| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Pylint 2.0:
* trailing-comma-tuple can be emitted for return statements as well.
* Fix a false positive inconsistent-return-statements message when exception is raised
inside an else statement.
* ImportFrom nodes correctly use the full name for the import sorting checks.
* [].extend and similar builtin operations don't emit dict-*-not-iterating with the Python 3 porting checker
* Add a check consider-using-dict-comprehension which is emitted if for dict initialization
the old style with list comprehensions is used.
* Add a check consider-using-set-comprehension which is emitted if for set initialization
the old style with list comprehensions is used.
* logging-not-lazy is emitted whenever pylint infers that a string is built with addition
* Add a check chained-comparison which is emitted if a boolean operation can be simplified
by chaining some of its operations.
e.g "a < b and b < c", can be simplified as "a < b < c".
* Add a check consider-using-in for comparisons of a variable against
multiple values with "==" and "or"s instead of checking if the variable
is contained "in" a tuple of those values.
* in is considered iterating context for some of the Python 3 porting checkers
* Add --ignore-none flag to control if pylint should warn about no-member where the owner is None
* Fix a false positive related to too-many-arguments and bounded __get__ methods
* mcs as the first parameter of metaclass's __new__ method was replaced by cls
* assignment-from-no-return considers methods as well.
* Support typing.TYPE_CHECKING for *unused-import* errors
* Inferred classes at a function level no longer emit invalid-name
when they don't respect the variable regular expression
* Added basic support for postponed evaluation of function annotations.
* Fix a bug with missing-kwoa and variadics parameters
* simplifiable-if-statement takes in account only when assigning to same targets
* Make len-as-condition test more cases, such as len() < 1 or len <= 0'
* Fix false-positive line-too-long message emission for
commented line at the end of a module
* Fix false-positive bad-continuation for with statements
* Don't warn about stop-iteration-return when using next() over itertools.count
* Add a check consider-using-get for unidiomatic usage of value/default-retrieval
for a key from a dictionary
* invalid-slice-index is not emitted when the slice is used as index for a complex object.
We only use a handful of known objects (list, set and friends) to figure out if
we should emit invalid-slice-index when the slice is used to subscript an object.
* Don't emit unused-import anymore for typing imports used in type comments.
* Add a new check 'useless-import-alias'.
* Add comparison-with-callable to warn for comparison with bare callable, without calling it.
* Don't warn for missing-type-doc and/or missing-return-type-doc, if type
annotations exist on the function signature for a parameter and/or return type.
* Add --exit-zero option for continuous integration scripts to more
easily call Pylint in environments that abort when a program returns a
non-zero (error) status code.
* Warn if the first argument of an instance/ class method gets assigned
* New check comparison-with-itself to check comparison between same value.
* Add a new warning, 'logging-fstring-interpolation', emitted when f-string
is used within logging function calls.
* Don't show 'useless-super-delegation' if the subclass method has different type annotations.
* Add unhashable-dict-key check.
* Don't warn that a global variable is unused if it is defined by an import
* Skip wildcard import check for __init__.py.
* The Python 3 porting mode can now run with Python 3 as well.
* too-few-public-methods is not emitted for dataclasses.
* New verbose mode option, enabled with --verbose command line flag, to
display of extra non-checker-related output. It is disabled by default.
* undefined-loop-variable takes in consideration non-empty iterred objects before emitting
* Add support for numpydoc optional return value names.
* singleton-comparison accounts for negative checks
* Add a check consider-using-in for comparisons of a variable against
multiple values with "==" and "or"s instead of checking if the variable
is contained "in" a tuple of those values.
* defaultdict and subclasses of dict are now handled for dict-iter-* checks
* logging-format-interpolation also emits when f-strings are used instead of % syntax.
* Don't trigger misplaced-bare-raise when the raise is in a finally clause
* Add a new check, possibly-unused-variable.
This is similar to unused-variable, the only difference is that it is
emitted when we detect a locals() call in the scope of the unused variable.
The locals() call could potentially use the said variable, by consuming
all values that are present up to the point of the call. This new check
allows to disable this error when the user intentionally uses locals()
to consume everything.
* no-else-return accounts for multiple cases
The check was a bit overrestrictive because we were checking for
return nodes in the .orelse node. At that point though the if statement
can be refactored to not have the orelse. This improves the detection of
other cases, for instance it now detects TryExcept nodes that are part of
the .else branch.
