Node / public / Retipedia rns://4cf8a0651c4d73cacd0f93ac1d95e80a/public/Retipedia

󰉖 Files • 󱌣 Work (0) • 󰋚 Commits (22) • 󰘬 Branches (1) • 󰓼 Tags (0) • 󰋑 Thanks (6) •  Stats

• Formatting overhaul to add anchors, tables, inline citations and better page layouts

• Support for multiple .zim archives

• Add full Gutenberg parser

• Parser improvements

• Server-side caching

• Option to read pages in sections for low-bandwidth users

• And more

!A screenshot of Retipedia running in NomadNet

This lets you host a NomadNet node that serves searchable versions of multiple sites from .zim archives provided by the Kiwix project

1. Install the python-libzim package for reading .zim archives

T383838pip install libzim

2. Install the BeautifulSoup HTML parser package for converting the HTML from each .zim entry to Micron

T383838pip install beautifulsoup4

3. Move the contents of the git repo to your NomadNet T383838storage/pages/ directory

4. The page folder is detected automatically when the files live under T383838storage/pages. You only need to adjust T383838root_folder in T383838settings.py for other layouts

5. Ensure all micron files for the project are executable

6. Point T383838zims_dir in T383838settings.py at a folder containing one or more T383838.zim archives. A single T383838archive_path is still supported as a fallback

7. Run T383838python3 generate_meta.py once to scan the archives and write the per-archive metadata sidecars into the T383838zims/ folder. Re-run it whenever you add new archives. Only archives with a sidecar are listed, so delete T383838zims/<name>.zim.meta to hide an archive

The T383838.mu pages start with T383838#!/bin/python3, so NomadNet runs them with the system Python by default. If you installed T383838libzim and T383838beautifulsoup4 into a virtual environment instead, pages will crash with T383838FileNotFoundError. To point every page at the environment's interpreter, run this from inside that environment

T383838python3 generate_meta.py --fix-shebangs

To choose an interpreter explicitly, pass it after T383838= and use quotes if it contains spaces

T383838python3 generate_meta.py --fix-shebangs=/home/user/miniconda3/envs/nomad/bin/python3

T383838python3 generate_meta.py "--fix-shebangs=/usr/bin/env python3"

To go back to the default just run it with T383838--fix-shebangs=/bin/python3

You can download .zim archives provided by the Kiwix project here

https://browse.library.kiwix.org

The recommended .zim is T383838wikipedia_en_all_nopic Wikipedia archives in every language are supported

For image support, download the T383838_maxi Wikipedia instead of the T383838nopic Wikipedia .zim archive

Wikipedia

https://browse.library.kiwix.org/#lang=&q=&category=wikipedia

Gutenberg

https://ebookfoundation.org/openzim.html

iFixIt

https://browse.library.kiwix.org/#lang=&q=&category=iFixit

StackExchange (all languages)

https://browse.library.kiwix.org/#lang=eng&category=stack_exchange&q=

MedlinePlus, health information from the National Institutes of Health (NIH)

https://browse.library.kiwix.org/#lang=&q=medline

Retipedia currently supports Wikipedia in all languages, Gutenberg, StackExchange in all languages, iFixIt and MedlinePlus archives, and also implements a generic .zim parser for anything else. PRs are welcome

!A screenshot of iFixIt in Retipedia

!A screenshot of Wikipedia maxi in Retipedia

iFixIt and Wikipedia archives with pictures (maxi) can show their images in NomadNet 1.4.0 or newer on terminals with Kitty graphics support such as Konsole, iTerm2, Ghostty and Kitty.

Set T383838images = True in T383838settings.py to turn it on. Readers only download images when their own NomadNet T383838image_loading setting allows it, and is safe for low-bandwidth links.

Images are extracted once into the T383838images/ folder and served from there. You can clear that cache with

T282828