Parsing & conversion of a PDF file to plain text / user defined format depends on its configuration. Solr does follow this mechanism before indexing it , try this link to for further details,
http://www.lucidimagination.com/search/?q=Can++the+PDF+file+be++parsed+and+converted