2014-07-05 22:49:34 +04:00
|
|
|
#!/usr/bin/env python
|
2020-07-06 18:39:31 +03:00
|
|
|
from setuptools import Extension, setup, find_packages
|
2014-07-05 22:49:34 +04:00
|
|
|
import sys
|
2020-04-29 13:51:49 +03:00
|
|
|
import platform
|
2020-11-23 13:15:11 +03:00
|
|
|
import numpy
|
2015-10-13 01:31:59 +03:00
|
|
|
from distutils.command.build_ext import build_ext
|
2015-12-13 13:49:17 +03:00
|
|
|
from distutils.sysconfig import get_python_inc
|
2020-02-12 01:46:18 +03:00
|
|
|
from pathlib import Path
|
2020-02-25 17:46:39 +03:00
|
|
|
import shutil
|
2020-02-12 01:46:18 +03:00
|
|
|
from Cython.Build import cythonize
|
|
|
|
from Cython.Compiler import Options
|
2020-07-20 15:58:04 +03:00
|
|
|
import os
|
|
|
|
import subprocess
|
2015-12-13 13:49:17 +03:00
|
|
|
|
|
|
|
|
2020-02-25 17:46:39 +03:00
|
|
|
ROOT = Path(__file__).parent
|
|
|
|
PACKAGE_ROOT = ROOT / "spacy"
|
2018-12-04 02:06:42 +03:00
|
|
|
|
2015-12-13 13:49:17 +03:00
|
|
|
|
2020-02-12 01:46:18 +03:00
|
|
|
# Preserve `__doc__` on functions and classes
|
|
|
|
# http://docs.cython.org/en/latest/src/userguide/source_files_and_compilation.html#compiler-options
|
|
|
|
Options.docstrings = True
|
2015-12-13 13:49:17 +03:00
|
|
|
|
2017-05-03 21:10:59 +03:00
|
|
|
PACKAGES = find_packages()
|
2015-12-13 13:49:17 +03:00
|
|
|
MOD_NAMES = [
|
2022-04-01 10:02:06 +03:00
|
|
|
"spacy.training.alignment_array",
|
2020-09-09 11:31:03 +03:00
|
|
|
"spacy.training.example",
|
2018-11-27 00:04:35 +03:00
|
|
|
"spacy.parts_of_speech",
|
|
|
|
"spacy.strings",
|
|
|
|
"spacy.lexeme",
|
|
|
|
"spacy.vocab",
|
|
|
|
"spacy.attrs",
|
2022-09-08 11:38:07 +03:00
|
|
|
"spacy.kb.candidate",
|
|
|
|
"spacy.kb.kb",
|
|
|
|
"spacy.kb.kb_in_memory",
|
2020-07-31 00:30:54 +03:00
|
|
|
"spacy.ml.parser_model",
|
2018-11-27 00:04:35 +03:00
|
|
|
"spacy.morphology",
|
2020-07-22 14:42:59 +03:00
|
|
|
"spacy.pipeline.dep_parser",
|
2022-03-28 12:13:50 +03:00
|
|
|
"spacy.pipeline._edit_tree_internals.edit_trees",
|
2019-03-07 12:45:55 +03:00
|
|
|
"spacy.pipeline.morphologizer",
|
2020-07-22 14:42:59 +03:00
|
|
|
"spacy.pipeline.multitask",
|
|
|
|
"spacy.pipeline.ner",
|
|
|
|
"spacy.pipeline.pipe",
|
2020-10-08 22:33:49 +03:00
|
|
|
"spacy.pipeline.trainable_pipe",
|
2020-07-22 14:42:59 +03:00
|
|
|
"spacy.pipeline.sentencizer",
|
|
|
|
"spacy.pipeline.senter",
|
|
|
|
"spacy.pipeline.tagger",
|
2020-07-31 00:30:54 +03:00
|
|
|
"spacy.pipeline.transition_parser",
|
|
|
|
"spacy.pipeline._parser_internals.arc_eager",
|
|
|
|
"spacy.pipeline._parser_internals.ner",
|
|
|
|
"spacy.pipeline._parser_internals.nonproj",
|
|
|
|
"spacy.pipeline._parser_internals._state",
|
|
|
|
"spacy.pipeline._parser_internals.stateclass",
|
|
|
|
