You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
ansible/setup.py

355 lines
13 KiB
Python

from __future__ import print_function
import json
import os
import os.path
import re
import sys
import warnings
from collections import defaultdict
from distutils.command.build_scripts import build_scripts as BuildScripts
from distutils.command.sdist import sdist as SDist
try:
from setuptools import setup, find_packages
from setuptools.command.build_py import build_py as BuildPy
from setuptools.command.install_lib import install_lib as InstallLib
from setuptools.command.install_scripts import install_scripts as InstallScripts
except ImportError:
print("Ansible now needs setuptools in order to build. Install it using"
" your package manager (usually python-setuptools) or via pip (pip"
" install setuptools).", file=sys.stderr)
sys.exit(1)
13 years ago
sys.path.insert(0, os.path.abspath('lib'))
from ansible.release import __version__, __author__
SYMLINK_CACHE = 'SYMLINK_CACHE.json'
def _find_symlinks(topdir, extension=''):
"""Find symlinks that should be maintained
Maintained symlinks exist in the bin dir or are modules which have
aliases. Our heuristic is that they are a link in a certain path which
point to a file in the same directory.
Install ansible-test (#60718) * Install ansible-test Modify the install script to install ansible-test and its supporting code. Alternative to #60701 that doesn't change package_dir ansible for fear that it might regress https://github.com/ansible/ansible/issues/10437 Also: * No longer use package_data. Everything in the package dirs is going to be installed. Anything that shouldn't be installed needs to be moved elsewhere. * modify the algorithm to store symlinks which are in the same tree instead of same directory * Add ansible_test files to package-data sanity test * MANIFEST.in cleanups * Add lib/ansible/config/*.yml * Make most things in code directories (lib/ansible and test/lib/ansible_test/) use explicit file extensions instead of wildcards for maintainability * Exclude common file extensions that we don't want included in the code directories * Change package-data test to be more complete * Now compares the repository, sdist, and install * Compares both that everything in the sdist is in the repo and everything in the install is in the sdist in addition to comparing that everything in the repo that we want is in the install * Leave out test artifacts Only include the directory structure for test/results and test/cache not any files that may have been generated by test runs Remove test/utils files from the sdist as these are only needed for our CI cleanup of docs in MANIFEST.in; getting rid of build files. * Add the ability to output sdist and snapshot to specific directory * Add a warning about modifying the heuristic to setup.py * Address generated files * Use make snapshot instead of sdist to generate changelog and man pages and make sure they're included * Ignore both the test/utils and generated test files (results, cache) * Deal with Python3 __pycache__ byte code caches * Don't check documentation, that isn't built for the sdist * Restructure for clarity * Add cli web docs to make clean This was causing problems when attempting to test that the sdist didn't have extra files * Fix bug constructing python names from __pycache__ names * Create a clean repo to work from * Exclude test/legacy and be more explicit on extensions * Exclude the legacy directory from sdist
5 years ago
.. warn::
We want the symlinks in :file:`bin/` that link into :file:`lib/ansible/*` (currently,
:command:`ansible`, :command:`ansible-test`, and :command:`ansible-connection`) to become
real files on install. Updates to the heuristic here *must not* add them to the symlink
cache.
