BlueWolf wrote:Hmm, perhaps it could cache the content to the local server/synch when first installed. Then only pull/update content if a basic date comparison doesn't match!? Just a performance idea to work around the load.
But generally, I hadn't even thought to take that into consideration, Martijn. Good call, as always.
Well, when you work in a big web environment like I do, you tend to think of nasty things that might happen. 
Anyway, a basic (date) comparison might work. Maybe some sort of hash value based on the page contents, dunno. Whatever comparison system is thought of, it'd have to be a server side thing since you'd only want to download the current date / hash / crc value or whatever, not the whole page.
A completely different approach (well sort of) would be to look at the possibility of dokuwiki creating some sort of local, static export. If possible, that cron job could be run once a day or even two days (the docs don't change that often after all) to create the static html pages. A second part of the cron job could create a simple list of pagename <=> hash/crc value combinations.
The plugin could then download that relatively small list and check the hash values it stored from previous updates to the most recent hash values. Upon detecting a difference, it'd download the relavant static page(s).
But this would require dokuwiki to generate static exports and include a 1-2 day delay in documentation updates.
Anyway, just some thoughts to take in. 
Wolf CMS founder and lead developer
Please always check the
Support forums and
Wiki before asking. (
My Ohloh account.)
Like Wolf CMS? Consider making a
financial contribution.