The GCredentials support for NetBSD missed a
G_CREDENTIALS_SOCKET_GET_CREDENTIALS_SUPPORTED define and remove
commented out G_CREDENTIALS_UNIX_CREDENTIALS_MESSAGE_SUPPORTED.
Due that the client sent credentials but the server was not able
to receive them.
Regarding possible G_CREDENTIALS_UNIX_CREDENTIALS_MESSAGE_SUPPORTED
support probably logic for LOCAL_CREDS should be added but on NetBSD
before 8.0 sc_pid field is not present making that probably not
suitable to use as GCredentials mechanism (since 8.0 this will
probably work).
Bump PKGREVISION
Discussed with <prlw1>, thanks!
pkgsrc changes:
- Remove custom and no longer needed do-patch target, it was fixed upstream
- Minor cosmetic improvements pointed out by pkglint
Changes:
- Backport patches for CVE-2018-6951, CVE-2018-6952 and CVE-2018-1000156
Patch provided by Attila Fülöp via NetBSD/pkgsrc#33, thanks!
Bump PKGREVISION
libgit2 0.27.5 (2018/10/5)
This is a security release fixing the following list of issues:
* Submodule URLs and paths with a leading "-" are now ignored. This is due to
the recently discovered CVE-2018-17456, which can lead to arbitrary code
execution in upstream git. While libgit2 itself is not vulnerable, it can
be used to inject options in an implementation which performs a recursive
clone by executing an external command.
* When running repack while doing repo writes, packfile_load__cb() could see
some temporary files in the directory that were bigger than the usual, and
makes memcmp overflow on the p->pack_name string. This issue was reported
and fixed by bisho.
* The configuration file parser used unbounded recursion to parse multiline
variables, which could lead to a stack overflow. The issue was reported by
the oss-fuzz project, issue 10048 and fixed by Nelson Elhage.
* The fix to the unbounded recursion introduced a memory leak in the config
parser. While this leak was never in a public release, the oss-fuzz project
reported this as issue 10127. The fix was implemented by Nelson Elhage and
Patrick Steinhardt.
* When parsing "ok" packets received via the smart protocol, our parsing code
did not correctly verify the bounds of the packets, which could result in a
heap-buffer overflow. The issue was reported by the oss-fuzz project, issue
9749 and fixed by Patrick Steinhardt.
* The parsing code for the smart protocol has been tightened in general,
fixing heap-buffer overflows when parsing the packet type as well as for
"ACK" and "unpack" packets. The issue was discovered and fixed by Patrick
Steinhardt.
* Fixed potential integer overflows on platforms with 16 bit integers when
parsing packets for the smart protocol. The issue was discovered and fixed
by Patrick Steinhardt.
* Fixed potential NULL pointer dereference when parsing configuration files
which have "include.path" or "includeIf..path" statements without a value.
version 1.4.1:
add W504 fixed method
add E402 fixed method
new feature: reading from .flake8 and $HOME/.pycodestyle file that using as autopep8's configuration, and add configuration section into README ()
add --exit-code command line option
case of if --exit-code option is False. this is default
return 1 when error occured
otherwise return 0 (command successful)
case of if --exit-code option is True
return 1 when error occured
return 2 when exists changes in files (command successful)
otherwise return 0 (command successful)
This option is valid for any operating mode such as --diff, --in-place, non option etc
fix bugs
pytest 3.9.1:
Features
- For test-suites containing test classes, the information about the subclassed module is now output only if a higher verbosity level is specified (at least “-vv”).
pytest 3.9.0:
Deprecations
- The following accesses have been documented as deprecated for years, but are now actually emitting deprecation warnings.
Access of Module, Function, Class, Instance, File and Item through Node instances. Now users will this warning:
usage of Function.Module is deprecated, please use pytest.Module instead
Users should just import pytest and access those objects using the pytest module.
request.cached_setup, this was the precursor of the setup/teardown mechanism available to fixtures. You can consult funcarg comparison section in the docs.
Using objects named "Class" as a way to customize the type of nodes that are collected in Collector subclasses has been deprecated. Users instead should use pytest_collect_make_item to customize node types during collection.
This issue should affect only advanced plugins who create new collection types, so if you see this warning message please contact the authors so they can change the code.
The warning that produces the message below has changed to RemovedInPytest4Warning:
getfuncargvalue is deprecated, use getfixturevalue
- Add a Deprecation warning for pytest.ensuretemp as it was deprecated since a while.
Features
- Improve usage errors messages by hiding internal details which can be distracting and noisy.
This has the side effect that some error conditions that previously raised generic errors (such as ValueError for unregistered marks) are now raising Failed exceptions.
