orion/docs/backend/overview.md

# Backend Development

Guide for developing backend features, services, and API endpoints.

## Overview

The Orion backend is built with FastAPI and follows a service-oriented architecture pattern. This guide covers backend development practices, patterns, and technical references.

## Backend Structure

```
app/
├── api/                    # API routes (REST endpoints)
│   ├── v1/                # Version 1 API
│   │   ├── admin/        # Admin API endpoints
│   │   ├── store/       # Store API endpoints
│   │   └── shop/         # Shop API endpoints
│   └── main.py           # API router configuration
│
├── routes/               # Page routes (HTML)
│   ├── admin_pages.py   # Admin page routes
│   ├── store_pages.py  # Store page routes
│   └── shop_pages.py    # Shop page routes
│
├── services/            # Business logic layer
│   ├── admin_service.py
│   ├── product_service.py
│   ├── order_service.py
│   └── ...
│
├── core/                # Core functionality
│   ├── database.py     # Database connection
│   ├── config.py       # Configuration
│   └── lifespan.py     # Application lifecycle
│
└── exceptions/          # Custom exceptions
    ├── base.py
    └── handler.py
```

## Development Workflow

### 1. Creating a New API Endpoint

**Steps**:
1. Define Pydantic schema in `models/schema/`
2. Create service method in `app/services/`
3. Add API route in `app/api/v1/`
4. Write tests in `tests/`

**Example** - Adding a new product endpoint:

```python
# Step 1: Schema (models/schema/product.py)
from pydantic import BaseModel

class ProductCreate(BaseModel):
    name: str
    description: str
    price: float
    store_id: int

class ProductResponse(BaseModel):
    id: int
    name: str
    description: str
    price: float
    store_id: int

    class Config:
        from_attributes = True

# Step 2: Service (app/services/product_service.py)
from sqlalchemy.orm import Session
from models.database.product import Product

class ProductService:
    def create_product(self, db: Session, data: ProductCreate) -> Product:
        product = Product(**data.dict())
        db.add(product)
        db.commit()
        db.refresh(product)
        return product

product_service = ProductService()

# Step 3: API Route (app/api/v1/shop/products.py)
from fastapi import APIRouter, Depends
from sqlalchemy.orm import Session
from app.core.database import get_db
from app.services.product_service import product_service
from app.modules.catalog.schemas import ProductCreate, ProductResponse

router = APIRouter(prefix="/products")

@router.post("/", response_model=ProductResponse)
async def create_product(
    data: ProductCreate,
    db: Session = Depends(get_db)
):
    return product_service.create_product(db, data)
```

### 2. Creating a Service

Services contain business logic and should be reusable across routes.

**Pattern**:
```python
# app/services/example_service.py
from sqlalchemy.orm import Session
from typing import List, Optional
import logging

logger = logging.getLogger(__name__)

class ExampleService:
    """Service for handling example operations."""

    def get_items(
        self,
        db: Session,
        store_id: int,
        skip: int = 0,
        limit: int = 100
    ) -> List[Item]:
        """Get items for a store with pagination."""
        try:
            items = db.query(Item).filter(
                Item.store_id == store_id
            ).offset(skip).limit(limit).all()

            logger.info(f"Retrieved {len(items)} items for store {store_id}")
            return items

        except Exception as e:
            logger.error(f"Error retrieving items: {e}")
            raise

    def create_item(
        self,
        db: Session,
        store_id: int,
        data: ItemCreate
    ) -> Item:
        """Create a new item."""
        try:
            item = Item(
                store_id=store_id,
                **data.dict()
            )
            db.add(item)
            db.commit()
            db.refresh(item)

            logger.info(f"Created item {item.id} for store {store_id}")
            return item

        except Exception as e:
            logger.error(f"Error creating item: {e}")
            db.rollback()
            raise

# Singleton instance
example_service = ExampleService()
```

### 3. Database Operations

**Best Practices**:

```python
# ✅ Good - Use service layer
@router.get("/products")
async def get_products(
    store_id: int,
    db: Session = Depends(get_db)
):
    products = product_service.get_products(db, store_id)
    return {"products": products}

# ❌ Bad - Database queries in route handler
@router.get("/products")
async def get_products(
    store_id: int,
    db: Session = Depends(get_db)
):
    products = db.query(Product).filter(
        Product.store_id == store_id
    ).all()
    return {"products": products}
```

**Always Scope to Store**:
```python
# ✅ Good - Scoped to store
products = db.query(Product).filter(
    Product.store_id == request.state.store_id
).all()

# ❌ Bad - Not scoped, security risk!
products = db.query(Product).all()
```

## Dependency Injection

FastAPI's dependency system provides clean code organization:

```python
from fastapi import Depends
from sqlalchemy.orm import Session
from app.core.database import get_db
from middleware.auth import auth_manager
from models.database.user import User

@router.get("/protected")
async def protected_endpoint(
    current_user: User = Depends(auth_manager.get_current_user),
    db: Session = Depends(get_db)
):
    # current_user is automatically injected
    # db session is automatically injected
    return {"user": current_user.username}
```

