Do you own the wiki? If not, then you will need to curl in the content, and index the content that way. If you do own the wiki, then just fetch the data from the database and index it there.... or if its a file based wiki then just get your scripts pointed at those files and index them.

In other words, you really don't need to spider the site with Lucene, that can be done with either another script (CURL) or form another source (DB, File System - if you have access to them)