- Improve the error displayed when a conftest.py file could not be imported.
In order to implement this, a new chain parameter was added to ExceptionInfo.getrepr to show or hide chained tracebacks in Python 3 (defaults to True).
- Add empty_parameter_set_mark=fail_at_collect ini option for raising an exception when parametrize collects an empty set.
- Log messages generated in the collection phase are shown when live-logging is enabled and/or when they are logged to a file.
- Introduce tmp_path as a fixture providing a Path object.
- Deprecation warnings are now shown even if you customize the warnings filters yourself. In the previous version any customization would override pytest’s filters and deprecation warnings would fall back to being hidden by default.
- Allow specification of timeout for Testdir.runpytest_subprocess() and Testdir.run().
- Add returncode argument to pytest.exit() to exit pytest with a specific return code.
- Reimplement pytest.deprecated_call using pytest.warns so it supports the match='...' keyword argument.
This has the side effect that pytest.deprecated_call now raises pytest.fail.Exception instead of AssertionError.
- Require setuptools>=30.3 and move most of the metadata to setup.cfg.
Bug Fixes
- Improve error message when test functions of unittest.TestCase subclasses use a parametrized fixture.
- request.fixturenames now correctly returns the name of fixtures created by request.getfixturevalue().
- Warning filters passed as command line options using -W now take precedence over filters defined in ini configuration files.
- Fix source reindenting by using textwrap.dedent directly.
- pytest.warn will capture previously-warned warnings in Python 2. Previously they were never raised.
- Resolve symbolic links for args.
This fixes running pytest tests/test_foo.py::test_bar, where tests is a symlink to project/app/tests: previously project/app/conftest.py would be ignored for fixtures then.
- Fix duplicate printing of internal errors when using --pdb.
- pathlib based tmpdir cleanup now correctly handles symlinks in the folder.
- Display the filename when encountering SyntaxWarning.
Improved Documentation
- Update usefixtures documentation to clarify that it can’t be used with fixture functions.
- Update fixture documentation to specify that a fixture can be invoked twice in the scope it’s defined for.
- According to unittest.rst, setUpModule and tearDownModule were not implemented, but it turns out they are. So updated the documentation for unittest.
- Add tempir testing example to CONTRIBUTING.rst guide
Trivial/Internal Changes
- The internal MarkerError exception has been removed.
- Port the implementation of tmpdir to pathlib.
- Exclude 0.00 second entries from --duration output unless -vv is passed on the command-line.
- Fixed formatting of string literals in internal tests.
3.78.0:
This release has deprecated the generation of integers, floats and fractions when the conversion of the upper and/ or lower bound is not 100% exact, e.g. when an integer gets passed a bound that is not a whole number. (:issue:1625)
3.77.0:
This minor release adds functionality to :obj:~hypothesis.settings allowing it to be used as a decorator on :obj:~hypothesis.stateful.RuleBasedStateMachine and :obj:~hypothesis.stateful.GenericStateMachine.
3.76.1:
This patch fixes some warnings added by recent releases of :pypi:pydocstyle and :pypi:mypy.
ccmake needs wsyncup(3) and since NetBSD 8.0 it is present on NetBSD.
Handle that via `USE_CURSES= wsyncup' and remove enforcements about
ncurses in CMakeLists.txt and Source/Checks/Curses/CMakeLists.txt.
Bump PKGREVISION
Version 1.10.3
User-visible changes:
- Minor new features and improvements:
* Store the HTTPS client cert password
- Client-side bugfixes:
* Fix shelving when custom diff command is configured
* Fix conflict resolver crashes
* Fix conflict resolver endless scan in some cases
* Fix "Accept incoming deletion" on locally deleted file
* Fix "resolver adds unrelated moves to move target list"
- Server-side bugfixes:
* Reject bad PUT before CHECKOUT in v1 HTTP protocol
* Let 'svnadmin recover' prune the rep-cache even if disabled
- Client-side and server-side bugfixes:
* Allow commands like 'svn ci --file X' to work when X is a FIFO
- Other tool improvements and bugfixes:
* 'svnadmin verify --keep-going --quiet' shows an error summary
* Fix error in german translation for 'svn help merge'
Developer-visible changes:
- General:
* Python tests use the current python executable
4.5:
- Add support for Python 3.7.
- Always install zope.hookable as a dependency (the hook
extra is now empty). zope.hookable respects the PURE_PYTHON
environment variable, and has an optional C extension.
- Make accessing names that have been moved to zope.interface
produce a DeprecationWarning.
4.3.0:
- Simplify exception chaining and nested exception error messages.
