Installation

Last Updated: 2021-11-04

Hadfield uses GitHub workflows to generate a new Docker image every time the main branch is updated in the git repository. The Docker image is the recommended and official method of using Hadfield.

Configuration
Docker
Docker Compose

Configuration

Hadfield looks for a TOML configuration file at the path provided by CONFIGURATION_PATH and if it cannot find one then for a configuration.toml in the working directory.

The following is a regular subset of configuration options available:

key	description	required	default
`server.host`	The address for the server to bind to	false	All available
`server.port`	The port for the server to bind to	false	`7000`
`gopher.host`	The address to display on menu entries	false	The value of `server.host`
`gopher.server`	The port to display on menu entries	false	The value of `server.port`
`gopher.documentRoot`	The path of the TOML document directory*	true
`gopher.assetRoot`	The location of the non-document file directory*	true
`gopher.capabilites.*`	Properties for `caps.txt`	false

^{*note: these values are relative to the working directory}

An example configuration.toml might look like:

[server]
	port = 70
[gopher]
	host = "julianwebb.ca"
	port = 70
	documentRoot = "data/documents"
	assetRoot = "data/assets"
	[gopher.capabilites]
		ServerName = "Julian's Server"

for per document configuration see Usage

Docker

Hadfield can be run via the following command:

docker run -p 70:7000 -e CONFIGURATION_PATH=/app/data/configuration.toml -v ${PWD}/data:/app/data ghcr.io/julianwebb/hadfield

By default Hadfield uses port 7000 and listens on every available address.

See the previous section for information on configuration

Docker Compose

Example Docker Compose:

version: "3"

services:
  hadfield:
    image: ghcr.io/julianwebb/hadfield
    ports:
      - "70:7000"
    environment:
      CONFIGURATION_PATH: "/app/data/configuration.toml"
    volumes:
      - "./data:/app/data"
    restart: "unless-stopped"