* Mon Jun 22 2015 firstname.lastname@example.org
- update to 0.10.4:
- Re-release of 0.10.3 with packaging artifacts manually removed.
- Re-release of 0.10.2 without packaging artifacts.
- Fixed issue where ``empty`` could break third-party libraries that relied on
keyword arguments (pull request ``#675``)
- Improved ``Rule.empty`` by providing a ```get_empty_kwargs`` to allow setting
custom kwargs without having to override entire ``empty`` method. (pull
- Fixed ```extra_files``` parameter for reloader to not cause startup
to crash when included in server params
- Using `MultiDict` when building URLs is now not supported again. The behavior
introduced several regressions.
- Fix performance problems with stat-reloader (pull request ``#715``).
- Fixed regression with multiple query values for URLs (pull request ``#667``).
- Fix issues with eventlet's monkeypatching and the builtin server (pull
- Changed the error handling of and improved testsuite for the caches in
- Fixed a bug on Python 3 when creating adhoc ssl contexts, due to `sys.maxint`
not being defined.
- Fixed a bug on Python 3, that caused
:func:`~werkzeug.serving.make_ssl_devcert` to fail with an exception.
- Added exceptions for 504 and 505.
- Added support for ChromeOS detection.
- Added UUID converter to the routing system.
- Added message that explains how to quit the server.
- Fixed a bug on Python 2, that caused ``len`` for
:class:`werkzeug.datastructures.CombinedMultiDict` to crash.
- Added support for stdlib pbkdf2 hmac if a compatible digest
- Ported testsuite to use ``py.test``.
- Minor optimizations to various middlewares (pull requests ``#496`` and
- Use stdlib ``ssl`` module instead of ``OpenSSL`` for the builtin server
(issue ``#434``). This means that OpenSSL contexts are not supported anymore,
but instead ``ssl.SSLContext`` from the stdlib.
- Allow protocol-relative URLs when building external URLs.
- Fixed Atom syndication to print time zone offset for tz-aware datetime
objects (pull request ``#254``).
- Improved reloader to track added files and to recover from broken
sys.modules setups with syntax errors in packages.
- ``cache.RedisCache`` now supports arbitrary ``**kwargs`` for the redis
- ``werkzeug.test.Client`` now uses the original request method when resolving
307 redirects (pull request ``#556``).
- ``werkzeug.datastructures.MIMEAccept`` now properly deals with mimetype
parameters (pull request ``#205``).
- ``werkzeug.datastructures.Accept`` now handles a quality of ``0`` as
intolerable, as per RFC 2616 (pull request ``#536``).
- ``werkzeug.urls.url_fix`` now properly encodes hostnames with ``idna``
encoding (issue ``#559``). It also doesn't crash on malformed URLs anymore
- ``werkzeug.routing.MapAdapter.match`` now recognizes the difference between
the path ``/`` and an empty one (issue ``#360``).
- The interactive debugger now tries to decode non-ascii filenames (issue
- Increased default key size of generated SSL certificates to 1024 bits (issue
- Added support for specifying a ``Response`` subclass to use when calling
- ``werkzeug.test.EnvironBuilder`` now doesn't use the request method anymore
to guess the content type, and purely relies on the ``form``, ``files`` and
``input_stream`` properties (issue ``#620``).
- Added Symbian to the user agent platform list.
- Fixed make_conditional to respect automatically_set_content_length
- Unset ``Content-Length`` when writing to response.stream (issue ``#451``)
- ``wrappers.Request.method`` is now always uppercase, eliminating
inconsistencies of the WSGI environment (issue ``647``).
- ``routing.Rule.empty`` now works correctly with subclasses of ``Rule`` (pull
- Made map updating safe in light of concurrent updates.
- Allow multiple values for the same field for url building (issue ``#658``).
- Fix unicode problems in ``werkzeug.debug.tbtools``.
- Fix Python 3-compatibility problems in ``werkzeug.posixemulation``.
