Changelog for python312-Werkzeug-3.0.4-ku.2.6.noarch.rpm :

* Mon Aug 26 2024 John Paul Adrian Glaubitz - Update to 3.0.4
* Restore behavior where parsing `multipart/x-www-form-urlencoded` data with invalid UTF-8 bytes in the body results in no form data parsed rather than a 413 error. :issue:`2930`
* Improve ``parse_options_header`` performance when parsing unterminated quoted string values. :issue:`2904`
* Debugger pin auth is synchronized across threads/processes when tracking failed entries. :issue:`2916`
* Dev server handles unexpected `SSLEOFError` due to issue in Python < 3.13. :issue:`2926`
* Debugger pin auth works when the URL already contains a query string. :issue:`2918`
* Tue May 07 2024 Daniel Garcia - Update to 3.0.3:
* Only allow ``localhost``, ``.localhost``, ``127.0.0.1``, or the specified hostname when running the dev server, to make debugger requests. Additional hosts can be added by using the debugger middleware directly. The debugger UI makes requests using the full URL rather than only the path. :ghsa:`2g68-c3qc-8985` (CVE-2024-34069, bsc#1223979)
* Make reloader more robust when ``\"\"`` is in ``sys.path``. :pr:`2823`
* Better TLS cert format with ``adhoc`` dev certs. :pr:`2891`
* Inform Python < 3.12 how to handle ``itms-services`` URIs correctly, rather than using an overly-broad workaround in Werkzeug that caused some redirect URIs to be passed on without encoding. :issue:`2828`
* Type annotation for ``Rule.endpoint`` and other uses of ``endpoint`` is ``Any``. :issue:`2836`- Update to 3.0.2:
* Ensure setting ``merge_slashes`` to ``False`` results in ``NotFound`` for repeated-slash requests against single slash routes. :issue:`2834`
* Fix handling of ``TypeError`` in ``TypeConversionDict.get()`` to match ``ValueError``. :issue:`2843`
* Fix ``response_wrapper`` type check in test client. :issue:`2831`
* Make the return type of ``MultiPartParser.parse`` more precise. :issue:`2840`
* Raise an error if converter arguments cannot be parsed. :issue:`2822`
* Fri Oct 27 2023 Steve Kowalik - Update to 3.0.1:
* Fix slow multipart parsing for large parts potentially enabling DoS attacks. (CVE-2023-46136, bsc#1216581)
* Remove previously deprecated code.
* Deprecate the ``__version__`` attribute. Use feature detection, or ``importlib.metadata.version(\"werkzeug\")``, instead.
* ``generate_password_hash`` uses scrypt by default.
* Add the ``\"werkzeug.profiler\"`` item to the WSGI ``environ`` dictionary passed to `ProfilerMiddleware`\'s `filename_format` function. It contains the ``elapsed`` and ``time`` values for the profiled request.
* Explicitly marked the PathConverter as non path isolating.
* Mon Sep 25 2023 Steve Kowalik - Update to 2.3.7:
* Use ``flit_core`` instead of ``setuptools`` as build backend.
* Fix parsing of multipart bodies. Adjust index of last newline in data start.
* ``_plain_int`` and ``_plain_float`` strip whitespace before type enforcement.
* Fix empty file streaming when testing.
* Clearer error message when URL rule does not start with slash.
* ``Accept`` ``q`` value can be a float without a decimal part.- Drop captialisation again.
* Mon Jun 19 2023 Antonio Larrosa - Update to 2.3.6:
* FileStorage.content_length does not fail if the form data did not provide a value.- Update to 2.3.5:
* Python 3.12 compatibility.
* Fix handling of invalid base64 values in Authorization.from_header.
* The debugger escapes the exception message in the page title.
* When binding routing.Map, a long IDNA server_name with a port does not fail encoding.
* iri_to_uri shows a deprecation warning instead of an error when passing bytes.
* When parsing numbers in HTTP request headers such as Content-Length, only ASCII digits are accepted rather than any format that Python’s int and float accept.- Update to 2.3.4:
* Authorization.from_header and WWWAuthenticate.from_header detects tokens that end with base64 padding (=).
* Remove usage of warnings.catch_warnings.
* Remove max_form_parts restriction from standard form data parsing and only use if for multipart content.
* Response will avoid converting the Location header in some cases to preserve invalid URL schemes like itms-services.- Update to 2.3.3:
* Fix parsing of large multipart bodies. Remove invalid leading newline, and restore parsing speed.
