pr-dataclass-integration.md

feat: add dataclass integration

Summary

Added support for Python dataclasses. Use @dynamodb_model on any dataclass to get DynamoDB persistence.

Usage

from dataclasses import dataclass
from pydynox import DynamoDBClient, dynamodb_model

client = DynamoDBClient(region="us-east-1")

@dynamodb_model(table="users", hash_key="pk", range_key="sk", client=client)
@dataclass
class User:
    pk: str
    sk: str
    name: str
    age: int = 0

# Save
user = User(pk="USER#1", sk="PROFILE", name="John", age=30)
user.save()

# Get
user = User.get(pk="USER#1", sk="PROFILE")

# Update
user.update(name="Jane")

# Delete
user.delete()

Changes

• integrations/dataclass.py - dataclass support
• integrations/functions.py - unified decorator that detects dataclass vs Pydantic
• integrations/_base.py - shared code
• docs/guides/dataclass.md - documentation
• Exported dynamodb_model from main __init__.py

Tests

• 17 unit tests
• 5 integration tests

All passing.

Why?

• Zero dependencies (dataclasses are built into Python 3.7+)
• Some users don't need Pydantic validation
• Simpler for basic use cases

Closes #68