- Backport fix of fatal typo for ``ImmutableList`` (issue ``#492``).
- Make creation of the cache dir for ``FileSystemCache`` atomic (issue
- Use native strings for memcached keys to work with Python 3 client (issue
- Fix charset detection for ``werkzeug.debug.tbtools.Frame`` objects (issues
``#547`` and ``#532``).
- Fix ``AttributeError`` masking in ``werkzeug.utils.import_string`` (issue
- Explicitly shut down server (issue ``#519``).
- Fix timeouts greater than 2592000 being misinterpreted as UNIX timestamps in
``werkzeug.contrib.cache.MemcachedCache`` (issue ``#533``).
- Fix bug where ``werkzeug.exceptions.abort`` would raise an arbitrary subclass
of the expected class (issue ``#422``).
- Fix broken ``jsrouting`` (due to removal of ``werkzeug.templates``)
- ``werkzeug.urls.url_fix`` now doesn't crash on malformed URLs anymore, but
returns them unmodified. This is a cheap workaround for ``#582``, the proper
fix is included in version 0.10.
- The repr of ``werkzeug.wrappers.Request`` doesn't crash on non-ASCII-values
anymore (pull request ``#466``).
- Fix bug in ``cache.RedisCache`` when combined with ``redis.StrictRedis``
object (pull request ``#583``).
- The ``qop`` parameter for ``WWW-Authenticate`` headers is now always quoted,
as required by RFC 2617 (issue ``#633``).
- Fix bug in ``werkzeug.contrib.cache.SimpleCache`` with Python 3 where add/set
may throw an exception when pruning old entries from the cache (pull request
* Fri Jul 18 2014 email@example.com
- Update to 0.9.6
- Added a safe conversion for IRI to URI conversion and use that
internally to work around issues with spec violations for
protocols such as ``itms-service``.
- Update to 0.9.5
- Forward charset argument from request objects to the environ
- Fixed error handling for missing boundaries in multipart data.
- Fixed session creation on systems without ``os.urandom()``.
- Fixed pluses in dictionary keys not being properly URL encoded.
- Fixed a problem with deepcopy not working for multi dicts.
- Fixed a double quoting issue on redirects.
- Fixed a problem with unicode keys appearing in headers on 2.x.
- Fixed a bug with unicode strings in the test builder.
- Fixed a unicode bug on Python 3 in the WSGI profiler.
- Fixed an issue with the safe string compare function on
Python 2.7.7 and Python 3.4.
* Thu Oct 24 2013 firstname.lastname@example.org
- Require python-setuptools instead of distribute (upstreams merged)
* Tue Sep 03 2013 email@example.com
- update to 0.9.4:
- Fixed an issue with Python 3.3 and an edge case in cookie parsing.
- Fixed decoding errors not handled properly through the WSGI
- Fixed URI to IRI conversion incorrectly decoding percent signs.
- Restored beahvior of the ``data`` descriptor of the request class to pre 0.9
behavior. This now also means that ``.data`` and ``.get_data()`` have
different behavior. New code should use ``.get_data()`` always.
In addition to that there is now a flag for the ``.get_data()`` method that
controls what should happen with form data parsing and the form parser will
honor cached data. This makes dealing with custom form data more consistent.
- Added `unsafe` parameter to :func:`~werkzeug.urls.url_quote`.
- Fixed an issue with :func:`~werkzeug.urls.url_quote_plus` not quoting
- Ported remaining parts of :class:`~werkzeug.contrib.RedisCache` to
- Ported remaining parts of :class:`~werkzeug.contrib.MemcachedCache` to
- Fixed a deprecation warning in the contrib atom module.
- Fixed a regression with setting of content types through the
headers dictionary instead with the content type parameter.
- Use correct name for stdlib secure string comparision function.
- Fixed a wrong reference in the docstring of
- Fixed an `AttributeError` that sometimes occurred when accessing the
- Fixed an issue with integers no longer being accepted in certain
parts of the routing system or URL quoting functions.
