pat2pdf

This script fetches the pages of a U.S. patent document from the USPTO patent database and converts them into a single PDF file. To find patents use the USPTO advanced search page. pat2pdf is released under the terms of the GPL.

For a free web-based interface to pat2pdf visit www.pat2pdf.org. Not affiliated with www.pat2pdf.com, a web site providing a similar service for a fee.

freePat is a Windows program loosely based on this script. It also supports international patents.

Why did I do this

To help open source developers who increasingly find themselves facing software patent problems. Actually, the real reason is because I am a cheap bastard who doesn't want to pay for downloading a PDF but I thought it would be a good place to make a point about the increasing abuse of software patents.

Is it legal?

The USPTO search site allows downloading a PDF file for a fee and this script lets you to get about the same result for free. However, the USPTO site clearly states that all patent data is in the public domain and may be used freely for any purpose so there is no reason why this should be illegal. Compare this with the conditions of use for Delphion's patent database (formerly IBM patent server) which do not allow pulling images from the site. But be nice and avoid overloading the USPTO site with too many automated requests.

Requirements

pat2pdf should run on virtually any unix-like system. It is reported to run on Linux, FreeBSD, IRIX and Mac OS/X. It also works just fine under Windows if you have Cygwin installed.

Changelog

0.7 - first public release

0.8 - improved error checking

0.9 - fixed error on bash 2.x

0.95 - support for non-utility patents (from Juhana Sadeharju)

0.97 - fix comma-separated numbers

1.00 - DNS name is gone, use numeric IP address instead (by Richard Ellis)

1.01 - Compatible with more Unixes: no longer uses -c switch to head(1) and /dev/stdin (thanks to Roger Cole)

1.02 - No longer depends on /dev/stderr, workaround for ".pdf.pdf" extension, fix to verify() function (thanks to Daniel Webb)

1.03 - Use new DNS name of search site

1.10 - Changed to work with new layout of USPTO site. Thanks to Thomas Bøge!

1.20 - Supports applications as well as issued patents. Thanks to Thomas Bøge again!

1.21 - Script will run regardless of where bash is installed, as long as it's in the search path.

1.22 - Fix USPTO image server name for patent applications.

1.23 - Fixed to work with changed USPTO site. Thanks to Mark Kahrs!.

1.30 - No longer requires GhostScript, only libtiff

1.31 - Improved error checking

1.32 - Fixed to work with changed USPTO site. Thanks again to Mark Kahrs!.

Download

Download the script here. Don't forget to chmod +x it.

Contact

If you have any questions, comments or improvements email Oren Tirosh oren@>nospam<hishome.net.