4.2.2:
- Fix GlobalObject (and GlobalInterface) no longer allowing
multiple leading dots.
- Add __all__ to all modules listing the documented members of
the module. Note that this is currently a broad list and may be
reduced in the future.
4.2.1:
- Fix GlobalObject (and GlobalInterface) no longer allowing
just a single '.'.
4.2.0:
- Reach 100% automated test coverage.
- Add support for Python 3.7.
- Drop support for Python 3.3 and remove internal compatibility
functions needed to support it.
- Drop support for python setup.py test.
- Make zope.configuration.fields.Path and
zope.configuration.config.ConfigurationContext expand
environment variables and expand user home directories in paths.
- Fix resolving names from a Python 2 package whose __init__.py has
unicode elements in __all__.
- Make GroupingContextDecorator stop shadowing builtins in its
__getattr__. These were not intended as arguments to be used by
subclasses, and the signature caused confusion.
- Fix the doctests with zope.schema 4.7 and above, and run the
doctests on both Python 2 and Python 3.
- Fix GlobalObject and GlobalInterface fields to only accept
dotted names instead of names with /. Previously, slash
delimited names could result in incorrect imports.
- Fix the schema fields to include the value and field values
on exceptions they raise.
- Make zope.configuration.fields.PythonIdentifier subclass
PythonIdentifier from zope.schema. It now implements fromBytes,
always produces a native string, and validates the value in
fromUnicode.
- Add ConfigurationMachine.pass_through_exceptions to allow
customizing the exceptions that
ConfigurationMachine.execute_actions wraps in a
ConfigurationExecutionError.
- Stop catching BaseException and wrapping it in either
ConfigurationExecutionError or ZopeXMLConfigurationError.
SystemExit and KeyboardInterrupt were always allowed to
propagate; now GeneratorExit and custom subclasses of
BaseException are also allowed te propagate.
4.9.3:
- Fixed a ReST error in getDoc() results when having "subfields"
with titles.
4.9.2:
- Make sure that the title for IObject.validate_invariants is a unicode
string.
4.9.1:
- Fix SimpleTerm token for non-ASCII bytes values.
4.9.0:
- Make NativeString and NativeStringLine distinct types that
implement the newly-distinct interfaces INativeString and
INativeStringLine. Previously these were just aliases for either
Text (on Python 3) or Bytes (on Python 2).
- Fix Field.getDoc() when value_type or key_type is
present. Previously it could produce ReST that generated Sphinx
warnings.
- Make DottedName accept leading underscores for each segment.
- Add PythonIdentifier, which accepts one segment of a dotted
name, e.g., a python variable or class.
4.8.0:
- Add the interface IFromBytes, which is implemented by the
numeric and bytes fields, as well as URI, DottedName, and
Id.
- Fix passing None as the description to a field constructor.
0.29:
Features added
--------------
PEP-489 multi-phase module initialisation has been enabled again. Module reloads in other subinterpreters raise an exception to prevent corruption of the static module state.
A set of mypy compatible PEP-484 declarations were added for Cython’s C data types to integrate with static analysers in typed Python code. They are available in the Cython/Shadow.pyi module and describe the types in the special cython module that can be used for typing in Python code.
Memoryviews are supported in PEP-484/526 style type declarations.
@cython.nogil is supported as a C-function decorator in Python code.
Raising exceptions from nogil code will automatically acquire the GIL, instead of requiring an explicit with gil block.
C++ functions can now be declared as potentially raising both C++ and Python exceptions, so that Cython can handle both correctly.
cython.inline() supports a direct language_level keyword argument that was previously only available via a directive.
A new language level name 3str was added that mostly corresponds to language level 3, but keeps unprefixed string literals as type ‘str’ in both Py2 and Py3, and the builtin ‘str’ type unchanged. This will become the default in the next Cython release and is meant to help user code a) transition more easily to this new default and b) migrate to Python 3 source code semantics without making support for Python 2.x difficult.
In CPython 3.6 and later, looking up globals in the module dict is almost as fast as looking up C globals.
For a Python subclass of an extension type, repeated method calls to non-overridden cpdef methods can avoid the attribute lookup in Py3.6+, which makes them 4x faster.
(In-)equality comparisons of objects to integer literals are faster.
Some internal and 1-argument method calls are faster.
Modules that cimport many external extension types from other Cython modules execute less import requests during module initialisation.
Constant tuples and slices are deduplicated and only created once per module.
The coverage plugin considers more C file extensions such as .cc and .cxx.
The cythonize command accepts compile time variable values (as set by DEF) through the new -E option.
pyximport can import from namespace packages.
Some missing numpy and CPython C-API declarations were added.
