motormongo 0.1.15

An asynchronous Object Document Mapper (ODM) for MongoDB built on top of Motor.

Description	Badge
PyPI - Version
Downloads
PyPI License
GitHub Contributors
Code Coverage
Code Style
Documentation

Author: Patrick Prunty.

motormongo is an Object Document Mapper (ODM) for MongoDB built on top of motor, the MongoDB recommended asynchronous Python driver for MongoDB Python applications, designed to work with Tornado or asyncio and enable non-blocking access to MongoDB.

Asynchronous operations in a backend system, built using FastAPI for example, enhances performance and scalability by enabling non-blocking, concurrent handling of multiple I/O requests, leading to more efficient use of server resources, by forcing the CPU usage on the backend server's main thread to be maximized across concurrent requests. For more low-level details on the advantages of asynchronous motormongo over existing MongoDB ODMs, such as mongoengine see here.

The interface for instantiating Document classes follows similar logic to mongoengine, enabling ease-of-transition and migration from mongoengine to the asynchronous motormongo.

Check out the documentation site here.

1. Installation
2. Quickstart
3. Why use motormongo?
4. motormongo Fields
5. CRUD classmethods
6. CRUD instance methods
7. Aggregation Operations
8. Polymorphism and Inheritance
9. Pooling Options and Configuration
10. FastAPI integration
11. License

Installation

To install motormongo, you can use pip inside your virtual environment:

python -m pip install motormongo

Alternatively, to install motormongo into your poetry environment:

poetry add motormongo

Quickstart

Step 1. Create a motormongo client:

import asyncio
from motormongo import DataBase


async def init_db():
    # This 'connect' method needs to be called inside of an async function
    await DataBase.connect(uri="<mongo_uri>", db="<mongo_database>")


if __name__ == "__main__":
    asyncio.run(init_db())

or, in a FastAPI application:

from fastapi import FastAPI
from motormongo import DataBase

app = FastAPI()


@app.on_event("startup")
async def startup_db_client():
    await DataBase.connect(uri="<mongo_uri>", db="<mongo_database>")


@app.on_event("shutdown")
async def shutdown_db_client():
    await DataBase.close()

The mongo_uri should look something like this:

mongodb+srv://<username>:<password>@<cluster>.mongodb.net

and database should be the name of an existing MongoDB database in your MongoDB instance.

For details on how to set up a local or cloud MongoDB database instance, see here.

You can also specify and pass pooling_options to the Motor on the DataBase.connect() method, like so:

import asyncio
from motormongo import DataBase

# Example pooling options
pooling_options = {
    'maxPoolSize': 50,
    'minPoolSize': 10,
    'maxIdleTimeMS': 30000,
    'waitQueueTimeoutMS': 5000,
    'connectTimeoutMS': 10000,
    'socketTimeoutMS': 20000
}


async def init_db():
    # This 'connect' method needs to be called inside of an async function
    await DataBase.connect(uri="<mongo_uri>", db="<mongo_database>", **pooling_options)


if __name__ == "__main__":
    asyncio.run(init_db())

See Pooling Options Configuration section for more details.

Step 2. Define a motormongo Document:

Define a motormongo User document:

import re
import bcrypt
from motormongo import Document, BinaryField, StringField


def hash_password(password: str) -> bytes:
    # Example hashing function
    return bcrypt.hashpw(password.encode('utf-8'), salt=bcrypt.gensalt())


class User(Document):
    username = StringField(help_text="The username for the user", min_length=3, max_length=50)
    email = StringField(help_text="The email for the user", regex=re.compile(r'^\S+@\S+\.\S+$'))  # Simple email regex
    password = BinaryField(help_text="The hashed password for the user", hash_function=hash_password)

    def verify_password(self, password: str) -> bool:
        """ Utility function which can be used to validate user's salted password later...
        
        ex.     user = await User.find_one({"_id": request.user_id})
                is_authenticated = user.verify_password(request.password)
        """
        return bcrypt.checkpw(password.encode("utf-8"), self.password)

    class Meta:
        collection = "users"  # < If not provided, will default to class name (ex. User->user, UserDetails->user_details)
        created_at_timestamp = True  # < Provide a DateTimeField for document creation
        updated_at_timestamp = True  # < Provide a DateTimeField for document updates

