Products API

Centralized API for accessing release and support lifecycle information for all Canonical products.

Why

Canonical products have different release cadences and support models. This API provides a single source of truth for release dates and support lifecycles.

Local development

The simplest way to run the API locally is using dotrun:

docker-compose up -d
dotrun

Once the server has started, you can visit http://0.0.0.0:8040/ in your browser.

After you close the server with <ctrl>+c you should run docker-compose down to stop the database.

Prerequisites

• Docker
• Python 3.10+
• dotrun (see https://github.com/canonical/dotrun)

Install dependencies

dotrun install

If you need to run migrations locally, use a host-local virtualenv:

python3 -m venv .venv
source .venv/bin/activate
pip install --upgrade pip setuptools==69.5.1
pip install -r requirements.txt

Run the API

docker-compose up -d
dotrun serve

The API will start at http://0.0.0.0:8040/ with a health endpoint at /v1/health.

On Linux, DATABASE_URL in .env uses localhost.

On macOS and Windows, Docker networking can behave differently. If the API cannot connect to Postgres, create a local override file .env.local and set:

DATABASE_URL=postgresql://postgres:pw@host.docker.internal:5433/products_db

Database migrations (local)

Run migrations from the host virtualenv (not through dotrun exec):

source .venv/bin/activate
set -a && source .env && set +a
export FLASK_APP=webapp.app
docker-compose up -d postgres
flask db upgrade

Useful checks:

flask db current
flask db heads

Note: avoid dotrun exec flask db ... for local migrations, as that runtime may use a different network/path context than the host .env values.

Testing

dotrun test

On Linux, TEST_DATABASE_URL in .env uses localhost.

On macOS and Windows, Docker networking can behave differently. If tests cannot connect to Postgres, create a local override file .env.local and set:

TEST_DATABASE_URL=postgresql://postgres:pw@host.docker.internal:5433/test_products_db

Linting and formatting

dotrun lint-python  # Run linters
dotrun format-python  # Auto-format Python code

Project layout

• app.py – WSGI entrypoint for production servers
• webapp/app.py – Flask application using FlaskBase
• requirements.txt – Python dependencies
• package.json – Node.js scripts and development tasks