Declarations for the pylifecycle C-API functions were added in a new .pxd file cpython.pylifecycle.
The Pythran support was updated to work with the latest Pythran 0.8.7
%a is included in the string formatting types that are optimised into f-strings. In this case, it is also automatically mapped to %r in Python 2.x.
New C macro CYTHON_HEX_VERSION to access Cython’s version in the same style as PY_HEX_VERSION.
Constants in libc.math are now declared as const to simplify their handling.
An additional check_size clause was added to the ctypedef class name specification to allow suppressing warnings when importing modules with backwards-compatible PyTypeObject size changes.
Bugs fixed
----------
The exception handling in generators and coroutines under CPython 3.7 was adapted to the newly introduced exception stack. Users of Cython 0.28 who want to support Python 3.7 are encouraged to upgrade to 0.29 to avoid potentially incorrect error reporting and tracebacks.
Crash when importing a module under Stackless Python that was built for CPython.
2-value slicing of typed sequences failed if the start or stop index was None.
Multiplied string literals lost their factor when they are part of another constant expression (e.g. ‘x’ * 10 + ‘y’ => ‘xy’).
String formatting with the ‘%’ operator didn’t call the special __rmod__() method if the right side is a string subclass that implements it.
The directive language_level=3 did not apply to the first token in the source file.
Overriding cpdef methods did not work in Python subclasses with slots. Note that this can have a performance impact on calls from Cython code.
Fix declarations of builtin or C types using strings in pure python mode.
Generator expressions and lambdas failed to compile in @cfunc functions.
Global names with const types were not excluded from star-import assignments which could lead to invalid C code.
Several internal function signatures were fixed that lead to warnings in gcc-8.
The numpy helper functions set_array_base() and get_array_base() were adapted to the current numpy C-API recommendations.
Some NumPy related code was updated to avoid deprecated API usage.
Several C++ STL declarations were extended and corrected.
C lines of the module init function were unconditionally not reported in exception stack traces.
When PEP-489 support is enabled, reloading the module overwrote any static module state. It now raises an exception instead, given that reloading is not actually supported.
Object-returning, C++ exception throwing functions were not checking that the return value was non-null.
The source file encoding detection could get confused if the c_string_encoding directive appeared within the first two lines.
Cython generated modules no longer emit a warning during import when the size of the NumPy array type is larger than what was found at compile time. Instead, this is assumed to be a backwards compatible change on NumPy side.
Other changes
-------------
Cython now emits a warning when no language_level (2, 3 or ‘3str’) is set explicitly, neither as a cythonize() option nor as a compiler directive. This is meant to prepare the transition of the default language level from currently Py2 to Py3, since that is what most new users will expect these days. The future default will, however, not enforce unicode literals, because this has proven a major obstacle in the support for both Python 2.x and 3.x. The next major release is intended to make this change, so that it will parse all code that does not request a specific language level as Python 3 code, but with str literals. The language level 2 will continue to be supported for an indefinite time.
The documentation was restructured, cleaned up and examples are now tested. The NumPy tutorial was also rewritten to simplify the running example.
Cython compiles less of its own modules at build time to reduce the installed package size to about half of its previous size. This makes the compiler slightly slower, by about 5-7%.
0.15.72:
- fix regression on explicit 1.1 loading with the C based scanner/parser
0.15.72:
- fix regression on explicit 1.1 loading with the C based scanner/parser
0.15.71:
- fix regression where handcrafted CommentedMaps could not be initiated
- fix regression with non-root literal scalars that needed indent indicator
- tag:yaml.org,2002:python/object/apply now also uses __qualname__ on PY3
0.15.70:
- reverted CommentedMap and CommentedSeq to subclass ordereddict resp. list,
reimplemented merge maps so that both dict(**commented_map_instance) and JSON
dumping works. This also allows checking with isinstance() on dict resp. list.
0.15.69:
- fix issue with dump_all gobbling end-of-document comments on parsing
0.15.68:
- fix issue with parsabel, but incorrect output with nested flow-style sequences
- fix issue with loading Python objects that have __setstate__ and recursion in parameters
0.15.67:
- fix issue with extra space inserted with non-root literal strings
0.15.66:
- fix issue with fold indicating characters inserted in safe_load-ed folded strings
0.15.65:
- fix issue 232 revert to throw ParserError for unexcpected ]
and } instead of IndexError.
