| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Currently, this will fail because setuptools.build_meta.build_wheel
assumes that no wheels already exist in the `dist/` directory.
See GH #1671
|
| | |
| | |
| | | |
Checking for new implementation solving issue #1740
|
| | | |
|
| | | |
|
| | |
| | |
| | |
| | | |
characters.
|
| | | |
|
| | | |
|
| | | |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This fixes GH #1682 by porting the pkg_resources requirement parsing
logic into setuptools.build_meta, so that all valid requirement
specifiers passed to setup_requires will be added to the
get_requires_for_build_* function outputs.
Fixes GH #1682
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Per GH #1682, with setuptools.build_meta we are not properly handling
the situation where setup_requires is actually a newline-delimited
string rather than a list, which is supported by setup.py interface.
This adds several failing (and some passing) tests for how
setup_requires is handled by setuptools.build_meta.
|
| | | |
|
| | |
| | |
| | |
| | |
| | | |
Parametrize the test to check different versions of pip
(including master) are correctly supported.
|
| | | |
|
|/ / |
|
| |
| |
| |
| |
| |
| | |
Many projects can get away with an empty `setup.py` and use *only* the
declarative `setup.cfg`. With the new PEP 517 backend, we can supply a
default empty `setup.py` if one is not provided.
|
| | |
|
| |
| |
| |
| |
| |
| | |
Rather than exposing a top-level module for the legacy backend, we will
move the legacy backend into the `setuptools.build_meta` module and
specify it using the module:object syntax.
|
| |
| |
| |
| |
| |
| |
| | |
This is part of the solution to GH #1642, it is a
backwards-compatibility backend that can be used as a default PEP 517
backend for projects that use setuptools but haven't opted in to
build_meta.
|
| |
| |
| |
| |
| |
| | |
This runs all build_meta tests, plus a test that it is possible to
import from the directory containing `setup.py` when using the
build_meta_legacy backend.
|
| |
| |
| |
| |
| |
| | |
Failing test adapted from PR #1643
Co-authored-by: Tzu-ping Chung <uranusjr@gmail.com>
|
| | |
|
|\ \ |
|
| | | |
|
| | | |
|
|\ \ \
| |/ / |
|
| |\ \
| | | |
| | | |
| | | |
| | | | |
benoit-pierre/fix_889_and_non-ascii_in_setup.cfg_take_2
improve encoding handling for `setup.cfg`
|
| | |\ \
| | | |/ |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Support the same mechanism as for Python sources for declaring
the encoding to be used when reading `setup.cfg` (see PEP 263),
and return the results of reading it as Unicode.
Fix #1062 and #1136.
|
| | | | |
|
| | |/
| |/| |
|
| | | |
|
| | | |
|
| |\ \
| | | |
| | | | |
Disallow files for license inputs
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Both the old and new approaches are deeply unsatisfying to me, but
without reworking how these test commands are run, I think this is
about as close as we can get to enforcing that this specific call
raises ValueError.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
The ability to handle files was originally added and documented based on
a misunderstanding of what the `license` field should include. The field
should be the name of the license, not the full text.
It is likely that anyone actually using this was outputing malformed
PKG-INFO files, because most license files contain newlines.
See GH issue #1551
|
| |/ / |
|
| | |
| | |
| | |
| | |
| | | |
This both prevents the warning from surfacing during the tests and
ensures that no regressions occur.
|
| | | |
|
| |\ \
| | | |
| | | | |
added unittest for _download_git
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | | |
Prior to this patch, if the author or author_email were omitted from
`setup`, a malformed `PKG-INFO` would be created.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This comment is not used anywhere and `platform.dist()` is deprecated.
See CPython PR #10414: https://github.com/python/cpython/pull/10414
and bpo-35186: https://bugs.python.org/issue35186
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
The `upload.show_response` feature was not added until Python 3. Rather
than backport it, it is now enabled only if supported.
This also adds a "smoke test" for the feature.
|
| | | |
| | | |
| | | |
| | | |
| | | | |
This fixes uploads when bdist_rpm or bdist_dumb are the command, both of
which insert a comment about what platform they are built for.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This fixes an issue where `distutils.spawn.spawn` was not available in
the ported upload_file, which is only used when signing the data.
This also adds a test that the gpg signature command is invoked and
included in the uploaded data.
|