SQLAlchemy & Flask-SQLAlchemy

Functional example for a process to upload data into a relational database and serve it from a Flask app.

I didn't found documentation about how to user a SQLAlchemy model from a Flask API while keeping the advantages of using flask-sqlalchemy to query that model.

Here is what I found out after some reading and trying.

System requirements to run the example

• Docker compose (for the db)
• Python>=3.6

Setting up the environment

1. Create a virtual environment with python>=3.6

2. Start the database container by running

   ./commands/run.sh

3. Install requirements

   pip install -r requirements.txt

Common observations

• The example configuration is defined in the src.core.config.ini file. The configuration for both parts (loading the database and serving the data) is defined in one file because, although the application was designed to work as decoupled as possible, having different config files is, in my opinion, error prone.

Generating the dataset

The example comes with a small dataset in src.batch_loader.data.user_data.csv . In src.batch_loader.data.__init__.py you will finde a script to generate that dataset. Feel free to edit it, but be aware that altering the script could cause errors if the model is not altered accordingly.

To generate a dataset of the same structure but with different lenghts, run:

python runner.py mock-dataset 1000

being 1000 in this case the number of rows.

Loading the database

Notes

• Database data is never dropped on initialization. If the loading script is run more than once with the same data, it will exit with the corresponding exception.

Running the script

> python runner.py batch

Serving loaded data

Starting the web server

> python runner.py server

Endpoints

/users/

/users/:id

Status Codes

Status Code	Description
200	Existing users measurement retrieved successfully
404	Unable to found requested user

Contact Info

Issues are welcome :) , both for questions and improvement suggestions.