On 10/04/07, Tels nospam-abuse@bloodgate.com wrote:
Well, or write an extension that implements his idea, e.g.:
* upon upload, run the document through an index-generator (per file type) * add that index-text to some searchable index, or store it in the mysql
I've thrown some very experimental initial work on such an extension into Subversion; trunk/extensions/FileSearch. It introduces a framework for defining "Extractors", which are capable of extracting indexable content from uploaded files, and adds this into the search index for the image page itself when a file is uploaded.
Searching seems to work (the reference implementation provided is for plain text file uploads, which are obviously straightforward), but result display leaves a lot to be desired. I anticipate rewriting the whole thing to provide cleaner, better-looking integration into MediaWiki at some point soonish.
Rob Church