Pydantic Mongo

A Python library that offers an easy-to-use Repository pattern for MongoDB, supporting both synchronous and asynchronous operations. It simplifies working with databases by providing a clear interface for CRUD (Create, Read, Update, Delete) operations using Pydantic models. With built-in data validation and serialization from Pydantic, it helps manage your MongoDB data safely.

Read the documentation

Features

• Asynchronous and Synchronous support
• Pydantic models integration
• Type-safe MongoDB operations
• Cursor-based pagination

Installation

pip install pydantic-mongo

Usage Examples

Defining Models and Repository

from bson import ObjectId
from pydantic import BaseModel
from pydantic_mongo import AbstractRepository, PydanticObjectId
from pymongo import MongoClient
from typing import Optional, List

# Define your models
class Foo(BaseModel):
   count: int
   size: float = None

class Bar(BaseModel):
   apple: str = 'x'
   banana: str = 'y'

class Spam(BaseModel):
   # PydanticObjectId is an alias to Annotated[ObjectId, ObjectIdAnnotation]
   id: Optional[PydanticObjectId] = None
   foo: Foo
   bars: List[Bar]

# Create a repository
class SpamRepository(AbstractRepository[Spam]):
   class Meta:
      collection_name = 'spams'

# Connect to database
client = MongoClient("mongodb://localhost:27017")
database = client["example"]
repo = SpamRepository(database)

Creating and Saving Documents

# Create a new document
spam = Spam(foo=Foo(count=1, size=1.0), bars=[Bar()])

# Create a document with predefined ID
spam_with_predefined_id = Spam(
   id=ObjectId("611827f2878b88b49ebb69fc"),
   foo=Foo(count=2, size=2.0),
   bars=[Bar()]
)

# Save a single document
repo.save(spam)  # spam.id is now set to an ObjectId

# Save multiple documents
repo.save_many([spam, spam_with_predefined_id])

Querying Documents

# Find by ID
result = repo.find_one_by_id(spam.id)

# Find by ID using string
result = repo.find_one_by_id(ObjectId('611827f2878b88b49ebb69fc'))
assert result.foo.count == 2

# Find one by custom query
result = repo.find_one_by({'foo.count': 1})

# Find multiple documents by query
results = repo.find_by({'foo.count': {'$gte': 1}})

Pagination

# Get first page
edges = repo.paginate({'foo.count': {'$gte': 1}}, limit=10)

# Get next page using the last cursor
more_edges = repo.paginate(
    {'foo.count': {'$gte': 1}}, 
    limit=10, 
    after=list(edges)[-1].cursor
)

Deleting Documents

# Delete a document
repo.delete(spam)

# Delete by ID
repo.delete_by_id(ObjectId("..."))

Async Support

For asynchronous applications, you can use AsyncAbstractRepository which provides the same functionality as AbstractRepository but with async/await support:

from pymongo import AsyncMongoClient
from pydantic import BaseModel
from pydantic_mongo import AsyncAbstractRepository

class User(BaseModel):
    id: str
    name: str
    email: str

class UserRepository(AsyncAbstractRepository[User]):
    class Meta:
        collection_name = 'users'

# Initialize database connection
client = AsyncMongoClient('mongodb://localhost:27017')
database = client["mydb"]

# Create repository instance
user_repo = UserRepository(database)

# Example usage
user = User(name='John Doe', email='john@example.com')
await user_repo.save(user)

user = await user_repo.find_one_by_id(user_id)

License

MIT License