setup.py

import os
from skbuild import setup

PACKAGE_NAME = "dedup_cpp_core"
CORE_MODULE_NAME = "_core"
SRC_DIR = "src"


init_py_path = os.path.join(SRC_DIR, PACKAGE_NAME, "__init__.py")


os.makedirs(os.path.dirname(init_py_path), exist_ok=True)

EXPOSED_FUNCTIONS = [
    "deduplicate_cpp",
    "get_chunks_and_hashes",
    "get_CDC_hashes_cpp",
    "find_duplicates_cpp",
    "get_document_simhash_performance",
    "process_texts_for_signatures_cpp",
    "clean_texts_with_bad_hashes_cpp"
]


with open(init_py_path, "w") as f:
    f.write(f"# This file is automatically generated by setup.py\n\n")
    f.write(f"from .{CORE_MODULE_NAME} import (\n")
    for func in EXPOSED_FUNCTIONS:
        f.write(f"    {func},\n")
    f.write(f")\n\n")

    f.write(f"__all__ = [\n")
    for func in EXPOSED_FUNCTIONS:
        f.write(f"    \"{func}\",\n")
    f.write(f"]\n")

print(f"--- Generated __init__.py at: {init_py_path} ---")

setup(
    name=PACKAGE_NAME,
    version="0.1.0",
    description="A C++ accelerated deduplication module",
    author="Jinming Hu",

    packages=[PACKAGE_NAME],
    package_dir={'': SRC_DIR},
    
    cmake_install_dir=os.path.join(SRC_DIR, PACKAGE_NAME),

    package_data={PACKAGE_NAME: ['*.*']},
    include_package_data=True,

    cmake_minimum_required_version='3.12',
    has_ext_modules=lambda: True,
)