"spacy.pipeline._parser_internals.transition_system",
|
2020-12-13 04:08:32 +03:00
|
|
|
"spacy.pipeline._parser_internals._beam_utils",
|
2018-11-27 00:04:35 +03:00
|
|
|
"spacy.tokenizer",
|
2020-11-03 18:24:38 +03:00
|
|
|
"spacy.training.align",
|
2020-09-09 11:31:03 +03:00
|
|
|
"spacy.training.gold_io",
|
2018-11-27 00:04:35 +03:00
|
|
|
"spacy.tokens.doc",
|
|
|
|
"spacy.tokens.span",
|
|
|
|
"spacy.tokens.token",
|
2021-01-14 09:30:41 +03:00
|
|
|
"spacy.tokens.span_group",
|
|
|
|
"spacy.tokens.graph",
|
2019-03-07 16:34:54 +03:00
|
|
|
"spacy.tokens.morphanalysis",
|
2018-11-27 00:04:35 +03:00
|
|
|
"spacy.tokens._retokenize",
|
2019-02-07 11:42:25 +03:00
|
|
|
"spacy.matcher.matcher",
|
|
|
|
"spacy.matcher.phrasematcher",
|
|
|
|
"spacy.matcher.dependencymatcher",
|
2018-11-27 00:04:35 +03:00
|
|
|
"spacy.symbols",
|
|
|
|
"spacy.vectors",
|
2017-06-05 13:45:29 +03:00
|
|
|
]
|
2018-11-27 00:04:35 +03:00
|
|
|
COMPILE_OPTIONS = {
|
|
|
|
"msvc": ["/Ox", "/EHsc"],
|
|
|
|
"mingw32": ["-O2", "-Wno-strict-prototypes", "-Wno-unused-function"],
|
|
|
|
"other": ["-O2", "-Wno-strict-prototypes", "-Wno-unused-function"],
|
2016-02-05 16:43:52 +03:00
|
|
|
}
|
2021-01-14 09:30:41 +03:00
|
|
|
LINK_OPTIONS = {"msvc": ["-std=c++11"], "mingw32": ["-std=c++11"], "other": []}
|
2020-02-12 01:46:18 +03:00
|
|
|
COMPILER_DIRECTIVES = {
|
|
|
|
"language_level": -3,
|
|
|
|
"embedsignature": True,
|
|
|
|
"annotation_typing": False,
|
|
|
|
}
|
2020-02-25 17:46:39 +03:00
|
|
|
# Files to copy into the package that are otherwise not included
|
|
|
|
COPY_FILES = {
|
|
|
|
ROOT / "setup.cfg": PACKAGE_ROOT / "tests" / "package",
|
|
|
|
ROOT / "pyproject.toml": PACKAGE_ROOT / "tests" / "package",
|
|
|
|
ROOT / "requirements.txt": PACKAGE_ROOT / "tests" / "package",
|
|
|
|
}
|
2015-10-13 01:31:59 +03:00
|
|
|
|
2016-02-05 16:43:52 +03:00
|
|
|
|
2020-02-12 01:46:18 +03:00
|
|
|
def is_new_osx():
|
2020-04-29 13:51:49 +03:00
|
|
|
"""Check whether we're on OSX >= 10.7"""
|
2020-02-12 01:46:18 +03:00
|
|
|
if sys.platform != "darwin":
|
|
|
|
return False
|
2020-04-29 13:51:49 +03:00
|
|
|
mac_ver = platform.mac_ver()[0]
|
|
|
|
if mac_ver.startswith("10"):
|
2020-07-01 17:46:56 +03:00
|
|
|
minor_version = int(mac_ver.split(".")[1])
|
2020-02-12 01:46:18 +03:00
|
|
|
if minor_version >= 7:
|
|
|
|
return True
|
|
|
|
else:
|
|
|
|
return False
|
2020-04-29 13:51:49 +03:00
|
|
|
return False
|
2016-02-05 16:43:52 +03:00
|
|
|
|
2016-12-20 13:05:06 +03:00
|
|
|
|
2018-12-04 02:06:42 +03:00
|
|
|
if is_new_osx():
|
|
|
|
# On Mac, use libc++ because Apple deprecated use of
|
|
|
|
# libstdc
|
|
|
|
COMPILE_OPTIONS["other"].append("-stdlib=libc++")
|
|
|
|
LINK_OPTIONS["other"].append("-lc++")