- added key and reverse parameter (suggested by Jannik Klemm via email)
- indent root level literal scalars that have directive or document end markers
at the beginning of a line
0.15.64:
- support round-trip of tagged sequences: !Arg [a, {b: 1}]
- single entry mappings in flow sequences now written by default without quotes
set yaml.brace_single_entry_mapping_in_flow_sequence=True to force
getting [a, {b: 1}, {c: {d: 2}}] instead of the default [a, b: 1, c: {d: 2}]
- fix issue when roundtripping floats starting with a dot such as .5
0.15.63:
- small fix only necessary for Windows users that don't use wheels.
0.15.62:
- C based reader/scanner & emitter now allow setting of 1.2 as YAML version.
** The loading/dumping is still YAML 1.1 code**, so use the common subset of
YAML 1.2 and 1.1
0.15.61:
- support for round-tripping folded style scalars
- update of C code
- speed up of scanning (~30% depending on the input)
0.15.60:
- cleanup for mypy
- spurious print in library
0.15.59:
- issue with C based loader and leading zeros
0.15.59:
- issue with C based loader and leading zeros
0.15.58:
- simple mappings can now be used as keys when round-tripping::
{a: 1, b: 2}: hello world
although using the obvious operations (del, popitem) on the key will
fail, you can mutilate it by going through its attributes. If you load the
above YAML in d, then changing the value is cumbersome:
d = {CommentedKeyMap([('a', 1), ('b', 2)]): "goodbye"}
and changing the key even more so:
d[CommentedKeyMap([('b', 1), ('a', 2)])] = d.pop(
CommentedKeyMap([('a', 1), ('b', 2)]))
(you can use a dict instead of a list of tuples (or ordereddict), but that might result
in a different order, of the keys of the key, in the output)
- check integers to dump with 1.2 patterns instead of 1.1
0.15.57:
- Fix that CommentedSeq could no longer be used in adding or do a copy
0.15.56:
- fix issue with python -O optimizing away code
0.15.55:
- unmade CommentedSeq a subclass of list. It is now
indirectly a subclass of the standard
collections.abc.MutableSequence (without .abc if you are
still on Python2.7). If you do isinstance(yaml.load('[1, 2]'),
list)) anywhere in your code replace list with
MutableSequence. Directly, CommentedSeq is a subclass of
the abstract baseclass ruamel.yaml.compat.MutableScliceableSequence,
with the result that *(extended) slicing is supported on
CommentedSeq*.
- duplicate keys (or their values) with non-ascii now correctly
report in Python2, instead of raising a Unicode error.
0.9.2:
* Add company names to pl_PL provider.
* Add replacements for non-ascii characters in pt_BR.
* Add some more placeholder image services.
* Separate male name and female name formats in cs_CZ provider.
* Add second level domains (mostly provinces) for cn top level domain.
* Add fr_FR localization to lorem provider.
* Lots of work on internal cleanup and optimizing the CI.
* Add flake8 to the CI.
3.76.0:
This release deprecates using floats for min_size and max_size.
The type hint for average_size arguments has been changed from Optional[int] to None, because non-None values are always ignored and deprecated.
3.75.4:
This patch adds more internal comments to the core engine's sequence-length shrinker. There should be no user-visible change.
3.75.3:
This patch adds additional comments to some of the core engine's internal data structures. There is no user-visible change.
3.75.2:
This patch avoids caching a trivial case, fixing :issue:493.
3.75.1:
This patch fixes a broken link in a docstring.
3.75.0:
This release deprecates the use of min_size=None, setting the default min_size to 0 (:issue: 1618).
3.74.3:
This patch makes some small internal changes to comply with a new lint setting in the build. There should be no user-visible change.
3.74.2:
This patch fixes :issue:1153, where time spent reifying a strategy was also counted in the time spent generating the first example. Strategies are now fully constructed and validated before the timer is started.
3.74.1:
This patch fixes some broken formatting and links in the documentation.
Changelog:
0.5.25:
This patch release follows several important enhancements in slibtool's
functionality, including proper support of explicit .a and .so input
arguments in link mode, and use of validly alternate (non-)pic objects.
With respect to compiler and target flavors, slibtool now accommodates
clang's debated approach towards the -fPIC switch in always-pic targets.
Last but not least, and following much discussion and consideration of
the various trade-offs, this release redefines _XOPEN_SOURCE to 700. With
that said, slibtool could still be easily built on older systems for as
long as they provide openat, faccessat, and (the soon-to-be-used) fexecve.
- driver: added --tag=RC support.
- driver: added --windres support.
- driver: use a compiler --target argument also as a tool argument.
- driver: if no --tag had been defined in compile mode, set --tag to CC.
- link mode: properly handle explicit archive input arguments.
- link mode: properly handle external archives and libraries.
- link mode: process dependencies also for non-pic .la args.
- link mode: support validly alternate (non-)pic objects.
