bakery-ia/services/sales/scripts/demo/seed_demo_sales.py

#!/usr/bin/env python3
"""
Demo Sales Seeding Script for Sales Service
Creates realistic historical sales data for demo template tenants

This script runs as a Kubernetes init job inside the sales-service container.
It populates the template tenants with historical sales data.

Usage:
    python /app/scripts/demo/seed_demo_sales.py

Environment Variables Required:
    SALES_DATABASE_URL - PostgreSQL connection string for sales database
    INVENTORY_DATABASE_URL - PostgreSQL connection string for inventory database (to lookup products)
    DEMO_MODE - Set to 'production' for production seeding
    LOG_LEVEL - Logging level (default: INFO)
"""

import asyncio
import uuid
import sys
import os
from datetime import datetime, timezone, timedelta
from pathlib import Path
import random
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, text
import structlog

from app.models.sales import SalesData

# Configure logging
structlog.configure(
    processors=[
        structlog.stdlib.add_log_level,
        structlog.processors.TimeStamper(fmt="iso"),
        structlog.dev.ConsoleRenderer()
    ]
)

logger = structlog.get_logger()

# Fixed Demo Tenant IDs (must match tenant service)
DEMO_TENANT_SAN_PABLO = uuid.UUID("a1b2c3d4-e5f6-47a8-b9c0-d1e2f3a4b5c6")
DEMO_TENANT_LA_ESPIGA = uuid.UUID("b2c3d4e5-f6a7-48b9-c0d1-e2f3a4b5c6d7")


# Hardcoded product IDs from ingredientes_es.json (finished products)
PRODUCT_IDS = {
    "PRO-BAG-001": "20000000-0000-0000-0000-000000000001",  # Baguette Tradicional
    "PRO-CRO-001": "20000000-0000-0000-0000-000000000002",  # Croissant de Mantequilla
    "PRO-PUE-001": "20000000-0000-0000-0000-000000000003",  # Pan de Pueblo
    "PRO-NAP-001": "20000000-0000-0000-0000-000000000004",  # Napolitana de Chocolate
}

# Sample product SKUs and their typical sales patterns
SAN_PABLO_PRODUCTS = [
    {"sku": "PRO-BAG-001", "name": "Baguette Tradicional", "avg_qty": 80, "variance": 15, "price": 1.20},
    {"sku": "PRO-CRO-001", "name": "Croissant de Mantequilla", "avg_qty": 50, "variance": 10, "price": 1.50},
    {"sku": "PRO-PUE-001", "name": "Pan de Pueblo", "avg_qty": 20, "variance": 5, "price": 3.50},
    {"sku": "PRO-NAP-001", "name": "Napolitana de Chocolate", "avg_qty": 35, "variance": 8, "price": 1.80},
]

LA_ESPIGA_PRODUCTS = [
    {"sku": "PRO-BAG-001", "name": "Baguette Tradicional", "avg_qty": 500, "variance": 80, "price": 0.90},
    {"sku": "PRO-CRO-001", "name": "Croissant de Mantequilla", "avg_qty": 300, "variance": 50, "price": 1.10},
    {"sku": "PRO-PUE-001", "name": "Pan de Pueblo", "avg_qty": 100, "variance": 20, "price": 2.80},
    {"sku": "PRO-NAP-001", "name": "Napolitana de Chocolate", "avg_qty": 200, "variance": 40, "price": 1.40},
]


def get_product_by_sku(tenant_id: uuid.UUID, sku: str, product_name: str):
    """
    Get tenant-specific product ID using hardcoded base IDs (no database lookup needed)

    Args:
        tenant_id: Tenant UUID
        sku: Product SKU code
        product_name: Product name

    Returns:
        Tuple of (product_id, product_name) or (None, None) if not found
    """
    if sku not in PRODUCT_IDS:
        return None, None

