sqlalchemy-api-handler
SQLAlchemy-Api-Handler is an extension that adds support for handling apis with sqlalchemy. It helps to handle models with :
- humanized ids once it is jsonified,
- throwing api errors for some casting of value during the save time,
- dictification of the model objects into jsonified ones.
- It also gives an activate method to help you better handle offline operational transforms, based on the PostgreSQL-Audit Activity model.
Installing
Install and update using pip:
$ pip install -U SQLAlchemy-Api-HandlerA Simple Example
Suppose a request POST /users {'email': 'marx.foo@plop.fr', name: 'Marx Foo'} :
from flask import Flask, jsonify, request
from sqlalchemy_api_handler import ApiHandler
app = Flask(__name__)
app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite:///example.sqlite'
db = SQLAlchemy(app)
ApiHandler.set_db(db)
class User(ApiHandler, db.Model):
email = db.Column(db.String, unique=True, nullable=False)
name = db.Column(db.String, unique=True, nullable=False)
@app.route('/users', methods=['POST'])
def post_user():
user = User(**request.form)
ApiHandler.save(user)
return jsonify(as_dict(user))The success result will have stored a user object at, let's say id = 32, and so will fetch an object at humanized id = humanize(32), ie
{'id': 'EA', 'email': 'marx.foo@plop.fr', name: 'Marx Foo'}
Playing with nesting data
Suppose a request GET /offers
from flask import Flask, jsonify, request
from sqlalchemy.orm import relationship
from sqlalchemy_api_handler import ApiHandler
app = Flask(__name__)
app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite:///example.sqlite'
db = SQLAlchemy(app)
ApiHandler.set_db(db)
class Venue(ApiHandler, db.Model):
address = db.Column(db.String, unique=True, nullable=False)
name = db.Column(db.String, unique=True, nullable=False)
class Offer(ApiHandler, db.Model):
name = db.Column(db.String, unique=True, nullable=False)
venueId = db.Column(db.BigInteger,
db.ForeignKey('venue.id'),
nullable=False,
index=True)
venue = relationship('Venue',
foreign_keys=[venueId],
backref='offers')
class Stock(ApiHandler, db.Model):
available = db.Column(db.Integer, nullable=False)
offerId = db.Column(db.BigInteger,
db.ForeignKey('offer.id'),
index=True,
nullable=False)
offer = relationship('Offer',
foreign_keys=[offerId],
backref='stocks')
venue = Venue(address='Somewhere I belong', name='MyVenue')
offer = Offer(name='MyOffer')
stock = Stock(available=10)
stock.offer = offer
offer.venue = venue
ApiHandler.save(stock)
offer_includes = [
'stocks',
{
'key': 'venue',
'includes': [
'-address'
]
}
]
@app.route('/offers', methods=['GET'])
def get_offers():
offers = Offer.query.all()
return jsonify(as_dict(offers, includes=offer_includes))The success will return
[
{
'id': 'AE',
'name': 'MyOffer',
'stocks': [
{
'available': 10,
'id': 'AE'
}
],
'venue': {
'name': 'MyVenue'
}
}
]
Activity
If you need to manage operation transforms of your entities (typically with a collaborative app working offline) :
from flask import Flask, jsonify, request
from sqlalchemy.orm import relationship
from sqlalchemy_api_handler import ApiHandler
from sqlalchemy_api_handler.mixins import ActivityMixin, \
HasActivitiesMixin
app = Flask(__name__)
app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite:///example.sqlite'
db = SQLAlchemy(app)
ApiHandler.set_db(db)
versioning_manager.init(db.Model)
class Activity(ActivityMixin,
ApiHandler,
versioning_manager.activity_cls):
__table_args__ = {'extend_existing': True}
id = versioning_manager.activity_cls.id
ApiHandler.set_activity(Activity)
class Venue(ApiHandler,
db.Model,
HasActivitiesMixin):
address = db.Column(db.String, unique=True, nullable=False)
name = db.Column(db.String, unique=True, nullable=False)
@app.route('/__activities__', methods=['POST'])
def create_activities():
activities = [Activity(**a) for a in request.json]
ApiHandler.activate(*activities)
return jsonify([as_dict(activity) for activity in activities]), 201The request POST /activities helps you to sync your backend with the last state of pushed activities. For example, if the data posted is :
[
{
"dateCreated": "2020-08-05T08:34:06.415000Z" ,
"entityIdentifier": "4039fb61-f085-43c4-a2ed-e5e97c5dcebc",
"modelName": "Venue",
"patch": { "address": "22 rue de la Loire", "name": "MyVenue" }
},
{
"dateCreated": "2020-08-06T09:34:06.415000Z" ,
"entityIdentifier": "4039fb61-f085-43c4-a2ed-e5e97c5dcebc",
"modelName": "Venue",
"patch": { "name": "MyVenueChanged" }
}
]
It will then end to create a venue instance:
venue = Venue.query.filter_by(name='MyVenueChanged').first()
print(as_dict(venue, includes=['__activities__'])){
'__activities__': [
{
'dateCreated': '2020-08-05T08:34:06.415000Z',
'entityIdentifier': '4039fb61-f085-43c4-a2ed-e5e97c5dcebc',
'id': 'BA',
'modelName': 'Venue',
'patch': {
'address': '22 rue de la Loire',
'name': 'MyVenue'
},
'verb': 'insert'
},
{
'dateCreated': '2020-08-06T09:34:06.415000Z',
'entityIdentifier': '4039fb61-f085-43c4-a2ed-e5e97c5dcebc',
'id': 'BF',
'modelName': 'Venue',
'patch': {
'name': 'MyVenueChanged'
},
'verb': 'update'
}
],
'activityIdentifier': '2020-08-05T08:34:06.415000Z',
'address': "22 rue de la Loire",
'id': 'AE',
'name': 'MyVenueChanged'
}
Celery
Sometimes it's hard to track celery tasks from the celery cli itself. And you may not want use flower when you have one exposed port by application, like in scalingo classic services. Therefore, ApiHandler can also synchronise the tasks to be stored in your postgres db, then you can build your Tasks Manager Dashboard front easy as you want, querying from classic postgres api routes. See how the example is build in the api folder.
Links
-
Issue tracker: https://github.com/betagouv/sqlalchemy-api-handler/issues
-
SQLAlchemy: https://www.sqlalchemy.org
Deploy
First, make sure that the deploy environment is started:
./sqlaah startIn a second tab, then:
-
Change the version into sqlalchemy_api_handler/init.py
-
Pre publish:
./sqlaah prepublish- Publish:
./sqlaah publish
