Skip to main content

Introduction

Build Coverage Release Codebase Support

Frictionless Repository is a Github Action created for continuous data validation. It is a minimalistic integration between Github Actions and Frictionless Data. In this introduction we will show how it works and describe the projects it relies on. Let's get started and here is a example of a validation report provided by Frictionless Repository:

Example

How It Works#

On every commit to your repository there will be run a validation process to find tabular errors and other problems in your data. We created a demo video to introduce a basic workflow of continious data validation using Frictionless Repository:

Github Actions#

Github Actions is a continuous integration service. If you're not familiar with Github Actions we really recommend you to watch a short talk given by Grant R. Vousden-Dishington on csv,conf,v6:

Frictionless Data#

Frictionless Data is a comprehensive data software and standards project covering many aspects working with data. Frictionless Repository uses a Python framework to validate data and a report component to show the validation results:

Frictionless Repository can be described by this simple flow:

  • you add a Frictionless Repository step to their workflow on Github
  • Frictionless Framework validates your data and saves the result as a workflow's artifact
  • Frictionless Components fetch and render this validation report

Frictionless Repository is completely server-less so it doesn't rely on any third-party hardware except for Github infrastructure. There is no vendor-lock or something like this a you can fork this project and run it on Github differently if needed.

Last updated on by roll