New wiki parser function - please review

I have added another simple parser function to the mediawiki extension I wrote a while ago. It is meant to replace the manual duplicated command lines we have on various pages, that only differ in the ENV vars. The purpose is to keep the page markup neat while producing the best usable output.

You can see it in action on this page:
http://wiki.diaspora-project.org/wiki/User:Raven24/XtrpltTest

And here is the source code of the parser function:


Note: This discussion was imported from Loomio. Click here to view the original discussion.

While I appreciate your effort, I’m not sure this is the way we need to go. Automatically generating the variations helps with maintaining the docs, it doesn’t remove the current bad UX though, which I consider a way more important issue.

Picking the right case from a big “bla” or “bal” or “lba” list is very difficult, especially if they look similar and you actually have no idea what they do. I think we need a way to generate multiple versions of a page from a single source. This way the entry point can be a list of “I want to Install Diaspora with a) MySQL b) PostgreSQL on a) Debian, b) Ubuntu…” and then show one page to read, without any “If you’re using bla then… otherwise do…” stuff.

how about something like they have in the selenium docs: http://docs.seleniumhq.org/docs/

  • a choice of “flavor” at the top of the page, and the snippets get shown/hidden based on what is selected.

Could work, though I’m not sure if we can make it clear that you have to choose one option for each of our five categories (RAILS_ENV, DB, RVM, OS, Release/Head).