Frequently Asked Questions

Which is the biggest site powered by Waliki?

It’s the `Python Argentina Community's wiki < >`_ . It was migrated to Waliki from MoinMoin in March 2015.

It has more than 1000 pages and few active users.

Does Waliki scale?

May be, but huge wiki site are not the Waliki’s target.

My main concern about the Waliki’s “scalability” is on how many concurrent users may it support and how slow is to save a page..

The “git commit per edition” is cool, I’m happy with the merge-instead-block approach for concurrent editions, but it could be a boottleneck for a high traffic wiki.

I guess it could be improved in the future, using libgit2 instead of plain system calls to the git cli.

Can Waliki render math?

Sure! Both reStructuredText and Markdown play well with Mathjax. As the Mathjax’s assets (javascript file) are huge, it’s disable by default. To enable it you need add waliki.context_processors.settings to TEMPLATE_CONTEXT_PROCESSORS and set WALIKI_USE_MATHJAX to True in your settings:


Does it use some cache?

Yes, it uses the builtin django’s cache framework . The cache is invalidated automatically when a page is modified. The default cache timeout is 1 day, but you can override it setting WALIKI_CACHE_TIMEOUT (in seconds). For example:

WALIKI_CACHE_TIMEOUT = 3600    # 1 hour cache
Can I have user pages?

Of course, but this feature isn’t built-in. As a user page is a regular page editable exclusively by an user, you can hook a simple code to create an ACLRule instance to the signal raised when a new User is signed up.

Check the code of this implementation in the Python Argentina’s site.