Files
linux-odyssey/README.md

200 lines
6.0 KiB
Markdown

# Linux Odyssey
[![Production](https://github.com/linux-odyssey/linux-odyssey/actions/workflows/production.yml/badge.svg)](https://github.com/linux-odyssey/linux-odyssey/actions/workflows/production.yml)
[![Staging](https://github.com/linux-odyssey/linux-odyssey/actions/workflows/staging.yml/badge.svg)](https://github.com/linux-odyssey/linux-odyssey/actions/workflows/staging.yml)
## TLDR
bun install
cp .env.sample .env
docker compose up -d
bun run dev
## Requirements
- Bun: v1.3+
- Docker
## Known Issues
- MongoDB requires x86 architecture to run for v5+ versions. You can track [this issue](https://github.com/linux-odyssey/linux-odyssey/issues/166) for updates.
Windows user please refer to [docs/windows-setup.md](docs/windows-setup.md) for setup instructions.
## Installation
git clone https://github.com/lancatlin/linux-odyssey.git
cd linux-odyssey
bun install
## Project Structures
1. **app**: frontend
2. **server**: backend
3. **quests**: tutorials scripts for each quest
4. **packages**: other packages
1. **analytics**: admin dashboard to analyze collected data
2. **constants**: some shared config values for app and server
3. **container**: the script run in container to capture users' command
4. **file-graph**: the algorithm to construct and modify file graph, used in both app and server
5. **models**: definitions of MondoDB collections (data schema)
6. **utils**: other utility code shared between packages
7. **game (in the future)**: core logic of game playing, will be extracted from server
## Configuration
The configuration is stored in the `.env` file. You can copy from the sample file.
cp .env.sample .env
## Docker Setup
### Docker compose files explaination
1. **docker-compose.yml**: The one used for development. It sets up MongoDB and builds base quest image.
2. **docker-compose.prod.yml**: The one used for production deployment. It has a Nginx service, a backend, a database, an analytics service, and creates isolated networks for player containers.
3. **docker-compose.testing.yml**: Similar to production config, but sets up an one-time environment for end-to-end testing.
### Common Development Setup
This sets up a development database, and running development server for frontend and backend.
Pull the latest images:
docker compose pull
Start the containers:
docker compose up -d
This will start the db. it will automatically restarts by Docker.
Building quest images:
docker compose build base
### Build Quest Images in Details
In each quest folder, there is a `game.yml` file, which is the quest definition.
If you want to include additional files, put them in a `home` folder. The files in `home` will be copied to the user's home on container creation automatically.
quests/
├── Dockerfile # Base image
├── entrypoint.sh # Entrypoint for the container
├── discover
│ └── game.yml
├── helloworld
│ ├── Dockerfile # Overwrite the image
│ ├── game.yml
│ └── home # Include additional files
│ └── forgotten_scroll.txt
├── read
│ └── game.yml
└── spell
├── Dockerfile
├── game.yml
└── home
└── ancient_scroll.txt
An optional `Dockerfile` can overwrite the base image. You can use this to install additional dependencies or modify the environment.
Customized `entrypoint.sh` for each quest is not yet supported.
To build all quests:
bun run build:quests
The base image is built with docker-compose, so you can use `docker compose build base` to rebuild it. Once you rebuild the base image, you have to rebuild all quests.
### Optional: Mounting Host Directory
To update the home content of a quest, simply re-create a new session will do. The home content is mounted and copied on container creation.
To further improve the development experience, we can mount the host directory to the container in the pattern:
/path/to/linux-odyssey/quests/${quest_id}/home:/home/commander
You can edit the files on your host, and the changes will be reflected in the container. This is disabled by default. To enable it, you can set the following environment variables:
MOUNT_QUEST=true
### Optional: Mounting terminal service & cmd-hook
To make the development of terminal services & cmd-hook smoother, we can enable `MOUNT_CLI` to mount the local binaries into container.
MOUNT_CLI
And you can run `air` inside of `services/terminal-service` or `services/cmd-hook` to get auto recompiling.
## Development
Copy `.env.sample` to `.env` and customize it for your needs.
Run dev server (frontend, backend):
bun run dev
To develop analytics dashboard:
bun run analytics
The default username and password is `alex` and `Alex1234`.
## Testing
Unit tests:
bun run test
Run Cypress locally, using the current development containers:
docker compose up -d
bun run test:e2e
Run Cypress in container, create brand-new containers along with it:
./docker-scripts.sh test
# or with building
./docker-scripts.sh test --build
# tearing off the containers
./docker-scripts.sh down
## Deployment
Create production env
cp .env.prod.sample .env.prod
Edit `.env.prod`
BASE_URL=https://yourdomain.com # IMPORTANT to make login work
SECRET_KEY=xxxx # Change it to something safe!
Start the service
# Build the images (only allowed for maintainers)
docker compose -f docker-compose.prod.yml build
docker compose -f docker-compose.prod.yml push
# Server side
docker compose -f docker-compose.prod.yml pull
bun run build:quests
docker compose -f docker-compose.prod.yml up -d
### Necessary Configuration
To enable social login, you should have OAuth client token in a `.env` file:
You can copy from `.env.sample`
**YOU MUST CHANGE THE SECRET_KEY** in `.env` file.
SECRET_KEY=your-secret-key
You can generate a random key by running
bun run setup