Multi-project Python

Local development

1. Install PostgreSQL, Python and Poetry

   brew install postgresql@15 pyenv
   pyenv install 3.10
   # Add pyenv to the path
   pyenv shell 3.10
   curl -sSL https://install.python-poetry.org | python3 -

2. Install Python dependencies

   make install

3. Set up database

   make db/reset migrate

4. Verify types and run tests

   make test

5. Check PEP-8 compliance and code format

   make check

6. Configure environment variables

   cp .env.example .env
   vi .env

7. Run app

   make run

PyCharm setup

1. In Settings → Project: multiproject-python → Python Interpreter set the Python interpreter for the top-level project to be the existing Python interpreter created via the command line.
2. Attach each Python subproject to the IDE project with File → Open... → Attach project.... Note: On Open..., navigate to subproject, click Open and when prompted, click Attach. Repeat for each subproject.
3. In Settings → Project: multiproject-python → Python Interpreter set the Python interpreter for each Python subproject to be the existing Python interpreter created via the command line.
4. Manually set up dependencies between Python subprojects in Settings → Project: multiproject-python → Project Dependencies. This is clunky, and PyCharm should be able to infer this information from the subprojects' pyproject.toml files. Watch the open issue in YouTrack for developments.
5. In Settings → Tools → Black configure PyCharm to use the Black formatter on code reformat and on save.

Run locally with Docker

cd applications/starter_app
poetry export --without-hashes -f requirements.txt > ../../requirements.txt
cd ../..
sed -i.backup "s?$(pwd)?/workspace?" requirements.txt
pack build multiproject-python --builder=gcr.io/buildpacks/builder:v1
docker run -p 8080:8080 multiproject-python