Hi Folks,

I'm seeking a linux software solution to something I'd like to do. I run ubuntu but I'm not tied to that or at least I can compile what I need. I have about 1200 pdf documents that are scans of a non-profits brochures and other miscellaneous publications. The are a mix of text and images but all have text. I'd like to see if there is a software that can ingest the documents, create a searchable index and serve up the documents over a local network to pc's, and android tablets, perhaps even android phones. By serve it doesn't need to render them it can pipe the data raw and I can use a local clients software to view the documents. I've worked with commercial search engines and repositories before like Verity and Autonomy and they would work well but I'd like to find and open source solution to this because this is for personal use not business. Just thinking of doing this local on a headless ubuntu server and no need for it to be accessible over the web.

Thanks in advance for your suggestions.
Spit

You could try Recoll or Calibre.

Calibre does a good job of indexing metadata, and can serve its library to remote clients. It doesn't do full-text search natively, but here is a plugin (which I haven't tried) intended to do just that, though it looks like it takes a bit of work to get it going.

The plugin uses Recoll, which - from a quick look at its home page - looks like it should be able to do the job on its own.

I would assume that the PDFs are images which means the text is also just graphics and not searchable. I have not played with recoll but it appears to use pdftotext which will not work for text as part of an image. There is OCR software but depending on the quality of the scanned documents and type of text could be a lot of work.


https://help.ubuntu.com/community/OCR

Some more suggestions for search tools.
http://www.linux.com/news/software/a...gines-compared