bakery-ia/services/orders/scripts/demo/seed_demo_orders.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Demo Orders Seeding Script for Orders Service
Creates realistic orders with order lines for demo template tenants

This script runs as a Kubernetes init job inside the orders-service container.
"""

import asyncio
import uuid
import sys
import os
import json
import random
from datetime import datetime, timezone, timedelta
from pathlib import Path
from decimal import Decimal

# Add app to path
sys.path.insert(0, str(Path(__file__).parent.parent.parent))

from sqlalchemy.ext.asyncio import AsyncSession, create_async_engine
from sqlalchemy.orm import sessionmaker
from sqlalchemy import select
import structlog

from app.models.order import CustomerOrder, OrderItem
from app.models.customer import Customer
from app.models.enums import OrderStatus, PaymentMethod, PaymentStatus, DeliveryMethod

# Add shared path for demo utilities
sys.path.insert(0, str(Path(__file__).parent.parent.parent.parent))
from shared.utils.demo_dates import BASE_REFERENCE_DATE

# Configure logging
logger = structlog.get_logger()

# Base demo tenant IDs
DEMO_TENANT_PROFESSIONAL = uuid.UUID("a1b2c3d4-e5f6-47a8-b9c0-d1e2f3a4b5c6")  # Individual bakery


def load_orders_config():
    """Load orders configuration from JSON file"""
    config_file = Path(__file__).parent / "pedidos_config_es.json"
    if not config_file.exists():
        raise FileNotFoundError(f"Orders config file not found: {config_file}")

    with open(config_file, 'r', encoding='utf-8') as f:
        return json.load(f)


def load_customers_data():
    """Load customers data from JSON file"""
    customers_file = Path(__file__).parent / "clientes_es.json"
    if not customers_file.exists():
        raise FileNotFoundError(f"Customers file not found: {customers_file}")

    with open(customers_file, 'r', encoding='utf-8') as f:
        data = json.load(f)
        return data.get("clientes", [])


def calculate_date_from_offset(offset_days: int) -> datetime:
    """Calculate a date based on offset from BASE_REFERENCE_DATE"""
    return BASE_REFERENCE_DATE + timedelta(days=offset_days)


# Model uses simple strings, no need for enum mapping functions
# (OrderPriority, DeliveryType don't exist in enums.py)


def weighted_choice(choices: list) -> dict:
    """Make a weighted random choice from list of dicts with 'peso' key"""
    total_weight = sum(c.get("peso", 1.0) for c in choices)
    r = random.uniform(0, total_weight)

    cumulative = 0
    for choice in choices:
        cumulative += choice.get("peso", 1.0)
        if r <= cumulative:
            return choice

    return choices[-1]


def generate_order_number(tenant_id: uuid.UUID, index: int) -> str:
    """Generate a unique order number"""
    tenant_prefix = "SP" if tenant_id == DEMO_TENANT_PROFESSIONAL else "LE"
    return f"ORD-{tenant_prefix}-{BASE_REFERENCE_DATE.year}-{index:04d}"


async def generate_orders_for_tenant(
    db: AsyncSession,
    tenant_id: uuid.UUID,
    tenant_name: str,
    config: dict,
    customers_data: list
):
    """Generate orders for a specific tenant"""
    logger.info(f"Generating orders for: {tenant_name}", tenant_id=str(tenant_id))

    # Check if orders already exist
    result = await db.execute(
        select(CustomerOrder).where(CustomerOrder.tenant_id == tenant_id).limit(1)
    )
    existing = result.scalar_one_or_none()

    if existing:
        logger.info(f"Orders already exist for {tenant_name}, skipping seed")
        return {"tenant_id": str(tenant_id), "orders_created": 0, "order_lines_created": 0, "skipped": True}

    # Get customers for this tenant
    result = await db.execute(
        select(Customer).where(Customer.tenant_id == tenant_id)
    )
    customers = list(result.scalars().all())

    if not customers:
        logger.warning(f"No customers found for {tenant_name}, cannot generate orders")
        return {"tenant_id": str(tenant_id), "orders_created": 0, "order_lines_created": 0, "error": "no_customers"}

    orders_config = config["configuracion_pedidos"]
    total_orders = orders_config["total_pedidos_por_tenant"]

    orders_created = 0
    lines_created = 0

    for i in range(total_orders):
        # Select random customer
        customer = random.choice(customers)

