I just answered my own question. The program is pdftotext, part of the xpdf package available for unix machines. It is very cool and very fast. terry Kapil Thangavelu wrote:

Terry Kerr wrote:

...

Hi,

I need to be able to index the text within pdf files. I assume I will somehow use PrincipiaSearchSource, but I need to know how to get the text out of the pdf when it is uploaded to the ZODB. Has anyone done this before? Are there any packages around that I can use that run in python or at least on a linux box that I can pipe to and from?

terry

from xml2pdf there are a multitude of ways in python

XSLT - check out the ibm.com/developer xmlzone they have an article in the education lib for transforming xml to pdf.

platypus packages from http://www.reportlab.com/

they might give you some help in going the other way..

as for implementation...

looking at a pdf in a text viewer it appears to be formating text and encoded display strings.

you could write a subclass of file, which read its content upon upload stripping the formatting string and decoding the display strings and storing that as a property to be indexed.

Kapil

-- Terry Kerr (terry@adroit.net) Adroit Internet Solutions Pty Ltd (www.adroit.net) Phone: +613 9563 4461 Fax: +613 9563 3856 Mobile: +61 414 938 124 ICQ: 79303381