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

import argparse
import json
import sys
from collections import Counter
from pathlib import Path
from typing import Any

from sqlalchemy import select

SERVER_DIR = Path(__file__).resolve().parents[1]
SRC_DIR = SERVER_DIR / "src"
if str(SRC_DIR) not in sys.path:
    sys.path.insert(0, str(SRC_DIR))

from app.db.session import get_session_factory  # noqa: E402
from app.models.financial_record import ExpenseClaim  # noqa: E402
from app.services.expense_claim_status_registry import (  # noqa: E402
    is_known_approval_stage,
    is_known_claim_status,
    normalize_expense_claim_state,
)


def main() -> None:
    parser = argparse.ArgumentParser(description="Audit expense claim status consistency.")
    parser.add_argument("--sample-limit", type=int, default=20)
    args = parser.parse_args()

    session_factory = get_session_factory()
    with session_factory() as db:
        claims = list(
            db.scalars(
                select(ExpenseClaim).order_by(
                    ExpenseClaim.claim_no.asc(),
                    ExpenseClaim.created_at.asc(),
                )
            ).all()
        )
        payload = audit_claims(claims, sample_limit=max(args.sample_limit, 0))
        print(json.dumps(payload, ensure_ascii=False, indent=2))


def audit_claims(claims: list[ExpenseClaim], *, sample_limit: int) -> dict[str, Any]:
    status_counts: Counter[str] = Counter()
    stage_counts: Counter[str] = Counter()
    status_stage_counts: Counter[str] = Counter()
    doc_type_counts: Counter[str] = Counter()
    unknown_statuses: Counter[str] = Counter()
    unknown_stages: Counter[str] = Counter()
    normalization_counts: Counter[str] = Counter()
    samples: list[dict[str, Any]] = []

    for claim in claims:
        status = str(claim.status or "").strip()
        stage = str(claim.approval_stage or "").strip()
        doc_type = _doc_type(claim)
        status_counts[status or "<empty>"] += 1
        stage_counts[stage or "<empty>"] += 1
        status_stage_counts[f"{status or '<empty>'} | {stage or '<empty>'}"] += 1
        doc_type_counts[doc_type] += 1

        if not is_known_claim_status(status):
            unknown_statuses[status or "<empty>"] += 1
        if not is_known_approval_stage(stage):
            unknown_stages[stage or "<empty>"] += 1

        normalized = normalize_expense_claim_state(
            status,
            stage,
            claim_no=claim.claim_no,
            expense_type=claim.expense_type,
        )
        if normalized.changed:
            key = (
                f"{status or '<empty>'}/{stage or '<empty>'}"
                f" -> {normalized.status}/{normalized.approval_stage}"
            )
            normalization_counts[key] += 1
            if len(samples) < sample_limit:
                samples.append(
                    {
                        "claim_no": claim.claim_no,
                        "doc_type": doc_type,
                        "status": status,
                        "approval_stage": stage,
                        "normalized_status": normalized.status,
                        "normalized_approval_stage": normalized.approval_stage,
                        "status_code": normalized.status_code,
                    }
                )

    return {
        "claim_count": len(claims),
        "doc_type_counts": dict(doc_type_counts),
        "status_counts": dict(status_counts),
        "approval_stage_counts": dict(stage_counts),
        "status_stage_counts": dict(status_stage_counts),
        "unknown_statuses": dict(unknown_statuses),
        "unknown_approval_stages": dict(unknown_stages),
        "normalization_needed": sum(normalization_counts.values()),
        "normalization_counts": dict(normalization_counts),
        "normalization_samples": samples,
    }


def _doc_type(claim: ExpenseClaim) -> str:
    claim_no = str(claim.claim_no or "").strip().upper()
    expense_type = str(claim.expense_type or "").strip().lower()
    if claim_no.startswith(("AP-", "APP-")) or expense_type.endswith("_application"):
        return "application"
    if str(claim.project_code or "").strip().upper() == "SIM-DEMO":
        return "sim_reimbursement"
    return "reimbursement"


if __name__ == "__main__":
    main()