Step 3: Create a MongoDB document using the User class

import bcrypt

await User.insert_one(
    {
        "username": "johndoe",
        "email": "[e-mail]",
        "password": "password123"
        # < hash_functon will hash the string literal password and store binary field in the database
    }
)

Step 4: Validate user was created in your MongoDB collection

You can do this by using MongoDB compass GUI, or alternatively, add a query to find all documents in the user collection after doing the insert in step 3:

users = User.find_many({})
if users:
    print("User collection contains the following documents:")
    for user in users:
        print(user.to_dict())
else:
    print("User collection failed to update! Check your MongoDB connection details and try again!")

Step 5: Put all the code above into one file and run it

python main.py

or in a FastAPI application:

uvicorn main:app --reload

Please refer to FastAPI Documentation for more details on how to get setup with FastAPI.

Congratulations 🎉

You've successfully created your first motormongo Object Document Mapper class. 🥳

The subsequent sections detail the datatype fields provided by motormongo, as well as the CRUD operations available on the classmethods and object instance methods of a motormongo document.

If you wish to get straight into how to integrate motormongo with your FastAPI application, skip ahead to the FastAPI Integration section.

Why use motormongo?

Integrating motormongo, an asynchronous Object-Document Mapper (ODM) for MongoDB, into a backend built with an asynchronous web framework like FastAPI, enhances the system's ability to handle I/O-bound operations efficiently. Using the await keyword with motormongo operations allows the event loop to manage concurrent requests effectively, freeing up the main thread to handle other tasks while waiting for database operations to complete.

Understanding the Efficiency of Asynchronous Operations

Traditional Approach with pymongo and mongoengine (Synchronous)

Typically, a web server handling multiple requests that involve fetching documents from MongoDB would face bottlenecks with synchronous database operations:

from fastapi import FastAPI, HTTPException
from mongoengine import connect, Document, StringField
from fastapi.encoders import jsonable_encoder

connect(db="testdb", host="mongodb://localhost:27017/", alias="default")


class User(Document):
    name = StringField(required=True)


app = FastAPI()


@app.get("/get_data/")
async def get_data(name: str):
    user = User.objects(name=name).first()
    if user:
        return jsonable_encoder(user.to_mongo().to_dict())
    else:
        raise HTTPException(status_code=404, detail="User not found")

In this setup, operations such as User.objects(name=name).first() block the thread until completion, hindering the server's ability to process other requests concurrently, leading to inefficient resource use and potential performance bottlenecks.

Asynchronous Approach with motor and motormongo

Switching to an asynchronous ODM like motormongo allows FastAPI to handle database operations without blocking:

from fastapi import FastAPI
from motormongo import DataBase, Document, StringField


class User(Document):
    name = StringField(required=True)


app = FastAPI()


@app.on_event("startup")
async def startup_db_client():
    await DataBase.connect(uri="mongodb://localhost:27017/", db="testdb")


@app.get("/get_data/")
async def get_data(name: str):
    user = await User.find_one({'name': name})
    if user:
        return user.to_dict()
    else:
        return {"message": "User not found"}

Using await with motormongo operations such as User.find_one() allows the application to perform non-blocking database operations. This asynchronous model is particularly advantageous for I/O-bound applications, allowing the server to handle multiple requests efficiently by utilizing Python's asyncio.

Maximizing Performance with Multiple FastAPI Workers

To further enhance the performance of FastAPI applications utilizing motormongo, deploying multiple worker processes can significantly increase the application's ability to handle high volumes of concurrent requests:

• Scalability: Deploying FastAPI with multiple workers enables the application to scale across multiple CPU cores, offering better handling of concurrent requests by running multiple instances of the application, each in its own process.
• Resource Utilization: More workers mean that the application can utilize more system resources, effectively distributing the load and preventing any single worker from becoming a bottleneck.
• Deployment Strategy: Use an ASGI server like uvicorn with the --workers option to specify the number of worker processes. For example, uvicorn app:app --workers 4 would run the application with four worker processes.