- Fixed an issue with `url_quote` not producing the right escape
codes for single digit codepoints.
- Fixed an issue with :class:`~werkzeug.wsgi.SharedDataMiddleware` not
reading the path correctly and breaking on etag generation in some
- Properly handle `Expect: 100-continue` in the development server
to resolve issues with curl.
- Automatically exhaust the input stream on request close. This should
fix issues where not touching request files results in a timeout.
- Fixed exhausting of streams not doing anything if a non-limited
stream was passed into the multipart parser.
- Raised the buffer sizes for the multipart parser.
- Added support for :meth:`~werkzeug.wsgi.LimitedStream.tell`
on the limited stream.
- :class:`~werkzeug.datastructures.ETags` now is nonzero if it
contains at least one etag of any kind, including weak ones.
- Added a workaround for a bug in the stdlib for SSL servers.
- Improved SSL interface of the devserver so that it can generate
certificates easily and load them from files.
- Refactored test client to invoke the open method on the class
for redirects. This makes subclassing more powerful.
- :func:`werkzeug.wsgi.make_chunk_iter` and
:func:`werkzeug.wsgi.make_line_iter` now support processing of
iterators and streams.
- URL generation by the routing system now no longer quotes
- URL fixing now no longer quotes certain reserved characters.
- The :func:`werkzeug.security.generate_password_hash` and
check functions now support any of the hashlib algorithms.
- `wsgi.get_current_url` is now ascii safe for browsers sending
non-ascii data in query strings.
- improved parsing behavior for :func:`werkzeug.http.parse_options_header`
- added more operators to local proxies.
- added a hook to override the default converter in the routing
- The description field of HTTP exceptions is now always escaped.
Use markup objects to disable that.
- Added number of proxy argument to the proxy fix to make it more
secure out of the box on common proxy setups. It will by default
no longer trust the x-forwarded-for header as much as it did
- Added support for fragment handling in URI/IRI functions.
- Added custom class support for :func:`werkzeug.http.parse_dict_header`.
- Renamed `LighttpdCGIRootFix` to `CGIRootFix`.
- Always treat `+` as safe when fixing URLs as people love misusing them.
- Added support to profiling into directories in the contrib profiler.
- The escape function now by default escapes quotes.
- Changed repr of exceptions to be less magical.
- Simplified exception interface to no longer require environmnts
to be passed to recieve the response object.
- Added sentinel argument to IterIO objects.
- Added pbkdf2 support for the security module.
- Added a plain request type that disables all form parsing to only
leave the stream behind.
- Removed support for deprecated `fix_headers`.
- Removed support for deprecated `header_list`.
- Removed support for deprecated parameter for `iter_encoded`.
- Removed support for deprecated non-silent usage of the limited
- Removed support for previous dummy `writable` parameter on
the cached property.
- Added support for explicitly closing request objects to close
- Conditional request handling or access to the data property on responses no
longer ignores direct passthrough mode.
- Removed werkzeug.templates and werkzeug.contrib.kickstart.
- Changed host lookup logic for forwarded hosts to allow lists of
hosts in which case only the first one is picked up.
- Added `wsgi.get_query_string`, `wsgi.get_path_info` and
`wsgi.get_script_name` and made the `wsgi.pop_path_info` and
`wsgi.peek_path_info` functions perform unicode decoding. This
was necessary to avoid having to expose the WSGI encoding dance
on Python 3.
- Added `content_encoding` and `content_md5` to the request object's
common request descriptor mixin.
- added `options` and `trace` to the test client.
- Overhauled the utilization of the input stream to be easier to use
and better to extend. The detection of content payload on the input
side is now more compliant with HTTP by detecting off the content
type header instead of the request method. This also now means that
the stream property on the request class is always available instead
of just when the parsing fails.
- Added support for using :class:`werkzeug.wrappers.BaseResponse` in a with
- Changed `get_app_iter` to fetch the response early so that it does not
fail when wrapping a response iterable. This makes filtering easier.
