https://actionbroker.herokuapp.com/

There is a makefile to help with deploying the softare to Heroku.

The tests are weak, but they run using pytest.

Postgres is a dependency and uses environment variables to connect to a development Docker instance or the production Heroku database.

The day of the sale, the production database is upgraded to a paid for version. Immediately after the sale the DB and web worker are downgraded.

Data is cleared out between years, except for patrons. See the makefile for loading data from TSV files.

docker compose up to get all the environments up and running -- Actually, using a local docker instance of PG and then running the dev server from PyCharm is probably better. These are already configured in the PyCharm IDE.

heroku pg:backups:capture and heroku pg:backups:download <b00?> to get the production data locally

pg_restore -d ab -h 127.0.0.1 -p 5432 -U ab -v --no-owner latest.dump to restore

manage.py dumpdata auction.patron > patron_list.json and manage.py loaddata patron_list.json and manage.py reset_all_buyer_nubmers to dump the list of patrons ( addresses, etc) and then reload it and remove all the buyer numbers -- see also make local_new

To push the local database to the remote heroku database: heroku pg:reset postgresql-triangular-11982 heroku pg:push postgres://ab@127.0.0.1:5432/ab postgresql-triangular-11982