By leveraging motormongo with FastAPI, developers can build backend systems capable of handling asynchronous I/O-bound operations efficiently. This setup not only improves the application's responsiveness and throughput by utilizing the asynchronous capabilities of Python's asyncio but also maximizes performance through the strategic deployment of multiple workers. Together, these strategies enable the creation of highly scalable, efficient, and modern web applications.

motormongo Fields

motormongo supports a variety of field types to accurately define the schema of your MongoDB documents. Each field type is designed to handle specific data types and validations:

• BinaryField: Stores binary data, useful for storing encoded or hashed data like passwords.
• BooleanField: Stores boolean values (True or False).
• DateTimeField: Manages date and time, with options for automatically setting current date/time on creation or update.
• EmbeddedDocumentField: For fields that should contain values from a predefined enumeration.
• EnumField: For fields that should contain values from a predefined enumeration.
• FloatField: Handles floating-point numbers, with options to specify minimum and maximum values.
• GeoJSONField: Manages geographical data in GeoJSON format, with an option to return data as JSON.
• IntegerField: Manages integer data, allowing specifications for minimum and maximum values.
• ListField: Handles lists of items, which can be of any type.
• ReferenceField: Creates a reference to another document.
• StringField: Handles string data with options for minimum and maximum length, and regex validation.

All fields have the following default parameters:

• default: Specifies the default value for the field if no value is provided. This parameter can be a static value or a callable object. The callable object is useful for dynamic values like generating timestamps or unique identifiers.
• required: A boolean indicating whether the field is mandatory. If set to True, the document cannot be saved without providing a value for this field.
• unique: A boolean specifying if the field value should be unique across the collection. This is crucial for fields like email addresses or usernames.

BinaryField

The BinaryField is designed for storing binary data within a database. It offers capabilities for encoding, hashing, and decoding data, making it versatile for handling various types of binary data, including but not limited to encrypted or hashed content.

Parameters:

• hash_function: (Optional) A callable that hashes input data. The function should have a type annotation to indicate whether it expects a str or bytes as input. This annotation is crucial as it dictates whether the BinaryField should encode the string before hashing. If the annotation indicates str, the field will pass the string directly to the hash_function. If bytes, the BinaryField will encode the string (using the provided encode function or default UTF-8 encoding) before hashing.
• return_decoded: (Optional) A boolean indicating whether to decode binary data when it is retrieved from the database. If set to True, the stored binary data will be decoded back into a string using the provided decode function or default UTF-8 decoding. This is useful for data that was encoded but not hashed, as hashed data cannot be meaningfully decoded.
• encode: (Optional) A function to encode a string to bytes before storage. If not provided, the class defaults to UTF-8 encoding. This function is used when the input data is a string and needs to be stored as binary data, or before hashing if the hash_function expects bytes.
• decode: (Optional) A function to decode bytes back to a string when data is retrieved from the database. This parameter is only relevant if return_decoded is True. If not provided, the class defaults to UTF-8 decoding.

Important: For the hash_function to work correctly with the BinaryField, it must include type annotations for its parameters. This enables the BinaryField to determine the correct processing strategy (i.e., whether to encode the string before hashing).

Example Usage:

from motormongo import Document, BinaryField, StringField
import bcrypt


# Hash function with type annotation indicating it expects a 'str'
def hash_password(password: str) -> bytes:
    return bcrypt.hashpw(password.encode('utf-8'), bcrypt.gensalt())


class User(Document):
    username = StringField(min_length=3, max_length=50)
    # Note: 'hash_function' requires a type annotation
    password = BinaryField(hash_function=hash_password, return_decoded=False)

    def verify_password(self, password: str) -> bool:
        # Verifies if the provided password matches the stored hash
        return bcrypt.checkpw(password.encode("utf-8"), self.password)


# Creating a user instance with a hashed password
user = User(username="johndoe", password="secret")
inserted_user = await user.save()

# Authentication checks
is_authenticated = inserted_user.verify_password("wrongpassword")  # Expected to return False
is_authenticated = inserted_user.verify_password("secret")  # Expected to return True

BooleanField

The BooleanField is used for storing boolean values (True or False). It ensures that the data stored in this field is strictly boolean.