- Introduced `get_data` and `set_data` methods for responses.
- Introduced `get_data` for requests.
- Soft deprecated the `data` descriptors for request and response objects.
- Added `as_bytes` operations to some of the headers to simplify working
with things like cookies.
- Made the debugger paste tracebacks into github's gist service as
* Mon Mar 12 2012 firstname.lastname@example.org
- Update to version 0.8.3:
- Fixed another issue with :func:`werkzeug.wsgi.make_line_iter`
where lines longer than the buffer size were not handled
- Restore stdout after debug console finished executing so
that the debugger can be used on GAE better.
- Fixed a bug with the redis cache for int subclasses
(affects bool caching).
- Fixed an XSS problem with redirect targets coming from
- Changes from version 0.8.2:
- Fixed a problem with request handling of the builtin server
not repsonding to socket errors properly.
- The routing request redirect exception's code attribute is now
- Fixed a bug with shutdowns on Windows.
- Fixed a few unicode issues with non-ascii characters being
hardcoded in URL rules.
- Fixed two property docstrings being assigned to fdel instead
- Fixed an issue where CRLF line endings could be split into two
by the line iter function, causing problems with multipart file
* Thu Nov 10 2011 email@example.com
- Update to version 0.8.1:
* Fixed an issue with the memcache not working properly.
* Fixed an issue for Python 2.7.1 and higher that broke
copying of multidicts with :func:`copy.copy`.
* Changed hashing methodology of immutable ordered multi dicts
for a potential problem with alternative Python implementations.
- Changes from version 0.8:
* Removed data structure specific KeyErrors for a general
* Documented :meth:`werkzeug.wrappers.BaseRequest._load_form_data`.
* The routing system now also accepts strings instead of
dictionaries for the `query_args` parameter since we're only
passing them through for redirects.
* Werkzeug now automatically sets the content length immediately when
the :attr:`~werkzeug.wrappers.BaseResponse.data` attribute is set
for efficiency and simplicity reasons.
* The routing system will now normalize server names to lowercase.
* The routing system will no longer raise ValueErrors in case the
configuration for the server name was incorrect. This should make
deployment much easier because you can ignore that factor now.
* Fixed a bug with parsing HTTP digest headers. It rejected headers
with missing nc and nonce params.
* Proxy fix now also updates wsgi.url_scheme based on X-Forwarded-Proto.
* Added support for key prefixes to the redis cache.
* Added the ability to supress some auto corrections in the wrappers
that are now controlled via `autocorrect_location_header` and
`automatically_set_content_length` on the response objects.
* Werkzeug now uses a new method to check that the length of incoming
data is complete and will raise IO errors by itself if the server
fails to do so.
* :func:`~werkzeug.wsgi.make_line_iter` now requires a limit that is
not higher than the length the stream can provide.
* Refactored form parsing into a form parser class that makes it possible
to hook into individual parts of the parsing process for debugging and
* For conditional responses the content length is no longer set when it
is already there and added if missing.
* Immutable datastructures are hashable now.
* Headers datastructure no longer allows newlines in values to avoid
header injection attacks.
* Made it possible through subclassing to select a different remote
addr in the proxy fix.
* Added stream based URL decoding. This reduces memory usage on large
transmitted form data that is URL decoded since Werkzeug will no longer
load all the unparsed data into memory.
* Memcache client now no longer uses the buggy cmemcache module and
supports pylibmc. GAE is not tried automatically and the dedicated
class is no longer necessary.
* Redis cache now properly serializes data.
* Removed support for Python 2.4
- Changes from version 0.7.2:
* Fixed a CSRF problem with the debugger.
* The debugger is now generating private pastes on lodgeit.
* If URL maps are now bound to environments the query arguments
are properly decoded from it for redirects.
* Tue Sep 27 2011 firstname.lastname@example.org
- Package renamed to python-Werkzeug
- Update to version 0.7.1:
* Fixed a problem with newer versions of IPython
* Disabled pyinotify based reloader which does not work reliably.
