Feature #17238


Searching parsed wiki content - Cache option for Wiki pages

Added by Fabian Strasser about 10 years ago. Updated almost 10 years ago.

Target version:
Start date:
Due date:
% Done:


Estimated time:


I was wondering if it would be possible to modify the search to optionally search parsed wiki sites and not the raw source. This should always be optional (checkbox?) because sometimes it might be necessary to search the pages source and not the parsed output.


I use macros a lot, like the built-in

along with others from Plugins that allow me to easily display tickets inside a wiki page.

My problem is that using the search, wiki pages that contain those tickets (or mainly Macros) aren't found as only the source code is searched.

Example wiki source

h1. Test issue summary

This site contains all information [...]


Let's assume that above wiki page will list 7 tickets (using the Redmine Wiki Lists plugin) that all contain "Searchterm" inside their topic. I can find that wiki page only when searching for "Searchterm" but not when I use any other term that is part of those tickets' topics.

If for example a tickets' topic is "Product Z has major issue with raw input" and I refer to that ticket with the term "Product Z" inide the wiki, I only find the wiki page with that term, but not when I search for "input".
It would be necessary to parse the wiki sites and then search them to find the term in question.

I am however afraid that this might lead to serious loading times and even more issues, especially with a bigger wiki since every page has to be parsed. A work-around might be a cache that keeps wiki sites availabe in a parsed state, ready for delivery. And when a page is altered the cache for that site gets renewed.

This would also speed up the overall site performance in my opinion.

Anyone besides me interested in such a feature? Good idea? Bad idea? Why?

Actions #1

Updated by Fabian Strasser almost 10 years ago

On another note: If such a feature would be implemented this would also be the resolution to another problem:

When you export a wiki site that uses macros to PDF the output doesn't show the parsed state but only the source-code which isn't useful at all. Here we could also provide the cached, rendered site.


Also available in: Atom PDF