hedgedoc/docs/content/dev/getting-started.md
Erik Michelson 305b6f5910 misc: add turbo monorepo util
Co-authored-by: Tilman Vatteroth <git@tilmanvatteroth.de>
Co-authored-by: David Mehren <git@herrmehren.de>
Co-authored-by: Erik Michelson <github@erik.michelson.eu>
Signed-off-by: Erik Michelson <github@erik.michelson.eu>
Signed-off-by: David Mehren <git@herrmehren.de>
Signed-off-by: Tilman Vatteroth <git@tilmanvatteroth.de>
2023-02-07 21:38:40 +01:00

7.4 KiB

Getting started

To run HedgeDoc 2.0 you need three components: the backend, the frontend and the reverse proxy.

Backend and Frontend are included in the HedgeDoc repo. The reverse proxy can be chosen by preference. For development, we recommend caddy and the provided configuration.

Quick guide for development setup

This describes the easiest way to start a local development environment. For other deployments follow the description below. To run HedgeDoc 2.0 you need three components: the backend, the frontend and the reverse proxy.

Backend and Frontend are included in the HegdeDoc repo. The reverse proxy can be chosen by preference. For development, we recommend caddy and the provided configuration.

  1. Clone our repository and go into its directory
    git clone https://github.com/hedgedoc/hedgedoc.git
    cd hedgedoc
    
  2. Install Node.js (at least Node 14, we recommend Node 18) and Yarn
  3. Install Caddy (select one of the two options)
    • Download and place the caddy binary in hedgedoc/dev-reverse-proxy. Ensure it is executable with chmod +x caddy. Users of macOS may need to run xattr -d com.apple.quarantine ./caddy to lift the quarantine for executables from the internet.
    • Install Caddy using your package manager
  4. Install the dependencies in repo root directory with yarn install
  5. Goto hedgedoc/backend directory with cd backend
  6. Create the .env config file by copying the example: cp .env.example .env
  7. Add a value to HD_SESSION_SECRET in the .env file. This can be any string, which has to be a secure password for production but can be set to simple string for debug purpose.
  8. Execute the following lines
    echo "HD_AUTH_LOCAL_ENABLE_LOGIN=true" >> .env
    echo "HD_AUTH_LOCAL_ENABLE_REGISTER=true" >> .env
    
  9. Go back into the root directory with cd ..
  10. Run yarn start:dev

    This will execute the backend, frontend and reverse proxy at once

  11. Use your browser to go to http://localhost:8080

More detailed development setup

The following sections describe a more detailed setup of all components.

Preconditions

If you want to run HedgeDoc in dev mode some preconditions have to be met.

  1. Make sure that NodeJS is installed. You need at least Node 14 (we recommend Node 18).
  2. Make sure that Yarn is installed.
  3. Clone this repo (e.g. git clone https://github.com/hedgedoc/hedgedoc.git hedgedoc)
  4. Go into the cloned directory

Installing the dependencies

Because we use Yarn workspaces, Yarn collects the dependencies of all packages automatically in one central top-level node_modules folder. To install the dependencies execute yarn install at the top level of the cloned repository. Execute this command ONLY there. There is no need to execute the install-command for every package. It's important to use Yarn. We don't support npm or any other package manager and using anything else than Yarn won't work.

Build the commons package

Some code is shared by backend and frontend. This code lives in the `commons package and needs to be built so frontend and backend can import it. This only needs to be done once, except if you've changed code in the commons package.

  1. Go into the commons directory.
  2. Execute yarn build to build the commons package.

Setting up the Backend

Note: The backend can be mocked instead of starting it for real. This is useful, if you just want to work on the frontend. See the "Mocked backend" section below.

  1. Go into the backend directory.
  2. Create an environment file. We recommend to use the example file by running cp .env.example .env You can modify this file according to the configuration documentation.
  3. Make sure that you've set HD_SESSION_SECRET in your .env file. Otherwise, the backend won't start.

    In dev mode you don't need a secure secret. So use any value. If you want to generate a secure session secret you can use e.g. openssl rand -hex 16 | sed -E 's/(.*)/HD_SESSION_SECRET=\1/' >> .env.

  4. Make sure that HD_BASE_URL in .env is set to the base url where HedgeDoc should be available. In local dev environment this is most likely http://localhost:8080.
  5. Start the backend by running yarn start:dev for dev mode or yarn start for production.

Setting up the frontend

The frontend can be run in four different ways. The development mode compiles everything on demand. So the first time you open a page in the browser it may take some time. See here for a more detailed description of the environment variables for the frontend. A special configuration isn't necessary but keep in mind that you execute all commands from within the frontend directory.

Mocked backend

To start the development mode, run yarn start:dev:mock . This task will run the frontend in mock-mode, meaning instead of running a real backend, the frontend mocks the backend. This way you can work on frontend functionality without starting up the full development environment. The app should run now and be available under http://localhost:3001 in your browser. In development mode the app will autoload changes you make to the code.

With local backend

To start the development mode with an actual HedgeDoc backend use yarn start:dev instead. This task will automatically set HD_BASE_URL to http://localhost:8080.

Production mode

Use yarn build to build the app in production mode and save it into the .next folder. The production build is minimized and optimized for best performance. Don't edit the generated files in the .next folder in any way!

You can run the production build using the built-in server with yarn start. You MUST provide the environment variable HD_BASE_URL with protocol, domain and (if needed) subdirectory path ( e.g. http://localhost:3001/) so the app knows under which URL the frontend is available in the browser.

If you use the production build then make sure that you set the environment variable HD_BASE_URL to the same value as HD_BASE_URL in the backend.

Production mock build

It is also possible to create a production build that uses the emulated backend by using yarn build:mock. This is usually not needed except for demonstration purposes like https://hedgedoc.dev.

Running backend and frontend together

To use backend and frontend together in development mode you'll need a local reverse proxy that combines both services under one URL origin. We recommend to use our pre-configured Caddy configuration.

Running the reverse proxy

  1. Download the latest version of Caddy from the Caddy website or alternatively install it using your package manager. You don't need any plugin. Place the downloaded binary in the directory dev-reverse-proxy. Don't forget to mark the file as executable using chmod +x caddy. Users of macOS may need to run xattr -d com.apple.quarantine ./caddy to lift the quarantine for executables from the internet.
  2. Start Caddy using ./caddy run (if you downloaded the binary manually) or caddy run (if you installed Caddy via a package manager).
  3. Open your browser on http://localhost:8080