Skip to main content

aiomongodel

https://travis-ci.org/ilex/aiomongodel.svg?branch=master Documentation Status

An asynchronous ODM similar to PyMODM on top of Motor an asynchronous Python MongoDB driver. Works on Python 3.5 and up. Some features such as asynchronous comprehensions require at least Python 3.6. aiomongodel can be used with asyncio as well as with Tornado.

Usage of session requires at least MongoDB version 4.0.

Install

Install aiomongodel using pip:

pip install aiomongodel

Documentation

Read the docs.

Getting Start

Modeling

To create a model just create a new model class, inherit it from aiomongodel.Document class, list all the model fields and place a Meta class with model meta options. To create a subdocument, create a class with fields and inherit it from aiomongodel.EmbeddedDocument.

# models.py

from datetime import datetime

from pymongo import IndexModel, DESCENDING

from aiomongodel import Document, EmbeddedDocument
from aiomongodel.fields import (
    StrField, BoolField, ListField, EmbDocField, RefField, SynonymField,
    IntField, FloatField, DateTimeField, ObjectIdField)

class User(Document):
    _id = StrField(regex=r'[a-zA-Z0-9_]{3, 20}')
    is_active = BoolField(default=True)
    posts = ListField(RefField('models.Post'), default=lambda: list())
    quote = StrField(required=False)

    # create a synonym field
    name = SynonymField(_id)

    class Meta:
        collection = 'users'

class Post(Document):
    # _id field will be added automatically as
    # _id = ObjectIdField(defalut=lambda: ObjectId())
    title = StrField(allow_blank=False, max_length=50)
    body = StrField()
    created = DateTimeField(default=lambda: datetime.utcnow())
    views = IntField(default=0)
    rate = FloatField(default=0.0)
    author = RefField(User, mongo_name='user')
    comments = ListField(EmbDocField('models.Comment'), default=lambda: list())

    class Meta:
        collection = 'posts'
        indexes = [IndexModel([('created', DESCENDING)])]
        default_sort = [('created', DESCENDING)]

class Comment(EmbeddedDocument):
    _id = ObjectIdField(default=lambda: ObjectId())
    author = RefField(User)
    body = StrField()

# `s` property of the fields can be used to get a mongodb string name
# to use in queries
assert User._id.s == '_id'
assert User.name.s == '_id'  # name is synonym
assert Post.title.s == 'title'
assert Post.author.s == 'user'  # field has mongo_name
assert Post.comments.body.s == 'comments.body'  # compound name

CRUD

from motor.motor_asyncio import AsyncIOMotorClient

async def go(db):
    # create model's indexes
    await User.q(db).create_indexes()

    # CREATE
    # create using save
    # Note: if do_insert=False (default) save performs a replace
    # with upsert=True, so it does not raise if _id already exists
    # in db but replace document with that _id.
    u = await User(name='Alexandro').save(db, do_insert=True)
    assert u.name == 'Alexandro'
    assert u._id == 'Alexandro'
    assert u.is_active is True
    assert u.posts == []
    assert u.quote is None
    # using query
    u = await User.q(db).create(name='Ihor', is_active=False)

    # READ
    # get by id
    u = await User.q(db).get('Alexandro')
    assert u.name == 'Alexandro'
    # find
    users = await User.q(db).find({User.is_active.s: True}).to_list(10)
    assert len(users) == 2
    # using for loop
    users = []
    async for user in User.q(db).find({User.is_active.s: False}):
        users.append(user)
    assert len(users) == 1
    # in Python 3.6 an up use async comprehensions
    users = [user async for user in User.q(db).find({})]
    assert len(users) == 3

    # UPDATE
    u = await User.q(db).get('Ihor')
    u.is_active = True
    await u.save(db)
    assert (await User.q(db).get('Ihor')).is_active is True
    # using update (without data validation)
    # object is reloaded from db after update.
    await u.update(db, {'$push': {User.posts.s: ObjectId()}})

    # DELETE
    u = await User.q(db).get('Ihor')
    await u.delete(db)


loop = asyncio.get_event_loop()
client = AsyncIOMotorClient(io_loop=loop)
db = client.aiomongodel_test
loop.run_until_complete(go(db))

Validation

Use model’s validate method to validate model’s data. If there are any invalid data an aiomongodel.errors.ValidationError will raise.

class Model(Document):
    name = StrField(max_length=7)
    value = IntField(gt=5, lte=13)
    data = FloatField()

def go():
    m = Model(name='xxx', value=10, data=1.6)
    # validate data
    # should not raise any error
    m.validate()

    # invalid data
    # note that there are no errors while creating
    # model with invalid data
    invalid = Model(name='too long string', value=0)
    try:
        invalid.validate()
    except aiomongodel.errors.ValidationError as e:
        assert e.as_dict() == {
            'name': 'length is greater than 7',
            'value': 'value should be greater than 5',
            'data': 'field is required'
        }

        # using translation - you can translate messages
        # to your language or modify them
        translation = {
            "field is required": "This field is required",
            "length is greater than {constraint}": ("Length of the field "
                                                    "is greater than "
                                                    "{constraint} characters"),
            # see all error messages in ValidationError docs
            # for missed messages default messages will be used
        }
        assert e.as_dict(translation=translation) == {
            'name': 'Length of the field is greater than 7 characters',
            'value': 'value should be greater than 5',
            'data': 'This field is required'
        }