* The cookie Path attribute is set to / by default again, to prevent clients from falling back to RFC 6265’s default-path behavior.- Update to 2.3.2:
* Parse the cookie Expires attribute correctly in the test client.
* max_content_length can only be enforced on streaming requests if the server sets wsgi.input_terminated.- Update to 2.3.1:
* Percent-encode plus (+) when building URLs and in test requests.
* Cookie values don’t quote characters defined in RFC 6265.
* Include pyi files for datastructures type annotations.
* Authorization and WWWAuthenticate objects can be compared for equality.- Update to 2.3.0:
* Drop support for Python 3.7.
* Remove previously deprecated code.
* Passing bytes where strings are expected is deprecated, as well as the charset and errors parameters in many places. Anywhere that was annotated, documented, or tested to accept bytes shows a warning. Removing this artifact of the transition from Python 2 to 3 removes a significant amount of overhead in instance checks and encoding cycles. In general, always work with UTF-8, the modern HTML, URL, and HTTP standards all strongly recommend this.
* Deprecate the werkzeug.urls module, except for the uri_to_iri and iri_to_uri functions. Use the urllib.parse library instead.
* Update which characters are considered safe when using percent encoding in URLs, based on the WhatWG URL Standard.
* Update which characters are considered safe when using percent encoding for Unicode filenames in downloads.
* Deprecate the safe_conversion parameter of iri_to_uri. The Location header is converted to IRI using the same process as everywhere else.
* Deprecate werkzeug.wsgi.make_line_iter and make_chunk_iter.
* Use modern packaging metadata with pyproject.toml instead of setup.cfg.
* Request.get_json() will raise a 415 Unsupported Media Type error if the Content-Type header is not application/json, instead of a generic 400.
* A URL converter’s part_isolating defaults to False if its regex contains a /.
* A custom converter’s regex can have capturing groups without breaking the router.
* The reloader can pick up arguments to python like -X dev, and does not require heuristics to determine how to reload the command. Only available on Python >= 3.10.
* The Watchdog reloader ignores file opened events. Bump the minimum version of Watchdog to 2.3.0.
* When using a Unix socket for the development server, the path can start with a dot.
* Increase default work factor for PBKDF2 to 600,000 iterations.
* parse_options_header is 2-3 times faster. It conforms to RFC 9110, some invalid parts that were previously accepted are now ignored.
* The is_filename parameter to unquote_header_value is deprecated.
* Deprecate the extra_chars parameter and passing bytes to quote_header_value, the allow_token parameter to dump_header, and the cls parameter and passing bytes to parse_dict_header.
* Improve parse_accept_header implementation. Parse according to RFC 9110. Discard items with invalid q values.
* quote_header_value quotes the empty string.
* dump_options_header skips None values rather than using a bare key.
* dump_header and dump_options_header will not quote a value if the key ends with an asterisk
*.
* parse_dict_header will decode values with charsets.
* Refactor the Authorization and WWWAuthenticate header data structures. + Both classes have type, parameters, and token attributes. The token attribute supports auth schemes that use a single opaque token rather than key=value parameters, such as Bearer. + Neither class is a dict anymore, although they still implement getting, setting, and deleting auth[key] and auth.key syntax, as well as auth.get(key) and key in auth. + Both classes have a from_header class method. parse_authorization_header and parse_www_authenticate_header are deprecated. + The methods WWWAuthenticate.set_basic and set_digest are deprecated. Instead, an instance should be created and assigned to response.www_authenticate. + A list of instances can be assigned to response.www_authenticate to set multiple header values. However, accessing the property only returns the first instance.
* Refactor parse_cookie and dump_cookie. + parse_cookie is up to 40% faster, dump_cookie is up to 60% faster. + Passing bytes to parse_cookie and dump_cookie is deprecated. The dump_cookie charset parameter is deprecated. + dump_cookie allows domain values that do not include a dot ., and strips off a leading dot. + dump_cookie does not set path=\"/\" unnecessarily by default.
* Refactor the test client cookie implementation. + The cookie_jar attribute is deprecated. http.cookiejar is no longer used for storage. + Domain and path matching is used when sending cookies in requests. The domain and path parameters default to localhost and /. + Added a get_cookie method to inspect cookies. + Cookies have decoded_key and decoded_value attributes to match what the app sees rather than the encoded values a client would see. + The first positional server_name parameter to set_cookie and delete_cookie is deprecated. Use the domain parameter instead. + Other parameters to delete_cookie besides domain, path, and value are deprecated.