* Added two new checks, invalid-envvar-value and invalid-envvar-default.
The former is trigger whenever pylint detects that environment variable manipulation
functions uses a different type than strings, while the latter is emitted whenever
the said functions are using a default variable of different type than expected.
* Add a check consider-using-join for concatenation of strings using str.join(sequence)
* Add a check consider-swap-variables for swapping variables with tuple unpacking
* Add new checker try-except-raise that warns the user if an except handler block
has a raise statement as its first operator. The warning is shown when there is
a bare raise statement, effectively re-raising the exception that was caught or the
type of the exception being raised is the same as the one being handled.
* Don't crash on invalid strings when checking for logging-format-interpolation
* Exempt __doc__ from triggering a redefined-builtin
__doc__ can be used to specify a docstring for a module without
passing it as a first-statement string.
* Fix false positive bad-whitespace from function arguments with default
values and annotations
* Fix stop-iteration-return false positive when next builtin has a
default value in a generator
* Fix emission of false positive no-member message for class with "private" attributes whose name is mangled.
* Fixed a crash which occurred when Uninferable wasn't properly handled in stop-iteration-return
* Use the proper node to get the name for redefined functions
* Don't crash when encountering bare raises while checking inconsistent returns
* Fix a false positive inconsistent-return-statements message when if statement is inside try/except.
* Fix a false positive inconsistent-return-statements message when while loop are used.
* Correct column number for whitespace conventions.
Previously the column was stuck at 0
* Fix unused-argument false positives with overshadowed variable in
dictionary comprehension.
* Fix false positive inconsistent-return-statements message when never
returning functions are used (i.e sys.exit for example).
* Fix error when checking if function is exception, as in bad-exception-context.
* Fix false positive inconsistent-return-statements message when a
function is defined under an if statement.
* New useless-return message when function or method ends with a "return" or
"return None" statement and this is the only return statement in the body.
* Fix false positive inconsistent-return-statements message by
avoiding useless exception inference if the exception is not handled.
* Fix bad thread instantiation check when target function is provided in args.
* Fixed false positive when a numpy Attributes section follows a Parameters
section
* Fix incorrect file path when file absolute path contains multiple path_strip_prefix strings.
* Fix false positive undefined-variable for lambda argument in class definitions
* Add of a new checker that warns the user if some messages are enabled or disabled
by id instead of symbol.
* Suppress false-positive not-callable messages from certain
staticmethod descriptors
* Fix indentation handling with tabs
* Fix false-positive bad-continuation error
* Fix false positive unused-variable in lambda default arguments
* Updated the default report format to include paths that can be clicked on in some terminals (e.g. iTerm).
* Fix inline def behavior with too-many-statements checker
* Fix KeyError raised when using docparams and NotImplementedError is documented.
* Fix 'method-hidden' raised when assigning to a property or data descriptor.
* Fix emitting useless-super-delegation when changing the default value of keyword arguments.
* Expand ignored-argument-names include starred arguments and keyword arguments
* Fix false-postive undefined-variable in nested lambda
* Fix false-positive bad-whitespace message for typing annoatations
with ellipses in them
|
|
astroid 2.0:
* String representation of nodes takes in account precedence and associativity rules of operators.
* Fix loading files with modutils.load_from_module when
the path that contains it in sys.path is a symlink and
the file is contained in a symlinked folder.
* Reworking of the numpy brain dealing with numerictypes
(use of inspect module to determine the class hierarchy of
numpy.core.numerictypes module)
* Added inference support for starred nodes in for loops
* Support unpacking for dicts in assignments
* Add support for inferring functools.partial
* Inference support for dict.fromkeys
* int() builtin is inferred as returning integers.
* str() builtin is inferred as returning strings.
* DescriptorBoundMethod has the correct number of arguments defined.
* Improvement of the numpy numeric types definition.
* Subclasses of *property* are now interpreted as properties
* AsStringRegexpPredicate has been removed.
Use transform predicates instead of it.
* Switched to using typed_ast for getting access to type comments
As a side effect of this change, some nodes gained a new type_annotation attribute,
which, if the type comments were correctly parsed, should contain a node object
with the corresponding objects from the type comment.
* typing.X[...] and typing.NewType are inferred as classes instead of instances.
