added tests for SpiderState extension
[scrapy.git] / setup.py
blob2072b37df52a121cb5fc6617992f3583cca492f2
1 # Scrapy setup.py script
3 # It doesn't depend on setuptools, but if setuptools is available it'll use
4 # some of its features, like package dependencies.
6 from __future__ import with_statement
8 from distutils.command.install_data import install_data
9 from distutils.command.install import INSTALL_SCHEMES
10 from subprocess import Popen, PIPE
11 import os
12 import sys
14 class osx_install_data(install_data):
15 # On MacOS, the platform-specific lib dir is /System/Library/Framework/Python/.../
16 # which is wrong. Python 2.5 supplied with MacOS 10.5 has an Apple-specific fix
17 # for this in distutils.command.install_data#306. It fixes install_lib but not
18 # install_data, which is why we roll our own install_data class.
20 def finalize_options(self):
21 # By the time finalize_options is called, install.install_lib is set to the
22 # fixed directory, so we set the installdir to install_lib. The
23 # install_data class uses ('install_data', 'install_dir') instead.
24 self.set_undefined_options('install', ('install_lib', 'install_dir'))
25 install_data.finalize_options(self)
27 if sys.platform == "darwin":
28 cmdclasses = {'install_data': osx_install_data}
29 else:
30 cmdclasses = {'install_data': install_data}
32 def fullsplit(path, result=None):
33 """
34 Split a pathname into components (the opposite of os.path.join) in a
35 platform-neutral way.
36 """
37 if result is None:
38 result = []
39 head, tail = os.path.split(path)
40 if head == '':
41 return [tail] + result
42 if head == path:
43 return result
44 return fullsplit(head, [tail] + result)
46 # Tell distutils to put the data_files in platform-specific installation
47 # locations. See here for an explanation:
48 # http://groups.google.com/group/comp.lang.python/browse_thread/thread/35ec7b2fed36eaec/2105ee4d9e8042cb
49 for scheme in INSTALL_SCHEMES.values():
50 scheme['data'] = scheme['purelib']
52 # Compile the list of packages available, because distutils doesn't have
53 # an easy way to do this.
54 packages, data_files = [], []
55 root_dir = os.path.dirname(__file__)
56 if root_dir != '':
57 os.chdir(root_dir)
59 def is_not_module(filename):
60 return os.path.splitext(f)[1] not in ['.py', '.pyc', '.pyo']
62 for scrapy_dir in ['scrapy', 'scrapyd']:
63 for dirpath, dirnames, filenames in os.walk(scrapy_dir):
64 # Ignore dirnames that start with '.'
65 for i, dirname in enumerate(dirnames):
66 if dirname.startswith('.'): del dirnames[i]
67 if '__init__.py' in filenames:
68 packages.append('.'.join(fullsplit(dirpath)))
69 data = [f for f in filenames if is_not_module(f)]
70 if data:
71 data_files.append([dirpath, [os.path.join(dirpath, f) for f in data]])
72 elif filenames:
73 data_files.append([dirpath, [os.path.join(dirpath, f) for f in filenames]])
75 # Small hack for working with bdist_wininst.
76 # See http://mail.python.org/pipermail/distutils-sig/2004-August/004134.html
77 if len(sys.argv) > 1 and sys.argv[1] == 'bdist_wininst':
78 for file_info in data_files:
79 file_info[0] = '\\PURELIB\\%s' % file_info[0]
81 scripts = ['bin/scrapy']
82 if os.name == 'nt':
83 scripts.append('extras/scrapy.bat')
85 if os.environ.get('SCRAPY_VERSION_FROM_HG'):
86 rev = Popen(["hg", "tip", "--template", "{rev}"], stdout=PIPE).communicate()[0]
87 with open('scrapy/__init__.py', 'a') as f:
88 f.write("\n__version__ = '.'.join(map(str, version_info)) + '.%s'" % rev)
89 elif os.environ.get('SCRAPY_VERSION_FROM_GIT'):
90 rev = Popen("git log --oneline | wc -l", shell=True, stdout=PIPE).communicate()[0]
91 with open('scrapy/__init__.py', 'a') as f:
92 f.write("\n__version__ = '.'.join(map(str, version_info)) + '.%s'" % rev.strip())
93 version = __import__('scrapy').__version__
95 setup_args = {
96 'name': 'Scrapy',
97 'version': version,
98 'url': 'http://scrapy.org',
99 'description': 'A high-level Python Screen Scraping framework',
100 'long_description': 'Scrapy is a high level scraping and web crawling framework for writing spiders to crawl and parse web pages for all kinds of purposes, from information retrieval to monitoring or testing web sites.',
101 'author': 'Scrapy developers',
102 'maintainer': 'Pablo Hoffman',
103 'maintainer_email': 'pablo@pablohoffman.com',
104 'license': 'BSD',
105 'packages': packages,
106 'cmdclass': cmdclasses,
107 'data_files': data_files,
108 'scripts': scripts,
109 'classifiers': [
110 'Programming Language :: Python',
111 'Programming Language :: Python :: 2.5',
112 'Programming Language :: Python :: 2.6',
113 'Programming Language :: Python :: 2.7',
114 'License :: OSI Approved :: BSD License',
115 'Operating System :: OS Independent',
116 'Development Status :: 5 - Production/Stable',
117 'Intended Audience :: Developers',
118 'Environment :: Console',
119 'Topic :: Software Development :: Libraries :: Application Frameworks',
120 'Topic :: Software Development :: Libraries :: Python Modules',
121 'Topic :: Internet :: WWW/HTTP',
125 try:
126 from setuptools import setup
127 except ImportError:
128 from distutils.core import setup
129 else:
130 setup_args['install_requires'] = ['Twisted>=2.5', 'w3lib', 'pyOpenSSL']
131 if sys.version_info < (2, 6):
132 setup_args['install_requires'] += ['simplejson']
133 try:
134 import libxml2
135 except ImportError:
136 setup_args['install_requires'] += ['lxml']
138 setup(**setup_args)