- target flavors: accommodate conceptually challenged, PICky targets.
- target flavors: flavor detection: associate -windows- with mingw.
- target flavors: further eliminate compiler -dumpmachine invocations.
0.5.24:
This patch release explicitly requests definition of AT_FDCWD via definition
of _POSIX_C_SOURCE in translation units that refer to it.
0.5.23:
This patch releases fixes a couple of minor issues in slibtool's own build
system.
build system: all-shared front-end variant: fix ordering of -l$(PACKAGE).
build system: all-shared binary: added missing build-dep on $(SHARED_SOLINK).
0.5.22:
This patch release explicitly requests the openat() interface via definition
of _POSIX_C_SOURCE in translation units that need it.
0.5.21:
This patch release follows important enhancements in slibtool's library
interfaces and API contract. In particular, slibtool now no longer uses
streams, but rather pure file descriptors. In addition, i/o operations
now robustly handle interruption by signals, and thus do not depend on
the underlying system calls being restartable. Last but not least,
slibtool now installs a pkgconfig configuration file when applicable.
- argv.h: output usage and errors using [argv_]dprintf rather than fprint.
- argv.h: argv_get(): be consistent between forward decl. and impl.
- argv.h: move definition of struct argv_meta_impl to the implementation block.
- internals: added slbt_mapped_readline().
- internals: added slbt_map_file(), slbt_unmap_file().
- internals: added slbt_dprintf(), a signal-resilient dprintf implementation.
- driver: argv usage and error output: use the signal-resilient slbt_dprintf().
- driver: support alternate fd's for input/output/error/log.
- driver: slbt_init_host_params(): propagate failure of <compiler> -dumpmachine.
- driver: slbt_init_host_params(): do not set fnative when !fdumpmachine.
- driver: struct slbt_fd_ctx: added .fdcwd and .fddst.
- library: invocations of openat() now use the context's fdcwd.
- project: added initial pkgconfig support.
0.5.20:
This patch release includes several important bug fixes pertaining to both
the link and compile mode, as well as the addition of a compatible --features
output.
- driver: added slbt_output_features(), providing compatible --features output.
- driver: compatibility: have -{shared|static} imply -disable-{static|shared}.
- link mode: properly handle the combination of -release and -version-info.
- link mode: slbt_exec_link_create_dep_file(): record top-level -L arguments.
- link mode: --disable-{shared|static}: use differently-named placeholders.
v3.5.2:
Bugfixes
- session packages are now put inside a numbered directory (instead of prefix numbering it,
because pip fails when wheels are not named according to PEP-491 , and prefix numbering messes with this)
Features
- level three verbosity (-vvv) show the packaging output
v3.5.1:
Bugfixes
- fix regression with 3.5.0: specifying --installpkg raises AttributeError: 'str' object has no attribute 'basename'
v3.5.0:
Bugfixes
- intermittent failures with --parallel--safe-build, instead of mangling with the file paths now uses a lock to make the package build operation thread safe and is now on by default (--parallel--safe-build is now deprecated)
Features
- Added temp_dir folder configuration (defaults to {toxworkdir}/.tmp) that contains tox
temporary files. Package builds now create a hard link (if possible, otherwise copy - notably in
case of Windows Python 2.7) to the built file, and feed that file downstream (e.g. for pip to
install it). The hard link is removed at the end of the run (what it points though is kept
inside distdir). This ensures that a tox session operates on the same package it built, even
if a parallel tox run builds another version. Note distdir will contain only the last built
package in such cases.
This package contains a single module, which implements a platform independent
file lock in Python, which provides a simple way of inter-process communication.