* If request.max_content_length is set, it is checked immediately when accessing the stream, and while reading from the stream in general, rather than only during form parsing.
* The development server, which must not be used in production, will exhaust the request stream up to 10GB or 1000 reads. This allows clients to see a 413 error if max_content_length is exceeded, instead of a “connection reset” failure.
* The development server discards header keys that contain underscores _, as they are ambiguous with dashes - in WSGI.
* secure_filename looks for more Windows reserved file names.
* Update type annotation for best_match to make default parameter clearer.
* Multipart parser handles empty fields correctly.
* The Map charset parameter and Request.url_charset property are deprecated. Percent encoding in URLs must always represent UTF-8 bytes. Invalid bytes are left percent encoded rather than replaced.
* The Request.charset, Request.encoding_errors, Response.charset, and Client.charset attributes are deprecated. Request and response data must always use UTF-8.
* Header values that have charset information only allow ASCII, UTF-8, and ISO-8859-1.
* Update type annotation for ProfilerMiddleware stream parameter.
* Use postponed evaluation of annotations.
* The development server escapes ASCII control characters in decoded URLs before logging the request to the terminal.
* The FormDataParser parse_functions attribute and get_parse_func method, and the invalid application/x-url-encoded content type, are deprecated.
* generate_password_hash supports scrypt. Plain hash methods are deprecated, only scrypt and pbkdf2 are supported.- Remove patch which was made obsolete by upstream:
* moved_root.patch
* Fri Apr 21 2023 Dirk Müller - add sle15_python_module_pythons (jsc#PED-68)
* Thu Apr 13 2023 Matej Cepl - Make calling of %{sle15modernpython} optional.
* Mon Mar 13 2023 Dirk Müller - update to 2.2.3 (bsc#1208283, CVE-2023-25577):
* drops 0001-limit-the-maximum-number-of-multipart-form-parts.patch in older dists
* Ensure that URL rules using path converters will redirect with strict slashes when the trailing slash is missing.
* Type signature for ``get_json`` specifies that return type is not optional when ``silent=False``.
* ``parse_content_range_header`` returns ``None`` for a value like ``bytes
*/-1`` where the length is invalid, instead of raising an ``AssertionError``.
* Address remaining ``ResourceWarning`` related to the socket used by ``run_simple``.
* Remove ``prepare_socket``, which now happens when creating the server.
* Update pre-existing headers for ``multipart/form-data`` requests with the test client.
* Fix handling of header extended parameters such that they are no longer quoted.
* ``LimitedStream.read`` works correctly when wrapping a stream that may not return the requested size in one ``read`` call.
* A cookie header that starts with ``=`` is treated as an empty key and discarded, rather than stripping the leading ``==``.
* Specify a maximum number of multipart parts, default 1000, after which a ``RequestEntityTooLarge`` exception is raised on parsing. This mitigates a DoS attack where a larger number of form/file parts would result in disproportionate resource use.
* Tue Sep 13 2022 Ben Greiner - Clean some unused python2 python36 code from specfile- Move MarkupSafe to runtime requirement. Versioned. This is checked in multibuild test flavor as build requirement.