"""
symlinks = defaultdict(list)
for base_path, dirs, files in os.walk(topdir):
for filename in files:
filepath = os.path.join(base_path, filename)
if os.path.islink(filepath) and filename.endswith(extension):
target = os.readlink(filepath)
Install ansible-test (#60718) * Install ansible-test Modify the install script to install ansible-test and its supporting code. Alternative to #60701 that doesn't change package_dir ansible for fear that it might regress https://github.com/ansible/ansible/issues/10437 Also: * No longer use package_data. Everything in the package dirs is going to be installed. Anything that shouldn't be installed needs to be moved elsewhere. * modify the algorithm to store symlinks which are in the same tree instead of same directory * Add ansible_test files to package-data sanity test * MANIFEST.in cleanups * Add lib/ansible/config/*.yml * Make most things in code directories (lib/ansible and test/lib/ansible_test/) use explicit file extensions instead of wildcards for maintainability * Exclude common file extensions that we don't want included in the code directories * Change package-data test to be more complete * Now compares the repository, sdist, and install * Compares both that everything in the sdist is in the repo and everything in the install is in the sdist in addition to comparing that everything in the repo that we want is in the install * Leave out test artifacts Only include the directory structure for test/results and test/cache not any files that may have been generated by test runs Remove test/utils files from the sdist as these are only needed for our CI cleanup of docs in MANIFEST.in; getting rid of build files. * Add the ability to output sdist and snapshot to specific directory * Add a warning about modifying the heuristic to setup.py * Address generated files * Use make snapshot instead of sdist to generate changelog and man pages and make sure they're included * Ignore both the test/utils and generated test files (results, cache) * Deal with Python3 __pycache__ byte code caches * Don't check documentation, that isn't built for the sdist * Restructure for clarity * Add cli web docs to make clean This was causing problems when attempting to test that the sdist didn't have extra files * Fix bug constructing python names from __pycache__ names * Create a clean repo to work from * Exclude test/legacy and be more explicit on extensions * Exclude the legacy directory from sdist
5 years ago
if target.startswith('/'):
# We do not support absolute symlinks at all
continue
if os.path.dirname(target) == '':
link = filepath[len(topdir):]
if link.startswith('/'):
link = link[1:]
symlinks[os.path.basename(target)].append(link)
Install ansible-test (#60718) * Install ansible-test Modify the install script to install ansible-test and its supporting code. Alternative to #60701 that doesn't change package_dir ansible for fear that it might regress https://github.com/ansible/ansible/issues/10437 Also: * No longer use package_data. Everything in the package dirs is going to be installed. Anything that shouldn't be installed needs to be moved elsewhere. * modify the algorithm to store symlinks which are in the same tree instead of same directory * Add ansible_test files to package-data sanity test * MANIFEST.in cleanups * Add lib/ansible/config/*.yml * Make most things in code directories (lib/ansible and test/lib/ansible_test/) use explicit file extensions instead of wildcards for maintainability * Exclude common file extensions that we don't want included in the code directories * Change package-data test to be more complete * Now compares the repository, sdist, and install * Compares both that everything in the sdist is in the repo and everything in the install is in the sdist in addition to comparing that everything in the repo that we want is in the install * Leave out test artifacts Only include the directory structure for test/results and test/cache not any files that may have been generated by test runs Remove test/utils files from the sdist as these are only needed for our CI cleanup of docs in MANIFEST.in; getting rid of build files. * Add the ability to output sdist and snapshot to specific directory * Add a warning about modifying the heuristic to setup.py * Address generated files * Use make snapshot instead of sdist to generate changelog and man pages and make sure they're included * Ignore both the test/utils and generated test files (results, cache) * Deal with Python3 __pycache__ byte code caches * Don't check documentation, that isn't built for the sdist * Restructure for clarity * Add cli web docs to make clean This was causing problems when attempting to test that the sdist didn't have extra files * Fix bug constructing python names from __pycache__ names * Create a clean repo to work from * Exclude test/legacy and be more explicit on extensions * Exclude the legacy directory from sdist
5 years ago
else:
# Count how many directory levels from the topdir we are
levels_deep = os.path.dirname(filepath).count('/')
# Count the number of directory levels higher we walk up the tree in target
target_depth = 0
for path_component in target.split('/'):
if path_component == '..':
target_depth += 1
# If we walk past the topdir, then don't store
if target_depth >= levels_deep:
break
else:
target_depth -= 1
else:
# If we managed to stay within the tree, store the symlink
link = filepath[len(topdir):]
if link.startswith('/'):
link = link[1:]
symlinks[target].append(link)
return symlinks
def _cache_symlinks(symlink_data):
with open(SYMLINK_CACHE, 'w') as f:
json.dump(symlink_data, f)
def _maintain_symlinks(symlink_type, base_path):
"""Switch a real file into a symlink"""
try:
# Try the cache first because going from git checkout to sdist is the
# only time we know that we're going to cache correctly
with open(SYMLINK_CACHE, 'r') as f:
symlink_data = json.load(f)
except (IOError, OSError) as e:
# IOError on py2, OSError on py3. Both have errno
if e.errno == 2:
# SYMLINKS_CACHE doesn't exist. Fallback to trying to create the
# cache now. Will work if we're running directly from a git
# checkout or from an sdist created earlier.
Install ansible-test (#60718) * Install ansible-test Modify the install script to install ansible-test and its supporting code. Alternative to #60701 that doesn't change package_dir ansible for fear that it might regress https://github.com/ansible/ansible/issues/10437 Also: * No longer use package_data. Everything in the package dirs is going to be installed. Anything that shouldn't be installed needs to be moved elsewhere. * modify the algorithm to store symlinks which are in the same tree instead of same directory * Add ansible_test files to package-data sanity test * MANIFEST.in cleanups * Add lib/ansible/config/*.yml * Make most things in code directories (lib/ansible and test/lib/ansible_test/) use explicit file extensions instead of wildcards for maintainability * Exclude common file extensions that we don't want included in the code directories * Change package-data test to be more complete * Now compares the repository, sdist, and install * Compares both that everything in the sdist is in the repo and everything in the install is in the sdist in addition to comparing that everything in the repo that we want is in the install * Leave out test artifacts Only include the directory structure for test/results and test/cache not any files that may have been generated by test runs Remove test/utils files from the sdist as these are only needed for our CI cleanup of docs in MANIFEST.in; getting rid of build files. * Add the ability to output sdist and snapshot to specific directory * Add a warning about modifying the heuristic to setup.py * Address generated files * Use make snapshot instead of sdist to generate changelog and man pages and make sure they're included * Ignore both the test/utils and generated test files (results, cache) * Deal with Python3 __pycache__ byte code caches * Don't check documentation, that isn't built for the sdist * Restructure for clarity * Add cli web docs to make clean This was causing problems when attempting to test that the sdist didn't have extra files * Fix bug constructing python names from __pycache__ names * Create a clean repo to work from * Exclude test/legacy and be more explicit on extensions * Exclude the legacy directory from sdist
5 years ago
library_symlinks = _find_symlinks('lib', '.py')
library_symlinks.update(_find_symlinks('test/lib'))
symlink_data = {'script': _find_symlinks('bin'),
Install ansible-test (#60718) * Install ansible-test Modify the install script to install ansible-test and its supporting code. Alternative to #60701 that doesn't change package_dir ansible for fear that it might regress https://github.com/ansible/ansible/issues/10437 Also: * No longer use package_data. Everything in the package dirs is going to be installed. Anything that shouldn't be installed needs to be moved elsewhere. * modify the algorithm to store symlinks which are in the same tree instead of same directory * Add ansible_test files to package-data sanity test * MANIFEST.in cleanups * Add lib/ansible/config/*.yml * Make most things in code directories (lib/ansible and test/lib/ansible_test/) use explicit file extensions instead of wildcards for maintainability * Exclude common file extensions that we don't want included in the code directories * Change package-data test to be more complete * Now compares the repository, sdist, and install * Compares both that everything in the sdist is in the repo and everything in the install is in the sdist in addition to comparing that everything in the repo that we want is in the install * Leave out test artifacts Only include the directory structure for test/results and test/cache not any files that may have been generated by test runs Remove test/utils files from the sdist as these are only needed for our CI cleanup of docs in MANIFEST.in; getting rid of build files. * Add the ability to output sdist and snapshot to specific directory * Add a warning about modifying the heuristic to setup.py * Address generated files * Use make snapshot instead of sdist to generate changelog and man pages and make sure they're included * Ignore both the test/utils and generated test files (results, cache) * Deal with Python3 __pycache__ byte code caches * Don't check documentation, that isn't built for the sdist * Restructure for clarity * Add cli web docs to make clean This was causing problems when attempting to test that the sdist didn't have extra files * Fix bug constructing python names from __pycache__ names * Create a clean repo to work from * Exclude test/legacy and be more explicit on extensions * Exclude the legacy directory from sdist
5 years ago
'library': library_symlinks,
}
# Sanity check that something we know should be a symlink was
# found. We'll take that to mean that the current directory
# structure properly reflects symlinks in the git repo
if 'ansible-playbook' in symlink_data['script']['ansible']:
_cache_symlinks(symlink_data)
else:
raise RuntimeError(
"Pregenerated symlink list was not present and expected "
"symlinks in ./bin were missing or broken. "
"Perhaps this isn't a git checkout?"