- Changes from version 0.7.0:
* Add support for python-libmemcached to the Werkzeug cache abstraction
* improved url_decode and url_encode performance.
* fixed an issue where the SharedDataMiddleware could cause an
internal server error on weird paths when loading via pkg_resources.
* fixed an URL generation bug that caused URLs to be invalid if a
generated component contains a colon.
* werkzeug.import_string now works with partially set up
* disabled automatic socket swiching for IPv6 on the development
server due to problems it caused.
* Werkzeug no longer overrides the Date header when creating a
conditional HTTP response.
* The routing system provides a method to retrieve the matching
methods for a given path.
* The routing system now accepts a parameter to change the encoding
* The local manager can now accept custom ident functions in the
constructor that are forwarded to the wrapped local objects.
* url_unquote_plus now accepts unicode strings again.
* fixed an issues with the filesystem session support's prune
function and concurrent usage.
* fixed a problem with external URL generation discarding the port.
* added support for pylibmc to the Werkzeug cache abstraction layer.
* fixed an issue with the new multipart parser that happened when
a linkebreak happend to be on the chunk limit.
* cookies are now set properly if ports are in use. A runtime error
is raised if one tries to set a cookie for a domain without a dot.
* fixed an issue with Template.from_file not working for file
* reloader can now use inotify to track reloads. This requires the
pyinotify library to be installed.
* See more in file CHANGES...
- Generate HTML documentation with Sphinx
- Don't package PKG-INFO
- BuildRequire python-distribute instead of python-setuptools
* Thu Apr 14 2011 email@example.com
- Add spec file license header
- Use py_requires
- Moved changelog from spec to changes file
- Corrected RPM groups
* Wed Apr 13 2011 firstname.lastname@example.org
- Update to 0.6.2
* Fri Mar 05 2010 email@example.com
- Updating because upstream release of Werkzeug 0.6
* Tue Aug 25 2009 firstname.lastname@example.org
- Initial package
* Wed Aug 14 2019 email@example.com
- Add 0001-unique-debugger-pin-in-Docker-containers.patch (bsc#1145383, CVE-2019-14806)
When running the development server in Docker, the debugger security pin is
now unique per container.
* Tue Aug 08 2017 firstname.lastname@example.org
- update to 0.12.2:
- Fix regression: Pull request ``#892`` prevented Werkzeug from correctly
logging the IP of a remote client behind a reverse proxy, even when using
- Fix a bug in `safe_join` on Windows.
* Tue Apr 04 2017 email@example.com
- update for singlespec
- update to 0.12.1
* deprecate werkzeug.script
* Use `inspect.getfullargspec` internally when available as
`inspect.getargspec` is gone in 3.6
* Added support for status code 451 and 423
* Improved the build error suggestions. In particular only if
someone stringifies the error will the suggestions be calculated.
* Added support for uWSGI's caching backend.
* Fix a bug where iterating over a `FileStorage` would result in an infinite
* Datastructures now inherit from the relevant baseclasses from the
`collections` module in the stdlib. See #794.
* Add support for recognizing NetBSD, OpenBSD, FreeBSD, DragonFlyBSD platforms
in the user agent string.
* Recognize SeaMonkey browser name and version correctly
* Recognize Baiduspider, and bingbot user agents
* If `LocalProxy`'s wrapped object is a function, refer to it with __wrapped__
* The defaults of ``generate_password_hash`` have been changed to more secure
ones, see pull request ``#753``.
* Add support for encoding in options header parsing, see pull request
* ``test.Client`` now properly handles Location headers with relative URLs, see
pull request ``#879``.
* When `HTTPException` is raised, it now prints the description, for easier
* Werkzeug's dict-like datastructures now have ``view``-methods under Python 2,
see pull request ``#968``.
* Fix a bug in ``MultiPartParser`` when no ``stream_factory`` was provided
during initialization, see pull request ``#973``.