* Mon Sep 12 2022 Yogalakshmi Arunachalam - test failed due to markupsafe module missing Included markupsafe module
* Fri Sep 09 2022 Yogalakshmi Arunachalam - Update to 2.2.2:
* Fix router to restore the 2.1 strict_slashes == False behaviour whereby leaf-requests match branch rules and vice versa. #2489
* Fix router to identify invalid rules rather than hang parsing them, and to correctly parse / within converter arguments. #2489
* Update subpackage imports in werkzeug.routing to use the import as syntax for explicitly re-exporting public attributes. #2493
* Parsing of some invalid header characters is more robust. #2494
* When starting the development server, a warning not to use it in a production deployment is always shown. #2480
* LocalProxy.__wrapped__ is always set to the wrapped object when the proxy is unbound, fixing an issue in doctest that would cause it to fail. #2485
* Address one ResourceWarning related to the socket used by run_simple. #2421- Update to Version 2.2.1:
* Fix router so that /path/ will match a rule /path if strict slashes mode is disabled for the rule. #2467
* Fix router so that partial part matches are not allowed i.e. /2df does not match /. #2470
* Fix router static part weighting, so that simpler routes are matched before more complex ones. #2471
* Restore ValidationError to be importable from werkzeug.routing. #2465- Update to Version 2.2.0
* Deprecated get_script_name, get_query_string, peek_path_info, pop_path_info, and extract_path_info. #2461
* Remove previously deprecated code. #2461
* Add MarkupSafe as a dependency and use it to escape values when rendering HTML. #2419
* Added the werkzeug.debug.preserve_context mechanism for restoring context-local data for a request when running code in the debug console. #2439
* Fix compatibility with Python 3.11 by ensuring that end_lineno and end_col_offset are present on AST nodes. #2425
* Add a new faster matching router based on a state machine. #2433
* Fix branch leaf path masking branch paths when strict-slashes is disabled. #1074
* Names within options headers are always converted to lowercase. This matches RFC 6266 that the case is not relevant. #2442
* AnyConverter validates the value passed for it when building URLs. #2388
* The debugger shows enhanced error locations in tracebacks in Python 3.11. #2407
* Added Sans-IO is_resource_modified and parse_cookie functions based on WSGI versions. #2408
* Added Sans-IO get_content_length function. #2415
* Don’t assume a mimetype for test responses. #2450
* Type checking FileStorage accepts os.PathLike. #2418
* Fri Jul 29 2022 Torsten Gruner - enable multibuild for test
* Wed May 11 2022 Dirk Müller - update to 2.1.2:
* The development server does not set ``Transfer-Encoding: chunked`` for 1xx, 204, 304, and HEAD responses. :issue:`2375`
* Response HTML for exceptions and redirects starts with ```` and ````. :issue:`2390`
* Fix ability to set some ``cache_control`` attributes to ``False``. :issue:`2379`
* Disable ``keep-alive`` connections in the development server, which are not supported sufficiently by Python\'s ``http.server``. :issue:`2397`- drop 2402-dev_server.patch (upstream)
* Thu Apr 28 2022 Matej Cepl - Replace no-network-testing.patch with the upstream solution 2402-dev_server.patch from gh#pallets/werkzeug#2402.- Add moved_root.patch to make test test_exclude_patterns with different PYTHONPATH.
* Tue Apr 19 2022 Matej Cepl - Update to 2.1.1: - ResponseCacheControl.s_maxage converts its value to an int, like max_age. - Drop support for Python 3.6. - Using gevent or eventlet requires greenlet>=1.0 or PyPy>=7.3.7. werkzeug.locals and contextvars will not work correctly with older versions. - Remove previously deprecated code. - Remove the non-standard shutdown function from the WSGI environ when running the development server. See the docs for alternatives. - Request and response mixins have all been merged into the Request and Response classes. - The user agent parser and the useragents module is removed. The user_agent module provides an interface that can be subclassed to add a parser, such as ua-parser. By default it only stores the whole string. - The test client returns TestResponse instances and can no longer be treated as a tuple. All data is available as properties on the response. - Remove locals.get_ident and related thread-local code from locals, it no longer makes sense when moving to a contextvars-based implementation. - Remove the python -m werkzeug.serving CLI. - The has_key method on some mapping datastructures; use key in data instead. - Request.disable_data_descriptor is removed, pass shallow=True instead. - Remove the no_etag parameter from Response.freeze(). - Remove the HTTPException.wrap class method. - Remove the cookie_date function. Use http_date instead. - Remove the pbkdf2_hex, pbkdf2_bin, and safe_str_cmp functions. Use equivalents in hashlib and hmac modules instead. - Remove the Href class. - Remove the HTMLBuilder class. - Remove the invalidate_cached_property function. Use del obj.attr instead. - Remove bind_arguments and validate_arguments. Use Signature.bind() and inspect.signature() instead. - Remove detect_utf_encoding, it’s built-in to json.loads. - Remove format_string, use