Upgrade to release 0.10.2:
- Added toml.dump example to Quick Tutorial
- Skip numpy tests when numpy is not available
- Bug fix: ids were based on the string, rather than the actual
object which causes incorrect detection of circular references
- Show meaningful error message for wrong case booleans
- Bug fix: broken quoted section name starts with dot
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 35.5.0:
- Added support for CAN FD in monitor.py cli tool
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 0.3.36:
- Updated docs section regarding max_years_between_matches to
be more shorter and hopefully more relevant
- Don't install tests
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 5.1.0:
- Support transactions in FanoutCache (probably a bad idea)
- Prevent cache shard attribute access when unsafe
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 20.3.0:
- attr.define(), attr.frozen(), attr.mutable(), and attr.field()
remain provisional.
- attr.s() now has a field_transformer hook that is called for all
Attributes and returns a (modified or updated) list of Attribute
instances. attr.asdict() has a value_serializer hook that can
change the way values are converted. Both hooks are meant to
help with data (de-)serialization workflows.
- kw_only=True now works on Python 2.
- raise from now works on frozen classes on PyPy.
- attr.asdict() and attr.astuple() now treat frozensets like sets
with regards to the retain_collection_types argument.
- The type stubs for attr.s() and attr.make_class() are not
missing the collect_by_mro argument anymore.
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 1.1.4:
- Fixed regression in read_csv() raising a ValueError when names
was of type dict_keys
- Fixed regression in read_csv() with more than 1M rows and
specifying a index_col argument
- Fixed regression where attempting to mutate a DateOffset object
would no longer raise an AttributeError
- Fixed regression where DataFrame.agg() would fail with TypeError
when passed positional arguments to be passed on to the
aggregation function
- Fixed regression in RollingGroupby with sort=False not being
respected
- Fixed regression in Series.astype() converting None to "nan"
when casting to string
- Fixed regression in Series.rank() method failing for read-only
data
- Fixed regression in RollingGroupby causing a segmentation fault
with Index of dtype object
- Fixed regression in DataFrame.resample(...).apply(...)() raised
AttributeError when input was a DataFrame and only a Series was
evaluated
- Fixed regression in DataFrame.groupby(..).std() with nullable
integer dtype
- Fixed regression in PeriodDtype comparing both equal and unequal
to its string representation
- Fixed regression where slicing DatetimeIndex raised AssertionError
on irregular time series with pd.NaT or on unsorted indices
- Fixed regression in certain offsets (pd.offsets.Day() and below)
no longer being hashable
- Fixed regression in StataReader which required chunksize to be
manually set when using an iterator to read a dataset
- Fixed regression in setitem with DataFrame.iloc() which raised
error when trying to set a value while filtering with a boolean
list
- Fixed regression in setitem with a Series getting aligned before
setting the values
- Fixed regression in MultiIndex.is_monotonic_increasing returning
wrong results with NaN in at least one of the levels
- Fixed regression in inplace arithmetic operation on a Series not
updating the parent DataFrame
- Fixed bug causing groupby(...).sum() and similar to not preserve
metadata
- Fixed bug in Series.isin() and DataFrame.isin() raising a
ValueError when the target was read-only
- Fixed bug in GroupBy.fillna() that introduced a performance
regression after 1.0.5
- Fixed bug in DataFrame.info() was raising a KeyError when the
DataFrame has integer column names
- Fixed bug in DataFrameGroupby.apply() would drop a
CategoricalIndex when grouped on
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 4.6.2:
- Include tests using MANIFEST.in instead of setup.py
- Switch to GitHub Actions
- Don't publish to test pypi - we'd have to bump the version
every commit
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 1.4.0:
- Fixed tab completion crash on Windows
- Changed how multiline doc string help is formatted to match
style of other help messages
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 7.19.0:
- Fix to restore the ability to specify more than one extension
using command line flags when using traitlets 5.0
- Docs docs formatting that make the install commands work on zsh
- Always display the last frame in tracebacks even if hidden with
__traceback_hide__
- Avoid an issue where a callback can be registered multiple times.
- Avoid an issue in debugger mode where frames changes could be
lost.
- Never hide the frames that invoke a debugger, even if marked as
hidden by __traceback_hide__
- Fix calling the debugger in a recursive manner
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 6.1:
- Windows support has been improved. Tornado is now compatible
with the proactor event loop (which became the default in Python
3.8) by automatically falling back to running a selector in a
second thread. This means that it is no longer necessary to
explicitly configure a selector event loop, although doing so
may improve performance. This does not change the fact that
Tornado is significantly less scalable on Windows than on other
platforms.
- Binary wheels are now provided for Windows, MacOS, and Linux
(amd64 and arm64).
- This is the last release of Tornado to support Python 3.5.
Future versions will require Python 3.6 or newer
License-Update: Checksum change, lisense remains "Apache-2.0"
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 1.3.1:
- allow to avoid linking against VC2014_1 on windows
- do not mark move constructor / assignment operator of expression
as noexcept. This is to circumvent a suspected bug in the GCC
compiler in the manylinux1 image.
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 1.0.0:
- Drop support for Python 2.7 and pypy
- Now DateDataParser.get_date_data() returns a DateData object
instead of a dict
- From now wrong settings are not silenced and raise
SettingValidationError
- Now dateparser.parse() is deterministic and doesn't try previous
locales. Also, DateDataParser.get_date_data() doesn't try the
previous locales by default
- Remove the 'base-formats' parser
- Extract the 'no-spaces-time' parser from the 'absolute-time'
parser and make it an optional parser
- Remove numeral_translation_data
- Remove the undocumented SKIP_TOKENS_PARSER and FUZZY settings
- Remove support for using strings in date_formats
- The undocumented ExactLanguageSearch class has been moved to the
private scope and some internal methods have changed
- Changes in dateparser.utils: normalize_unicode() doesn't accept
bytes as input and convert_to_unicode has been deprecated
- Add Python 3.9 support
- Detect hours separated with a period/dot
- Add support for "decade"
- Add support for the hijri calendar in Python ≥ 3.6
- New logo!
- Improve the README and docs
- Fix the "calendars" extra
- Fix leap years when PREFER_DATES_FROM is set
- Fix STRICT_PARSING setting in no-spaces-time parser
- Consider RETURN_AS_TIME_PERIOD setting for relative-time parser
- Parse the 24hr time format with meridian info
- Other small improvements
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 2020.10.28:
- Added Apache 2.0 licence file to the files in the git repository
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 0.6.0:
- aiofiles is now tested on ppc64le.
- Added name and mode properties to async file objects.
- Fixed a DeprecationWarning internally.
- Python 3.9 support and tests.
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 1.3.0:
- add c++ benchmarks and run them on CIs
- modernize the c++ code by using more c++11 features
- introduce move semantic in some c++ constructors to improve
performances
- add support for Python 3.9
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 1.14.1:
- bugfix where setting _ok_code to not include 0, but 0 was the
exit code
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 3.3.3:
- Unify all block-level tags
- Fix issue where some empty elements would have text rendered
as None when using md_in_html
- Avoid catastrophic backtracking in hr regex
- Fix hr HTML handling
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 3.2:
- SECURITY ISSUE: Attempted to make RSA PKCS#1v1.5 decryption more
constant time, to protect against Bleichenbacher vulnerabilities.
Due to limitations imposed by our API, we cannot completely
mitigate this vulnerability and a future release will contain a
new API which is designed to be resilient to these for contexts
where it is required. Credit to Hubert Kario for reporting the
issue. CVE-2020-25659
- Support for OpenSSL 1.0.2 has been removed. Users on older
version of OpenSSL will need to upgrade.
- Added basic support for PKCS7 signing (including SMIME) via
:class:`~cryptography.hazmat.primitives.serialization.pkcs7.PKCS7SignatureBuilder`.
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 3.7.0:
- Drop support for Python 3.5, only 3.6 or newer is supported now
- Add support for SSD1351 128x96 display
- Pin luma.core to 1.x.y line only, in anticipation of performance
improvements in upcoming major release
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 3.7.1:
- Fixed a type error caused by the conditional import of Protocol
- Server doesn't send Content-Length for 1xx or 204
- Fix run_app typing
- Always require typing_extensions library
- Fix a variable-shadowing bug causing ThreadedResolver.resolve
to return the resolved IP as the hostname in each record, which
prevented validation of HTTPS connections
- Added annotations to all public attributes
- Fix flaky test_when_timeout_smaller_second
- Ensure sending a zero byte file does not throw an exception
- Fix a bug in web.run_app() about Python version checking on
Windows
License-Update: Update year
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 3.4.0:
- Path.joinpath now takes arbitrary positional arguments and no
longer accepts add as a keyword argument.
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 1.17.3:
- Drop support for Python 3.5, only 3.6 or newer is supported now
- Add missing cmdline interfaces: "noop" & "gpio_cs_spi"
- Legacy proportional font wrapper raises user-friendly error
message when character is not in the font table
License-Update: Update year
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 2.7.1:
- Add Ext type value validation to Ext class and
ext_serializable() decorator
- Change string formatting from % to .format() throughout codebase
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 5.7.3:
- [FreeBSD] add support for Process.rlimit().
- [BSD] add support for Process.environ()
- [UNIX] net_if_stats()'s isup also checks whether the NIC is
running (meaning Wi-Fi or ethernet cable is connected)
- [Linux] improved battery detection and charge "secsleft"
calculation
- [Linux] physical cpu_count() result is incorrect on systems with
more than one CPU socket
- [macOS] Process.exe() may raise FileNotFoundError if process is
still alive but the exe file which launched it got deleted
- [macOS] fix missing include for getpagesize()
- [Windows] Process.open_files() may cause a segfault due to a
NULL pointer
- [Linux] sensors_battery(): if percent can be determined but not
the remaining values, still return a result instead of None.
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 3.2:
- SECURITY ISSUE: Attempted to make RSA PKCS#1v1.5 decryption more
constant time, to protect against Bleichenbacher vulnerabilities.
Due to limitations imposed by our API, we cannot completely
mitigate this vulnerability and a future release will contain a
new API which is designed to be resilient to these for contexts
where it is required. Credit to Hubert Kario for reporting the
issue. CVE-2020-25659
- Support for OpenSSL 1.0.2 has been removed. Users on older
version of OpenSSL will need to upgrade.
- Added basic support for PKCS7 signing (including SMIME) via
:class:`~cryptography.hazmat.primitives.serialization.pkcs7.PKCS7SignatureBuilder`.
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Upgrade to release 4.6.1:
- lxml.html.InputGetter supports __len__() to count the number of
input fields.
- lxml.html.InputGetter has a new .items() method to ease
processing all input fields.
- lxml.html.InputGetter.keys() now returns the field names in
document order.
- GH-309: The API documentation is now generated using
sphinx-apidoc.
- C14N 2.0 serialisation failed for unprefixed attributes when a
default namespace was defined.
- TreeBuilder.close() raised AssertionError in some error cases
where it should have raised XMLSyntaxError. It now raises a
combined exception to keep up backwards compatibility, while
switching to XMLSyntaxError as an interface.
- A vulnerability was discovered in the HTML Cleaner by Yaniv
Nizry, which allowed JavaScript to pass through. The cleaner
now removes more sneaky "style" content.
Signed-off-by: Leon Anavi <leon.anavi@konsulko.com>
Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