        # Determine temporal distribution
        rand_temporal = random.random()
        cumulative = 0
        temporal_category = None

        for category, details in orders_config["distribucion_temporal"].items():
            cumulative += details["porcentaje"]
            if rand_temporal <= cumulative:
                temporal_category = details
                break

        if not temporal_category:
            temporal_category = orders_config["distribucion_temporal"]["completados_antiguos"]

        # Calculate order date
        offset_days = random.randint(
            temporal_category["offset_dias_min"],
            temporal_category["offset_dias_max"]
        )
        order_date = calculate_date_from_offset(offset_days)

        # Select status based on temporal category (use strings directly)
        status = random.choice(temporal_category["estados"])

        # Select priority (use strings directly)
        priority_rand = random.random()
        cumulative_priority = 0
        priority = "normal"
        for p, weight in orders_config["distribucion_prioridad"].items():
            cumulative_priority += weight
            if priority_rand <= cumulative_priority:
                priority = p
                break

        # Select payment method (use strings directly)
        payment_method_choice = weighted_choice(orders_config["metodos_pago"])
        payment_method = payment_method_choice["metodo"]

        # Select delivery type (use strings directly)
        delivery_type_choice = weighted_choice(orders_config["tipos_entrega"])
        delivery_method = delivery_type_choice["tipo"]

        # Calculate delivery date (1-7 days after order date typically)
        delivery_offset = random.randint(1, 7)
        delivery_date = order_date + timedelta(days=delivery_offset)

        # Select delivery time
        delivery_time = random.choice(orders_config["horarios_entrega"])

        # Generate order number
        order_number = generate_order_number(tenant_id, i + 1)

        # Select notes
        notes = random.choice(orders_config["notas_pedido"]) if random.random() < 0.6 else None

        # Create order (using only actual model fields)
        order = CustomerOrder(
            id=uuid.uuid4(),
            tenant_id=tenant_id,
            order_number=order_number,
            customer_id=customer.id,
            status=status,
            order_type="standard",
            priority=priority,
            order_date=order_date,
            requested_delivery_date=delivery_date,
            confirmed_delivery_date=delivery_date if status != "pending" else None,
            actual_delivery_date=delivery_date if status in ["delivered", "completed"] else None,
            delivery_method=delivery_method,
            delivery_address={"address": customer.address_line1, "city": customer.city, "postal_code": customer.postal_code} if customer.address_line1 else None,
            payment_method=payment_method,
            payment_status="paid" if status in ["delivered", "completed"] else "pending",
            payment_terms="immediate",
            subtotal=Decimal("0.00"),  # Will calculate
            discount_percentage=Decimal("0.00"),  # Will set
            discount_amount=Decimal("0.00"),  # Will calculate
            tax_amount=Decimal("0.00"),  # Will calculate
            delivery_fee=Decimal("0.00"),
            total_amount=Decimal("0.00"),  # Will calculate
            special_instructions=notes,
            order_source="manual",
            sales_channel="direct",
            created_at=order_date,
            updated_at=order_date
        )

        db.add(order)
        await db.flush()  # Get order ID

        # Generate order lines
        num_lines = random.randint(
            orders_config["lineas_por_pedido"]["min"],
            orders_config["lineas_por_pedido"]["max"]
        )

        # Select random products
        selected_products = random.sample(
            orders_config["productos_demo"],
            min(num_lines, len(orders_config["productos_demo"]))
        )

        subtotal = Decimal("0.00")

        for line_num, product in enumerate(selected_products, 1):
            quantity = random.randint(
                orders_config["cantidad_por_linea"]["min"],
                orders_config["cantidad_por_linea"]["max"]
            )