Parameters:

• There are no specific parameters unique to BooleanField other than those inherited from the base Field class.

from motormongo import Document, BooleanField, StringField


class Product(Document):
    name = StringField(min_length=1, max_length=100)
    is_available = BooleanField(default=False)


# Create a product indicating its availability
product = Product(name="Gadget", is_available=True)
await product.save()

DateTimeField

The DateTimeField handles date and time values, with options to automatically update these values on document creation or modification.

Parameters:

• auto_now: Automatically update the field to the current datetime when the document is saved.
• auto_now_add: Automatically set the field to the current datetime when the document is created.
• datetime_formats: List of string formats to parse datetime strings.

Example Usage:

from motormongo import Document, DateTimeField


class Event(Document):
    start_time = DateTimeField(auto_now_add=True)


# Create an event with the current start time
event = Event()
await event.save()

EmbeddedDocumentField

The EmbeddedDocumentField is used for embedding documents within a document, supporting nested document structures. This field allows you to include complex data structures as part of your document.

Parameters:

• document_type: The class of the embedded document, which must be a subclass of EmbeddedDocument, BaseModel from Pydantic, or dict representation of the EmbeddedDocument.

Example Usage:

from motormongo import Document, EmbeddedDocument, EmbeddedDocumentField, StringField
from pydantic import BaseModel


class Address(EmbeddedDocument):
    street = StringField()
    city = StringField()


class User(Document):
    name = StringField()
    address = EmbeddedDocumentField(document_type=Address)


class PydanticAddress(BaseModel):
    street: str
    city: str


# Create a user with an embedded address document
user = User(name="John Doe", address={"street": "123 Elm St", "city": "Springfield"})
# user = User(name="John Doe", address=Address(street="123 Elm St", city="Springfield")) # Also works
# user = User(name="John Doe", address=PydanticAddress(street="123 Elm St", city="Springfield")) # Also works
await user.save()

EnumField

The EnumField is designed to store enumerated values, allowing for validation against a predefined set of options.

Parameters:

• enum: The enumeration class that defines valid values for the field.

Example Usage:

import enum
from motormongo import Document, EnumField


class UserStatus(enum.Enum):
    ACTIVE = 'active'
    INACTIVE = 'inactive'
    BANNED = 'banned'


class User(Document):
    status = EnumField(enum=UserStatus)


# Create a user and set their status using the EnumField
user = User(status=UserStatus.ACTIVE)
# user = User(status="active") # Also works
await user.save()

FloatField

The FloatField handles floating-point numbers, with options to specify minimum and maximum values.

Parameters:

• min_value: (Optional) The minimum allowable value.
• max_value: (Optional) The maximum allowable value.

Example Usage:

from motormongo import Document, FloatField


class Measurement(Document):
    temperature = FloatField(min_value=-273.15)  # Absolute zero constraint


# Record a temperature measurement
measurement = Measurement(temperature=25.5)
await measurement.save()

GeoJSONField

The GeoJSONField is designed for storing geographical coordinates in GeoJSON format.

Parameters:

• return_as_list: (Optional) If True, returns the coordinates as a [longitude, latitude] list instead of a GeoJSON object.

Example Usage:

from motormongo import Document, GeoJSONField


class Location(Document):
    point = GeoJSONField()


# Create a location point
location = Location(point={"type": "Point", "coordinates": [-73.856077, 40.848447]})  # Could also use 
# location = Location(point=[-73.856077, 40.848447]) # This would also work
await location.save()

IntegerField

The IntegerField is used for storing integer values, with optional validation for minimum and maximum values.

Parameters:

• min_value: (Optional) The minimum allowable value.
• max_value: (Optional) The maximum allowable value.

Example Usage:

from motormongo import Document, IntegerField


class Product(Document):
    quantity = IntegerField(min_value=0)


# Create a product with quantity validation
product = Product(quantity=10)
await product.save()

ListField

The ListField is used for storing a list of items, optionally validating the type of items in the list.

Parameters:

• field: (Optional) A Field instance specifying the type of items in the list.

Example Usage:

from motormongo import Document, ListField, StringField


class ShoppingList(Document):
    items = ListField(field=StringField())


# Create a shopping list with string items
shopping_list = ShoppingList(items=["Milk", "Eggs", "Bread"])
await shopping_list.save()

ReferenceField

The ReferenceField is used to create a reference to another document, typically for creating relationships between collections.

Parameters:

• document: The class of the document to which the field references.