## Error Handling

**Custom Exceptions**:
```python
from app.exceptions import (
    ResourceNotFoundException,
    ValidationException,
    InsufficientPermissionsException
)

@router.get("/products/{product_id}")
async def get_product(
    product_id: int,
    db: Session = Depends(get_db)
):
    product = db.query(Product).filter(Product.id == product_id).first()

    if not product:
        raise ResourceNotFoundException(f"Product {product_id} not found")

    return product
```

**Exception Handlers** are configured globally in `app/exceptions/handler.py`

## Testing

### Unit Tests

```python
# tests/unit/services/test_product_service.py
import pytest
from app.services.product_service import ProductService

def test_create_product(db_session):
    service = ProductService()
    data = ProductCreate(
        name="Test Product",
        price=29.99,
        store_id=1
    )

    product = service.create_product(db_session, data)

    assert product.id is not None
    assert product.name == "Test Product"
    assert product.store_id == 1
```

### Integration Tests

```python
# tests/integration/test_product_api.py
def test_create_product_endpoint(client, auth_headers):
    response = client.post(
        "/api/v1/products",
        json={
            "name": "Test Product",
            "price": 29.99,
            "store_id": 1
        },
        headers=auth_headers
    )

    assert response.status_code == 200
    data = response.json()
    assert data["name"] == "Test Product"
```

## Code Quality

### Linting

```bash
# Run linters
make lint

# Auto-fix formatting
make format
```

### Type Checking

```bash
# Run mypy
make lint
```

**Use Type Hints**:
```python
from typing import List, Optional
from sqlalchemy.orm import Session

def get_items(
    db: Session,
    store_id: int,
    limit: Optional[int] = None
) -> List[Item]:
    query = db.query(Item).filter(Item.store_id == store_id)

    if limit:
        query = query.limit(limit)

    return query.all()
```

## Database Migrations

**Creating a Migration**:
```bash
make migrate-create message="add_product_table"
```

**Applying Migrations**:
```bash
make migrate-up
```

**See**: [Database Migrations](../development/migration/database-migrations.md)

## API Documentation

FastAPI automatically generates API documentation:

- **Swagger UI**: http://localhost:8000/docs
- **ReDoc**: http://localhost:8000/redoc
- **OpenAPI JSON**: http://localhost:8000/openapi.json

**Document your endpoints**:
```python
@router.post(
    "/products",
    response_model=ProductResponse,
    summary="Create a new product",
    description="Creates a new product for the authenticated store",
    responses={
        201: {"description": "Product created successfully"},
        400: {"description": "Invalid product data"},
        401: {"description": "Not authenticated"},
        403: {"description": "Not authorized to create products"}
    }
)
async def create_product(
    data: ProductCreate,
    current_user: User = Depends(auth_manager.require_store),
    db: Session = Depends(get_db)
):
    """
    Create a new product.

    - **name**: Product name (required)
    - **description**: Product description
    - **price**: Product price (required)
    - **store_id**: Store ID (required)
    """
    return product_service.create_product(db, current_user.id, data)
```

## Configuration

**Environment Variables** (`.env`):
```bash
# Database
DATABASE_URL=postgresql://user:password@localhost:5432/dbname

# JWT
JWT_SECRET_KEY=your-secret-key
JWT_EXPIRE_MINUTES=30

# Application
ENVIRONMENT=development
DEBUG=true
LOG_LEVEL=INFO
```

**Accessing Config**:
```python
from app.core.config import settings

database_url = settings.database_url
secret_key = settings.jwt_secret_key
```

## Logging

**Use Structured Logging**:
```python
import logging

logger = logging.getLogger(__name__)

def process_order(order_id: int):
    logger.info(f"Processing order", extra={"order_id": order_id})

    try:
        # Process order
        logger.info(f"Order processed successfully", extra={"order_id": order_id})

    except Exception as e:
        logger.error(
            f"Order processing failed",
            extra={"order_id": order_id, "error": str(e)},
            exc_info=True
        )
        raise
```

## Performance

### Database Query Optimization

```python
# ✅ Good - Use joins to avoid N+1 queries
from sqlalchemy.orm import joinedload

products = db.query(Product).options(
    joinedload(Product.category),
    joinedload(Product.store)
).filter(Product.store_id == store_id).all()

# ❌ Bad - N+1 query problem
products = db.query(Product).filter(
    Product.store_id == store_id
).all()

for product in products:
    category = product.category  # Additional query!
```

### Caching

```python
from functools import lru_cache

@lru_cache(maxsize=100)
def get_store_settings(store_id: int) -> dict:
    # Expensive operation cached in memory
    return db.query(StoreSettings).filter(
        StoreSettings.store_id == store_id
    ).first()
```

## Security Best Practices

1. **Always validate input** with Pydantic schemas
2. **Always scope queries** to store/user
3. **Use parameterized queries** (SQLAlchemy ORM does this)
4. **Never log sensitive data** (passwords, tokens, credit cards)
5. **Use HTTPS** in production
6. **Implement rate limiting** on sensitive endpoints
7. **Validate file uploads** (type, size, content)
8. **Sanitize user input** before rendering in templates

## Next Steps

- [Middleware Reference](middleware-reference.md) - Technical API documentation
- [Architecture Overview](../architecture/overview.md) - System architecture
- [Database Migrations](../development/migration/database-migrations.md) - Migration guide
- [Testing Guide](../testing/testing-guide.md) - Testing practices