Data extraction

I need a Lucene or SOLR-based implementation that does the following: 1) A web-based UI to fetch a web document or a PDF from a defined source (eg. a URL) 2) Allows the user to define what to look for (i.e. keywords, phrases) 3) Scan the document for specific information, eg. <Field A>, <Field B>, etc. 4) Extracts the information and outputs it to a CSV file I have attached a sample PDF file. The script would: a) Go to the Apple website (<[login to view URL]>). It would click on each link: - [[login to view URL]][1] - <[login to view URL]> etc. etc. Each link has a downloadable PDF document, from which I would like to extract info. b) It would allow me to define what to look for. For Apple, I want to extract: - iPhone and Related Products and Services - Units, Revenue for the various periods (Q2 2012, Q32011 and Q32012) - iPad and Related Products and Services - Units, Revenue for the various periods (Q2 2012, Q32011 and Q32012) c) The output of the search will be a CSV file