This library serves as a wrapper around Google's Firestore. It implements validation, type checking facilities, and relational (work in progress) model logic.
When you want a simple abstraction layer on top of Firestore to help you keep track of entities and their relationships, the way you would achieve it with an ORM. The library's API borrows several design paradigms from Django's ORM, because I like Django's ORM.
pip install fsmodels
You will need to set up your environment in the usual way to have access to your resources on Google Cloud Platform.
import os
os.environ['GOOGLE_APPLICATION_CREDENTIALS'] = '/path/to/credentials.json'
If you do not have this variable set, model instances will log a warning, and nothing will actually work with Firestore.
from fsmodels.models import Model
from fsmodels.fields import Field
def is_valid_string(value):
return value is None or isinstance(value, str), f'value must be str, cannot be {value}'
class User(Model):
username = Field(required=True)
password = Field(required=True)
first_name = Field(required=True, default='Billy')
last_name = Field(validation=is_valid_string)
created_date = Field(default=time.time)
my_user = User(username='user1', password='goodpassword')
my_user.save() # automatically generates a new id and creates this user in firestore.
my_user.retrieve() # returns a dict of the record in firestore
my_user.delete() # deletes the record in firestore with my_user.id
my_user2 = User(username='user2') # validation functions are not called on __init__ by defalt
my_user2.save() # raises a validation error; password is a required field.
user = User(id='my_id')
# retrieve ignores everything about the local model instance other than the id
user.retrieve(overwrite_local=True)
print(user.to_dict()) # see what was in firestore
user.first_name = 'a different name'
user.save()
user = User(id='my_id')
user.delete()
import uuid
from fsmodels.models import Model
from fsmodels.fields import Field, ModelField, IDField
def uuid_as_str():
return str(uuid.uuid4())
class Profile(Model):
id = IDField(required=True, default=uuid_as_str)
first_name = Field(required=True)
last_name = Field(required=True)
class User(Model):
id = IDField(required=True, default=uuid_as_str)
username = Field(required=True)
password = Field(required=True)
profile = ModelField(Profile, required=True)
profile = Profile(first_name='Billy', last_name='Mayes')
# will save both the profile and the user
user = User(username='bmayes', password='plaintextpassword', profile=profile)
print(user.retrieve())
#{'profile': {'id': 'bd3ca41a-b6c4-4249-ac48-eb05db79bb3d',
# 'first_name': 'Billy',
# 'last_name': 'Mayes'},
# 'password': 'plaintextpassword',
# 'username': 'bmayes',
# 'profile_id': 'bd3ca41a-b6c4-4249-ac48-eb05db79bb3d',
# 'id': '1e586d79-f2c0-4618-a7f7-95308a54298e'}
print(profile.retrieve())
#{'user_id': '1e586d79-f2c0-4618-a7f7-95308a54298e',
# 'first_name': 'Billy',
# 'last_name': 'Mayes',
Planned.
# model_name in error messages and collection name in firestore is my_model1
class MyModel1(Model):
first_name = Field(required=True)
last_name = Field(required=True)
class Meta:
model_name = 'Profiles'
# model_name in error messages and collection name in firestore is MY-MODEL-2
class MyModel2(Model):
first_name = Field(required=True)
last_name = Field(required=True)
class Meta:
model_name = 'MY-MODEL-2'
# model_name in error messages is my_model3 and collection name in firestore is mY-mODeL-3
class MyModel3(Model):
first_name = Field(required=True)
last_name = Field(required=True)
class Meta:
collection = 'mY-mODeL-3'
class MyModel1(Model):
first_name = Field(required=True)
last_name = Field(required=True)
m = MyModel1()
# same as google.cloud.firestore
m.firestore
# same as google.cloud.firestore.Client()
m.db
# same as google.cloud.firestore.Client().collection(m._collection) where m._collection in this case is my_model1
m.collection
# you can do stuff like
for record in m.collection.get():
print(record.to_dict())