Querying

async def go(db):
    # find returns a cursor
    cursor = User.q(db).find({}, {'_id': 1}).skip(1).limit(2)
    async for user in cursor:
        print(user.name)
        assert user.is_active is None  # we used projection

    # find one
    user = await User.q(db).find_one({User.name.s: 'Alexandro'})
    assert user.name == 'Alexandro'

    # update
    await User.q(db).update_many(
        {User.is_active.s: True},
        {'$set': {User.is_active.s: False}})

    # delete
    await User.q(db).delete_many({})

Models Inheritance

A hierarchy of models can be built by inheriting one model from another. A aiomongodel.Document class should be somewhere in hierarchy for model adn aiomongodel.EmbeddedDocument for subdocuments. Note that fields are inherited but meta options are not.

class Mixin:
    value = IntField()

class Parent(Document):
    name = StrField()

class Child(Mixin, Parent):
    # also has value and name fields
    rate = FloatField()

class OtherChild(Child):
    # also has rate and name fields
    value = FloatField() # overwrite value field from Mixin

class SubDoc(Mixin, EmbeddedDocument):
    # has value field
    pass

Models Inheritance With Same Collection

class Mixin:
    is_active = BoolField(default=True)

class User(Mixin, Document):
    _id = StrField()
    role = StrField()
    name = SynonymField(_id)

    class Meta:
        collection = 'users'

    @classmethod
    def from_mongo(cls, data):
        # create appropriate model when loading from db
        if data['role'] == 'customer':
            return super(User, Customer).from_mongo(data)
        if data['role'] == 'admin':
            return super(User, Admin).from_mongo(data)

class Customer(User):
    role = StrField(default='customer', choices=['customer'])  # overwrite role field
    address = StrField()

    class Meta:
        collection = 'users'
        default_query = {User.role.s: 'customer'}

class Admin(User):
    role = StrField(default='admin', choices=['admin'])  # overwrite role field
    rights = ListField(StrField(), default=lambda: list())

    class Meta:
        collection = 'users'
        default_query = {User.role.s: 'admin'}

Transaction

from motor.motor_asyncio import AsyncIOMotorClient

async def go(db):
    # create collection before using transaction
    await User.create_collection(db)

    async with await db.client.start_session() as session:
        try:
            async with s.start_transaction():
                # all statements that use session inside this block
                # will be executed in one transaction

                # pass session to QuerySet
                await User.q(db, session=session).create(name='user')  # note session param
                # pass session to QuerySet method
                await User.q(db).update_one(
                    {User.name.s: 'user'},
                    {'$set': {User.is_active.s: False}},
                    session=session)  # note session usage
                assert await User.q(db, session).count_documents({User.name.s: 'user'}) == 1

                # session could be used in document crud methods
                u = await User(name='user2').save(db, session=session)
                await u.delete(db, session=session)

                raise Exception()  # simulate error in transaction block
         except Exception:
             # transaction was not committed
             assert await User.q(db).count_documents({User.name.s: 'user'}) == 0


loop = asyncio.get_event_loop()
client = AsyncIOMotorClient(io_loop=loop)
db = client.aiomongodel_test
loop.run_until_complete(go(db))

License

The library is licensed under MIT License.

Changelog

0.2.2 (2020-12-14)

Bump version of motor for python 3.11 compatibility.

Add tests workflow for GitHub Actions CI.

0.2.1 (2020-12-14)

Add verbose_name to Field for meta information.

Fix DecimalField’s issue to load field from float value.

0.2.0 (2018-09-12)

Move requirements to motor>=2.0.

Remove count method from MotorQuerySetCursor.

Add session support to MotorQuerySet and Document.

Add create_collection method to Document.

Fix __aiter__ of MotorQuerySetCursor for python 3.7.

Deprecate count method of MotorQuerySet.

Deprecate create method of Document.

0.1.0 (2017-05-19)

The first aiomongodel release.

Metadata

Release files for aiomongodel 0.2.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for aiomongodel 0.2.2
File Size Uploaded
aiomongodel-0.2.2.tar.gz 31.5 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for aiomongodel 0.2.2
File Interpreter ABI Platform
aiomongodel-0.2.2-py3-none-any.whl Python 3 none any Details

Total release size: 51.0 kB

Release files / aiomongodel-0.2.2.tar.gz

Download URL aiomongodel-0.2.2.tar.gz
Size 31.5 kB
Tags Source
SHA-256 checksum
How to use checksums
7166e6c722836a39fe98e9cf1fdb9ecb885c7679e46a8d70863b87a0c5331c68
BLAKE2b-256 checksum
How to use checksums
4e025179e1ea657db43d153c32cadf224c3383771fbd6729b3e6c03712d038b7
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/4.0.2 CPython/3.9.16

Release files / aiomongodel-0.2.2-py3-none-any.whl

Download URL aiomongodel-0.2.2-py3-none-any.whl
Size 19.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
91b3c300ca89aacd42c9f6df791f7480b04cb91d953e65adbc04c7416ec449b8
BLAKE2b-256 checksum
How to use checksums
3246a51326e69b50e185cae18d1f3620b0db1de835e2dbb1931738828da732f4
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/4.0.2 CPython/3.9.16

Release history Release notifications | RSS feed

This release

0.2.2 This release

2 release files

0.2.1

2 release files

0.2.0

2 release files

0.1.0

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page