* Disable autocorrect and spellchecker in the debugger middleware's Python
prompt, see pull request ``#994``.
* Don't redirect to slash route when method doesn't match, see pull request
* Fix a bug when using ``SharedDataMiddleware`` with frozen packages, see pull
* `Range` header parsing function fixed for invalid values ``#974``.
* Add support for byte Range Requests, see pull request ``#978``.
* Use modern cryptographic defaults in the dev servers ``#1004``.
* the post() method of the test client now accept file object through the data
* Color run_simple's terminal output based on HTTP codes ``#1013``.
* Fix self-XSS in debugger console, see ``#1031``.
* Fix IPython 5.x shell support, see ``#1033``.
* Thu Nov 17 2016 firstname.lastname@example.org
- Include in SLE 12 (FATE#320818, bsc#979331)
* Fri Sep 16 2016 email@example.com
- Fix download url.
* Thu Sep 15 2016 firstname.lastname@example.org
- update to version 0.11.11:
* Fix JSONRequestMixin for Python3. See #731
* Fix broken string handling in test client when passing
integers. See #852
* Fix a bug in "parse_options_header" where an invalid content type
starting with comma or semi-colon would result in an invalid
return value, see issue "#995".
* Fix a bug in multidicts when passing empty lists as values, see
* Fix a security issue that allows XSS on the Werkzeug debugger. See
- update to version 0.11.10:
* Fixed a bug that occurs when running on Python 2.6 and using a
broken locale. See pull request #912.
* Fixed a crash when running the debugger on Google App Engine. See
* Fixed an issue with multipart parsing that could cause memory
- Update to 0.11.9
- Corrected an issue that caused the debugger not to use the
machine GUID on POSIX systems.
- Corrected an Unicode error on Python 3 for the debugger's
- Corrected the timestamp verification in the pin debug code.
Without this fix the pin was remebered until too long.
- update to version 0.11.8:
* fixed a problem with the machine GUID detection code on OS X on
- changes from version 0.11.7:
* fixed a regression on Python 3 for the debugger.
- changes from version 0.11.6:
* werkzeug.serving: Still show the client address on bad requests.
* improved the PIN based protection for the debugger to make it
harder to brute force via trying cookies. Please keep in mind
that the debugger *is not intended for running on production
* increased the pin timeout to a week to make it less annoying for
people which should decrease the change that users disable the pin
* werkzeug.serving: Fix broken HTTP_HOST when path starts with
- update to version 0.11.5:
* werkzeug.serving: Fix crash when attempting SSL connection to HTTP
- update to version 0.11.4:
* Fixed werkzeug.serving not working from -m flag.
* Fixed incorrect weak etag handling.
- Rebase 0001_create_a_thread_to_reap_death_process.patch
- Split documentation into own subpackage to speed up build.
* Mon Feb 08 2016 email@example.com
- Add 0001_create_a_thread_to_reap_death_process.patch
* Mon Feb 08 2016 firstname.lastname@example.org
- update to 0.11.3:
- Added reloader_paths option to run_simple and other functions in
werkzeug.serving. This allows the user to completely override the
Python module watching of Werkzeug with custom paths.
- Many custom cached properties of Werkzeug?s classes are now
subclasses of Python?s property type (issue #616).
- bind_to_environ now doesn?t differentiate between implicit and
explicit default port numbers in HTTP_HOST (pull request #204).
- BuildErrors are now more informative. They come with a complete
sentence as error message, and also provide suggestions (pull
- Fix a bug in the user agent parser where Safari?s build number
instead of version would be extracted (pull request #703).
- Fixed issue where RedisCache set_many was broken for twemproxy,
which doesn?t support the default MULTI command (pull request
- mimetype parameters on request and response classes are now always
converted to lowercase.
- Changed cache so that cache never expires if timeout is 0. This
also fixes an issue with redis setex (issue #550)
- Werkzeug now assumes UTF-8 as filesystem encoding on Unix if
Python detected it as ASCII.
- New optional has method on caches.
