42 lines
1.3 KiB
Python
42 lines
1.3 KiB
Python
"""Загрузка документов из директории в векторное хранилище."""
|
|
from __future__ import annotations
|
|
|
|
import argparse
|
|
from pathlib import Path
|
|
|
|
from vector_store import add_document_to_store
|
|
|
|
SUPPORTED_SUFFIXES = {".txt", ".md", ".markdown"}
|
|
|
|
|
|
def load_directory(directory: str | Path) -> int:
|
|
root = Path(directory)
|
|
if not root.exists():
|
|
raise FileNotFoundError(f"Директория не найдена: {root}")
|
|
|
|
total_chunks = 0
|
|
for path in sorted(root.rglob("*")):
|
|
if not path.is_file() or path.suffix.lower() not in SUPPORTED_SUFFIXES:
|
|
continue
|
|
content = path.read_text(encoding="utf-8")
|
|
title = str(path.relative_to(root))
|
|
total_chunks += add_document_to_store(content, title)
|
|
print(f"Загружен: {title} ({len(content)} символов)")
|
|
return total_chunks
|
|
|
|
|
|
def main() -> None:
|
|
parser = argparse.ArgumentParser(description="Инициализация базы знаний из папки")
|
|
parser.add_argument(
|
|
"--dir",
|
|
default="./docs",
|
|
help="Путь к директории с .txt/.md файлами",
|
|
)
|
|
args = parser.parse_args()
|
|
chunks = load_directory(args.dir)
|
|
print(f"Готово. Всего чанков: {chunks}")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|