)
else:
raise
symlinks = symlink_data[symlink_type]
for source in symlinks:
for dest in symlinks[source]:
dest_path = os.path.join(base_path, dest)
if not os.path.islink(dest_path):
try:
os.unlink(dest_path)
except OSError as e:
if e.errno == 2:
# File does not exist which is all we wanted
pass
os.symlink(source, dest_path)
class BuildPyCommand(BuildPy):
def run(self):
BuildPy.run(self)
_maintain_symlinks('library', self.build_lib)
class BuildScriptsCommand(BuildScripts):
def run(self):
BuildScripts.run(self)
_maintain_symlinks('script', self.build_dir)
class InstallLibCommand(InstallLib):
def run(self):
InstallLib.run(self)
_maintain_symlinks('library', self.install_dir)
class InstallScriptsCommand(InstallScripts):
def run(self):
InstallScripts.run(self)
_maintain_symlinks('script', self.install_dir)
class SDistCommand(SDist):
def run(self):
# have to generate the cache of symlinks for release as sdist is the
# only command that has access to symlinks from the git repo
Install ansible-test (#60718) * Install ansible-test Modify the install script to install ansible-test and its supporting code. Alternative to #60701 that doesn't change package_dir ansible for fear that it might regress https://github.com/ansible/ansible/issues/10437 Also: * No longer use package_data. Everything in the package dirs is going to be installed. Anything that shouldn't be installed needs to be moved elsewhere. * modify the algorithm to store symlinks which are in the same tree instead of same directory * Add ansible_test files to package-data sanity test * MANIFEST.in cleanups * Add lib/ansible/config/*.yml * Make most things in code directories (lib/ansible and test/lib/ansible_test/) use explicit file extensions instead of wildcards for maintainability * Exclude common file extensions that we don't want included in the code directories * Change package-data test to be more complete * Now compares the repository, sdist, and install * Compares both that everything in the sdist is in the repo and everything in the install is in the sdist in addition to comparing that everything in the repo that we want is in the install * Leave out test artifacts Only include the directory structure for test/results and test/cache not any files that may have been generated by test runs Remove test/utils files from the sdist as these are only needed for our CI cleanup of docs in MANIFEST.in; getting rid of build files. * Add the ability to output sdist and snapshot to specific directory * Add a warning about modifying the heuristic to setup.py * Address generated files * Use make snapshot instead of sdist to generate changelog and man pages and make sure they're included * Ignore both the test/utils and generated test files (results, cache) * Deal with Python3 __pycache__ byte code caches * Don't check documentation, that isn't built for the sdist * Restructure for clarity * Add cli web docs to make clean This was causing problems when attempting to test that the sdist didn't have extra files * Fix bug constructing python names from __pycache__ names * Create a clean repo to work from * Exclude test/legacy and be more explicit on extensions * Exclude the legacy directory from sdist
5 years ago
library_symlinks = _find_symlinks('lib', '.py')
library_symlinks.update(_find_symlinks('test/lib'))
symlinks = {'script': _find_symlinks('bin'),
Install ansible-test (#60718) * Install ansible-test Modify the install script to install ansible-test and its supporting code. Alternative to #60701 that doesn't change package_dir ansible for fear that it might regress https://github.com/ansible/ansible/issues/10437 Also: * No longer use package_data. Everything in the package dirs is going to be installed. Anything that shouldn't be installed needs to be moved elsewhere. * modify the algorithm to store symlinks which are in the same tree instead of same directory * Add ansible_test files to package-data sanity test * MANIFEST.in cleanups * Add lib/ansible/config/*.yml * Make most things in code directories (lib/ansible and test/lib/ansible_test/) use explicit file extensions instead of wildcards for maintainability * Exclude common file extensions that we don't want included in the code directories * Change package-data test to be more complete * Now compares the repository, sdist, and install * Compares both that everything in the sdist is in the repo and everything in the install is in the sdist in addition to comparing that everything in the repo that we want is in the install * Leave out test artifacts Only include the directory structure for test/results and test/cache not any files that may have been generated by test runs Remove test/utils files from the sdist as these are only needed for our CI cleanup of docs in MANIFEST.in; getting rid of build files. * Add the ability to output sdist and snapshot to specific directory * Add a warning about modifying the heuristic to setup.py * Address generated files * Use make snapshot instead of sdist to generate changelog and man pages and make sure they're included * Ignore both the test/utils and generated test files (results, cache) * Deal with Python3 __pycache__ byte code caches * Don't check documentation, that isn't built for the sdist * Restructure for clarity * Add cli web docs to make clean This was causing problems when attempting to test that the sdist didn't have extra files * Fix bug constructing python names from __pycache__ names * Create a clean repo to work from * Exclude test/legacy and be more explicit on extensions * Exclude the legacy directory from sdist
5 years ago
'library': library_symlinks,
}
_cache_symlinks(symlinks)
SDist.run(self)
# Print warnings at the end because no one will see warnings before all the normal status
# output
if os.environ.get('_ANSIBLE_SDIST_FROM_MAKEFILE', False) != '1':
warnings.warn('When setup.py sdist is run from outside of the Makefile,'
' the generated tarball may be incomplete. Use `make snapshot`'
' to create a tarball from an arbitrary checkout or use'
' `cd packaging/release && make release version=[..]` for official builds.',
RuntimeWarning)
def read_file(file_name):
"""Read file and return its contents."""