string.Template instead. - Remove escape and unescape. Use MarkupSafe instead. - The multiple parameter of parse_options_header is deprecated. - Rely on PEP 538 and PEP 540 to handle decoding file names with the correct filesystem encoding. The filesystem module is removed. - Default values passed to Headers are validated the same way values added later are. - Setting CacheControl int properties, such as max_age, will convert the value to an int. - Always use socket.fromfd when restarting the dev server. - When passing a dict of URL values to Map.build, list values do not filter out None or collapse to a single value. Passing a MultiDict does collapse single items. This undoes a previous change that made it difficult to pass a list, or None values in a list, to custom URL converters. - run_simple shows instructions for dealing with “address already in use” errors, including extra instructions for macOS. - Extend list of characters considered always safe in URLs based on RFC 3986. - Optimize the stat reloader to avoid watching unnecessary files in more cases. The watchdog reloader is still recommended for performance and accuracy. - The development server uses Transfer-Encoding: chunked for streaming responses when it is configured for HTTP/1.1. - The development server uses HTTP/1.1, which enables keep-alive connections and chunked streaming responses, when threaded or processes is enabled. - cached_property works for classes with __slots__ if a corresponding _cache_{name} slot is added. - Refactor the debugger traceback formatter to use Python’s built-in traceback module as much as possible. - The TestResponse.text property is a shortcut for r.get_data(as_text=True), for convenient testing against text instead of bytes. - safe_join ensures that the path remains relative if the trusted directory is the empty string. - Percent-encoded newlines (%0a), which are decoded by WSGI servers, are considered when routing instead of terminating the match early. - The test client doesn’t set duplicate headers for CONTENT_LENGTH and CONTENT_TYPE. - append_slash_redirect handles PATH_INFO with internal slashes. - The default status code for append_slash_redirect is 308 instead of 301. This preserves the request body, and matches a previous change to strict_slashes in routing. - Fix ValueError: I/O operation on closed file. with the test client when following more than one redirect. - Response.autocorrect_location_header is disabled by default. The Location header URL will remain relative, and exclude the scheme and domain, by default. - Request.get_json() will raise a 400 BadRequest error if the Content-Type header is not application/json. This makes a very common source of confusion more visible.- Add no-network-testing.patch to mark all tests requiring network access (so they can be skipped by pytest test runner, gh#pallets/werkzeug#2393).
* Tue Feb 15 2022 Dirk Müller - update to 2.0.3:
* ``ProxyFix`` supports IPv6 addresses.
* Type annotation for ``Response.make_conditional``, ``HTTPException.get_response``, and ``Map.bind_to_environ`` accepts ``Request`` in addition to ``WSGIEnvironment`` for the first parameter.
* Fix type annotation for ``Request.user_agent_class``.
* Accessing ``LocalProxy.__class__`` and ``__doc__`` on an unbound proxy returns the fallback value instead of a method object.
* Redirects with the test client set ``RAW_URI`` and ``REQUEST_URI`` correctly.
* Sat Oct 16 2021 Dirk Müller - update to 2.0.2:
* Handle multiple tokens in ``Connection`` header when routing WebSocket requests.
* Set the debugger pin cookie secure flag when on https.
* Fix type annotation for ``MultiDict.update`` to accept iterable values :pr:`2142`
* Prevent double encoding of redirect URL when ``merge_slash=True`` for ``Rule.match``.
* ``CombinedMultiDict.to_dict`` with ``flat=False`` considers all component dicts when building value lists. :issue:`2189`
* ``send_file`` only sets a detected ``Content-Encoding`` if ``as_attachment`` is disabled to avoid browsers saving decompressed ``.tar.gz`` files.
* Fix type annotations for ``TypeConversionDict.get`` to not return an ``Optional`` value if both ``default`` and ``type`` are not ``None``.
* Fix type annotation for routing rule factories to accept ``Iterable[RuleFactory]`` instead of ``Iterable[Rule]`` for the ``rules`` parameter. :issue:`2183`
* Add missing type annotation for ``FileStorage.__getattr__``
* The debugger pin cookie is set with ``SameSite`` set to ``Strict`` instead of ``None`` to be compatible with modern browser security.
* Type annotations use ``IO[bytes]`` and ``IO[str]`` instead of ``BinaryIO`` and ``TextIO`` for wider type compatibility.
* Ad-hoc TLS certs are generated with SAN matching CN. :issue:`2158`
* Fix memory usage for locals when using Python 3.6 or pre 0.4.17 greenlet versions. :pr:`2212`
* Fix type annotation in ``CallbackDict``, because it is not utilizing a bound TypeVar. :issue:`2235`
* Fix setting CSP header options on the response. :pr:`2237`
* Fix an issue with with the interactive debugger where lines would not expand on click for very long tracebacks. :pr:`2239`
* The interactive debugger handles displaying an exception that does not have a traceback, such as from ``ProcessPoolExecutor``.