Example Usage:

from motormongo import Document, ReferenceField, StringField


class User(Document):
    name = StringField()


class Post(Document):
    author = ReferenceField(document=User)


# Create a user and a post referencing the user
user = User(name="John Doe")
await user.save()
post = Post(author=user)

To fetch the referenced document, you must await the coroutine returned by accessing the reference field. This operation asynchronously retrieves the related document instance from the database.

# Assuming `post` is an instance of the Post document with a reference to a User
# Fetch the user referenced by the post's author field
referenced_user = await post.author
if referenced_user:
    print("Referenced User:",
          referenced_user.to_dict())  # Should print, {'_id': '65d8bf2dad3fa2e9169d2f94', 'name': 'John Doe'}
else:
    print("User not found or failed to fetch.")

This example demonstrates how to access and asynchronously fetch the document referenced by a ReferenceField. The await keyword is crucial because the operation is asynchronous, involving a database query to retrieve the referenced document.

Note: Ensure that the fetching operation is performed within an asynchronous context, such as an async function. The ReferenceField provides a powerful way to manage relationships between documents, enabling complex data models with interconnected documents.

StringField

The StringField is used for storing string data in a document. It supports validation for minimum and maximum length and can enforce a specific regex pattern.

Parameters:

• min_length: (Optional) The minimum length of the string.
• max_length: (Optional) The maximum length of the string.
• regex: (Optional) A regex pattern that the string must match.

Example Usage:

from motormongo import Document, StringField


class UserProfile(Document):
    username = StringField(min_length=3, max_length=50)
    bio = StringField(max_length=200, regex=r'^[A-Za-z0-9 ]*$')  # Alphanumeric and space only


# Create a user profile with validation
profile = UserProfile(username="user123", bio="I love coding.")
await profile.save()

Class methods

Operations

The following class methods are supported by motormongo's Document class:

CRUD Type	Operation
Create	insert_one(document: dict, **kwargs) -> Document
Create	insert_many(documents: List[dict]) -> Tuple[List[Document], Any]
Read	find_one(query: dict, **kwargs) -> Document
Read	find_many(query: dict, limit: int = None, return_as_list: bool = True **kwargs) -> List[Document]
Update	update_one(query: dict, update_fields: dict) -> Document
Update	update_many(query: dict, update_fields: dict) -> Tuple[List[Document], int]
Delete	delete_one(query: dict, **kwargs) -> bool
Delete	delete_many(query: dict, **kwargs) -> int
Mixed	find_one_or_create(query: dict, defaults: dict) -> Tuple[Document, bool]
Mixed	find_one_and_replace(query: dict, replacement: dict) -> Document
Mixed	find_one_and_delete(query: dict) -> Document
Mixed	find_one_and_update_empty_fields(query: dict, update_fields: dict) -> Tuple[Document, bool]

All examples below assume User is a subclass of motormongo provided Document class.

Create

insert_one(document: dict, **kwargs) -> Document

Inserts a single document into the database.

user = await User.insert_one({
    "name": "John",
    "age": 24,
    "alive": True
})

Alternatively, using **kwargs:

user = await User.insert_one(
    name="John",
    age=24,
    alive=True)

And similarly, with a dictionary:

user_document = {
    "name": "John",
    "age": 24,
    "alive": True
}
user = await User.insert_one(**user_document)

insert_many(List[document]) -> tuple[List['Document'], Any]

users, user_ids = await User.insert_many(
    [
        {
            "name": "John",
            "age": 24,
            "alive": True
        },
        {
            "name": "Mary",
            "age": 2,
            "alive": False
        }
    ]
)

or

docs_to_insert = [{"name": "Alice", "age": 30}, {"name": "Bob", "age": 25}]
inserted_docs, inserted_ids = await User.insert_many(docs_to_insert)

Read

find_one(query, **kwargs) -> Document

user = await User.find_one(
    {
        "_id": "655fc281c440f677fa1e117e"
    }
)

Alternatively, using **kwargs:

user = await User.find_one(_id="655fc281c440f677fa1e117e")

