[plug] Creating multiple-page PDFs from scanned images [SOLVED]

[Gavin Chester]

On Thu, 2007-01-18 at 17:19 +0900, Alex Nordstrom wrote:
> A while ago, I posed a question about scanning multiple-page documents 
> into PDF documents. For the completeness of the archives, here's what I 
> came up with, in the hope that someone might find it useful.
> 
> I was having problems using convert from ImageMagick to convert images 
> to PDF. Upgrading ImageMagick, getting another 512 megs of RAM, using 
> input already converted to PDF documents, using certain convert 
> options, or a combination of these factors seems to have solved the 
> problem.
> 
> I hacked up the attached Perl script to perform the batch scanning. It 
> takes a single optional argument: the directory in which to place the 
> scanned images, named 001.pdf, 002.pdf, 003.pdf and so on in the 
> directory given. If no directory is provided, the current working 
> directory is used.

I hate to sound like I'm 'raining on your parade', but I did just this
sort of operation again today in a way that I find far easier than you
describe.  I thought I had given my solution to this problem the first
time you posted, but I'm too lazy/busy to verify that in the archives.
Anyway, you may prefer your solution to the one I proffered ;-)

The way I found to do it requires only two tools and no scripts: xsane
and ghostscript.  How it works is Xsane will scan multiple pages direct
to pdf, either with flat bed or automatic document feeder. It will also
automatically increment the pages as xx01.pdf, xx02.pdf, etc.  Then you
fire up a terminal and enter a ghostscript command to stitch all the
pdfs together into one file (xsane is supposed to do this itself, but I
have not got it to work).  The url telling how to join multiple pdfs is
here:
<http://software.newsforge.com/article.pl?sid=04/06/09/1844259>

I'm sure I wrote all this the first time you posted, so maybe you just
don't like the sound of that solution ;-) 

Gavin