* Sat Jun 19 2021 Michael Ströder - skip building for Python 2.x- updated upstream project URL- Update to 2.0.1
* Version 2.0.1 - Fix type annotation for send_file max_age callable. Don’t pass pathlib.Path to max_age. #2119 - Mark top-level names as exported so type checking understands imports in user projects. #2122 - Fix some types that weren’t available in Python 3.6.0. #2123 - cached_property is generic over its return type, properties decorated with it report the correct type. #2113 - Fix multipart parsing bug when boundary contains special regex characters. #2125 - Type checking understands that calling headers.get with a string default will always return a string. #2128 - If HTTPException.description is not a string, get_description will convert it to a string. #2115
* Version 2.0.0 - Drop support for Python 2 and 3.5. #1693 - Deprecate utils.format_string(), use string.Template instead. #1756 - Deprecate utils.bind_arguments() and utils.validate_arguments(), use Signature.bind() and inspect.signature() instead. #1757 - Deprecate utils.HTMLBuilder. #1761 - Deprecate utils.escape() and utils.unescape(), use MarkupSafe instead. #1758 - Deprecate the undocumented python -m werkzeug.serving CLI. #1834 - Deprecate the environ[\"werkzeug.server.shutdown\"] function that is available when running the development server. #1752 - Deprecate the useragents module and the built-in user agent parser. Use a dedicated parser library instead by subclassing user_agent.UserAgent and setting Request.user_agent_class. #2078 - Remove the unused, internal posixemulation module. #1759 - All datetime values are timezone-aware with tzinfo=timezone.utc. This applies to anything using http.parse_date: Request.date, .if_modified_since, .if_unmodified_since; Response.date, .expires, .last_modified, .retry_after; parse_if_range_header, and IfRange.date. When comparing values, the other values must also be aware, or these values must be made naive. When passing parameters or setting attributes, naive values are still assumed to be in UTC. #2040 - Merge all request and response wrapper mixin code into single Request and Response classes. Using the mixin classes is no longer necessary and will show a deprecation warning. Checking isinstance or issubclass against BaseRequest and BaseResponse will show a deprecation warning and check against Request or Response instead. #1963 - JSON support no longer uses simplejson if it’s installed. To use another JSON module, override Request.json_module and Response.json_module. #1766 - Response.get_json() no longer caches the result, and the cache parameter is removed. #1698 - Response.freeze() generates an ETag header if one is not set. The no_etag parameter (which usually wasn’t visible anyway) is no longer used. #1963 - Add a url_scheme argument to build() to override the bound scheme. #1721 - Passing an empty list as a query string parameter to build() won’t append an unnecessary ?. Also drop any number of None items in a list. [#1992] - When passing a Headers object to a test client method or EnvironBuilder, multiple values for a key are joined into one comma separated value. This matches the HTTP spec on multi-value headers. [#1655] - Setting Response.status and status_code uses identical parsing and error checking. #1658, #1728 - MethodNotAllowed and RequestedRangeNotSatisfiable take a response kwarg, consistent with other HTTP errors. #1748 - The response generated by Unauthorized produces one WWW-Authenticate header per value in www_authenticate, rather than joining them into a single value, to improve interoperability with browsers and other clients. #1755 - If parse_authorization_header can’t decode the header value, it returns None instead of raising a UnicodeDecodeError. #1816 - The debugger no longer uses jQuery. #1807 - The test client includes the query string in REQUEST_URI and RAW_URI. #1781 - Switch the parameter order of default_stream_factory to match the order used when calling it. #1085 - Add send_file function to generate a response that serves a file. Adapted from Flask’s implementation. #265, #1850 - Add send_from_directory function to safely serve an untrusted path within a trusted directory. Adapted from Flask’s implementation. #1880 - send_file takes download_name, which is passed even if as_attachment=False by using Content-Disposition: inline. download_name replaces Flask’s attachment_filename. #1869 - send_file sets conditional=True and max_age=None by default. Cache-Control is set to no-cache if max_age is not set, otherwise public. This tells browsers to validate conditional requests instead of using a timed cache. max_age=None replaces Flask’s cache_timeout=43200. [#1882] - send_file can be called with etag=\"string\" to set a custom ETag instead of generating one. etag replaces Flask’s add_etags. #1868 - send_file sets the Content-Encoding header if an encoding is returned when guessing mimetype from download_name. #3896 - Update