setup.py

# -*- coding: utf-8 -*-
from setuptools import setup, find_packages, Extension
import os
import sys

__version__ = None #placeholder, will be filled by exec
with open('estnltk/__about__.py', 'r') as about_file:
    exec(about_file.read())
assert __version__ is not None, 'Reading version number from file failed'

os.environ['CC'] = 'g++'
os.environ['CXX'] = 'g++'

def get_sources(src_dir='src', ending='.cpp'):
    """Function to get a list of files ending with `ending` in `src_dir`."""
    return [os.path.join(src_dir, fnm) for fnm in os.listdir(src_dir) if fnm.endswith(ending)]

# define directories for vabamorf source directories
dirs = ['etana', 'etyhh', 'fsc', 'json', 'proof', 'estnltk']
src_dirs = [os.path.join('src', d) for d in dirs]

# define a list of C++ source files
lib_sources = []
vabamorf_src = os.path.join('src', 'estnltk', 'vabamorf.cpp')
for d in src_dirs:
    lib_sources.extend(get_sources(d))
lib_sources.remove(vabamorf_src) # we add it later as the first file to compile
                                 # less time to wait, when working with vabamorf wrapper

# define directories for vabamorf include directories
dirs.append(os.path.join('fsc', 'fsjni'))
include_dirs = [os.path.join('include', d) for d in dirs]

# define the vabamorf SWIG wrapper generator interface file
swig_interface = os.path.join('estnltk', 'vabamorf', 'vabamorf.i')
swig_opts = ['-builtin']

# Python 3 specific configuration
extra = {}
if sys.version_info[0] == 3:
    swig_opts.append('-py3')
swig_opts.append('-c++')

setup(
    name = "estnltk",

    #change the version in file `estnltk/__about__.py`
    version = __version__,

    packages = find_packages(),
    include_package_data=True,
    package_data = {
        'estnltk': ['corpora/arvutustehnika_ja_andmetootlus/*.xml', 'corpora/*.json', 'java-res/*.*'],
        'estnltk.vabamorf': ['dct/*.dct'],
        'estnltk.estner': ['gazetteer/*', 'models/py2_default/*', 'models/py3_default/*'],
        'estnltk.wordnet': ['*.cnf', 'data/*.txt', 'data/*.soi', 'data/*.cnf', 'data/scripts/*.py'],
        'estnltk.mw_verbs': ['res/*'],
        'estnltk.converters': ['*.mrf'],
        'estnltk.syntax': ['files/*']
    },

    author       = "University of Tartu",
    author_email = "siim.orasmaa@gmail.com, alex.tk.fb@gmail.com, tpetmanson@gmail.com, swen@math.ut.ee",
    description  = "Estnltk — open source tools for Estonian natural language processing",
    long_description = '''
Estnltk -- Open source tools for Estonian natural language processing
=====================================================================

Estnltk provides common natural language processing functionality such as paragraph, sentence and word tokenization,
morphological analysis, named entity recognition, etc. for the Estonian language.


    ''',
    license      = "GPLv2",
    url          = "https://github.com/estnltk/estnltk",
    ext_modules = [
        Extension('estnltk.vabamorf._vabamorf',
                  [swig_interface, vabamorf_src] + lib_sources,
                  swig_opts = swig_opts,
                  include_dirs=include_dirs)
        ],

    # we have fixed dependency versions to guarantee, what works
    # however, you can probably safely install newer versions of the dependencies
    install_requires = [
        'six>=1.9.0',                       # helps to build Py2/Py3 compatible programs
        'nltk>=3.1',                        # NLTK mainly used for English
        'regex>=2015.07.19',                # improved Python regular expressions
        'pandas>=0.18',                     # Panel Data Analysis library for Python
        'python-crfsuite>=0.8.4',           # Conditional random fields library
        'cached-property>=1.2.0',           # Simple property for caching results
        'beautifulsoup4>=4.4.0',            # HTML parsing library
        'html5lib>=0.9999999',              # HTML parsing library
        'elasticsearch==2',                 # database support
        'pyahocorasick>=1.1.4',             # fast pattern matching NOTE: we might have to make this an optional dep if we fail to install it on windows.
        'unicodecsv>= 0.14.1',              # Read/write csv files in unicode
    ],

    classifiers = ['Intended Audience :: Developers',
                   'Intended Audience :: Education',
                   'Intended Audience :: Science/Research',
                   'Intended Audience :: Information Technology',
                   'Operating System :: OS Independent',
                   'Programming Language :: Python :: 2.7',
                   'Programming Language :: Python :: 3.5',
                   'Topic :: Scientific/Engineering',
                   'Topic :: Scientific/Engineering :: Artificial Intelligence',
                   'Topic :: Scientific/Engineering :: Information Analysis',
                   'Topic :: Text Processing',
                   'Topic :: Text Processing :: Linguistic']
)