with open(file_name, 'r') as f:
return f.read()
def read_requirements(file_name):
"""Read requirements file as a list."""
reqs = read_file(file_name).splitlines()
if not reqs:
raise RuntimeError(
"Unable to read requirements from the %s file"
"That indicates this copy of the source code is incomplete."
% file_name
)
return reqs
PYCRYPTO_DIST = 'pycrypto'
def get_crypto_req():
"""Detect custom crypto from ANSIBLE_CRYPTO_BACKEND env var.
pycrypto or cryptography. We choose a default but allow the user to
override it. This translates into pip install of the sdist deciding what
package to install and also the runtime dependencies that pkg_resources
knows about.
"""
crypto_backend = os.environ.get('ANSIBLE_CRYPTO_BACKEND', '').strip()
if crypto_backend == PYCRYPTO_DIST:
Cyptography pr 20566 rebase (#25560) Make pyca/cryptography the preferred backend for cryptographic needs (mainly vault) falling back to pycrypto pyca/cryptography is already implicitly a dependency in many cases through paramiko (2.0+) as well as the new openssl_publickey module, which requires pyOpenSSL 16.0+. Additionally, pyca/cryptography is an optional dep for better performance with vault already. This commit leverages cryptography's padding, constant time comparisons, and CBC/CTR modes to reduce the amount of code ansible needs to maintain. * Handle wrong password given for VaultAES format * Do not display deprecation warning for cryptography on python-2.6 * Namespace all of the pycrypto imports and always import them Makes unittests better and the code less likely to get stupid mistakes (like using HMAC from cryptogrpahy when the one from pycrypto is needed) * Add back in atfork since we need pycrypto to reinitialize its RNG just in case we're being used with old paramiko * contrib/inventory/gce: Remove spurious require on pycrypto (cherry picked from commit 9e16b9db275263b3ea8d1b124966fdebfc9ab271) * Add cryptography to ec2_win_password module requirements * Fix python3 bug which would pass text strings to a function which requires byte strings. * Attempt to add pycrypto version to setup deps * Change hacking README for dual pycrypto/cryptography * update dependencies for various CI scripts * additional CI dockerfile/script updates * add paramiko to the windows and sanity requirement set This is needed because ansible lists it as a requirement. Previously the missing dep wasn't enforced, but cryptography imports pkg_resources so you can't ignore a requirement any more * Add integration test cases for old vault and for wrong passwords * helper script for manual testing of pycrypto/cryptography * Skip the pycrypto tests so that users without it installed can still run the unittests * Run unittests for vault with both cryptography and pycrypto backend
8 years ago
# Attempt to set version requirements
return '%s >= 2.6' % PYCRYPTO_DIST
return crypto_backend or None
def substitute_crypto_to_req(req):
"""Replace crypto requirements if customized."""