            # Use base price with some variation
            unit_price = Decimal(str(product["precio_base"])) * Decimal(str(random.uniform(0.95, 1.05)))
            unit_price = unit_price.quantize(Decimal("0.01"))

            line_total = unit_price * quantity

            order_line = OrderItem(
                id=uuid.uuid4(),
                order_id=order.id,
                product_id=uuid.uuid4(),  # Generate placeholder product ID
                product_name=product["nombre"],
                product_sku=product["codigo"],
                quantity=Decimal(str(quantity)),
                unit_of_measure="each",
                unit_price=unit_price,
                line_discount=Decimal("0.00"),
                line_total=line_total,
                status="pending"
            )

            db.add(order_line)
            subtotal += line_total
            lines_created += 1

        # Apply order-level discount
        discount_rand = random.random()
        if discount_rand < 0.70:
            discount_percentage = Decimal("0.00")
        elif discount_rand < 0.85:
            discount_percentage = Decimal("5.00")
        elif discount_rand < 0.95:
            discount_percentage = Decimal("10.00")
        else:
            discount_percentage = Decimal("15.00")

        discount_amount = (subtotal * discount_percentage / 100).quantize(Decimal("0.01"))
        amount_after_discount = subtotal - discount_amount
        tax_amount = (amount_after_discount * Decimal("0.10")).quantize(Decimal("0.01"))
        total_amount = amount_after_discount + tax_amount

        # Update order totals
        order.subtotal = subtotal
        order.discount_percentage = discount_percentage
        order.discount_amount = discount_amount
        order.tax_amount = tax_amount
        order.total_amount = total_amount

        orders_created += 1

    await db.commit()
    logger.info(f"Successfully created {orders_created} orders with {lines_created} lines for {tenant_name}")

    return {
        "tenant_id": str(tenant_id),
        "orders_created": orders_created,
        "order_lines_created": lines_created,
        "skipped": False
    }


async def seed_all(db: AsyncSession):
    """Seed all demo tenants with orders"""
    logger.info("Starting demo orders seed process")

    # Load configuration
    config = load_orders_config()
    customers_data = load_customers_data()

    results = []

    # Seed Professional Bakery (merged from San Pablo + La Espiga)
    result_professional = await generate_orders_for_tenant(
        db,
        DEMO_TENANT_PROFESSIONAL,
        "Professional Bakery",
        config,
        customers_data
    )
    results.append(result_professional)

    total_orders = sum(r["orders_created"] for r in results)
    total_lines = sum(r["order_lines_created"] for r in results)

    return {
        "results": results,
        "total_orders_created": total_orders,
        "total_lines_created": total_lines,
        "status": "completed"
    }


async def main():
    """Main execution function"""
    # Get database URL from environment
    database_url = os.getenv("ORDERS_DATABASE_URL")
    if not database_url:
        logger.error("ORDERS_DATABASE_URL environment variable must be set")
        return 1

    # Ensure asyncpg driver
    if database_url.startswith("postgresql://"):
        database_url = database_url.replace("postgresql://", "postgresql+asyncpg://", 1)

    # Create async engine
    engine = create_async_engine(database_url, echo=False)
    async_session = sessionmaker(engine, class_=AsyncSession, expire_on_commit=False)

    try:
        async with async_session() as session:
            result = await seed_all(session)

            logger.info(
                "Orders seed completed successfully!",
                total_orders=result["total_orders_created"],
                total_lines=result["total_lines_created"],
                status=result["status"]
            )

            # Print summary
            print("\n" + "="*60)
            print("DEMO ORDERS SEED SUMMARY")
            print("="*60)
            for tenant_result in result["results"]:
                tenant_id = tenant_result["tenant_id"]
                orders = tenant_result["orders_created"]
                lines = tenant_result["order_lines_created"]
                skipped = tenant_result.get("skipped", False)
                status = "SKIPPED (already exists)" if skipped else f"CREATED {orders} orders, {lines} lines"
                print(f"Tenant {tenant_id}: {status}")
            print(f"\nTotal Orders: {result['total_orders_created']}")
            print(f"Total Order Lines: {result['total_lines_created']}")
            print("="*60 + "\n")

            return 0

    except Exception as e:
        logger.error(f"Orders seed failed: {str(e)}", exc_info=True)
        return 1
    finally:
        await engine.dispose()


if __name__ == "__main__":
    exit_code = asyncio.run(main())
    sys.exit(exit_code)