lasthourcloud/CONTRIBUTION.md

Contribution

First, thanks for considering to contribute to my project. It really means a lot! :)

You can ask for guidance anytime on our Discord server in the #contribution channel.

Setup your development environment

Container based development flow (recommended and the easiest)

All you need is to intall Docker Engine 20.11+ on your local machine and run pnpm dev:container. It will build the base image for Coolify and start the development server inside Docker. All required ports (3000, 3001) will be exposed to your host.

Github codespaces

If you have github codespaces enabled then you can just create a codespace and run pnpm dev to run your the dev environment. All the required dependencies and packages has been configured for you already.

Gitpod

1. Use container based development flow
2. Or setup your workspace manually:

Create a workspace from this repository, run pnpm install && pnpm db:push && pnpm db:seed and then pnpm dev. All the required dependencies and packages has been configured for you already.

Some packages, just pack are not installed in this way. You cannot test all the features. Please use the container based development flow.

Local Machine

At the moment, Coolify doesn't support Windows. You must use Linux or MacOS or consider using Gitpod or Github Codespaces.

Install all the prerequisites manually to your host system. If you would not like to install anything, I suggest to use the container based development flow.

• Due to the lock file, this repository is best with pnpm. I recommend you try and use pnpm because it is cool and efficient!
• You need to have Docker Engine installed locally.
• You need to have Docker Compose Plugin installed locally.
• You need to have GIT LFS Support installed locally.

Optional:

• To test Heroku buildpacks, you need pack binary installed locally.

Inside a Docker container

WIP

Setup Coolify

• Copy apps/api/.env.template to apps/api/.env.template and set the COOLIFY_APP_ID environment variable to something cool.

• pnpm install to install dependencies.

• pnpm db:push to o create a local SQlite database.

  This will apply all migrations at db/dev.db.

• pnpm db:seed seed the database.

• pnpm dev start coding.

Technical skills required

• Languages: Node.js / Javascript / Typescript
• Framework JS/TS: SvelteKit & Fastify
• Database ORM: Prisma.io
• Docker Engine API

Add a new service

Which service is eligable to add to Coolify?

The following statements needs to be true:

• Self-hostable
• Open-source
• Maintained (I do not want to add software full of bugs)

Create Prisma / Database schema for the new service.

All data that needs to be persist for a service should be saved to the database in cleartext or encrypted.

very password/api key/passphrase needs to be encrypted. If you are not sure, whether it should be encrypted or not, just encrypt it.

Update Prisma schema in src/apps/api/prisma/schema.prisma.

• Add new model with the new service name.
• Make a relationship with Service model.
• In the Service model, the name of the new field should be with low-capital.
• If the service needs a database, define a publicPort field to be able to make it's database public, example field name in case of PostgreSQL: postgresqlPublicPort. It should be a optional field.

Once done, create Prisma schema with pnpm db:push.

You may also need to restart Typescript Language Server in your IDE to get the new types.

Add available versions

Versions are hardcoded into Coolify at the moment and based on Docker image tags.

• Update supportedServiceTypesAndVersions function here

Include the new service in queries

At here in includeServices function add the new table name, so it will be included in all places in the database queries where it is required.

Define auto-generated fields

At here in configureServiceType function add the initial auto-generated details such as password, users etc, and the encryption process of secrets (if applicable).

Define input field details

At here add details about the input fields shown in the UI, so every component (API/UI) will know what to do with the values (decrypt/show it by default/readonly/etc).

Define the start process

• At here, define how the service should start. It could be complex and based on docker-compose definitions.

See startUmamiService() function as example.

• At here, add the new start service process to startService function.

Define the deletion process

Here in removeService add the database deletion process.

Custom logo

• At here add the service custom log as a Svelte component and export it here.

SVG is recommended, but you can use PNG as well. It should have the isAbsolute variable with the suitable CSS classes, primarily for sizing and positioning.

• At here add links to the documentation of the service.

Custom fields on the UI

By default the URL and name are shown on the UI. Everything else needs to be added here

If you need to show more details on the frontend, such as users/passwords, you need to add Svelte component here. For example, see other here.

Good job! 👏