dynantic 0.4.0

Pydantic ODM for DynamoDB based on Boto3

Dynantic

Type-safe DynamoDB ORM with Pydantic validation

Read the Documentation

---

Dynantic is a synchronous-first Python ORM for Amazon DynamoDB that combines Pydantic v2 validation with an elegant query DSL.

Features

• Pydantic v2 validation and type safety
• Metaclass-based DSL for elegant query building
• Comprehensive type support (datetime, UUID, Enum, Decimal, sets, etc.)
• Global Secondary Indexes (GSI)
• Polymorphic models for single-table design
• Conditional writes with SQLModel-like syntax
• Atomic updates without fetching first (add, set_if_not_exists, append)
• Projection queries (.values()) and server-side counts (.count())
• Batch operations with auto-chunking and retry
• ACID transactions across tables
• TTL support with automatic datetime/epoch conversion
• Auto-UUID with Key(auto=True) and INSERT-safe create()

Optimized for: AWS Lambda, serverless functions, FastAPI, batch jobs, and scripts.

Installation

pip install dynantic

Requirements: Python 3.10+ · pydantic >= 2.6.0 · boto3 >= 1.34.0

Quick Start

from datetime import datetime, timezone
from enum import Enum
from dynantic import DynamoModel, Key, SortKey

class UserStatus(Enum):
    ACTIVE = "active"
    INACTIVE = "inactive"

class User(DynamoModel):
    user_id: str = Key()
    email: str = SortKey()
    name: str
    status: UserStatus
    created_at: datetime
    balance: float
    tags: set[str]

    class Meta:
        table_name = "users"

# Create
user = User(
    user_id="user-123",
    email="[e-mail]",
    name="John Doe",
    status=UserStatus.ACTIVE,
    created_at=datetime.now(timezone.utc),
    balance=99.99,
    tags={"premium", "verified"}
)
user.save()

# Read
user = User.get("user-123", "[e-mail]")

# Update (atomic)
User.update("user-123", "[e-mail]") \
    .add(User.balance, 10.0) \
    .execute()

# Delete
User.delete("user-123", "[e-mail]")

Full documentation and guides

Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Write tests and ensure they pass (uv run pytest)
4. Run type checking (uv run mypy dynantic)
5. Submit a pull request

git clone https://github.com/Simi24/dynantic.git
cd dynantic
uv sync --dev
docker compose up -d  # Start LocalStack
uv run pytest

License

MIT License - see LICENSE for details.