):
@@ -498,7 +499,9 @@ def should_bypass_proxies(url):
     if no_proxy:
         # We need to check whether we match here. We need to see if we match
         # the end of the netloc, both with and without the port.
-        no_proxy = no_proxy.replace(' ', '').split(',')
+        no_proxy = (
+            host for host in no_proxy.replace(' ', '').split(',') if host
+        )
 
         ip = netloc.split(':')[0]
         if is_ipv4_address(ip):
@@ -536,36 +539,22 @@ def get_environ_proxies(url):
     else:
         return getproxies()
 
+def select_proxy(url, proxies):
+    """Select a proxy for the url, if applicable.
+
+    :param url: The url being for the request
+    :param proxies: A dictionary of schemes or schemes and hosts to proxy URLs
+    """
+    proxies = proxies or {}
+    urlparts = urlparse(url)
+    proxy = proxies.get(urlparts.scheme+'://'+urlparts.hostname)
+    if proxy is None:
+        proxy = proxies.get(urlparts.scheme)
+    return proxy
 
 def default_user_agent(name="python-requests"):
     """Return a string representing the default user agent."""
-    _implementation = platform.python_implementation()
-
-    if _implementation == 'CPython':
-        _implementation_version = platform.python_version()
-    elif _implementation == 'PyPy':
-        _implementation_version = '%s.%s.%s' % (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'
-
-    try:
-        p_system = platform.system()
-        p_release = platform.release()
-    except IOError:
-        p_system = 'Unknown'
-        p_release = 'Unknown'
-
-    return " ".join(['%s/%s' % (name, __version__),
-                     '%s/%s' % (_implementation, _implementation_version),
-                     '%s/%s' % (p_system, p_release)])
+    return '%s/%s' % (name, __version__)
 
 
 def default_headers():

requirements.txt

@@ -1,4 +1,4 @@
-py==1.4.12
-pytest==2.3.4
-pytest-cov==1.6
+py==1.4.30
+pytest==2.8.1
+pytest-cov==2.1.0
 wheel

test_requests.py

@@ -9,6 +9,7 @@ import os
 import pickle
 import unittest
 import collections
+import contextlib
 
 import io
 import requests
@@ -16,7 +17,9 @@ import pytest
 from requests.adapters import HTTPAdapter
 from requests.auth import HTTPDigestAuth, _basic_auth_str
 from requests.compat import (
-    Morsel, cookielib, getproxies, str, urljoin, urlparse, is_py3, builtin_str)
+    Morsel, cookielib, getproxies, str, urljoin, urlparse, is_py3,
+    builtin_str, OrderedDict
+    )
 from requests.cookies import cookiejar_from_dict, morsel_to_cookie
 from requests.exceptions import (ConnectionError, ConnectTimeout,
                                  InvalidSchema, InvalidURL, MissingSchema,
@@ -32,6 +35,11 @@ try:
 except ImportError:
     import io as StringIO
 
+try:
+    from multiprocessing.pool import ThreadPool
+except ImportError:
+    ThreadPool = None
+
 if is_py3:
     def u(s):
         return s
@@ -124,6 +132,13 @@ class RequestsTestCase(unittest.TestCase):
             "http://example.com/path?key=value#fragment", params={"a": "b"}).prepare()
         assert request.url == "http://example.com/path?key=value&a=b#fragment"
 
+    def test_params_original_order_is_preserved_by_default(self):
+        param_ordered_dict = OrderedDict((('z', 1), ('a', 1), ('k', 1), ('d', 1)))
+        session = requests.Session()
+        request = requests.Request('GET', 'http://example.com/', params=param_ordered_dict)
+        prep = session.prepare_request(request)
+        assert prep.url == 'http://example.com/?z=1&a=1&k=1&d=1'
+
     def test_mixed_case_scheme_acceptable(self):
         s = requests.Session()
         s.proxies = getproxies()
@@ -411,6 +426,21 @@ class RequestsTestCase(unittest.TestCase):
         r = requests.get(url, auth=auth)
         assert '"auth"' in r.request.headers['Authorization']
 
+    def test_DIGESTAUTH_THREADED(self):
+
+        auth = HTTPDigestAuth('user', 'pass')
+        url = httpbin('digest-auth', 'auth', 'user', 'pass')
+        session = requests.Session()
+        session.auth=auth
+
+        def do_request(i):
+            r = session.get(url)
+            assert '"auth"' in r.request.headers['Authorization']
+            return 1
+        if ThreadPool is not None:
+            pool = ThreadPool(processes=50)
+            pool.map(do_request, range(100))
+
     def test_POSTBIN_GET_POST_FILES(self):
 
         url = httpbin('post')
@@ -538,6 +568,17 @@ class RequestsTestCase(unittest.TestCase):
             method=u('POST'), url=httpbin('post'), files=files)
         assert r.status_code == 200
 
