PDF Labs


How to Collate Even and Odd Scanned PDF Pages

Shuffle the even and odd pages you get from scanning a two-sided document using PDFtk Server.

You scanned a two-sided document by first scanning all the odd pages and then scanning all the even pages. Now you have two PDFs: one of odd pages and one of even pages. And the even pages are in reverse order! How can you collate these pages into a single, right-reading document? These instructions will show you how to do this with our command-line tool PDFtk Server.

PDFtk Server is free (as in freedom) software that you can use to manipulate PDF documents. You can download free installers for Windows and Mac OS X from our site. Most Linux distributions maintain a "pdftk" binary package you can install.

PDFtk has a special feature that we added specifically to solve this problem of arranging scanned pages: shuffle. Say you have two PDFs: even_pages.pdf and odd_pages.pdf. Then you can collate them into a single document like this:

pdftk A=odd_pages.pdf B=even_pages.pdf shuffle A B output collated_pages.pdf

If your even pages are in reverse order, you can reverse its page range:

pdftk A=odd_pages.pdf B=even_pages.pdf shuffle A Bend-1 output collated_pages.pdf

The shuffle feature works by taking one page at a time from each of the input page ranges and assembling them into a new PDF. You specify these ranges after the shuffle keyword, and you can have more than two ranges. Our PDFtk Server manual explains how to describe page ranges — they are very flexible.

When using the Windows command-prompt, it helps to use drag-and-drop from the file manager: drag the input PDF file from the file manager onto the command-prompt, and its full pathname will appear at the prompt.

Shuffling can be used for other jobs, too.

Say you have a 1500-page PDF print job and you need to place a single-page PDF on the ‘back’ of each of these pages so that the final PDF print job is 3000 pages. First, turn the single-page ‘back’ into a 1500-page PDF. The following example isn’t the most elegant, but it illustrates the method:

pdftk A=back_page.pdf cat A A A A A A A A A A output back_x10.pdf

pdftk A=back_x10.pdf cat A A A A A A A A A A output back_x100.pdf

pdftk A=back_x100.pdf cat A A A A A output back_x500.pdf

pdftk A=back_x500.pdf cat A A A output back_x1500.pdf

Next, collate the print job ‘fronts’ with the ‘backs’:

pdftk A=fronts.pdf B=back_x1500.pdf shuffle A B output print_job.pdf

If one range runs out of pages, shuffle continues collating with the remaining ranges. So in the examples above you probably want to be sure that the input ranges all have the same number of pages.

Finally, sometimes you want to go the other direction and extract the even or odd pages from a PDF. You can combine them again using shuffle, later. You might do this if you wanted to apply a stamp to just the even pages of a PDF, for example.

Extract the even pages:

pdftk A=full_doc.pdf cat Aeven output even_pages.pdf

Extract the odd pages:

pdftk A=full_doc.pdf cat Aodd output odd_pages.pdf


Article Author: Sid Steward


About PDF Labs

Our mission is to make PDF easier to use.

PDF Labs is operated by Sid Steward, author of PDF Hacks (O’Reilly) and the popular PDF Toolkit.

Please contact Sid Steward by email. Include PDF Labs in the subject to ensure prompt delivery:


About PDF Hacks

PDF Hacks Cover

"If you like pdftk, then you’ll appreciate the other 25 programs and scripts I created for my book, PDF Hacks." –Sid


About PDFtk

For ten years, power users have relied on PDFtk for server-side processing of PDF documents. Our command-line PDFtk has been — and always will be — free (as in freedom) software.

Today we also offer a convenient graphical application: PDFtk Pro. Your purchase of PDFtk Pro will help fund the continued development of our GPL PDFtk.

Thank you for your support!