# ============================================
# Projet   : API-KA
# Fichier  : scripts/backfill_api_requests.py
# Node     : m3u96b
# Author   : Simon-Pierre Boucher
# Contact  : contact@spboucher.ai
# Date     : 2026-08-17
# ============================================
"""Rattrapage : importe l'historique des requêtes HTTP déjà journalisées dans
``logs/apika.log*`` (lignes JSON ``http_request`` du middleware) vers la table
``api_requests`` qui alimente la page /stats.

Idempotent : seules les lignes STRICTEMENT antérieures au plus ancien
enregistrement déjà présent dans ``api_requests`` sont insérées (à la première
exécution, tout l'historique des logs est repris ; aux suivantes, rien).

Usage : venv/bin/python -m scripts.backfill_api_requests
"""

from __future__ import annotations

import datetime
import json

from sqlalchemy import func, select

from src.api.reqstats import normalize_endpoint
from src.config import get_settings
from src.database.db import init_db, session_scope
from src.database.models import ApiRequest
from src.utils.logger import get_logger


def main() -> None:
    logger = get_logger("apika.backfill_requests")
    init_db()
    settings = get_settings()

    with session_scope() as session:
        cutoff = session.execute(select(func.min(ApiRequest.ts))).scalar()
    if cutoff is not None and cutoff.tzinfo is None:
        cutoff = cutoff.replace(tzinfo=datetime.UTC)
    if cutoff is None:
        cutoff = datetime.datetime.now(tz=datetime.UTC)

    entries: list[dict] = []
    parsed = skipped = 0
    for log_file in sorted(settings.logs_dir.glob("apika.log*")):
        with open(log_file, encoding="utf-8") as fh:
            for line in fh:
                line = line.strip()
                if not line or '"http_request"' not in line:
                    continue
                try:
                    rec = json.loads(line)
                except json.JSONDecodeError:
                    continue
                if rec.get("message") != "http_request":
                    continue
                parsed += 1
                try:
                    ts = datetime.datetime.fromisoformat(rec["timestamp"])
                except (KeyError, ValueError):
                    skipped += 1
                    continue
                if ts.tzinfo is None:
                    ts = ts.replace(tzinfo=datetime.UTC)
                if ts >= cutoff:
                    skipped += 1
                    continue
                entries.append(
                    {
                        "ts": ts,
                        "method": rec.get("method", "GET"),
                        "endpoint": normalize_endpoint(rec.get("path", "(autre)")),
                        "status": int(rec.get("status_code", 0)),
                        "duration_ms": round(float(rec.get("duration_ms", 0.0)), 2),
                    }
                )

    if entries:
        with session_scope() as session:
            session.bulk_insert_mappings(ApiRequest.__mapper__, entries)

    logger.info(
        "Backfill api_requests terminé",
        extra={"inserted": len(entries), "parsed": parsed, "skipped": skipped},
    )
    print(f"Backfill : {len(entries)} requêtes insérées ({parsed} lues, {skipped} ignorées)")


if __name__ == "__main__":
    main()