Note: The _id string datatype here is automatically converted to a BSON ObjectID, however, motormongo handles the scenario when a BSON ObjectId is passed as the _id datatype:

from bson import ObjectId

user = await User.find_one(
    {
        "_id": ObjectId("655fc281c440f677fa1e117e")
    }
)

find_many(query, limit, **kwargs) -> List[Document]

users = await User.find_many(age={"$gt": 40}, alive=False, limit=20)

or

query = {"age": {"$gt": 40}, "alive": False}
users = await User.find_many(**query, limit=20)

Update

update_one(query, updated_fields) -> Document

updated_user = await User.update_one(
    {
        "_id": "655fc281c440f677fa1e117e"
    },
    {
        "name": "new_name",
        "age": 30
    }
)

or

query_criteria = {"name": "old_name"}
update_data = {"name": "updated_name"}
updated_user = await User.update_one(query_criteria, update_data)

update_many(qeury, fields) -> Tuple[List[Any], int]

updated_users, modified_count = await User.update_many({'age': {'$gt': 40}}, {'category': 'senior'})

another example:

updated_users, modified_count = await User.update_many({'name': 'John Doe'}, {'$inc': {'age': 1}})

Delete

delete_one(query, **kwargs) -> bool

deleted = await User.delete_one({'_id': '507f191e810c19729de860ea'})

Alternatively, using **kwargs:

deleted = await User.delete_one(name='John Doe')

delete_many(query, **kwargs) -> int

deleted_count = await User.delete_many({'age': {'$gt': 40}})

Another example:

# Delete all users with a specific status
deleted_count = await User.delete_many({'status': 'inactive'})

Alternatively, using **kwargs:

deleted_count = await User.delete_many(status='inactive')

Mixed

find_one_or_create(query, defaults) -> Tuple['Document', bool]

user, created = await User.find_one_or_create({'username': 'johndoe'}, defaults={'age': 30})

find_one_and_replace(query, replacement) -> Document

replaced_user = await User.find_one_and_replace({'username': 'johndoe'}, {'username': 'johndoe', 'age': 35})

find_one_and_delete(query) -> Document

deleted_user = await User.find_one_and_delete({'username': 'johndoe'})

find_one_and_update_empty_fields(query, update_fields) -> Tuple['Document', bool]

updated_user, updated = await User.find_one_and_update_empty_fields(
    {'username': 'johndoe'},
    {'email': '[e-mail]', 'age': 30}
)

Instance methods

motormongo also supports the manipulation of fields on the object instance. This allows users to programmatically achieve the same operations listed above through the object instance itself.

Operations

The following are object instance methods are supported by motormongo's Document class:

CRUD Type	Operation
Create	save() -> None
Delete	delete() -> None

Note: All update operations can be manipulated on the fields in the Document class object itself.

user.save() -> None

# Find user by MongoDB _id
user = await User.find_one(
    {
        "_id": "655fc281c440f677fa1e117e"
    }
)
# If there age is greater than 80, make them dead
if user.age > 80:
    user.alive = False
# Persist update on User instance in MongoDB mongo
user.save()

In this example, User.find_one() returns an instance of User. If the age field is greater than 80, the alive field is set to false. The instance of the document in the MongoDB database is then updated by calling the .save() method on the User object instance.

Delete

user.delete() -> None

# Find all users where the user is not alive
users = await User.find_many(
    {
        "alive": False
    }
)
# Recursively delete all User instances in the users list who are not alive
for user in users:
    user.delete()

Aggregation

The aggregate class method is designed to perform aggregation operations on the documents within the collection. It allows the execution of a sequence of data aggregation operations defined by the pipeline parameter. This method can return the results either as a list of documents or as a cursor, based on the return_as_list flag.

Parameters:

• pipeline: A list of dictionaries defining the aggregation operations to be performed on the collection.
• return_as_list (optional): A boolean flag that determines the format of the returned results. If set to True, the method returns a list of documents. If False (default), it returns a cursor.

Returns:

• If return_as_list is True, returns a list of documents resulting from the aggregation pipeline.
• If return_as_list is False, returns a Cursor to iterate over the results.

Raises:

• ValueError: If an error occurs during the execution of the pipeline.

