I have sent the first useable version of the filesearch module to Merijn. I leave it up to him to decide how/if/when the module is released for testing by others.
Main features of the module is that the text content of the most popular binary formats (doc, docx, xls, xlsx, ppt, pptx, ods, sxc, odt, swx, odp, sxi, pdf, mp3, jpg) is extracted from files being oploaded and stored in the database. Mysql's full text indexer is taking care of indexing and serving search results.
This operation seems to be going very fast. On my development systems, it is not causing a noticeable lag during file upload, but stress tests could, of course, still reveal performance issues.
I am looking forward to your comments.