Skip to main content

Contributing Guide

So, you want to contribute to Discordeno? Fantastic!

I can appreciate the struggle and obstacles that are in the way to contributing to Discordeno when you are coming from another library. Discordeno is a very different kind of beast so it may be hard to go through it. So let's get started breaking it down. This guide will continue being updated with more Q&A as future contributors have questions.

Finding Something To Contribute To​

If you are in a position where you want to contribute but are not sure what you can contribute, look no further than the GitHub issues marked with the good first issue label. When an issue is marked with this label, it means that this is an easy issue to tackle and can be done by anyone.

To view all the issues with this label: https://github.com/discordeno/discordeno/issues?q=is%3Aissue+is%3Aopen+label%3Agood-first-issue

Setting Up Coding Environment To Contribute​

Alright then, by now you have something you want to contribute yourself or have found one of the issues you want to help close. Once your ready:

One Time Steps:

  • Install Node.js version 18+
  • Install Yarn (corepack enable)
  • Fork the repo.
  • Git clone the project on your local machine
  • Run yarn to install the dependencies

For Every Contribution:

  • Make a new branch
  • Make your code changes
  • Perform testing
  • Push to the repo and make a pull request

Testing​

For every contribution you make you can run the Unit and E2Etests to make sure all the code is still running correctly

  • yarn test:unit
  • yarn test:e2e*

* You need to setup some environment variables to run this script, see below

E2E environment requirement​

The scripts test:e2e requires some additional setup. This is because it will run a complete test for the entire library connecting directly to discord to verify that the code is working correctly and can correctly handle discord messages / api requirements

You can do this only once and don't need to do this every time:

  • Make a copy of the .env.example file and call it .env file.
  • Fill the variables in the file
    • Put a test bot's token in the DISCORD_TOKEN env.
    • For the Guild id, you should make a server and enable the Community setting and invite this test bot there.

Running test for Deno and Bun​

Since discordeno supports Node.js but also Deno and Bun you can run tests for them too:

  • Bun: yarn test:bun-unit
  • Deno: yarn test:deno-unit

Keep in mind that running those will require you to have Bun and/or Deno installed based on what you run.

Documentation / Website​

If you want to contribute to the documentation you need some additional setup

  • Run yarn to install the dependencies inside the root folder of the project
  • Run yarn release-build to build the entire project
  • Run yarn build:doc to generate the documentation from the comments in the code
  • Run yarn inside the website folder to install the dependencies

Done this you are ready to start editing the code, you will find the documentation under the website/docs folder with .md and .mdx. We use docusaurus to manage the site and it does treat all .md and .mdx file like .mdx ones, so you will need to make sure you don't use invalid syntax, you can check out the docusaurus documentation for more details.

In case you need to modify some React components you will find the code inside the website/src folder.

Understanding Repo setup​

Yarn​

As a package manager for the dependencies we use yarn on version 4 (berry). You can install it with the corepack enable command. Corepack is a feature that is built into node.js, however you might need to install it, in such cases you can run npm install -g corepack

Turborepo​

To manage the multiple packages we use a combination of yarn workspaces and turborepo. Turborepo is run by every script that you run from yarn in the root folder of the project, if you go inside a specific package, for example packages/bot it will run the local script and not turbo.

The workspaces are setup for all the folders inside packages. A few of the package are for the configuration of this repository, such as:

  • eslint-config-discordeno - Configures eslint (with the prettier plugin) for the entire turborepo project
  • tsconfig - Configures the base tsconfig.json for every package inside the turborepo project

Other node projects outside packages are not managed by turborepo and they require their own setup, an example is the website in website/ that requires you to install its dependencies manually outside from the main dependencies that are used with the packages.

lint-staged & Husky​

The project is setup to have Husky to install after you run yarn in the root folder of the project and will setup 2 hooks:

  • When you pull/merge: install the dependencies if they changed
  • Before commit: Running lint-staged

lint-staged is configured in the /package.json file and will run

  • Prettier for any changes you make to a source file, markdown file, html file or css file
  • Eslint with the "normal" configuration resolved based on eslint-config-discordeno for every file (outside website/)
  • Eslint with the configuration for the website for any source file you edit in the website/ directory

There will run automatically when git performs one of those actions (merge/commit), this might however slow down commit time as it requires some time to run list-staged

Yarn scripts​

There a bunch of scripts but we will list the most important ones

yarn build​

This will compile the TypeScript code into Javascript using SWC, it will not compile the type definitions (.d.ts) files.

note

This will clear the previus build file creating a clean build.

yarn build:type​

This will invoke the Typescript compiler (tsc) to compile the type definitions (.d.ts) files

yarn release-build​

This will run both the build script and build:type one.

yarn fmt​

This will format your code to match the rest of the codebase.

This will run every time you make a commit. If you disable this you should still run it manually before creating the pr

yarn lint​

This will do all the checking and linting to make sure your code is in a useable state. There is also one other key thing to note about this. When you change something in the @discordeno/types package, you can run yarn lint to rebuild all the types.

yarn test:unit​

This will run all the unit tests to make sure anything you changed, did not break any unexpected things.

yarn test:unit-coverage​

This will run the unit tests and give you the coverage of them both in the console and in a coverage folder

yarn test:deno-unit​

This will run all the unit tests like yarn test:unit but with Deno as the runtime instead of Node.js

note

This script requires Deno to be installed. See deno install page for details

yarn test:bun-unit​

This will run all the unit tests like yarn test:unit and yarn test:deno-unit but with Bun as the runtime instead of Node.js or Deno

note

This script requires Bun to be installed. See bun install page for details

yarn test:integration​

This will run integration tests. Integration tests do not connect to discord and use fake data to verify that everything is working correctly.

yarn test:test-type​

This will verify that the tests do not have type errors.

yarn test:e2e​

This will run all the end to end tests to make sure anything you changed, did not break any unexpected things.

This takes much longer as these tests actually run across the network using the bot token you provided in the .env file. It makes many requests to test the entire library. This should not be spammed as your bot token has rate limits. Use once your code changes are finalized to make sure that everything is working properly before pushing.

note

This requires you have setup the required environment variables, see above for more details

Notes​

  • If the contribution is relatively small go ahead and do it! If it is a larger change, I would highly recommend you read the guides we have on the website in order to understand our library better before undertaking such a big action.
  • If an issue exists for the contribution you want to make, leave a comment on it so no one else begins working on it as well. If an issue does NOT exist for the contribution, then create an issue first before working on it to get some feedback to decide whether it is worth doing.
  • Formatting and linting will be automatically handled by Husky and lint-staged. However, if it doesn't work you can format by doing yarn fmt and you can lint by doing yarn lint.
  • If you want to edit a docker app (rest proxy for example), the website or some examples you will need to run yarn to install the dependencies of those