+    def test_unicode_method_name_with_request_object(self):
+        files = {'file': open('test_requests.py', 'rb')}
+        s = requests.Session()
+        req = requests.Request(u("POST"), httpbin('post'), files=files)
+        prep = s.prepare_request(req)
+        assert isinstance(prep.method, builtin_str)
+        assert prep.method == "POST"
+
+        resp = s.send(prep)
+        assert resp.status_code == 200
+
     def test_custom_content_type(self):
         r = requests.post(
             httpbin('post'),
@@ -1052,6 +1093,13 @@ class RequestsTestCase(unittest.TestCase):
         assert 'application/json' in r.request.headers['Content-Type']
         assert {'life': 42} == r.json()['json']
 
+    def test_json_param_post_should_not_override_data_param(self):
+        r = requests.Request(method='POST', url='http://httpbin.org/post',
+                             data={'stuff': 'elixr'},
+                             json={'music': 'flute'})
+        prep = r.prepare()
+        assert 'stuff=elixr' == prep.body
+
     def test_response_iter_lines(self):
         r = requests.get(httpbin('stream/4'), stream=True)
         assert r.status_code == 200
@@ -1060,6 +1108,15 @@ class RequestsTestCase(unittest.TestCase):
         next(it)
         assert len(list(it)) == 3
 
+    def test_unconsumed_session_response_closes_connection(self):
+        s = requests.session()
+
+        with contextlib.closing(s.get(httpbin('stream/4'), stream=True)) as response:
+            pass
+
+        self.assertFalse(response._content_consumed)
+        self.assertTrue(response.raw.closed)
+
     @pytest.mark.xfail
     def test_response_iter_lines_reentrant(self):
         """Response.iter_lines() is not reentrant safe"""
@@ -1214,6 +1271,7 @@ class TestCaseInsensitiveDict(unittest.TestCase):
         del othercid['spam']
         assert cid != othercid
         assert cid == {'spam': 'blueval', 'eggs': 'redval'}
+        assert cid != object()
 
     def test_setdefault(self):
         cid = CaseInsensitiveDict({'Spam': 'blueval'})
@@ -1251,6 +1309,16 @@ class TestCaseInsensitiveDict(unittest.TestCase):
         assert frozenset(cid.keys()) == keyset
         assert frozenset(cid) == keyset
 
+    def test_copy(self):
+        cid = CaseInsensitiveDict({
+            'Accept': 'application/json',
+            'user-Agent': 'requests',
+        })
+        cid_copy = cid.copy()
+        assert cid == cid_copy
+        cid['changed'] = True
+        assert cid != cid_copy
+
 
 class UtilsTestCase(unittest.TestCase):
 
@@ -1297,6 +1365,15 @@ class UtilsTestCase(unittest.TestCase):
             'http://localhost.localdomain:5000/v1.0/') == {}
         assert get_environ_proxies('http://www.requests.com/') != {}
 
+    def test_select_proxies(self):
+        """Make sure we can select per-host proxies correctly."""
+        from requests.utils import select_proxy
+        proxies = {'http': 'http://http.proxy',
+                   'http://some.host': 'http://some.host.proxy'}
+        assert select_proxy('hTTp://u:p@Some.Host/path', proxies) == 'http://some.host.proxy'
+        assert select_proxy('hTTp://u:p@Other.Host/path', proxies) == 'http://http.proxy'
+        assert select_proxy('hTTps://Other.Host', proxies) is None
+
     def test_guess_filename_when_int(self):
         from requests.utils import guess_filename
         assert None is guess_filename(1)
@@ -1627,6 +1704,16 @@ def test_urllib3_retries():
     with pytest.raises(RetryError):
         s.get(httpbin('status/500'))
 
+
+def test_urllib3_pool_connection_closed():
+    s = requests.Session()
+    s.mount('http://', HTTPAdapter(pool_connections=0, pool_maxsize=0))
+
+    try:
+        s.get(httpbin('status/200'))
+    except ConnectionError as e:
+        assert u"HTTPConnectionPool(host='httpbin.org', port=80): Pool is closed." in str(e)
+
 def test_vendor_aliases():
     from requests.packages import urllib3
     from requests.packages import chardet