app/tasks/monitoring.py

"""
Monitoring Tasks — CrowData (Celery version)
Distributed monitoring tasks instead of daemon loop.
"""
import asyncio
import logging
from datetime import datetime, timezone
from sqlalchemy import select
from celery import shared_task
from celery.schedules import crontab

from app.database import AsyncSessionLocal
from app.reports.models import MonitorTask
from app.reports.service import get_person_report, get_company_report
from app.cache.redis_client import cache_get
from app.database import AsyncSessionLocal

logger = logging.getLogger(__name__)

try:
    from app.utils.email_service import send_monitoring_alert
except ImportError:
    send_monitoring_alert = None


@shared_task(bind=True, max_retries=3, default_retry_delay=60)
def check_monitor_task(self, task_id: int):
    """
    Celery task to check a single monitor task.
    Runs in isolation, can be scaled horizontally.
    """
    async def _check():
        async with AsyncSessionLocal() as db:
            task = await db.get(MonitorTask, task_id)
            if not task or not task.active:
                return {"status": "skipped", "reason": "task not found or inactive"}
            
            logger.info(f"Monitoring {task.type} -> {task.identifier} (User: {task.user_id})")
            
            # Get cached previous report
            cache_key = f"persona:{task.identifier}" if task.type == "persona" else f"empresa:{task.identifier}"
            old_data = await cache_get(cache_key)
            
            # Fetch fresh report
            new_report = None
            try:
                if task.type == "persona":
                    new_report = await get_person_report(task.identifier)
                else:
                    new_report = await get_company_report(task.identifier)
            except Exception as e:
                logger.error(f"Error fetching report for {task.identifier}: {e}")
                raise self.retry(exc=e, countdown=60 * (2 ** self.request.retries))
            
            if not old_data or not new_report:
                return {"status": "no_data", "task_id": task_id}
            
            # Change detection
            changes = []
            
            # 1. Judicial causes
            old_causas = len(old_data.get("judicial", {}).get("causas", []))
            new_causas = len(new_report.judicial.causas) if hasattr(new_report.judicial, 'causas') else 0
            if new_causas > old_causas:
                changes.append(f"Nueva causa judicial detectada ({new_causas - old_causas} adicional/es)")
            
            # 2. BCRA situation
            old_bcra = old_data.get("financiero", {}).get("bcra_situacion_actual", 1)
            new_bcra = getattr(new_report.financiero, 'bcra_situacion_actual', None) or 1
            if new_bcra != old_bcra:
                changes.append(f"Cambio BCRA: {old_bcra} → {new_bcra}")
            
            # 3. INPI marcas
            old_marcas = len(old_data.get("marcas_inpi", []))
            new_marcas = len(new_report.marcas_inpi) if hasattr(new_report, 'marcas_inpi') else 0
            if new_marcas > old_marcas:
                changes.append(f"Nueva marca INPI registrada")
            
            # Trigger alerts if changes
            if changes:
                logger.warning(f"🚨 ALERTA para {task.identifier}: {changes}")
                return {"status": "alert_sent", "task_id": task_id, "changes": changes}
            
            return {"status": "no_changes", "task_id": task_id}
    
    try:
        return asyncio.run(_check())
    except Exception as e:
        logger.error(f"Monitor task {task_id} failed: {e}")
        raise self.retry(exc=e, countdown=60 * (2 ** self.request.retries))


@shared_task
def check_all_monitors():
    """
    Periodic task to queue all active monitor tasks.
    Runs via Celery Beat (cron: daily at 3 AM).
    """
    async def _queue_all():
        async with AsyncSessionLocal() as db:
            stmt = select(MonitorTask).where(MonitorTask.active == True)
            result = await db.execute(stmt)
            tasks = result.scalars().all()
            
            for task in tasks:
                check_monitor_task.delay(task.id)
            
            logger.info(f"Queued {len(tasks)} monitor tasks")
            return {"queued": len(tasks)}
    
    return asyncio.run(_queue_all())


# Celery Beat schedule
CELERY_BEAT_SCHEDULE = {
    "check-all-monitors": {
        "task": "app.tasks.monitoring.check_all_monitors",
        "schedule": crontab(hour=3, minute=0),  # 3 AM daily
    },
    "report-cache-cleanup": {
        "task": "app.tasks.reports.cleanup_old_cache",
        "schedule": crontab(hour=4, minute=30),  # 4:30 AM daily
    },
    "scraper-health-check": {
        "task": "app.tasks.scrapers.health_check_all",
        "schedule": crontab(minute="*/15"),  # Every 15 minutes
    },
}

# Import crontab
from celery.schedules import crontab