This is a ZIM archive reader for browser extensions or add-ons, developed in HTML5/Javascript. You can get the extension from the Mozilla, Chrome and Edge extension stores (search for "Kiwix", or click on a badge below). There is a version implemented as an offline-first Progressive Web App (PWA) at https://moz-extension.kiwix.org/current/, primarily intended for use within the Mozilla Extension.
Once you have obtained an archive (see below), you can select it in Kiwix JS, and search for article titles. No further Internet access is required to read the archive's content. For example, you can have the entire content of Wikipedia in your own language inside your device (including images and audiovisual content) entirely offline. If your Internet access is expensive, intermittent, slow, unreliable, observed or censored, you can still have access to this amazing repository of knowledge, information and culture.
The reader also works with other content in the OpenZIM format: https://wiki.openzim.org/wiki/OpenZIM.
A proof-of-concept inspired and enabled by Hosting SQLite Databases on Github Pages and the ensuing Hacker News post. The compiled single-page app supports autocomplete for titles, automatic redirecting & other MediaWiki datasets like WikiQuote or Chinese Wikipedia. It makes no external API calls except to get Wikipedia's images.
Seems ideal for making offline copies of Wikipedia (or other Mediawiki installs, it's implied) available.
Search is disabled right now.
Github: https://github.com/segfall/static-wiki
SQLite copies of Wikipedia: https://www.kaggle.com/segfall/markdownlike-wikipedia-dumps-in-sqlite
Has instructions for turning an XML dump of Wikipedia into a SQLite database; unfortunately it uses node.js.
What to do when you see the dreaded "Call to undefined function filter_var()" error message in your Apache logs. It's actually easier that it sounds but I recommend matching the version of the PHP source you download to that installed from your distro's official package.
A Mediawiki extension that allows the creation of raw HTML snippets that can be used in wiki pages that do interesting and helpful things. Implemented by creating a namespace called Widget, into which new wiki pages containing the widget code in question can be written like any other wiki page. The API is documented on this page.
XowA does one thing and does it well: It lets you make a local backup of Wikipedia (probably any Mediawiki, really) to carry around with you or copy to removable media (like a USB key or DVD-ROMs) so you can give out copies and read it offline. Runs on Windows, Linux, and MacOSX. Portable - can be run from removable media. Even includes a search engine. Written in Java.
Documentation for the Mediawiki REST API.
Official Mediawiki Python module. Designed to make it easy to interface with a wiki and write tools to manipulate it. Specifically for writing bots.