PDF, the format. By Adobe (reference required)

slightly fedora (my current distro) centric description of the python or other pdf parsing libraries:

library	where	example	capabilities	availability	platforms
l poppler	yum install pypoppler

|| l pdfresurrect || yum install

l pdftools

pypoppler is a launchpad hosted project, missing many features compared to c-poppler.

poppler itself is pretty good, but some problems:

text selection is broken, especially for hebrew text.
- but pdftotext is actually much better.

Concrete Usage Examples

TODO

Extracting Data

http://www.757labs.com/projects/pdfresurrect/ - very basic (0.0.4), meant to track changes in pdf documents (history embedded in the pdf).

http://www.boddie.org.uk/david/Projects/Python/pdftools/ - can get number of pages, and each page, but no layout engine - cannot easily (without further manipulation, like regular expressions and worse) get the text (as the user sees it - individual words are easy).
pdfminer - http://pypi.python.org/pypi/pdfminer/20080906