    # Generate tenant-specific product ID (same as inventory seed script)
    base_product_id = uuid.UUID(PRODUCT_IDS[sku])
    tenant_int = int(tenant_id.hex, 16)
    product_id = uuid.UUID(int=tenant_int ^ int(base_product_id.hex, 16))

    return product_id, product_name


async def seed_sales_for_tenant(
    sales_db: AsyncSession,
    tenant_id: uuid.UUID,
    tenant_name: str,
    product_patterns: list,
    days_of_history: int = 90
) -> dict:
    """
    Seed sales data for a specific tenant

    Args:
        sales_db: Sales database session
        tenant_id: UUID of the tenant
        tenant_name: Name of the tenant (for logging)
        product_patterns: List of product sales patterns
        days_of_history: Number of days of historical data to generate

    Returns:
        Dict with seeding statistics
    """
    logger.info("─" * 80)
    logger.info(f"Seeding sales data for: {tenant_name}")
    logger.info(f"Tenant ID: {tenant_id}")
    logger.info(f"Days of history: {days_of_history}")
    logger.info("─" * 80)

    created_sales = 0
    skipped_sales = 0

    # Generate sales data for each day
    for days_ago in range(days_of_history, 0, -1):
        sale_date = datetime.now(timezone.utc) - timedelta(days=days_ago)

        # Skip some random days to simulate closures
        if random.random() < 0.05:  # 5% chance of being closed
            continue

        # For each product, generate sales
        for product_pattern in product_patterns:
            sku = product_pattern["sku"]
            product_name = product_pattern["name"]

            # Get tenant-specific product ID using hardcoded base IDs
            product_id, product_name = get_product_by_sku(tenant_id, sku, product_name)

            if not product_id:
                logger.warning(f"      ⚠️  Product not found: {sku}")
                continue

            # Check if sales record already exists
            result = await sales_db.execute(
                select(SalesData).where(
                    SalesData.tenant_id == tenant_id,
                    SalesData.inventory_product_id == product_id,
                    SalesData.date == sale_date
                )
            )
            existing = result.scalars().first()

            if existing:
                skipped_sales += 1
                continue

            # Calculate sales quantity with variance
            avg_qty = product_pattern["avg_qty"]
            variance = product_pattern["variance"]

            # Add weekly patterns (weekends sell more)
            weekday = sale_date.weekday()
            if weekday in [5, 6]:  # Saturday, Sunday
                multiplier = random.uniform(1.2, 1.5)
            else:
                multiplier = random.uniform(0.8, 1.2)

            quantity = max(0, int((avg_qty + random.uniform(-variance, variance)) * multiplier))

            if quantity == 0:
                continue

            # Calculate revenue
            unit_price = Decimal(str(product_pattern["price"]))
            revenue = Decimal(str(quantity)) * unit_price

            # Check if it's a weekend
            is_weekend = weekday in [5, 6]

            # Create sales record
            sales_record = SalesData(
                id=uuid.uuid4(),
                tenant_id=tenant_id,
                inventory_product_id=product_id,
                date=sale_date,
                quantity_sold=quantity,
                revenue=revenue,
                unit_price=unit_price,
                sales_channel="in_store",
                location_id="main",
                source="demo_seed",
                is_weekend=is_weekend,
                created_at=sale_date,
                updated_at=sale_date
            )

            sales_db.add(sales_record)
            created_sales += 1

    # Commit all changes for this tenant
    await sales_db.commit()

    logger.info(f"   📊 Created: {created_sales}, Skipped: {skipped_sales}")
    logger.info("")

    return {
        "tenant_id": str(tenant_id),
        "tenant_name": tenant_name,
        "sales_records_created": created_sales,
        "sales_records_skipped": skipped_sales,
        "days_of_history": days_of_history
    }


async def seed_sales(sales_db: AsyncSession):
    """
    Seed sales for all demo template tenants

    Args:
        sales_db: Sales database session

    Returns:
        Dict with overall seeding statistics
    """
    logger.info("=" * 80)
    logger.info("💰 Starting Demo Sales Seeding")
    logger.info("=" * 80)

    results = []

