>From: Andreas Gohr <andi@xxxxxxxxxxxxxx> > Yes. The snippet creation isn't very efficient speedwise so I restricted > it. If someone could come up with a faster (but UTF-8 aware) method it > would be great. Snippet caching, done at page rendering time. You'd make the snippet a bit longer, though, so you could use it in RSS feeds too, as long as the feeds still only show snippets from the most recent revision (even if we eventually show results from multiple revisions). You'd just truncate the snippet to a word boundary for search results -- or even pre-calculate that boundary to save even more time. You were baiting me, weren't you? ;-) ~joe -- DokuWiki mailing list - more info at http://wiki.splitbrain.org/wiki:mailinglist