Recently, the drunkenpacket had the need to unburden himself of a large supply of paper marked with various interesting (but obviously obsolete) information storage squiggles. Having serenditiptious access to a XEROX with a bulk sheet scanner attachment, the packet could not resist the obviously technical temptation: scan the informative squiggles (we think this is called printing or writing). Like all dp side projects, this one seemed to immediately explode with dependencies. Read on for all of the gory details…

Single Sided Scanning

First off, we soon noted that our paper was frequently squiggled on both sides. But our XEROX sheet scanner would only scan one side off each paper sheet as it drew down from the stack of paper in the sheet feeder.

The solution was simple: scan the document going one way, giving a page sequence 1, 3, 5, …, 2(n-1) - 1, 2n-1. etc. Then, because dp dislikes paper cuts and didn’t feel like inverting the order of the pages, we simply scanned the pages in reverse order from the end of the document so that the even pages have the order 2n, 2(n-1), …, 4, 2.

Time passed. Jams were handled. The air reeked of toner because the XEROX insists on printing scan status sheets. (We’re scanning to reduce paper right? So, let’s get a printed page summarizing the result of doing a scan. Right? And they call it The Document Company.)

Back at the Ranch

Meanwhile, back at the ranch (that would be the workstation actually) we find our mailspool jam-packed with documents. The XEROX has emailed us a PDF for each scan. Glad the mail server could handle those multiple 12MB PDF files. I’m sure the sysadmins are really loving dp about now. (Given the price of disk, why isn’t there a NFS/CIFS server on the XEROX?)

As for what’s in the email: one PDF/email, each one containing pages as discussed in the single sided section above. All with the same name. Please oh XEROX usability people, think about what one might want to do with the scanned documents: perhaps save them all into a directory? All at once. without re-naming each one of them.

Assembly

So, after saving each PDF, one at a time, and tediously renaming them, we have two PDFs: the odd pages in ascending order and the evens in descending order. We here at drunkenpacket assumed that merging these together would be easy.

We were wrong.

Preview doesn’t provide any convenient interface for merging, rearranging, splitting or otherwise organizing a giant PDF file of image scans.