Document object mapper for pydantic and pymongo
pip install pydantic-mongo
from bson import ObjectId
from pydantic import BaseModel
from pydantic_mongo import AbstractRepository, ObjectIdField
from pymongo import MongoClient
from typing import List
class Foo(BaseModel):
count: int
size: float = None
class Bar(BaseModel):
apple: str = 'x'
banana: str = 'y'
class Spam(BaseModel):
id: ObjectIdField = None
foo: Foo
bars: List[Bar]
class SpamRepository(AbstractRepository[Spam]):
class Meta:
collection_name = 'spams'
client = MongoClient("mongodb://localhost:27017")
database = client["example"]
spam = Spam(foo=Foo(count=1, size=1.0),bars=[Bar()])
spam_with_predefined_id = Spam(
id=ObjectId("611827f2878b88b49ebb69fc"),
foo=Foo(count=2, size=2.0),
bars=[Bar()]
)
spam_repository = SpamRepository(database=database)
# Insert / Update
spam_repository.save(spam)
# Insert / Update many items
spam_repository.save_many([spam, spam_with_predefined_id])
# Delete
spam_repository.delete(spam)
# Find One By Id
result = spam_repository.find_one_by_id(spam.id)
# Find One By Id using string if the id attribute is a ObjectIdField
result = spam_repository.find_one_by_id(ObjectId('611827f2878b88b49ebb69fc'))
# Find One By Query
result = spam_repository.find_one_by({'foo.count': 1})
# Find By Query
results = spam_repository.find_by({'foo.count': {'$gte': 1}})
# Paginate using cursor based pagination
edges = spam_repository.paginate({'foo.count': {'$gte': 1}}, limit=1)
more_edges = spam_repository.paginate({'foo.count': {'$gte': 1}}, limit=1, after=list(edges)[-1].cursor)
# Simple pagination
edges = spam_repository.paginate_simple({'foo.count': {'$gte': 1}}, limit=25, page=1)