|
|
|
|
# g++ (used by unix compiler on mac) links to libstdc++ as a default lib.
|
|
|
|
# See: https://stackoverflow.com/questions/1653047/avoid-linking-to-libstdc
|
|
|
|
LINK_OPTIONS["other"].append("-nodefaultlibs")
|
2016-02-05 16:43:52 +03:00
|
|
|
|
2016-12-20 13:05:06 +03:00
|
|
|
|
2016-04-19 20:50:42 +03:00
|
|
|
# By subclassing build_extensions we have the actual compiler that will be used which is really known only after finalize_options
|
|
|
|
# http://stackoverflow.com/questions/724664/python-distutils-how-to-get-a-compiler-that-is-going-to-be-used
|
2015-12-13 13:49:17 +03:00
|
|
|
class build_ext_options:
|
|
|
|
def build_options(self):
|
|
|
|
for e in self.extensions:
|
2016-11-06 13:58:26 +03:00
|
|
|
e.extra_compile_args += COMPILE_OPTIONS.get(
|
2018-11-27 00:04:35 +03:00
|
|
|
self.compiler.compiler_type, COMPILE_OPTIONS["other"]
|
|
|
|
)
|
2015-12-13 13:49:17 +03:00
|
|
|
for e in self.extensions:
|
2016-11-06 13:58:26 +03:00
|
|
|
e.extra_link_args += LINK_OPTIONS.get(
|
2018-11-27 00:04:35 +03:00
|
|
|
self.compiler.compiler_type, LINK_OPTIONS["other"]
|
|
|
|
)
|
2015-01-17 08:19:54 +03:00
|
|
|
|
|
|
|
|
2015-12-13 13:49:17 +03:00
|
|
|
class build_ext_subclass(build_ext, build_ext_options):
|
|
|
|
def build_extensions(self):
|
2022-07-08 20:21:17 +03:00
|
|
|
if self.parallel is None and os.environ.get("SPACY_NUM_BUILD_JOBS") is not None:
|
|
|
|
self.parallel = int(os.environ.get("SPACY_NUM_BUILD_JOBS"))
|
2015-12-13 13:49:17 +03:00
|
|
|
build_ext_options.build_options(self)
|
|
|
|
build_ext.build_extensions(self)
|
2015-01-17 08:19:54 +03:00
|
|
|
|
|
|
|
|
2020-07-02 18:10:27 +03:00
|
|
|
# Include the git version in the build (adapted from NumPy)
|
|
|
|
# Copyright (c) 2005-2020, NumPy Developers.
|
|
|
|
# BSD 3-Clause license, see licenses/3rd_party_licenses.txt
|
|
|
|
def write_git_info_py(filename="spacy/git_info.py"):
|
|
|
|
def _minimal_ext_cmd(cmd):
|
|
|
|
# construct minimal environment
|
|
|
|
env = {}
|
|
|
|
for k in ["SYSTEMROOT", "PATH", "HOME"]:
|
|
|
|
v = os.environ.get(k)
|
|
|
|
if v is not None:
|
|
|
|
env[k] = v
|
|
|
|
# LANGUAGE is used on win32
|
|
|
|
env["LANGUAGE"] = "C"
|
|
|
|
env["LANG"] = "C"
|
|
|
|
env["LC_ALL"] = "C"
|
|
|
|
out = subprocess.check_output(cmd, stderr=subprocess.STDOUT, env=env)
|
|
|
|
return out
|
|
|
|
|
|
|
|
git_version = "Unknown"
|
2020-07-20 15:58:04 +03:00
|
|
|
if Path(".git").exists():
|
2020-07-02 18:10:27 +03:00
|
|
|
try:
|
|
|
|
out = _minimal_ext_cmd(["git", "rev-parse", "--short", "HEAD"])
|
|
|
|
git_version = out.strip().decode("ascii")
|
2020-07-20 15:58:04 +03:00
|
|
|
except Exception:
|
2020-07-02 18:10:27 +03:00
|
|
|
pass
|
2020-07-20 15:58:04 +03:00
|
|
|
elif Path(filename).exists():
|
2020-07-02 18:10:27 +03:00
|
|
|
# must be a source distribution, use existing version file
|
|
|
|
try:
|
|
|
|
a = open(filename, "r")
|
|
|
|
lines = a.readlines()
|
|
|
|
git_version = lines[-1].split('"')[1]
|
2020-07-20 15:58:04 +03:00
|
|
|
except Exception:
|
