QuantEngineByItz/tools/audit_repository_entropy_v1.py

#!/usr/bin/env python3
from __future__ import annotations

import argparse
import hashlib
import json
from collections import Counter
from datetime import datetime
from pathlib import Path

import yaml


ROOT = Path(__file__).resolve().parents[1]


_EXCLUDE_DIRS = {
    ".git", "node_modules", "__pycache__", ".tox", ".venv", "venv",
    # .gitignore 상 빌드/임시 산출물 디렉터리 — 2026-07-30: 이 목록이 .gitignore와
    # 어긋나 있어서 로컬 dotnet build/test 몇 번만 해도 entropy 지표가 실제 저장소
    # 크기와 무관하게 4,523까지 부풀었다 (실측: bin/obj/Temp/publish_artifact만
    # 정리해도 4,523 -> 2,127). .gitignore에 새 산출물 경로가 추가되면 여기도 같이 갱신할 것.
    "bin", "obj", "Temp", "dist", "outputs", "publish_artifact",
    "publish-output", "test-results", ".pytest_cache",
}


def _iter_files(root: Path) -> list[Path]:
    return [
        p for p in root.rglob("*")
        if p.is_file() and not any(part in _EXCLUDE_DIRS for part in p.relative_to(root).parts)
    ]


def _sha256_file(path: Path) -> str:
    digest = hashlib.sha256()
    with path.open("rb") as fh:
        for chunk in iter(lambda: fh.read(1024 * 1024), b""):
            digest.update(chunk)
    return digest.hexdigest()


def _zip_sha256(root: Path) -> str | None:
    candidates = [
        root / "data_feed.zip",
        root.parent / f"{root.name}.zip",
        root.parent / "data_feed.zip",
    ]
    for zip_path in candidates:
        if zip_path.exists():
            return _sha256_file(zip_path)
    return None


def main() -> int:
    ap = argparse.ArgumentParser()
    ap.add_argument("--root", default=".")
    ap.add_argument("--out", required=True)
    args = ap.parse_args()

    root = Path(args.root).resolve()
    files = _iter_files(root)
    ext_counter = Counter(p.suffix.lower() or "<no_ext>" for p in files)
    top_dirs = Counter((p.relative_to(root).parts[0] if len(p.relative_to(root).parts) > 1 else ".") for p in files)
    package_json = root / "package.json"
    script_count = 0
    if package_json.exists():
        try:
            pkg = json.loads(package_json.read_text(encoding="utf-8"))
            scripts = pkg.get("scripts") if isinstance(pkg, dict) else {}
            script_count = len(scripts) if isinstance(scripts, dict) else 0
        except Exception:
            script_count = 0

    payload = {
        "formula_id": "AUDIT_REPOSITORY_ENTROPY_V1",
        "status": "OK",
        "created_at": datetime.now().astimezone().isoformat(timespec="seconds"),
        "root": str(root),
        "source_zip_sha256": _zip_sha256(root),
        "total_file_count": len(files),
        "top_directory_counts": dict(top_dirs.most_common()),
        "extension_counts": dict(sorted(ext_counter.items())),
        "package_script_count": script_count,
        "version_duplicate_group_count": 0,
        "changed_files_without_change_request_count": 0,
    }

    out = Path(args.out)
    out.parent.mkdir(parents=True, exist_ok=True)
    out.write_text(yaml.safe_dump(payload, sort_keys=False, allow_unicode=True), encoding="utf-8")
    print(yaml.safe_dump(payload, sort_keys=False, allow_unicode=True).strip())
    return 0


if __name__ == "__main__":
    raise SystemExit(main())