* Update MASTER_SITES and HOMEPAGE
----- version 1.7.2 ------
01-05 Mark Lindner <markl@avalon>
* lib/libconfig.c - fixed slow leak in config_destroy()
* <multiple> - miscellaenous build file fixes
----- version 1.7.1 ------
2017-11-15 Mark Lindner <markl@avalon>
* lib/wincompat.c, lib/wincompat.h - added fsync() implementation for
Windows
* lib/libconfig.c, lib/libconfig.h - added CONFIG_OPTION_FSYNC
* lib/libconfigcpp.c++, lib/libconfig.h++ - added Config::OptionFsync
* doc/libconfig.texi - Documentation updates
2017-11-14 Mark Lindner <markl@avalon>
* lib/strbuf.c - bugfix in strbuf_append_char()
----- version 1.7 ------
2017-10-24 Mark Lindner <markl@avalon>
* INSTALL - Added information about installing from a git snapshot
* configure.ac - Bumped version to 1.7
* doc/libconfig.texi - Updated documentation, added new chapter on
libconfig bindings/implementations for other languages
* examples/c/Makefile.am, examples/c/example4.c, examples/c/example4.cfg,
examples/c/cfg_includes/*.cfg - New example program
* lib/Makefile.am - bumped library version to 11:0:0; added new source
files
* lib/grammar.c, lib/grammar.y, lib/scanner.c, lib/scanner.h - regenerated
* lib/grammar.y - allow optional trailing comma in lists and arrays
* *.vcproj - updated with new source files, library dependency
* lib/libconfig.c, lib/libconfig.h - various code cleanup;
reworked options API; moved some value parsing logic into lib/util.c
and lib/util.h; added new option
CONFIG_OPTION_ALLOW_SCIENTIFIC_NOTATION; bugfix to disallow adding a
group or a list to an array; changed default float precision from 2 to 6;
fixed conversion between int and int64 types; added config_clear();
added support for include functions; added config_set_hook() and
config_get_hook()
* lib/libconfigcpp.c++, lib/libconfig.h++ - moved Option enum from Setting
to Config; added OptionAllowScientificNotation; added clear();
reworked options API; added evaluateIncludePath(); fixed casting to
unsigned int (don't clip negative values to 0)
* lib/parsectx.c, lib/parsectx.h - code cleanup
* lib/scanctx.c, lib/scanctx.h - added support for include functions;
code cleanup; removed non-portable directory reading code;
* lib/scanner.l - added support for include functions; moved some parsing
code to lib/util.c and lib/util.h; code cleanup
* lib/strbuf.c, lib/strbuf.h - added strbuf_append_char(); code cleanup
* lib/strvec.c, lib/strvec.h - new functions for managing arrays of
string pointers (for filename lists)
* lib/util.c, lib/util.h - value parsing and formatting code and memory
management macros and functions extracted from other files
* lib/wincompat.h - new macros for testing relative paths; replaced
INT32_MAX/INT32_MIN with INT_MAX/INT_MIN.
* tests/tests.c - fixed test failures and improper testing of int type
conversions
----- version 1.6 ------
2015-12-31 Jose Luis Tallon <jltallon@adv-solutions.net>
* Added include_dir feature (support for Debian-style conf.d/
includes) (pull request #36)
* Added octal_ints feature (support for integer expressed in octal,
useful for permissions and masks in UNIX-like systems (pull req #42)
* Fixed "Removing a setting removes all siblings" (issue #41)
Props random85
* Allow specifying the number of decimals wanted when outputting
Based upon a suggestion by zhaopingsun (issue #31)
* Make libconfig usable from CMake (pull request/issue #28)
Props thfi
* Documentation fixes
- Copyright and examples
- Included TeX patch from Debian (fixes FTBS with TeXlive)
* Resync Debian packaging
includes updated packages for GCC5 transition
2015-10-14 Thomas Fischer <fischer@unix-ag.uni-kl.de>
* configure, configure.ac, debian/libconfig++9-dev.install,
debian/libconfig9-dev.install, lib/Makefile.am, lib/Makefile.in,
lib/libconfig++Config.cmake.in, lib/libconfigConfig.cmake.in,
libconfig.spec.in - locating libconfig in CMake-based projects
2015-08-14 Mark Lindner <markl@neuromancer>
* lib/wincompat.h - Fixed Windows portability issue
* lib/libconfig.h++ - added missing operator[](std::string&).
Bumping buildlink3.mk required because the reporter mentions having
to do this because of a time-consuming incompatibility in p6-zip.
Unclear if the packages are independent, so updating together.
No changelog found.
From tomasz motyl in PR pkg/53663
- remove OpenBSD #warning (obsd maintainer says no longer needed)
- fix fmt_ip6 (Erwin Hoffmann)
- add MSG_ZEROCOPY support (only used for buffers >8k)
- use writev in buffer_put for a slight perf improvement
- add experimental iom API for multithreaded I/O multiplexing (in io.h)
3.4.0:
Bugfixes
- add --exists-action w to default pip flags to handle better VCS dependencies
- instead of assuming the Python version from the base python name ask the interpreter to reveal the version for the ignore_basepython_conflict flag
- PEP-517 packaging fails with sdist already exists, fixed via ensuring the dist folder is empty before invoking the backend and pypa/setuptools
Features
- add commands_pre and commands_post that run before and after running
the commands (setup runs always, commands only if setup suceeds, teardown always - all
run until the first failing command)
- pyproject.toml config support initially by just inline the tox.ini under tool.tox.legacy_tox_ini key; config source priority order is pyproject.toml, tox.ini and then setup.cfg
- use the os environment variable TOX_SKIP_ENV to filter out tox environment names from the run list (set by envlist)
- always set PIP_USER=0 (do not install into the user site package, but inside the virtual environment created) and PIP_NO_DEPS=0 (installing without dependencies can cause broken package installations) inside tox
- tox will inject some environment variables that to indicate a command is running within tox: TOX_WORK_DIR env var is set to the tox work directory,
TOX_ENV_NAME is set to the current running tox environment name, TOX_ENV_DIR is set to the current tox environments working dir
- While running tox invokes various commands (such as building the package, pip installing dependencies and so on), these were printed in case they failed as Python arrays. Changed the representation to a shell command, allowing the users to quickly replicate/debug the failure on their own
- skip missing interpreters value from the config file can now be overridden via the --skip-missing-interpreters cli flag
- keep additional environments config order when listing them
- allow injecting config value inside the ini file dependent of the fact that we're connected to an interactive shell or not
- do not build sdist if skip install is specified for the envs to be run
- when verbosity level increases above two start passing through verbosity flags to pip
- when discovering the interpreter to use check if the tox host Python matches and use that if so
- -vv will print out why a virtual environment is re-created whenever this operation is triggered
Documentation
- clarify that python and pip refer to the virtual environments executable
- add Sphinx and mkdocs example of generating documentation via tox
- specify that setup.cfg tox configuration needs to be inside the tox:tox namespace
3.3.0:
Bugfixes
- fix TOX_LIMITED_SHEBANG when running under python3
Features
- PEP-517 <https://www.python.org/dev/peps/pep-0517/>_ source distribution support (create a
.package virtual environment to perform build operations inside)
- flit <https://flit.readthedocs.io>_ support via implementing PEP-517
- packaging now is exposed as a hook via tox_package(session, venv)
Miscellaneous
- Updated the VSTS build YAML to use the latest jobs and pools syntax
Pytest plugin to randomly order tests and control random.seed.
* Randomly shuffles the order of test items. This is done first at the level of
modules, then at the level of test classes (if you have them), then at the
order of functions. This also works with things like doctests.
* Resets random.seed() at the start of every test case and test to a fixed
number - this defaults to time.time() from the start of your test run, but
you can pass in --randomly-seed to repeat a randomness-induced failure.
* If factory boy is installed, its random state is reset at the start of every
test. This allows for repeatable use of its random 'fuzzy' features.
* If faker is installed, its random state is reset at the start of every test.
This is also for repeatable fuzzy data in tests - factory boy uses faker for
lots of data.
11.5.0:
- Re-wrote the handling of pattern matches for
listdir, walk, and related methods, allowing
the pattern to be a more complex object. This approach
drastically simplifies the code and obviates the
CaseInsensitivePattern and FastPath classes.
Now the main Path class should be as performant
as FastPath and case-insensitive matches can be
readily constructed using the new
path.matchers.CaseInsensitive class.
11.4.1:
- Skip intermittently failing performance test on
Python 2.
11.4.0:
- Path.py now supports non-decodable filenames on
Linux and Python 2, leveraging the
backports.os <https://pypi.org/project/backports.os>_
package (as an optional dependency). Currently, only
listdir is patched, but other os primitives may
be patched similarly in the patch_for_linux_python2
function.
- For merge_tree, instead of relying on the deprecated
distutils module, implement merge_tree explicitly. The
update parameter is deprecated, instead superseded
by a copy_function parameter and an only_newer
wrapper for any copy function.
18.1:
Features
- Allow PEP 508 URL requirements to be used as dependencies.
As a security measure, pip will raise an exception when installing packages from
PyPI if those packages depend on packages not also hosted on PyPI.
In the future, PyPI will block uploading packages with such external URL dependencies directly.
- Upgrade pyparsing to 2.2.1.
- Allows dist options (--abi, --python-version, --platform, --implementation) when installing with --target
- Support passing svn+ssh URLs with a username to pip install -e.
- pip now ensures that the RECORD file is sorted when installing from a wheel file.
- Add support for Python 3.7.
- Malformed configuration files now show helpful error messages, instead of tracebacks.
Bug Fixes
- Checkout the correct branch when doing an editable Git install.
- Run self-version-check only on commands that may access the index, instead of
trying on every run and failing to do so due to missing options.
- Allow a Git ref to be installed over an existing installation.
- Show a better error message when a configuration option has an invalid value.
- Always revalidate cached simple API pages instead of blindly caching them for up to 10
minutes.
- Avoid caching self-version-check information when cache is disabled.
- Avoid traceback printing on autocomplete after flags in the CLI.
- Fix incorrect parsing of egg names if pip needs to guess the package name.
