"""READ-ONLY phase timing of the Agent pipeline on a real database (default MMMODAS ~21k products) against the local dev Hub. Reports wall time per phase and peak RSS.
    python scripts/measure_scale_agent.py [--db MMMODAS]"""
import argparse
import json
import os
import sys
import time
from pathlib import Path

ROOT = Path(__file__).resolve().parents[1]
sys.path[:0] = [str(ROOT), str(ROOT / "scripts"), str(ROOT / "agent" / "src")]
from dotenv import load_dotenv  # noqa: E402

load_dotenv(ROOT / ".env")
import local_completion as lc  # noqa: E402
from central.app.operational_state import classify_product_state, summarize_operational_states  # noqa: E402
from gti_fiscal_agent import firebird_reader, unified_fiscal  # noqa: E402
from gti_fiscal_agent.client import HubClient  # noqa: E402
from gti_fiscal_agent.runner import to_hub_product  # noqa: E402


def rss_mb():
    try:
        import psutil
        return round(psutil.Process().memory_info().rss / 1e6, 1)
    except Exception:  # noqa: BLE001
        return None


def main():
    ap = argparse.ArgumentParser(); ap.add_argument("--db", default="MMMODAS"); a = ap.parse_args()
    path = lc.allowlist()[a.db]; env = lc.env_for(path)
    out = {"database": a.db, "phases_seconds": {}}
    t = time.time(); products = firebird_reader.read_products(env); out["phases_seconds"]["firebird_read"] = round(time.time() - t, 2); out["products"] = len(products)
    t = time.time(); decisions = HubClient(os.environ.get("HUB_URL", lc.HUB), os.environ["AGENT_DEV_TOKEN"]).classify_batch([to_hub_product(p) for p in products]); out["phases_seconds"]["hub_classification"] = round(time.time() - t, 2)
    t = time.time(); states = [classify_product_state(p, d) for p, d in zip(products, decisions)]; summarize_operational_states(states); out["phases_seconds"]["operational_states"] = round(time.time() - t, 2)
    t = time.time(); conn = firebird_reader._connect(env)
    try:
        assessed = unified_fiscal.assess_from_connection(conn, {"crt": "1"}, rtc_override=None)
    finally:
        conn.close()
    out["phases_seconds"]["fiscal_rtc_pis_cofins_assessment"] = round(time.time() - t, 2)
    t = time.time(); payload = json.dumps({"operational": summarize_operational_states(states), "findings": len(assessed.get("findings", {}))}); out["phases_seconds"]["payload_generation"] = round(time.time() - t, 3)
    out["total_seconds"] = round(sum(out["phases_seconds"].values()), 2); out["peak_rss_mb"] = rss_mb(); out["products_per_second"] = round(len(products) / max(out["total_seconds"], 0.01), 1)
    print(json.dumps(out, indent=1))
    (ROOT / "docs" / "runs" / "final_corpus" / f"scale_agent_{a.db}.json").write_text(json.dumps(out, indent=1), encoding="utf-8")


if __name__ == "__main__":
    main()