- Fixed various bugs in parse_options_header (pull request #643).
- If the reloader is enabled the server will now open the socket in
the parent process if this is possible. This means that when the
reloader kicks in the connection from client will wait instead of
tearing down. This does not work on all Python versions.
- Implemented PIN based authentication for the debugger. This can
optionally be disabled but is discouraged. This change was
necessary as it has been discovered that too many people run the
debugger in production.
- Devserver no longer requires SSL module to be installed.
- Reloader: Correctly detect file changes made by moving temporary
files over the original, which is e.g. the case with PyCharm (pull
- Fix bool behavior of werkzeug.datastructures.ETags under Python 3
* Thu May 10 2018 email@example.com
- Make sure ssl is available
- Avoid problem with bytecode being overwritten in tests
* Thu Mar 08 2018 firstname.lastname@example.org
- Allows Recommends and Suggest in Fedora
* Tue Feb 27 2018 email@example.com
- Recommends only for SUSE
* Wed Jan 03 2018 firstname.lastname@example.org
* update copyright year
- update to version 0.14.1:
* Resolved a regression with status code handling in the integrated
- changes from version 0.14:
* HTTP exceptions are now automatically caught by
* Added support for edge as browser.
* Added support for platforms that lack SpooledTemporaryFile.
* Add support for etag handling through if-match
* Added support for the SameSite cookie attribute.
* Added werkzeug.wsgi.ProxyMiddleware
* Implemented has for NullCache
* get_multi on cache clients now returns lists all the time.
* Improved the watchdog observer shutdown for the reloader to not
crash on exit on older Python versions.
* Added support for filename* filename attributes according to RFC
* Resolved an issue where machine ID for the reloader PIN was not
read accurately on windows.
* Added a workaround for syntax errors in init files in the
* Added support for using the reloader with console scripts on
* The built-in HTTP server will no longer close a connection in
cases where no HTTP body is expected (204, 204, HEAD requests
* The EnvironHeaders object now skips over empty content type and
lengths if they are set to falsy values.
* Werkzeug will no longer send the content-length header on 1xx or
* Cookie values are now also permitted to include slashes and equal
signs without quoting.
* Relaxed the regex for the routing converter arguments.
* If cookies are sent without values they are now assumed to have an
empty value and the parser accepts this. Previously this could
have corrupted cookies that followed the value.
* The test Client and EnvironBuilder now support mimetypes like the
request object does.
* Added support for static weights in URL rules.
* Better handle some more complex reloader scenarios where sys.path
contained non directory paths.
* EnvironHeaders no longer raises weird errors if non string keys
are passed to it.
* Fri Dec 08 2017 email@example.com
* added CHANGES.rst and README.rst to %doc section
- update to version 0.13:
* Deprecate support for Python 2.6 and 3.3. CI tests will not run
for these versions, and support will be dropped completely in the
next version. (pallets/meta#24)
* Raise TypeError when port is not an integer. (#1088)
* Fully deprecate werkzeug.script. Use Click instead. (#1090)
* response.age is parsed as a timedelta. Previously, it was
incorrectly treated as a datetime. The header value is an integer
number of seconds, not a date string. (#414)
* Fix a bug in TypeConversionDict where errors are not propagated
when using the converter. (#1102)
* Authorization.qop is a string instead of a set, to comply with RFC
* An exception is raised when an encoded cookie is larger than, by
default, 4093 bytes. Browsers may silently ignore cookies larger
than this. BaseResponse has a new attribute max_cookie_size and
dump_cookie has a new argument max_size to configure this. (#780,
* Fix a TypeError in
* BaseResponse.calculate_content_length now correctly works for
Unicode responses on Python 3. It first encodes using
* Secure cookie contrib works with string secret key on Python
* Shared data middleware accepts a list instead of a dict of static
locations to preserve lookup order. (#1197)
* HTTP header values without encoding can contain single
* The built-in dev server supports receiving requests with chunked
transfer encoding. (#1198)