crypto_backend = get_crypto_req()
if crypto_backend is None:
return req
def is_not_crypto(r):
CRYPTO_LIBS = PYCRYPTO_DIST, 'cryptography'
return not any(r.lower().startswith(c) for c in CRYPTO_LIBS)
return [r for r in req if is_not_crypto(r)] + [crypto_backend]
Cyptography pr 20566 rebase (#25560) Make pyca/cryptography the preferred backend for cryptographic needs (mainly vault) falling back to pycrypto pyca/cryptography is already implicitly a dependency in many cases through paramiko (2.0+) as well as the new openssl_publickey module, which requires pyOpenSSL 16.0+. Additionally, pyca/cryptography is an optional dep for better performance with vault already. This commit leverages cryptography's padding, constant time comparisons, and CBC/CTR modes to reduce the amount of code ansible needs to maintain. * Handle wrong password given for VaultAES format * Do not display deprecation warning for cryptography on python-2.6 * Namespace all of the pycrypto imports and always import them Makes unittests better and the code less likely to get stupid mistakes (like using HMAC from cryptogrpahy when the one from pycrypto is needed) * Add back in atfork since we need pycrypto to reinitialize its RNG just in case we're being used with old paramiko * contrib/inventory/gce: Remove spurious require on pycrypto (cherry picked from commit 9e16b9db275263b3ea8d1b124966fdebfc9ab271) * Add cryptography to ec2_win_password module requirements * Fix python3 bug which would pass text strings to a function which requires byte strings. * Attempt to add pycrypto version to setup deps * Change hacking README for dual pycrypto/cryptography * update dependencies for various CI scripts * additional CI dockerfile/script updates * add paramiko to the windows and sanity requirement set This is needed because ansible lists it as a requirement. Previously the missing dep wasn't enforced, but cryptography imports pkg_resources so you can't ignore a requirement any more * Add integration test cases for old vault and for wrong passwords * helper script for manual testing of pycrypto/cryptography * Skip the pycrypto tests so that users without it installed can still run the unittests * Run unittests for vault with both cryptography and pycrypto backend
8 years ago
def read_extras():
"""Specify any extra requirements for installation."""
extras = dict()
extra_requirements_dir = 'packaging/requirements'
for extra_requirements_filename in os.listdir(extra_requirements_dir):
filename_match = re.search(r'^requirements-(\w*).txt$', extra_requirements_filename)
if not filename_match:
continue
extra_req_file_path = os.path.join(extra_requirements_dir, extra_requirements_filename)
try:
extras[filename_match.group(1)] = read_file(extra_req_file_path).splitlines()
except RuntimeError:
pass
return extras
def get_dynamic_setup_params():
"""Add dynamically calculated setup params to static ones."""
return {
# Retrieve the long description from the README
'long_description': read_file('README.rst'),
'install_requires': substitute_crypto_to_req(
read_requirements('requirements.txt'),
),
'extras_require': read_extras(),
}
static_setup_params = dict(
# Use the distutils SDist so that symlinks are not expanded
# Use a custom Build for the same reason
cmdclass={
'build_py': BuildPyCommand,
'build_scripts': BuildScriptsCommand,
'install_lib': InstallLibCommand,
'install_scripts': InstallScriptsCommand,
'sdist': SDistCommand,
},
name='ansible',
version=__version__,
description='Radically simple IT automation',
author=__author__,
author_email='info@ansible.com',
url='https://ansible.com/',
project_urls={
'Bug Tracker': 'https://github.com/ansible/ansible/issues',
'CI: Shippable': 'https://app.shippable.com/github/ansible/ansible',
'Code of Conduct': 'https://docs.ansible.com/ansible/latest/community/code_of_conduct.html',
'Documentation': 'https://docs.ansible.com/ansible/',
'Mailing lists': 'https://docs.ansible.com/ansible/latest/community/communication.html#mailing-list-information',
