/classified-ads-crud

A simple classified ads CRUD app built in Flask, PostgreSQL, SQLAlchemy, etc.

Primary LanguagePython

Classfied Ads CRUD API

Test playground with docker-compose, Flask and some SQLAlchemy.

Project structure copied from: https://flask-restful.readthedocs.io/en/latest/intermediate-usage.html#project-structure

Shipped with a docker-compose that can setup a PostgresSQL server with volume on ./posgres-data for persistance.

Setup:

$ vi .env.dev  # Modify the values correctly 
$ cp .env.dev .env  
$ docker-compose build
$ docker-compose up -d

or to run locally,

pyenv install 3.10.0 
# Make a virtual env out of it.
# TODO: Add docs to setup a virtual env.  
(venv-py-3.10)$ pip install -r requirements.txt 
(venv-py-3.10)$ # Optionally change .env.dev 
(venv-py-3.10)$ gunicorn -c gunicorn.config.py wsgi:app

The local version would talk to a temporary sqlite. You can change the settings by setting the following OS environment variables:

export SQLALCHEMY_DATABASE_URI=sqlite:////tmp/test.db

or, likewise for postgresql. The app uses pythonv3.10 Tech used:

  • Python-Flask for web services
  • Flask-RESTful for REST services (resources)
  • Flask-SQLAlchemy for talking to DBMS
  • marshmallow for serialization/de-serialization

APIs:

@TODO: Document this later in swagger or something.

  • A way to insert a new ad. An ad consists of a subject, a body, an optional price, and an email address.
curl --request POST \
  --url http://localhost:5000/api/v1/ads/ \
  --header 'Content-Type: application/json' \
  --data '{
	"subject": "Ad Name", 
	"body": "This is a very short body.", 
	"price": 50000, 
	"currency": "SEK", 
	"email": "teggggter@gmail.com"
}'
  • A way of fetching existing ads. It should be possible to sort the ads on the time they were inserted and by their price.
curl --request GET \
  --url 'http://localhost:5000/api/v1/ads/?sort=creation&orderBy=asc'

or

curl --request GET \
  --url 'http://localhost:5000/api/v1/ads/?sort=price&orderBy=asc'
  • A way of deleting a previously inserted ad.
curl --request DELETE \
  --url http://localhost:5000/api/v1/ads/1/

TODOS:

  • Tests need models imoprted inline, because of db_session being global in database.py. This has to be fixed properly.
  • Pagination on the listing API
  • Better handling of the Sessions (making it easier to test)
  • pytest do not want to run locally on shell, however runs on pycharm.
  • Handle database connect timeout correctly so that the app can wait until postgresql service is up.
  • Figure out if changes in the codebase is reflected correctly in the docker environment. We have a volume mount setup, but I could not test it.
  • Handle migrations. alembic is installed, but not setup.

Known issues:

  • Postgresql can error out on the first boot, due to some hba.conf missing reason. This is fixed by simply restarting docker-compose using
$ docker-compose restart
  • Web service restarts a bit until potsgresql service is up .
  • Problems running pytest locally via CLI.