Example Usage:

from yourmodule import YourDocumentClass

# Connect to the database (Assuming the database connection is already set up)
# Define an aggregation pipeline
pipeline = [
    {"$match": {"status": "active"}},
    {"$project": {"_id": 0, "username": 1, "status": 1}},
    {"$sort": {"username": 1}}
]

# Execute the aggregation without returning a list
cursor = await YourDocumentClass.aggregate(pipeline)
async for doc in cursor:
    print(doc)

# Execute the aggregation and return results as a list
docs_list = await YourDocumentClass.aggregate(pipeline, return_as_list=True)
print(docs_list)

Polymorphism and Inheritance

This part of the documentation provides an overview of implementing and using polymorphism and inheritance using the motormongo framework, enabling flexible and organized data models for various use cases.

Base Model: Item

The Item class serves as the base model for different types of items stored in a MongoDB collection. It defines common fields and methods that are shared across all item types.

from motormongo import Document, StringField, FloatField


class Item(Document):
    name = StringField()
    cost = FloatField()

Subclass Models

Subclasses of Item can introduce specific fields or override methods to cater to different item categories.

Book

A Book represents a specific type of Item with additional attributes related to books.

class Book(Item):
    title = StringField()
    author = StringField()
    isbn = StringField()

Electronics

An Electronics item represents electronic goods with attributes like warranty period and brand.

class Electronics(Item):
    warranty_period = StringField()  # E.g., "2 years"
    brand = StringField()

Usage

Creating and Inserting Items

To insert items into the database, use the insert_one method. The item's type is managed automatically.

# Insert a book
book = await Book.insert_one(title="1984", author="George Orwell", isbn="123456789", cost=20.0, name="Book")

# Insert an electronics item
electronics = await Electronics.insert_one(warranty_period="2 years", brand="TechBrand", cost=999.99, name="Laptop")

Querying Items

You can query items of any type using their base or specific models. Polymorphism allows retrieved instances to be of the correct subclass.

# Find a book by ISBN
found_book = await Book.find_one(isbn="123456789")

# Find an electronics item by brand
found_electronics = await Electronics.find_one(brand="TechBrand")

Polymorphic Behavior

The following operations are supported over the base Item Document class, enabling complex querying over base Item Document class and all of its subclasses (i.e Book and Electronics):

CRUD Type	Operation
Read	find_many(query: dict, limit: int = None, return_as_list: bool = True **kwargs) -> List[Document]
Update	update_many(query: dict, update_fields: dict) -> Tuple[List[Document], int]
Delete	delete_many(query: dict, **kwargs) -> int

As well as aggregate operations, see the Aggregation Operation section for more details.

Querying on the base Item model returns items of all types, correctly instantiated as their specific subclasses. See below for a logical example of polymorphic querying:

# Find all items with a cost over 50
expensive_items = await Item.find_many(cost={"$gt": 50})

for item in expensive_items:
    print(type(item))  # Prints the subclass (Book, Electronics, etc.)
    if isinstance(item, Book):
        print(f"Book: {item.title} by {item.author}")
    elif isinstance(item, Electronics):
        print(f"Electronics: {item.brand} with {item.warranty_period} warranty")

Pooling Options Configuration

In motormongo, you have the flexibility to customize the pooling options for the Motor client. This allows you to fine-tune the behavior of database connections according to your application's needs. Below are some of the parameters you can configure, along with their descriptions and example usage.

Configuration Parameters

• Max Pool Size: The maximum number of connections in the connection pool.
• Min Pool Size: The minimum number of connections in the connection pool.
• Max Idle Time: The maximum time (in milliseconds) a connection can remain idle in the pool before being closed.
• Wait Queue Timeout: The time (in milliseconds) a thread will wait for a connection to become available when the pool is exhausted.
• Connect Timeout: The time (in milliseconds) to wait for a connection to the MongoDB server to be established before timing out.
• Socket Timeout: The time (in milliseconds) to wait for a socket read or write to complete before timing out.