the defaults used by generate_password_hash. Increase PBKDF2 iterations to 260000 from 150000. Increase salt length to 16 from 8. Use secrets module to generate salt. #1935 - The reloader doesn’t crash if sys.stdin is somehow None. #1915 - Add arguments to delete_cookie to match set_cookie and the attributes modern browsers expect. #1889 - utils.cookie_date is deprecated, use utils.http_date instead. The value for Set-Cookie expires is no longer “-” delimited. #2040 - Use request.headers instead of request.environ to look up header attributes. #1808 - The test Client request methods (client.get, etc.) always return an instance of TestResponse. In addition to the normal behavior of Response, this class provides request with the request that produced the response, and history to track intermediate responses when follow_redirects is used. #763, #1894 - The test Client request methods takes an auth parameter to add an Authorization header. It can be an Authorization object or a (username, password) tuple for Basic auth. #1809 - Calling response.close() on a response from the test Client will close the request input stream. This matches file behavior and can prevent a ResourceWarning in some cases. #1785 - EnvironBuilder.from_environ decodes values encoded for WSGI, to avoid double encoding the new values. #1959 - The default stat reloader will watch Python files under non-system/virtualenv sys.path entries, which should contain most user code. It will also watch all Python files under directories given in extra_files. #1945 - The reloader ignores __pycache__ directories again. #1945 - run_simple takes exclude_patterns a list of fnmatch patterns that will not be scanned by the reloader. #1333 - Cookie names are no longer unquoted. This was against RFC 6265 and potentially allowed setting __Secure prefixed cookies. #1965 - Fix some word matches for user agent platform when the word can be a substring. #1923 - The development server logs ignored SSL errors. #1967 - Temporary files for form data are opened in rb+ instead of wb+ mode for better compatibility with some libraries. #1961 - Use SHA-1 instead of MD5 for generating ETags and the debugger pin, and in some tests. MD5 is not available in some environments, such as FIPS 140. This may invalidate some caches since the ETag will be different. [#1897] - Add Cross-Origin-Opener-Policy and Cross-Origin-Embedder-Policy response header properties. #2008 - run_simple tries to show a valid IP address when binding to all addresses, instead of 0.0.0.0 or ::. It also warns about not running the development server in production in this case. #1964 - Colors in the development server log are displayed if Colorama is installed on Windows. For all platforms, style support no longer requires Click. #1832 - A range request for an empty file (or other data with length 0) will return a 200 response with the empty file instead of a 416 error. #1937 - New sans-IO base classes for Request and Response have been extracted to contain all the behavior that is not WSGI or IO dependent. These are not a public API, they are part of an ongoing refactor to let ASGI frameworks use Werkzeug. #2005 - Parsing multipart/form-data has been refactored to use sans-io patterns. This should also make parsing forms with large binary file uploads significantly faster. #1788, #875 - LocalProxy matches the current Python data model special methods, including all r-ops, in-place ops, and async. __class__ is proxied, so the proxy will look like the object in more cases, including isinstance. Use issubclass(type(obj), LocalProxy) to check if an object is actually a proxy. #1754 - Local uses ContextVar on Python 3.7+ instead of threading.local. #1778 - request.values does not include form for GET requests (even though GET bodies are undefined). This prevents bad caching proxies from caching form data instead of query strings. #2037 - The development server adds the underlying socket to environ as werkzeug.socket. This is non-standard and specific to the dev server, other servers may expose this under their own key. It is useful for handling a WebSocket upgrade request. #2052 - URL matching assumes websocket=True mode for WebSocket upgrade requests. #2052 - Updated UserAgentParser to handle more cases. #1971 - werzeug.DechunkedInput.readinto will not read beyond the size of the buffer. #2021 - Fix connection reset when exceeding max content size. #2051 - pbkdf2_hex, pbkdf2_bin, and safe_str_cmp are deprecated. hashlib and hmac provide equivalents. #2083 - invalidate_cached_property is deprecated. Use del obj.name instead. #2084 - Href is deprecated. Use werkzeug.routing instead. #2085 - Request.disable_data_descriptor is deprecated. Create the request with shallow=True instead. #2085 - HTTPException.wrap is deprecated. Create a subclass manually instead. #2085
* Sun Jun 13 2021 Michael Ströder - skip building for Python 2.x
* Tue Jan 12 2021 Markéta Machová - Workaround pytest 6.2