2020-07-02 18:10:27 +03:00
|
|
|
pass
|
|
|
|
finally:
|
|
|
|
a.close()
|
|
|
|
|
|
|
|
text = """# THIS FILE IS GENERATED FROM SPACY SETUP.PY
|
|
|
|
#
|
|
|
|
GIT_VERSION = "%(git_version)s"
|
|
|
|
"""
|
|
|
|
a = open(filename, "w")
|
|
|
|
try:
|
2020-07-20 15:58:04 +03:00
|
|
|
a.write(text % {"git_version": git_version})
|
2020-07-02 18:10:27 +03:00
|
|
|
finally:
|
|
|
|
a.close()
|
|
|
|
|
|
|
|
|
2015-12-14 01:32:23 +03:00
|
|
|
def clean(path):
|
2020-02-12 01:46:18 +03:00
|
|
|
for path in path.glob("**/*"):
|
2020-06-26 20:34:12 +03:00
|
|
|
if path.is_file() and path.suffix in (".so", ".cpp", ".html"):
|
2020-02-12 01:46:18 +03:00
|
|
|
print(f"Deleting {path.name}")
|
|
|
|
path.unlink()
|
2015-12-14 01:32:23 +03:00
|
|
|
|
|
|
|
|
|
|
|
def setup_package():
|
2020-07-02 18:10:27 +03:00
|
|
|
write_git_info_py()
|
2018-11-27 00:04:35 +03:00
|
|
|
if len(sys.argv) > 1 and sys.argv[1] == "clean":
|
2020-02-25 17:46:39 +03:00
|
|
|
return clean(PACKAGE_ROOT)
|
2020-02-12 01:46:18 +03:00
|
|
|
|
2020-02-25 17:46:39 +03:00
|
|
|
with (PACKAGE_ROOT / "about.py").open("r") as f:
|
2020-02-12 01:46:18 +03:00
|
|
|
about = {}
|
|
|
|
exec(f.read(), about)
|
|
|
|
|
2020-02-25 17:46:39 +03:00
|
|
|
for copy_file, target_dir in COPY_FILES.items():
|
2020-02-25 18:01:58 +03:00
|
|
|
if copy_file.exists():
|
|
|
|
shutil.copy(str(copy_file), str(target_dir))
|
|
|
|
print(f"Copied {copy_file} -> {target_dir}")
|
2020-02-25 17:46:39 +03:00
|
|
|
|
2020-02-12 01:46:18 +03:00
|
|
|
include_dirs = [
|
|
|
|
numpy.get_include(),
|
2020-11-25 12:33:49 +03:00
|
|
|
get_python_inc(plat_specific=True),
|
2020-02-12 01:46:18 +03:00
|
|
|
]
|
|
|
|
ext_modules = []
|
2022-09-14 18:05:22 +03:00
|
|
|
ext_modules.append(
|
|
|
|
Extension(
|
|
|
|
"spacy.matcher.levenshtein",
|
|
|
|
[
|
|
|
|
"spacy/matcher/levenshtein.pyx",
|
|
|
|
"spacy/matcher/polyleven.c",
|
|
|
|
],
|
|
|
|
language="c",
|
|
|
|
include_dirs=include_dirs,
|
|
|
|
)
|
|
|
|
)
|
2020-02-12 01:46:18 +03:00
|
|
|
for name in MOD_NAMES:
|
|
|
|
mod_path = name.replace(".", "/") + ".pyx"
|
2021-01-30 12:12:42 +03:00
|
|
|
ext = Extension(
|
2022-07-08 20:21:17 +03:00
|
|
|
name,
|
|
|
|
[mod_path],
|
|
|
|
language="c++",
|
|
|
|
include_dirs=include_dirs,
|
|
|
|
extra_compile_args=["-std=c++11"],
|
2021-01-30 12:12:42 +03:00
|
|
|
)
|
2020-02-12 01:46:18 +03:00
|
|
|
ext_modules.append(ext)
|
|
|
|
print("Cythonizing sources")
|
|
|
|
ext_modules = cythonize(ext_modules, compiler_directives=COMPILER_DIRECTIVES)
|
|
|
|
|
|
|
|
setup(
|
2021-01-30 12:12:42 +03:00
|
|
|
name="spacy",
|
2020-02-12 01:46:18 +03:00
|
|
|
packages=PACKAGES,
|
|
|
|
version=about["__version__"],
|
|
|
|
ext_modules=ext_modules,
|
|
|
|
cmdclass={"build_ext": build_ext_subclass},
|
2020-11-25 12:33:49 +03:00
|
|
|
package_data={"": ["*.pyx", "*.pxd", "*.pxi"]},
|
2020-02-12 01:46:18 +03:00
|
|
|
)
|
2015-01-04 21:30:56 +03:00
|
|
|
|
|
|
|
|
2018-11-27 00:04:35 +03:00
|
|
|
if __name__ == "__main__":
|
2015-12-14 01:32:23 +03:00
|
|
|
setup_package()
|