Example Configuration

import asyncio
from motormongo import DataBase

# Example pooling options
pooling_options = {
    'maxPoolSize': 50,
    'minPoolSize': 10,
    'maxIdleTimeMS': 30000,
    'waitQueueTimeoutMS': 5000,
    'connectTimeoutMS': 10000,
    'socketTimeoutMS': 20000
}


async def init_db():
    # This 'connect' method needs to be called inside of an async function
    await DataBase.connect(uri="<mongo_uri>", db="<mongo_database>", **pooling_options)


if __name__ == "__main__":
    asyncio.run(init_db())

or in FastAPI:

from fastapi import FastAPI
from motormongo import DataBase

app = FastAPI()

# Example pooling options
pooling_options = {
    'maxPoolSize': 50,
    'minPoolSize': 10,
    'maxIdleTimeMS': 30000,
    'waitQueueTimeoutMS': 5000,
    'connectTimeoutMS': 10000,
    'socketTimeoutMS': 20000
}


@app.on_event("startup")
async def startup_db_client():
    await DataBase.connect(uri="<mongodb_uri>", db="<mongodb_db>", **pooling_options)

This configuration demonstrates how to set up motormongo with specific pooling options to optimize performance and resource utilization in high-throughput environments.

For more information, consult the official documentation:

• Motor: Asynchronous Python driver for MongoDB
• MongoDB Manual: Tuning Your Connection Pool Settings

FastAPI integration

motormongo can be easily integrated in FastAPI APIs to leverage the asynchronous ability of FastAPI. To leverage motormongo's ease-of-use, Pydantic model's should be created to represent the MongoDB motormongo Document as a Pydantic model. Below is a light-weight CRUD FastAPI application using motormongo:

from fastapi import FastAPI, HTTPException
from pydantic import BaseModel, Field
from motormongo import DataBase, Document, BinaryField, StringField
import re
import bcrypt


def hash_password(password: str) -> bytes:
    # Example hashing function
    return bcrypt.hashpw(password.encode('utf-8'), salt=bcrypt.gensalt())


class User(Document):
    username = StringField(help_text="The username for the user", min_length=3, max_length=50)
    email = StringField(help_text="The email for the user", regex=re.compile(r'^\S+@\S+\.\S+$'))  # Simple email regex
    password = BinaryField(help_text="The hashed password for the user", hash_function=hash_password)

    def verify_password(self, password: str) -> bool:
        return bcrypt.checkpw(password.encode("utf-8"), self.password)

    class Meta:
        collection = "users"  # < If not provided, will default to class name (ex. User->user, UserDetails->user_details)
        created_at_timestamp = True  # < Provide a DateTimeField for document creation
        updated_at_timestamp = True  # < Provide a DateTimeField for document updates


class UserModelRequest(BaseModel):
    username: str = Field(example="johndoe")
    email: str = Field(example="[e-mail]")
    password: str = Field(example="password123")


app = FastAPI()


@app.on_event("startup")
async def startup_db_client():
    await DataBase.connect(uri="<mongodb_uri>", db="<mongodb_db>")


@app.on_event("shutdown")
async def shutdown_db_client():
    await DataBase.close()


@app.post("/users/", status_code=201)
async def create_user(user: UserModelRequest):
    new_user = await User.insert_one(**user.dict())
    return new_user.to_dict()


@app.post("/user/auth", status_code=200)
async def is_authenticated(username: str, password: str):
    user = await User.find_one({"username": username})
    if not user:
        raise HTTPException(status_code=404, detail="User not found")
    if not user.verify_password(password):
        raise HTTPException(status_code=401, detail="Unauthorized")
    else:
        return "You are authenticated! You can see this!"


@app.get("/users")
async def get_users():
    users = await User.find_many()
    if not users:
        raise HTTPException(status_code=404, detail="User not found")
    return [user.to_dict() for user in users]


@app.get("/users/{user_id}")
async def get_user(user_id: str):
    user = await User.find_one({"_id": user_id})
    if not user:
        raise HTTPException(status_code=404, detail="User not found")
    return user.to_dict()


@app.put("/users/{user_id}", status_code=200)
async def update_user(user_id: str, user_data: UserModelRequest):
    updated_user = await User.update_one({"_id": user_id}, user_data.model_dump())
    if not updated_user:
        raise HTTPException(status_code=404, detail="User not found")
    return updated_user.to_dict()


@app.delete("/users/{user_id}", status_code=204)
async def delete_user(user_id: str):
    user = await User.find_one({"_id": user_id})
    if not user:
        raise HTTPException(status_code=404, detail="User not found")
    await user.delete()
    return {"status": "User deleted successfully"}

License

This project is licensed under the MIT License.