The use of the Python language for the purpose of supporting the KRIM teaching load planning.
Project developed with Python v3.8.6 (changed from Python v3.6.7)
Heroku deploy: http://shielded-ocean-48265.herokuapp.com (PostgreSQL)
http://gret.ct8.pl/ with MySQL DB
Local with SQLite DB
# create folder for project
mkdir root_folder
cd root_folder
# get copy of the application's files
git clone https://github.com/gretkierewicz/dissertation.git .
# build app with docker-compose
docker-compose build
# login with heroku and create app for deployment
heroku login
heroku create
# login to container
heroku container:login
# create postgresql DB (this creates DATABASE_URL env variable as well)
heroku addons:create heroku-postgresql:hobby-dev
# push and release container
heroku container:push web
heroku container:release web
# create DB tables for cars app
heroku run python manage.py migrate
# start app in web-browser
heroku open
- Modules / FIX for PUT method and nested classes
Minor
- Implement custom filters, sorting or search
- Change upload CSV files method for employees - it should first create employees without supervisors, and then save supervisors to it (so employees added from list could be set as supervisors for previously added employees)
- Errors output for nested JSON data import
- Tests