* Module.__path__ is now a list
It used to be a string containing the path, but it doesn't reflect the situation
on Python, where it is actually a list.
* Fix a bug with namespace package's __path__ attribute.
* Added brain tips for random.sample
* Add brain tip for issubclass builtin
* Fix submodule imports from six
* Fix missing __module__ and __qualname__ from class definition locals
* Fix a crash when __annotations__ access a parent's __init__ that does not have arguments
* Fix multiple objects sharing the same InferenceContext.path causing uninferable results
* Fix improper modification of col_offset, lineno upon inference of builtin functions
* Subprocess.Popen brain now knows of the args member
* add move_to_end method to collections.OrderedDict brain
* Include new hashlib classes added in python 3.6
* Fix RecursionError for augmented assign
* Add missing attrs special attribute
* Inference now understands the 'isinstance' builtin
* Stop duplicate nodes with the same key values
from appearing in dictionaries from dictionary unpacking.
* Fix contextlib.contextmanager inference for nested context managers
* Implement inference for len builtin
* Add qname method to Super object preventing potential errors in upstream
pylint
* Stop astroid from getting stuck in an infinite loop if a function shares
its name with its decorator
* Fix issue with inherited __call__ improperly inferencing self
* Fix __call__ precedence for classes with custom metaclasses
* Limit the maximum amount of interable result in an NodeNG.infer() call to
100 by default for performance issues with variables with large amounts of
possible values.
The max inferable value can be tuned by setting the max_inferable_values flag on
astroid.MANAGER.
|
|
0.7.0:
Unknown changes.
|
|
Changes since 1.8.1:
- Export language.RegisterPluralFunc
- Export language.Operand
- Update to CLDR 31.0.1
- Fix panic in goi18n command
- Export language.GetPluralSpec
|
|
Upstream changes:
0.50 Mon May 14 19:41:04 CEST 2018
* improve 'brief_usage()' output (patch from Ildar Shaimordanov)
|
|
|
|
Without it, I got:
No local packages or working download links found for setuptools-scm
|
|
v4.6.0:
Improve performance of the following functions for large datasets:
duplicates
sorted_uniq
sorted_uniq_by
union
union_by
union_with
uniq
uniq_by
uniq_with
xor
xor_by
xor_with
|
|
2.1.14:
Bug fixes.
|
|
|
|
0.8.17:
Add ein, itin and refactored ssn Provider for en_US.
Add job provier for zh_CN.
Add date_of_birth provider.
Add alpha-3 representation option for country-code provider.
|
|
0.15.43:
merge PR33: Python2.7 on Windows is narrow, but has no sysconfig.get_config_var('Py_UNICODE_SIZE').
register_class() now returns class
|
|
0.3:
Ditching Py2 due to pylint having done the same
|
|
3.0.1:
Bug fixes.
|
|
|
|
The typing_extensions module contains both backports of these changes
as well as experimental types that will eventually be added to the
typing module, such as Protocol.
|
|
0.2.12:
- More C90 (ANSI C) compliance.
- Prevent some compiling conflicts with other sources.
- Fix miscellaneous compiler warnings.
- Prevent trimming on extremely long dictionary path names.
|
|
Most importantly, this version adds support for OCaml 4.07. There are some
other minor extensions as well.
|
|
3.3 - Charlie the unicorn
* Use masquerade as compiler white-list.
* New --allow-private (the default) which allows non-global
* IP and IPv6 addresses.
* Cross-compilation support.
* Fix parsing of IPv6 addresses.
* Python 3, not python 2.
* Can build without python (and without pump mode or tests).
For those upgrading: you must run update-distcc-symlinks on every server machine, and add manually (see MASQUERADING of distcc(1)) those compilers it does not detect.
|
|
|
|
Godef, given an expression or a location in a source file, prints the
location of the definition of the symbol referred to.
