Skip to content

marktennyson/Flask-Tortoise

Repository files navigation

Flask-Tortoise

Flask-Tortoise is an extension for Flask that adds support for asynchronous Tortoise-ORM to your application with in-built migration system. It aims to simplify using Tortoise-ORM with Flask by providing useful defaults and extra helpers that make it easier to accomplish common tasks.

Tortoise-ORM is one of the best tool to interact with the database asynchronously. It's clean and Django type implementation provides you a better view and understanding of the database ORM system. Also you can use the Pydantic data module to write the error-less code.

The main aim of Tortoise-ORM is to provide the same service and api like the Django-ORM.

Installing

Install and update from PYPI:

pip install -U Flask-Tortoise

Install from source code:

git clone https://github.com/marktennyson/Flask-Tortoise
cd Flask-Tortoise && pip install .

Important links

Github Link: https://github.com/marktennyson/Flask-Tortoise
Official Documentation: https://marktennyson.github.io/Flask-Tortoise

Simple Examples for better understanding:

from flask import Flask, jsonify
from flask_tortoise import Tortoise
from random import choice


STATUSES = ["New", "Old", "Gone"]

app:"Flask" = Flask(__name__)
app.config['TORTOISE_ORM_DATABASE_URI'] = 'sqlite://db.sqlite3'

db:"Tortoise" = Tortoise(app)


class Users(db.Model):
    id = db.IntField(pk=True)
    status = db.CharField(20)

    def __str__(self):
        return f"User {self.id}: {self.status}"


class Workers(db.Model):
    id = db.IntField(pk=True)
    status = db.CharField(20)

    def __str__(self):
        return f"Worker {self.id}: {self.status}"

@app.get("/")
async def list_all():
    users = await Users.all()
    workers = await Workers.all()
    return jsonify(
        {"users": [str(user) for user in users], "workers": [str(worker) for worker in workers]}
    )


@app.get("/user")
async def add_user():
    user = await Users.create(status=choice(STATUSES))  # nosec
    return str(user)


@app.get("/worker")
async def add_worker():
    worker = await Workers.create(status=choice(STATUSES))  # nosec
    return str(worker)

@app.get("/get-worker")
async def get_worker():
    worker:"Workers" = await Workers.get(id=1)
    return str(worker.status)


if __name__ == '__main__':
    app.run(debug=True, port=8080)

If you save your models into a separate file than you have mention the file name on app config:

let's assume you have stores all of your models at models.py file.

models.py file
from flask_tortoise import Tortoise, Model, fields

db:"Tortoise" = Tortoise()


class Users(db.Model):
    id = db.IntField(pk=True)
    status = db.CharField(20)

    def __str__(self):
        return f"User {self.id}: {self.status}"


class Workers(db.Model):
    id = db.IntField(pk=True)
    status = db.CharField(20)

    def __str__(self):
        return f"Worker {self.id}: {self.status}"
app.py file:
from flask import Flask, jsonify
from models import *
from random import choice


STATUSES = ["New", "Old", "Gone"]

app:"Flask" = Flask(__name__)
app.config['TORTOISE_ORM_DATABASE_URI'] = 'sqlite://db.sqlite3'
app.config['TORTOISE_ORM_MODELS'] = "models" # if you have more than one models file then : ["models_1", "models_2", "models_3"]

db.init_app(app)

@app.get("/")
async def list_all():
    users = await Users.all()
    workers = await Workers.all()
    return jsonify(
        {"users": [str(user) for user in users], "workers": [str(worker) for worker in workers]}
    )


@app.get("/user")
async def add_user():
    user = await Users.create(status=choice(STATUSES))  # nosec
    return str(user)


@app.get("/worker")
async def add_worker():
    worker = await Workers.create(status=choice(STATUSES))  # nosec
    return str(worker)

@app.get("/get-worker")
async def get_worker():
    worker:"Workers" = await Workers.get(id=1)
    return str(worker.status)


if __name__ == '__main__':
    app.run(debug=True, port=8080)

Contributing:

We welcome all types of contributions. If you face any issue while using this module then please let us know by creating a github issue on the official github repo. If you have the solution for the issue raised by you or somebody else then please fork this repo and create a pull request with the main branch.

How to run this project on local machine:
  1. Fork this repo.
  2. Clone this repo on your local machine.
  3. create a virtual environment using python virtualenv module and activate it.
  4. now run python setup.py install.
  5. the above command will install the latest dev version of Flask-Tortoise on the virtual environment.

Contributor List:

License

MIT

Copyright (c) 2023 Aniket Sarkar(aniketsarkar@yahoo.com)