You can’t just ask that. Not without at least taking them out to a fancy dinner first. Maybe a little dancing.

Before anything, you’ll want to create a folder for the unix socket: mkdir /var/run/ytproxy and chown it to your reverse proxy’s user and group.

The Docker files:

services:
  piped-frontend:
    image: 1337kavin/piped-frontend:latest
    container_name: piped-frontend
    environment:
      BACKEND_HOSTNAME: $API_ENDPOINT
    depends_on:
      - piped
    restart: unless-stopped
    networks:
      - proxy

  piped-proxy:
    image: 1337kavin/piped-proxy:latest
    container_name: piped-proxy
    environment:
      - UDS=1
    volumes:
      - /var/run/ytproxy:/app/socket  # unix socket location
    user: 1000:1000
    restart: unless-stopped
    networks:
      - proxy

  piped:
    image: 1337kavin/piped:latest
    container_name: piped-backend
    volumes:
      - ./piped/config/config.properties:/app/config.properties:ro
    depends_on:
      - piped-db
    restart: unless-stopped
    networks:
      - backend_piped
      - proxy

  piped-db:
    image: pgautoupgrade/pgautoupgrade:16-alpine
    container_name: piped-db
    environment:
      - POSTGRES_DB=$DB_NAME
      - POSTGRES_USER=$DB_USER
      - POSTGRES_PASSWORD=$DB_PASS
    volumes:
      - ./piped/pgdb:/var/lib/postgresql/data
    restart: unless-stopped
    networks:
      - backend_piped

networks:
  backend_piped:

  proxy:
    external: true

API_ENDPOINT=  # no scheme prefix (ex. pipedapi.domain.example)
DB_NAME=
DB_USER=
DB_PASS=

To configure the reverse proxy (I use nginx), you can use Piped’s recommended files with your domains. The linked repo also contains a template for the required config.properties.

Let me know if you run into issues, I’ll be glad to help 🙂

I didn’t say it was, I’m just answering the dude’s question

So there’s 2 files that go along with the main script: a config file and a download list.

The download list file is literally a list of YT channel URLs and any time I want it to get new channels, I add to the list. Conversely if I get tired of watching a channel, I comment it out or just delete it. YT-DLP just traverses the list.

The config is where the magic happens. That dictates how the file is named, quality you download at, location it’s saved to, format, etc. You can incorporate SponsorBlock which is awesome and even DeArrow has an integration. I enjoy the logging feature so it doesn’t re-download vids it’s already snagged. Generally I’ll tell it to get every video from a channel that’s been posted in the last 15 days.

My destination for the files is a YouTube library folder I set up in Plex. Plex sees it just as another TV-like library and it pulls the metadata from the files, which are embedded by YT-DLP.

Hope this helps!

Can’t speak to his method, but the jellyfin media sever has a YouTube plugin called Fintube that uses the above downloader to integrate YouTube content.