pydantic_mongo_orm 0.0.0

ORM working with MongoDb and using Pydantic as a model.

Pydantic Mongo ORM

Pydantic Mongo ORM is a library that allows users to map Pydantic models to MongoDb.

Installation

pip install pydantic_mongo_orm

Usage

Models

PMO uses Pydantic models to represent data in database. To start using it, you need to create new model based on pydantic_mongo_orm.BaseModel. This class implements all the features from Pydantic but also adds some orm-specific functionality, e.g. adds read-only field id that represents document-id (_id) from MongoDb:

from pydantic_mongo_orm import BaseModel

class MyModel(BaseModel):
    greeting: str

my_model = MyModel(greeting='Hello World!')

print(my_model.greeting)  # >> 'Hello World!'
print(my_model.id)  # >> uuid.UUID4('2daaa0ac-501b-4018-9101-cc822a37c1a1')

You may also specify mongo_config for each model class that further modifies ORM behaviour:

from pydantic_mongo_orm import BaseModel, MongoConfigDict

class MyModel(BaseModel):
    greeting: str

    mongo_config = MongoConfigDict(collection='my_model')

Available options are:

• collection: Name of the collection where the model will be stored. By default, it's detected from the class name.
• objects_class: Class that will be used as an object storage for given model. By default, ObjectsStorage is used. Explained later

Storage

The base of PMO is Storage - class connected to MongoDb able to map models to objects and vice-versa.

Connect to MongoDb and save new instance of MyModel from previous example:

from pydantic_mongo_orm import Storage

storage = Storage(host='mongodb://127.0.0.1:27017', db_name='local')
storage.connect()

my_model = MyModel(greeting='Hello World!')
storage.save(my_model)

my_models = storage.find(MyModel, {'greeting': 'Hello World!'})
print(list(my_models))  # [MyModel greetings: 'Hello World!']

Bound objects

You may use Storage object directly but PMO offers a shorthand that simplifies working with the ORM. Each model class provides attribute objects that provides direct access to collection storing that particular model. To enable this feature, you must first bind the Storage to the BaseModel using .bind(). This example is equivalent with the previous one:

from pydantic_mongo_orm import Storage

storage = Storage(host='mongodb://127.0.0.1:27017', db_name='local')
storage.connect()
storage.bind()

my_model = MyModel(greeting='Hello World!')
my_model.save()

my_models = MyModel.objects.find({'greeting': 'Hello World!'})
print(list(my_models))  # [MyModel greetings: 'Hello World!']

By default, the objects attribute contains an instance of pydantic_mongo_orm.ObjectsStorage. It provides the same functionality as pydantic_mongo_orm.Storage. If you wish to customize behaviour of certain model - change how it's stored, add custom filters etc., you may extend the ObjectsStorage class, add new functionality and then configure it to be used for your models:

from pydantic_mongo_orm import BaseModel, ObjectsStorage, Storage, MongoConfigDict


class RectangleStorage(ObjectsStorage):
    def squares(self):
        return self.find({'$where': 'this.x == this.y'})

    
class Rectangle(BaseModel):
    mongo_config = MongoConfigDict(objects_class=RectangleStorage)

    length: int
    width: int

    
storage = Storage(host='mongodb://127.0.0.1:27017', db_name='local')
storage.connect()
storage.bind()

my_square = Rectangle(x=5, y=5)
my_square.save()

my_rectangle = Rectangle(x=5, y=10)
my_rectangle.save()

print(list(Rectangle.objects.squares()))  # >> [Rectangle x=5 y=5]