'Source Code': 'https://github.com/ansible/ansible',
},
license='GPLv3+',
# Ansible will also make use of a system copy of python-six and
# python-selectors2 if installed but use a Bundled copy if it's not.
python_requires='>=2.7,!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*',
Install ansible-test (#60718) * Install ansible-test Modify the install script to install ansible-test and its supporting code. Alternative to #60701 that doesn't change package_dir ansible for fear that it might regress https://github.com/ansible/ansible/issues/10437 Also: * No longer use package_data. Everything in the package dirs is going to be installed. Anything that shouldn't be installed needs to be moved elsewhere. * modify the algorithm to store symlinks which are in the same tree instead of same directory * Add ansible_test files to package-data sanity test * MANIFEST.in cleanups * Add lib/ansible/config/*.yml * Make most things in code directories (lib/ansible and test/lib/ansible_test/) use explicit file extensions instead of wildcards for maintainability * Exclude common file extensions that we don't want included in the code directories * Change package-data test to be more complete * Now compares the repository, sdist, and install * Compares both that everything in the sdist is in the repo and everything in the install is in the sdist in addition to comparing that everything in the repo that we want is in the install * Leave out test artifacts Only include the directory structure for test/results and test/cache not any files that may have been generated by test runs Remove test/utils files from the sdist as these are only needed for our CI cleanup of docs in MANIFEST.in; getting rid of build files. * Add the ability to output sdist and snapshot to specific directory * Add a warning about modifying the heuristic to setup.py * Address generated files * Use make snapshot instead of sdist to generate changelog and man pages and make sure they're included * Ignore both the test/utils and generated test files (results, cache) * Deal with Python3 __pycache__ byte code caches * Don't check documentation, that isn't built for the sdist * Restructure for clarity * Add cli web docs to make clean This was causing problems when attempting to test that the sdist didn't have extra files * Fix bug constructing python names from __pycache__ names * Create a clean repo to work from * Exclude test/legacy and be more explicit on extensions * Exclude the legacy directory from sdist
5 years ago
package_dir={'': 'lib',
'ansible_test': 'test/lib/ansible_test'},
packages=find_packages('lib') + find_packages('test/lib'),
include_package_data=True,
classifiers=[
'Development Status :: 5 - Production/Stable',
'Environment :: Console',
'Intended Audience :: Developers',
'Intended Audience :: Information Technology',
'Intended Audience :: System Administrators',
'License :: OSI Approved :: GNU General Public License v3 or later (GPLv3+)',
'Natural Language :: English',
'Operating System :: POSIX',
'Programming Language :: Python :: 2',
'Programming Language :: Python :: 2.7',
'Programming Language :: Python :: 3',
'Programming Language :: Python :: 3.5',
'Programming Language :: Python :: 3.6',
'Programming Language :: Python :: 3.7',
'Topic :: System :: Installation/Setup',
'Topic :: System :: Systems Administration',
'Topic :: Utilities',
],
scripts=[
'bin/ansible',
'bin/ansible-playbook',
'bin/ansible-pull',
'bin/ansible-doc',
'bin/ansible-galaxy',
'bin/ansible-console',
'bin/ansible-connection',
'bin/ansible-vault',
'bin/ansible-config',
'bin/ansible-inventory',
Install ansible-test (#60718) * Install ansible-test Modify the install script to install ansible-test and its supporting code. Alternative to #60701 that doesn't change package_dir ansible for fear that it might regress https://github.com/ansible/ansible/issues/10437 Also: * No longer use package_data. Everything in the package dirs is going to be installed. Anything that shouldn't be installed needs to be moved elsewhere. * modify the algorithm to store symlinks which are in the same tree instead of same directory * Add ansible_test files to package-data sanity test * MANIFEST.in cleanups * Add lib/ansible/config/*.yml * Make most things in code directories (lib/ansible and test/lib/ansible_test/) use explicit file extensions instead of wildcards for maintainability * Exclude common file extensions that we don't want included in the code directories * Change package-data test to be more complete * Now compares the repository, sdist, and install * Compares both that everything in the sdist is in the repo and everything in the install is in the sdist in addition to comparing that everything in the repo that we want is in the install * Leave out test artifacts Only include the directory structure for test/results and test/cache not any files that may have been generated by test runs Remove test/utils files from the sdist as these are only needed for our CI cleanup of docs in MANIFEST.in; getting rid of build files. * Add the ability to output sdist and snapshot to specific directory * Add a warning about modifying the heuristic to setup.py * Address generated files * Use make snapshot instead of sdist to generate changelog and man pages and make sure they're included * Ignore both the test/utils and generated test files (results, cache) * Deal with Python3 __pycache__ byte code caches * Don't check documentation, that isn't built for the sdist * Restructure for clarity * Add cli web docs to make clean This was causing problems when attempting to test that the sdist didn't have extra files * Fix bug constructing python names from __pycache__ names * Create a clean repo to work from * Exclude test/legacy and be more explicit on extensions * Exclude the legacy directory from sdist
5 years ago
'bin/ansible-test',
],
data_files=[],
# Installing as zip files would break due to references to __file__
zip_safe=False
13 years ago
)
def main():
"""Invoke installation process using setuptools."""
setup_params = dict(static_setup_params, **get_dynamic_setup_params())
ignore_warning_regex = (
r"Unknown distribution option: '(project_urls|python_requires)'"
)
warnings.filterwarnings(
'ignore',
message=ignore_warning_regex,
category=UserWarning,
module='distutils.dist',
)
setup(**setup_params)
warnings.resetwarnings()
if __name__ == '__main__':
main()