    # Seed for San Pablo (Traditional Bakery) - 90 days of history
    logger.info("")
    result_san_pablo = await seed_sales_for_tenant(
        sales_db,
        DEMO_TENANT_SAN_PABLO,
        "Panadería San Pablo (Traditional)",
        SAN_PABLO_PRODUCTS,
        days_of_history=90
    )
    results.append(result_san_pablo)

    # Seed for La Espiga (Central Workshop) - 90 days of history
    result_la_espiga = await seed_sales_for_tenant(
        sales_db,
        DEMO_TENANT_LA_ESPIGA,
        "Panadería La Espiga (Central Workshop)",
        LA_ESPIGA_PRODUCTS,
        days_of_history=90
    )
    results.append(result_la_espiga)

    # Calculate totals
    total_sales = sum(r["sales_records_created"] for r in results)
    total_skipped = sum(r["sales_records_skipped"] for r in results)

    logger.info("=" * 80)
    logger.info("✅ Demo Sales Seeding Completed")
    logger.info("=" * 80)

    return {
        "service": "sales",
        "tenants_seeded": len(results),
        "total_sales_created": total_sales,
        "total_skipped": total_skipped,
        "results": results
    }


async def main():
    """Main execution function"""

    logger.info("Demo Sales Seeding Script Starting")
    logger.info("Mode: %s", os.getenv("DEMO_MODE", "development"))
    logger.info("Log Level: %s", os.getenv("LOG_LEVEL", "INFO"))

    # Get database URL from environment
    sales_database_url = os.getenv("SALES_DATABASE_URL") or os.getenv("DATABASE_URL")
    if not sales_database_url:
        logger.error("❌ SALES_DATABASE_URL or DATABASE_URL environment variable must be set")
        return 1

    # Convert to async URLs if needed
    if sales_database_url.startswith("postgresql://"):
        sales_database_url = sales_database_url.replace("postgresql://", "postgresql+asyncpg://", 1)

    logger.info("Connecting to sales database")

    # Create engine and session
    sales_engine = create_async_engine(
        sales_database_url,
        echo=False,
        pool_pre_ping=True,
        pool_size=5,
        max_overflow=10
    )

    sales_session_maker = sessionmaker(
        sales_engine,
        class_=AsyncSession,
        expire_on_commit=False
    )

    try:
        async with sales_session_maker() as sales_session:
            result = await seed_sales(sales_session)

            logger.info("")
            logger.info("📊 Seeding Summary:")
            logger.info(f"   ✅ Tenants seeded: {result['tenants_seeded']}")
            logger.info(f"   ✅ Sales records created: {result['total_sales_created']}")
            logger.info(f"   ⏭️  Skipped: {result['total_skipped']}")
            logger.info("")

            # Print per-tenant details
            for tenant_result in result['results']:
                logger.info(
                    f"   {tenant_result['tenant_name']}: "
                    f"{tenant_result['sales_records_created']} sales records "
                    f"({tenant_result['days_of_history']} days)"
                )

            logger.info("")
            logger.info("🎉 Success! Sales history is ready for cloning.")
            logger.info("")
            logger.info("Sales data includes:")
            logger.info("  • 90 days of historical sales")
            logger.info("  • 4 product types per tenant")
            logger.info("  • Realistic weekly patterns (higher on weekends)")
            logger.info("  • Random variance and occasional closures")
            logger.info("")
            logger.info("Next steps:")
            logger.info("  1. Run seed jobs for other services (orders, production, etc.)")
            logger.info("  2. Verify sales data in database")
            logger.info("  3. Test demo session creation with sales cloning")
            logger.info("")

            return 0

    except Exception as e:
        logger.error("=" * 80)
        logger.error("❌ Demo Sales Seeding Failed")
        logger.error("=" * 80)
        logger.error("Error: %s", str(e))
        logger.error("", exc_info=True)
        return 1

    finally:
        await sales_engine.dispose()


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