Home

Awesome

Dataface

Build and manage data in a Postgres database with a spreadsheet-like interface. Demo.

screencast of a user editing a spreadsheet-like interface

Deploy to Heroku

Motivation

Ideally all data would be managed in a purpose-built application backed by a database, designed by a database expert, but IT departments have to prioritize what applications they build or buy. As a result, a lot of data ends up being managed in a spreadsheet or a Microsoft Access database. These tools are flexible and easy for non-IT staff to build, but IT departments often see them as sources of technical debt: they only support one user at a time, they’re single points of failure since they’re usually not backed up, and they’re difficult to integrate into other systems.

Dataface aims to be an alternative tool that IT departments can offer non-IT staff to empower them to easily manage their data in a system the IT department would support. For the non-IT staff, dataface is a spreadsheet-like interface for data that lets you create columns and rows, even linking columns between sheets. For the IT department, dataface is a vanilla PostgreSQL database with a REST API and web app on top. This way, the non-IT staff can get started building a database on their own while keeping it standard and portable under the hood, for when the time comes for it to graduate to a full-fledged application.

Roadmap

See the milestones for the major features roadmap.

Development

The docker-compose.yml file provides a postgres container, and an application container. To spin them up, install docker and run:

docker-compose up

Then navigate to localhost:9966 in the browser.

Testing

To test the client, run:

yarn test:client

To test the server, you'll need a throwaway postgres database running (the tests will wipe it clean afterwards). To run one using docker, use:

docker run -p 5434:5432 postgres

Once a postgres database is available, run the server tests while passing the DB_URL environment variable:

DB_URL="postgres://postgres:pwd@localhost:5434/postgres" yarn test:server

You can run both the client and server tests together with yarn test; just don't forget the DB_URL environment variable:

DB_URL="postgres://postgres:pwd@localhost:5434/postgres" yarn test