uptime-kuma/CONTRIBUTING.md

Project Info

First of all, thank you everyone who made pull requests for Uptime Kuma, I never thought GitHub Community can be that nice! And also because of this, I also never thought other people actually read my code and edit my code. It is not structed and commented so well, lol. Sorry about that.

The project was created with vite.js (vue3). Then I created a sub-directory called "server" for server part. Both frontend and backend share the same package.json.

The frontend code build into "dist" directory. The server uses "dist" as root. This is how production is working.

Can I create a pull request for Uptime Kuma?

Generally, if the pull request is working fine and it do not affect any existing logic, workflow and perfomance, I will merge to the master branch once it is tested.

If you are not sure, feel free to create an empty pull request draft first.

Pull Request Examples

✅ High - Medium Priority

• Add a new notification
• Add a chart
• Fix a bug

*️⃣ Requires one more reviewer

I do not have such knowledge to test it.

• Add k8s supports

*️⃣ Low Priority

It changed my current workflow and require further studies.

• Change my release approach

❌ Won't Merge

• Duplicated pull request
• Buggy
• Existing logic is completely modified or deleted
• A function that is completely out of scope

Project Styles

I personally do not like something need to learn so much and need to config so much before you can finally start the app.

For example, recently, because I am not a python expert, I spent a 2 hours to resolve all problems in order to install and use the Apprise cli. Apprise requires so many hidden requirements, I have to figure out myself how to solve the problems by Google search for my OS. That is painful. I do not want Uptime Kuma to be like this way, so:

• Easy to install for non-Docker users, no native build dependency is needed (at least for x86_64), no extra config, no extra effort to get it run
• Single container for Docker users, no very complex docker-composer file. Just map the volume and expose the port, then good to go
• All settings in frontend.
• Easy to use

Coding Styles

• Follow .editorconfig
• Follow eslint

Name convention

• Javascript/Typescript: camelCaseType
• SQLite: underscore_type
• CSS/SCSS: dash-type

Tools

• Node.js >= 14
• Git
• IDE that supports .editorconfig and eslint (I am using Intellji Idea)
• A SQLite tool (I am using SQLite Expert Personal)

Install dependencies

npm install --dev

For npm@7, you need --legacy-peer-deps

npm install --legacy-peer-deps --dev

Backend Dev

npm run start-server

# Or 

node server/server.js

It binds to 0.0.0.0:3001 by default.

Backend Details

It is mainly a socket.io app + express.js.

express.js is just used for serving the frontend built files (index.html, .js and .css etc.)

Frontend Dev

Start frontend dev server. Hot-reload enabled in this way. It binds to 0.0.0.0:3000.

npm run dev

PS: You can ignore those scss warnings, those warnings are from Bootstrap that I cannot fix.

You can use Vue Devtool Chrome extension for debugging.

After the frontend server started. It cannot connect to the websocket server even you have started the server. You need to tell the frontend that is a dev env by running this in DevTool console and refresh:

localStorage.dev = "dev";

So that the frontend will try to connect websocket server in 3001.

Alternately, you can specific NODE_ENV to "development".

Build the frontend

npm run build

Frontend Details

Uptime Kuma Frontend is a single page application (SPA). Most paths are handled by Vue Router.

The router in "src/main.js"

As you can see, most data in frontend is stored in root level, even though you changed the current router to any other pages.

The data and socket logic in "src/mixins/socket.js"

Database Migration

1. create patch{num}.sql in ./db/
2. update latestVersion in ./server/database.js

Unit Test

Yes, no unit test for now. I know it is very important, but at the same time my spare time is very limited. I want to implement my ideas first. I will go back to this in some points.