github-discord-webhooks

A serverless function running on Cloudflare Workers and D1 that delivers enhanced GitHub Workflow webhooks for Discord.

Features

• More information about the running Workflow, author and commit information

• Messages are edited as workflow events come through, reducing spam:

  

Installation

If you want to run and/or deploy this project for yourself, you will need the following:

• wrangler CLI
• Node.js LTS

1. Create a Discord Webhook and Secret

Create a Discord Webhook in the channel of your choice following the instructions here.

Copy .dev.vars.example to .dev.vars and update the DISCORD_WEBHOOK variable with your webhook URL. This value will be used for local development.

Create a secret for use in the deployed version:

$ wrangler secret put DISCORD_WEBHOOK
 ⛅️ wrangler 2.6.2
-------------------
Enter a secret value: *************************************************************************************************************************
🌀 Creating the secret for the Worker "github-webhooks-worker"

2. Create a D1 Database

NOTE: D1 is currently in Alpha, the following is subject to change

This project uses a D1 Database to track the state of the Discord message associated with a GitHub workflow run. To generate a database for the project, use the following:

$ wrangler d1 create webhook_state
...
✅ Successfully created DB 'webhook_state'!

Add the following to your wrangler.toml to connect to it from a Worker:

[[ d1_databases ]]
binding = "DB" # i.e. available in your Worker on env.DB
database_name = "webhook_state"
database_id = "48298b61-95b1-4705-b948-ebc87c434246"

Then, copy the wrangler.example.toml file into wrangler.toml, and paste in the suggested snippet to create your database binding.

Next, populate your local database with the schema:

$ wrangler d1 execute DB --local --file=./schema.sql
🌀 Mapping SQL input into an array of statements
🌀 Loading DB at .wrangler/state/d1/DB.sqlite3

And finally, populate the deployed database with the schema (by dropping the --local flag):

$ wrangler d1 execute DB --file=./schema.sql
🌀 Mapping SQL input into an array of statements
🌀 Parsing 2 statements
🌀 Executing on DB (48298b61-95b1-4705-b948-ebc87c434246):
🚣 Executed 2 commands in 41.074951000511646ms

More documentation on getting started with D1 can be found here.

4. Running locally

To run the project locally, use the start npm script:

$ npm run start

> github-webhooks-worker@0.0.0 start
> wrangler dev --persist --local
...
[mf:inf] Worker reloaded! (9.09KiB)
[mf:inf] Listening on 0.0.0.0:8787

5. Deploy to Cloudflare

To deploy to Cloudflare, use the deploy npm script:

$ npm run deploy

> github-webhooks-worker@0.0.0 deploy
> wrangler publish
...
Uploaded github-webhooks-worker (2.00 sec)
Published github-webhooks-worker (1.56 sec)