Fulltext search with rich media files

What's the best way to perform fulltext searches across rich media files? I'm trying to implement a system where the user could upload random files (.doc, .pdf, .jpg, ...) and down the line, he would be able to search for them based on the file contents or metadata.

I would appreciate some ideas on how to build this.

PS - I started looking into Lucene and Nutch but I think they do way more than what I need.

Tha开发者_如何学运维nk you.

You should take a look at Tika (http://lucene.apache.org/tika/) which is a toolkit for detecting and extracting metadata and structured text.