tldr translation progress

Generates and publishes a website, which shows the progress of translation of the tldr project.

This project is inspired by

• Extension Registration Wall of Superpowers
• Python 3 Wall of Superpowers

Building

Docker

If you've got Docker on your system, you need nothing else to build and run this project.

docker build -t ghcr.io/lukwebsforge/tldrprogress:latest .

Local installation

To build this project without Docker, you'll need the task runner Task and a Go SDK (>= 1.18).

The build instructions are defined in the Taskfile.yml. Install Task and run

task build

Configuration

You can configure certain details using environment variables.

Environment variable	Description	Required
GIT_NAME	The committers name	Yes
GIT_EMAIL	The committers email	Yes
SITE_REMOTE_URL	The Git SSH url of the deployment repository	Yes
CHECK_KNOWN_HOSTS	Checks the known_hosts file when connecting	No
SSH_KEY_PASSWORD	A password for the SSH key	No
DONT_PUBLISH	No changes will be committed & pushed	No
RUN_ONCE	Instantly updates & quits	No

You can either let the program generate a new SSH key pair, or you can use your own even if a password is required. This application doesn't support HTTP authentication.

The SSH key pair id_ed25519 and id_ed25519.pub in the folder keys will be used to access the tldr repository and publish changes to the deployment repository (SITE_REMOTE_URL). If the key pair is missing, it'll be generated during the startup of the application. The generated public key id_ed25519.pub will be printed to the console. I recommend adding the public key as a deployment key (with writing access) for the deployment repository.

Each day at midnight (UTC) the program will execute the update. This includes the steps of cloning / updating the tldr repository in the folder tldr and generating the static asset files and copying them to the upstream folder. Finally, changes in this folder will be committed and pushed to the upstream repository.

If you want to test your changes, I recommend you set the environment variables DONT_PUBLISH and RUN_ONCE and view the updated website locally before deploying it. The value of this environment variable doesn't matter, it just has be set.

Run

Docker

You can simply start a container which runs this program. The program will start the repository update at midnight (UTC) each day.

# Create a docker container with name tldrprogress, which will always restart (system reboot, error)
docker run -d --restart=always --name=tldrprogress \
 -e GIT_NAME=LukWebBuilder -e GIT_EMAIL=gitbuilder@lukweb.de \
 -e SITE_REMOTE_URL=git@github.com:LukWebsForge/tldri18n.git \
 ghcr.io/lukwebsforge/tldrprogress:latest
# Optional: Take a look at the logs to view the new public SSH key
docker logs tldrprogress

systemd

The program runs permanently, and start the repository update at midnight (UTC) each day. This repository contains a systemd configuration file, which can be used to keep it running.

If you want to use the systemd unit file

1. Create a user named tldr
2. Put the executable at the path /home/tldr/progress/update
3. Copy the files from the systemd folder into /etc/systemd/system/
4. Finally, run

# Load the new systemd unit file
systemctl daemon-reload
# Start the program on every system startup
systemctl enable tldrprogress.service
# Start the program now
sudo systemctl start tldrprogress.service
# Optional: View generated the public SSH key
cat /home/tldr/progress/keys/id_ed25519.pub

Contributing

If you spot a bug or got an idea for an improvement, free welcome to open a new issue or create a pull request.

Development tips

Working on the Go application

It's useful to set the environment variables DONT_PUBLISH and RUN_ONCE to any value (e.g. true).

Even, if you don't want to push anything, don't forget to add the public SSH key to a GitHub account or as a deployment key. Otherwise, GitHub repositories can't be cloned or updated.

Working on the React website

Run the Go application once to generate a data.json file. Copy this file to the resources/public folder. Now you can use HMR by starting the development web server with yarn run start.

Testing the application as a whole

Run the compiled artifact and start a local webserver by using npx serve upstream to inspect the generated website.