cmdclass在Pythons中做什么？

Question

我最近收到了一个请求，它增加了

class build_ext(_build_ext):
    'to install numpy'
    def finalize_options(self):
        _build_ext.finalize_options(self)
        # Prevent numpy from thinking it is still in its setup process:
        __builtins__.__NUMPY_SETUP__ = False
        import numpy
        self.include_dirs.append(numpy.get_include())

到我的setup.py，结果是：

from setuptools.command.build_ext import build_ext as _build_ext

try:
    from setuptools import setup
except ImportError:
    from distutils.core import setup


class build_ext(_build_ext):
    'to install numpy'
    def finalize_options(self):
        _build_ext.finalize_options(self)
        # Prevent numpy from thinking it is still in its setup process:
        __builtins__.__NUMPY_SETUP__ = False
        import numpy
        self.include_dirs.append(numpy.get_include())


config = {
    'cmdclass':{'build_ext':build_ext}, #numpy hack
    'setup_requires':['numpy'],         #numpy hack
    'name': 'nntoolkit',
    'version': '0.1.25',
    'author': 'Martin Thoma',
    'author_email': 'info@martin-thoma.de',
    'packages': ['nntoolkit'],
    'scripts': ['bin/nntoolkit'],
    'url': 'https://github.com/MartinThoma/nntoolkit',
    'license': 'MIT',
    'description': 'Neural Network Toolkit',
    'long_description': """...""",
    'install_requires': [
        "argparse",
        "theano",
        "nose",
        "natsort",
        "PyYAML",
        "matplotlib",
        "h5py",
        "numpy",
        "Cython"
    ],
    'keywords': ['Neural Networks', 'Feed-Forward', 'NN', 'MLP'],
    'download_url': 'https://github.com/MartinThoma/nntoolkit',
    'classifiers': ['Development Status :: 3 - Alpha'],
    'zip_safe': False,
    'test_suite': 'nose.collector'
}

setup(**config)

是干什么的呢？

文档只声明：

cmdclass:命令名到命令子类的映射(字典)

Answer 1

Numpy库是用C/C++编写的。因此，与其他包不同的是，它需要在实际调用它们之前进行编译。所以'build_ext‘只是编译它们。

博客详细信息：http://sadafnoor.me/blog/how-to-automate-numpy-installation-in-your-project-using-setuptool/

Answer 2

所有的骗局：

我发现另一个博客很好地解释了这一点：

https://niteo.co/blog/setuptools-run-custom-code-in-setup-py

对studioj也大喊大叫；这一页解释了同样的事情，但是对于distutils而不是setuptools (它们的定义非常相似)。

本质上，它们创建类来覆盖setuptools/distutils使用的各种命令(特别是它们的".run“命令)。

例如，如果要在setuptools中覆盖"build“命令，可以使用如下模式：

from setuptools import setup
from setuptools.command.build import build

class MyBuild(build):
    def run(self):
        # ---- some custom code here ----
        # important to do this instead of "super", since it hasn't been properly updated
        build.run(self) 

setup(
    # setup specifications here
    cmdclass={"build": MyBuild}
)