|
|
Upstream changes:
version 1.18 at 2018-07-02 09:18:10 +0000
-----------------------------------------
Change: 58133ae09a1087d1d0cf6e5eac3767961b2e2577
Author: Chris 'BinGOs' Williams <chris@bingosnet.co.uk>
Date : 2018-07-02 10:18:10 +0000
Added usessl option to enable SSL/TLS
|
|
Upstream changes:
1.20180705 2018-07-05 16:47:18-07:00 America/Los_Angeles
- added destructor to clear counters (choroboa PR #15, RT #71831)
|
|
NEW IN WAF 2.0.9
* Add dependencies on scriptlet outputs
* Made options optional for cython waftool
* Improve doxygen error handling
|
|
Upstream changes:
0.23 2018-06-26 00:00:13Z
- properly skip potentially-problematic tests when needed, due to circular
dependencies between Moose and Test::CleanNamespaces (RT#125678)
|
|
Version 1.22.0:
* unix: remove checksparse.sh
* win: fix mingw build error
* win: fix -Wunused-function warnings in thread.c
* unix,win: merge timers implementation
* win: fix pointer type in pipe.c
* win: fixing build for older MSVC compilers
* zos: clear poll events on every iteration
* zos: write-protect message queue
* zos: use correct pointer type in strnlen
* unix,win: merge handle flags
* doc: update Imran Iqbal's GitHub handle
* src: add new error apis to prevent memory leaks
* test: make test-condvar call uv_cond_wait
* fs: change position of uv_fs_lchown
|
|
|
|
|
|
Python client for the Apache Kafka distributed stream processing
system. kafka-python is designed to function much like the official
java client, with a sprinkling of pythonic interfaces (e.g., consumer
iterators).
This package replaces devel/py-kafka.
|
|
0.7.6:
Bug fixes
|
|
objects to. Clean up permissions before packaging.
|
|
0.19.5:
IMPROVEMENTS
Add porcelain.describe.
BUG FIXES
Fix regression in dulwich.porcelain.clone that prevented cloning of remote repositories.
Don’t leave around empty parent directories for removed refs.
0.19.4:
IMPROVEMENTS
Add porcelain.ls_files.
Add Index.items.
BUG FIXES
Avoid unicode characters (e.g. the digraph ij in my surname) in setup.cfg, since setuptools doesn’t deal well with them.
|
|
1.4.3:
* Catch SyntaxError from asyncio import on older versions of Python
|
|
0.18:
Add new Python 3.7 modules to the stdlib list, and support 3.7.
|
|
1.5:
- switch to setuptools_scm
- move to github
- fix up python compat matrix
- avoid dict iteration (fixes issue on python3)
|
|
0.28.4:
Bugs fixed
* Reallowing tp_clear() in a subtype of an @no_gc_clear extension type generated an invalid C function call to the (non-existent) base type implementation.
* Exception catching based on a non-literal (runtime) tuple could fail to match the exception.
* Compile fix for CPython 3.7.0a2.
|
|
|
|
A simple module that adds an extension for the pydocstyle tool to
flake8.
|
|
|
|
Guile-Gnome-Platform 2.16.5 - David Pirotte, 2017-06-02
=======================================================
* Changes since 2.16.4
Guile-Gnome is now compatible with Guile-2.2
But don't hail, this is a maintainance release, which actually
merely comment the build of the Corba and Gnome-VFS modules, due to
incompatible changes in Guile-2.2 (see their corresponding entry in
configure.ac for a better explanation):
This is reversible though, if you think you have the skill and some
free time to fix these two, please do! And get in touch with us of
course...
Dependencies
------------
Guile-Gnome now allows Guile-2.2, and requires Guile >= 2.0.14
|
|
Upstream changes:
1.706 2018-07-06 20:20:00-05:00 America/Chicago
[Fixed]
- The File, Stderr, and Stdout adapters now correctly allow being
set to the "emergency" log level threshold. Previously, trying to
only allow "emergency" log lines would result in all logs being
written (and a warning about an invalid log level being set).
Thanks @alabamapaul! [Github #74]
|
|
Upstream changes:
0.237 2018-07-06
. Don't load vars.pm
This drops compatibility with Perl versions before Perl 5.006.
Patch provided by Atoomic and guillemj
RT#132077
|
|
Additional functools in the spirit of stdlib’s functools.
|
|
1.5
Refresh packaging.
Use Python 3 syntax for new-style classes.
|
|
|
|
The Evolve Extension extends the "changeset evolution" features of
Mercurial core. It provides a set of commands to easily mutate history as
well as the topics extension.
|
|
|
|
By update of lang/ruby/files/update-gemspec.rb, dependencies changed in
gemspec.
Bump PKGREVISION.
|
|
|
|
2.0.5:
Bug fixes.
|
