这是一个适合我的 setup.py:
# pkg - A fancy software package
# Copyright (C) 2013 author (email)
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see http://www.gnu.org/licenses/gpl.html.
"""pkg: a software suite for
Hey look at me I'm a long description
But how long am I?
"""
from __future__ import division, print_function
#ideas for setup/f2py came from:
# -numpy setup.py: https://github.com/numpy/numpy/blob/master/setup.py 2013-11-07
# -winpython setup.py: http://code.google.com/p/winpython/source/browse/setup.py 2013-11-07
# -needing to use
# import setuptools; from numpy.distutils.core import setup, Extension:
# http://comments.gmane.org/gmane.comp.python.f2py.user/707 2013-11-07
# -wrapping FORTRAN code with f2py: http://www2-pcmdi.llnl.gov/cdat/tutorials/f2py-wrapping-fortran-code 2013-11-07
# -numpy disutils: http://docs.scipy.org/doc/numpy/reference/distutils.html 2013-11-07
# -manifest files in disutils:
# 'distutils doesn't properly update MANIFEST. when the contents of directories change.'
# https://github.com/numpy/numpy/blob/master/setup.py
# -if things are not woring try deleting build, sdist, egg directories and try again:
# https://stackoverflow.com/a/9982133/2530083 2013-11-07
# -getting fortran extensions to be installed in their appropriate sub package
# i.e. "my_ext = Extension(name = 'my_pack._fortran', sources = ['my_pack/code.f90'])"
# Note that sources is a list even if one file:
# http://numpy-discussion.10968.n7.nabble.com/f2py-and-setup-py-how-can-I-specify-where-the-so-file-goes-tp34490p34497.html 2013-11-07
# -install fortran source files into their appropriate sub-package
# i.e. "package_data={'': ['*.f95','*.f90']}# Note it's a dict and list":
# https://stackoverflow.com/a/19373744/2530083 2013-11-07
# -Chapter 9 Fortran Programming with NumPy Arrays:
# Langtangen, Hans Petter. 2013. Python Scripting for Computational Science. 3rd edition. Springer.
# -Hitchhikers guide to packaging :
# http://guide.python-distribute.org/
# -Python Packaging: Hate, hate, hate everywhere :
# http://lucumr.pocoo.org/2012/6/22/hate-hate-hate-everywhere/
# -How To Package Your Python Code:
# http://www.scotttorborg.com/python-packaging/
# -install testing requirements:
# https://stackoverflow.com/a/7747140/2530083 2013-11-07
import setuptools
from numpy.distutils.core import setup, Extension
import os
import os.path as osp
def readme(filename='README.rst'):
with open('README.rst') as f:
text=f.read()
f.close()
return text
def get_package_data(name, extlist):
"""Return data files for package *name* with extensions in *extlist*"""
#modified slightly from taken from http://code.google.com/p/winpython/source/browse/setup.py 2013-11-7
flist = []
# Workaround to replace os.path.relpath (not available until Python 2.6):
offset = len(name)+len(os.pathsep)
for dirpath, _dirnames, filenames in os.walk(name):
for fname in filenames:
if not fname.startswith('.') and osp.splitext(fname)[1] in extlist:
# flist.append(osp.join(dirpath, fname[offset:]))
flist.append(osp.join(dirpath, fname))
return flist
DOCLINES = __doc__.split("\n")
CLASSIFIERS = """\
Development Status :: 1 - Planning
License :: OSI Approved :: GNU Lesser General Public License v3 or later (LGPLv3+)
Programming Language :: Python :: 2.7
Topic :: Scientific/Engineering
"""
NAME = 'pkg'
MAINTAINER = "me"
MAINTAINER_EMAIL = "me@me.com"
DESCRIPTION = DOCLINES[0]
LONG_DESCRIPTION = "\n".join(DOCLINES[2:])#readme('readme.rst')
URL = "http://meeeee.mmemem"
DOWNLOAD_URL = "https://github.com/rtrwalker/geotecha.git"
LICENSE = 'GNU General Public License v3 or later (GPLv3+)'
CLASSIFIERS = [_f for _f in CLASSIFIERS.split('\n') if _f]
KEYWORDS=''
AUTHOR = "me"
AUTHOR_EMAIL = "me.com"
PLATFORMS = ["Windows"]#, "Linux", "Solaris", "Mac OS-X", "Unix"]
MAJOR = 0
MINOR = 1
MICRO = 0
ISRELEASED = False
VERSION = '%d.%d.%d' % (MAJOR, MINOR, MICRO)
INSTALL_REQUIRES=[]
ZIP_SAFE=False
TEST_SUITE='nose.collector'
TESTS_REQUIRE=['nose']
DATA_FILES = [(NAME, ['LICENSE.txt','README.rst'])]
PACKAGES=setuptools.find_packages()
PACKAGES.remove('tools')
PACKAGE_DATA={'': ['*.f95','*f90']}
ext_files = get_package_data(NAME,['.f90', '.f95','.F90', '.F95'])
ext_module_names = ['.'.join(osp.splitext(v)[0].split(osp.sep)) for v in ext_files]
EXT_MODULES = [Extension(name=x,sources=[y]) for x, y in zip(ext_module_names, ext_files)]
setup(
name=NAME,
version=VERSION,
maintainer=MAINTAINER,
maintainer_email=MAINTAINER_EMAIL,
description=DESCRIPTION,
long_description=LONG_DESCRIPTION,
url=URL,
download_url=DOWNLOAD_URL,
license=LICENSE,
classifiers=CLASSIFIERS,
author=AUTHOR,
author_email=AUTHOR_EMAIL,
platforms=PLATFORMS,
packages=PACKAGES,
data_files=DATA_FILES,
install_requires=INSTALL_REQUIRES,
zip_safe=ZIP_SAFE,
test_suite=TEST_SUITE,
tests_require=TESTS_REQUIRE,
package_data=PACKAGE_DATA,
ext_modules=EXT_MODULES,
)
要安装,在命令行我使用:
python setup.py install
python setup.py clean --all
我似乎唯一的问题是一个小问题。当我在 site-packages 中查找我的包时,它安装在 egg 文件夹C:\Python27\Lib\site-packages\pkg-0.1.0-py2.7-win32.egg\pkg
中。我看到的大多数其他软件包都有一个C:\Python27\Lib\site-packages\pkg
与 egg 文件夹分开的文件夹。有谁知道如何获得这种分离?
至于测试,安装后,我在命令行输入以下内容:
nosetests package_name -v
尝试调查python setup.py develop
(Python setup.py develop 与 install)不必在每次更改后安装包。
正如我在代码中评论的那样,我发现以下有用: