Puff Serverless

Project Goals & Use Case

The intent of Puff Serverless is to allow the deployment of hyper-scalable solution to centralize sign-on, access control, and unified billing for multiple organisations across multiple applications.

Technologies

We run JavaScript on Node-compatible Cloudflare Functions. APIs return full HTML via HTMX to update in-page. Data is stored within Postgres-like CoackroachDB accessed via Cloudflare Hyperdrive.

More can be read in Architecture.md

First time project setup

Node & NPM

Use the latest Node version. We recommend you Install NVM if you are not already using Node.

nvm install stable
nvm use stable

Postgres or CockroachDB

Note: SQL Schema can be found in the SQL folder, one file per table. users.sql should be imported first as it provides the foreign key for many other tables.

for Local Development

You can override the Hyperdrive connection strings by setting the following in .env:

WRANGLER_HYPERDRIVE_LOCAL_CONNECTION_STRING_HYPERDRIVE="postgres://user:password@localhost:5432/databasename"

for Production Deployment

TODO

Continuous Development

npm ci
npm run dev

Before pushing you may wish to run linting or allow autoformatting to run, or configure your editor to automatically use prettier.

npm run lint
npm run format

You may also need to update types:

npx wrangler types

Project Maintenance

Dependabot should update NPM and GitHub Actions with automatic pull requests.