Extracts the pages containing annotations from the given pdf file.
Requires pdftk. You may also need to install bcprov and java-commons-lang or similarly named packages if pdftk yells at you.
Usage:
python extract.py in.pdf out.pdf
Generates a new pdf file out.pdf containing only the pages with annotations from in.pdf.