In some cases, keyword indexing of PDF, DOCX or ODT need a special server configuration in order to work.
Before you start worrying about keyword indexing, you should take a look at the File Info box below the form when editing a file. It shows the keywords that will be used for searches. If you don’t find a list of keywords here this article gives some ingsights about requirements and troubleshooting.
PDF indexing relies on Ghostscript. This is a commandline tool (gs) that need to be installed on the server. The minimum version required is 9.05. Any older version before 9.05 does not support text extraction of PDF-files. WP-Filebase will let you know if this is the case (notice on the settings page). If you are unsure about gs availability or how to install it, ask your hosting provider. It might be necessary to run PHP in CGI-mode in order to run external programs.
Apart from that, WP-Filebase Pro also includes an experimental text extraction method, that does work even without Ghostscript. It tries to find GZIP-Compressed text blocks in the PDF file and extracts them.
Indexing of Microsoft Word .docx and Openc Office .odt documents requires the PHP unzip_file-function to be available. If you are experiencing a lack of keywords, ask your hosting provider to enable the PHP ZIP functions..