backend/scripts/preflight_env_check.py

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

import argparse
import json
import os
import sys
from pathlib import Path
from typing import Any, Dict, List

REPO_ROOT = Path(__file__).resolve().parents[2]
BACKEND_DIR = REPO_ROOT / "backend"
if str(BACKEND_DIR) not in sys.path:
    sys.path.insert(0, str(BACKEND_DIR))

from app.auth import load_auth_settings  # noqa: E402


def _is_placeholder_key(value: str) -> bool:
    v = str(value or "").strip()
    if not v:
        return True
    vu = v.upper()
    return vu.startswith("YOUR_KEY") or vu in {"<YOUR_OPENAI_API_KEY>", "YOUR_OPENAI_API_KEY", "REPLACE_ME"}


def _resolve_router_rag_paths() -> Dict[str, Path]:
    data_dir = Path(str(os.getenv("ROUTER_RAG_DATA_DIR", "") or "").strip() or (REPO_ROOT / "_RAG_Ready_KB_Organized"))
    chunks = Path(str(os.getenv("ROUTER_RAG_CHUNKS_PATH", "") or "").strip() or (data_dir / "04_ingestion" / "rag_ingestion_chunks.jsonl"))
    manifest = Path(str(os.getenv("ROUTER_RAG_MANIFEST_PATH", "") or "").strip() or (data_dir / "03_manifests" / "rag_manifest_organized.csv"))
    return {"data_dir": data_dir, "chunks": chunks, "manifest": manifest}


def _check_path(path: Path, *, file_required: bool = True) -> Dict[str, Any]:
    exists = path.exists()
    ok = exists and (path.is_file() if file_required else path.is_dir())
    return {"path": str(path), "exists": exists, "ok": ok}


def _build_report(args: argparse.Namespace) -> Dict[str, Any]:
    checks: Dict[str, Any] = {}
    failures: List[str] = []
    warnings: List[str] = []

    openai_key = str(os.getenv("OPENAI_API_KEY", "") or "").strip()
    openai_ok = bool(openai_key) and not _is_placeholder_key(openai_key)
    checks["openai"] = {
        "required": bool(args.require_openai),
        "present": bool(openai_key),
        "placeholder": _is_placeholder_key(openai_key) if openai_key else True,
        "ok": openai_ok or (not args.require_openai),
    }
    if args.require_openai and not openai_ok:
        failures.append("OPENAI_API_KEY is missing or placeholder.")

    auth = load_auth_settings()
    checks["auth"] = {
        "required": bool(auth.required),
        "enabled": bool(auth.enabled),
        "config_error": auth.config_error,
        "config_details": list(auth.config_details),
        "config_warnings": list(auth.config_warnings),
        "ok": (not auth.required) or bool(auth.enabled),
    }
    if auth.required and not auth.enabled:
        failures.append(auth.config_error or "Auth is required but Auth0 config is invalid.")
    if auth.config_warnings:
        warnings.extend(list(auth.config_warnings))

    repo_assets = {
        "routers_eos_eol_by_sku.csv": REPO_ROOT / "routers_eos_eol_by_sku.csv",
        "feb2026routers.csv": REPO_ROOT / "feb2026routers.csv",
        "FAQ_master_updated.csv": REPO_ROOT / "docs" / "faq" / "FAQ_master_updated.csv",
    }
    checks["required_assets"] = {name: _check_path(path, file_required=True) for name, path in repo_assets.items()}
    for name, data in checks["required_assets"].items():
        if not bool(data.get("ok")):
            failures.append(f"Missing required asset: {name}")

    rag = _resolve_router_rag_paths()
    checks["router_rag"] = {
        "data_dir": _check_path(rag["data_dir"], file_required=False),
        "chunks": _check_path(rag["chunks"], file_required=True),
        "manifest": _check_path(rag["manifest"], file_required=True),
        "strict": bool(args.fail_on_missing_router_rag),
    }
    if args.fail_on_missing_router_rag:
        if not checks["router_rag"]["chunks"]["ok"]:
            failures.append("Router RAG chunks file is missing.")
        if not checks["router_rag"]["manifest"]["ok"]:
            failures.append("Router RAG manifest file is missing.")
    else:
        if not checks["router_rag"]["chunks"]["ok"] or not checks["router_rag"]["manifest"]["ok"]:
            warnings.append("Router RAG ingestion artifacts are missing; router-doc retrieval may be degraded.")

    return {
        "ok": not failures,
        "failures": failures,
        "warnings": warnings,
        "checks": checks,
    }


def main() -> int:
    parser = argparse.ArgumentParser(description="Preflight environment/configuration checks for CI and release gates.")
    parser.add_argument("--require-openai", action="store_true", help="Fail if OPENAI_API_KEY is missing/placeholder.")
    parser.add_argument(
        "--fail-on-missing-router-rag",
        action="store_true",
        help="Fail if router RAG manifest/chunks files are missing.",
    )
    parser.add_argument("--out", default="", help="Optional output JSON path.")
    args = parser.parse_args()

    report = _build_report(args)
    payload = json.dumps(report, indent=2)
    if args.out:
        out_path = Path(args.out)
        out_path.parent.mkdir(parents=True, exist_ok=True)
        out_path.write_text(payload + "\n", encoding="utf-8")
        print(f"Wrote preflight report: {out_path}")
    print(payload)
    return 0 if report.get("ok") else 2


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