Re: DRM PDFs (and SED and ETD ...)

[Adam Hardy <adam.ant@cyberspaceroad.com>]

Johannes Wiedersich on 29/05/08 09:02, wrote:
> On 2008-05-28 20:36, Adam Hardy wrote:
> > Just thought I'd ask, even though my web searches don't reveal anything
> > but Adobe promises and work-arounds involving printing to postscript on
> > a windows box first - but is there software for linux to read DRM pdfs?
>
> I don't know, if I understand you correctly. If you just want to read a
> DRM pdf, there are various linux tools as suggested. I've had good luck
> with kpdf so far, but xpdf and others might work as well. Those tools
> actually ignore (or can be configured to ignore) most of the restrictions.

I haven't found any configuration on any package (xpdf, acroread, evince, pdftk, 
kpdf) for ignoring security. In fact you're the second person to say that, but 
the man pages don't contain any reference to overriding passwords. Are you sure?


> If just the DRM is so 'enhanced' that it won't work with those, there is
> also Adobe's acroread. 

That's the first package I tried - but Adobe haven't upgraded it yet to 
understand security, so it balks at loading the file at all.


> If a pdf is password protected, however, you still require the password.
> As mentioned, there are password crackers around, but those might
> fail/take a very long time, if the password is long and complex.

Tried pdfcrack, haven't found any others.

After looking at the website providing the pdf that I am trying to read, I 
realise I must be wrong in my assumption that this pdf file has a password - I 
just assumed it would. It's a test pdf provided by the British Library to let 
readers check that their set-up can handle the secure electronic delivery.

So the pdf must be written with instructions that link back to the British 
Library to control printing - they let you print once and once only (and you 
must be online of course for the security to verify with their website).

However evince and kpdf ask me for a password, and pdfcrack says it can't deal 
with this version.

So the British Library gives the reader the following ETD (xml), rather than the 
PDF,

http://edd.bl.uk/fulfill/ebx.etd?action=free&ordersource=live&bookid=DOI:SEDSuccess&gbdetect=true


and then Adobe Acrobat (or the new version, Digital Editions) grabs the PDF:

http://w2k-edd2.bl.uk/bookbytes/ebx_DOI_SEDSuccess.pdf

and checks operations against some security at the British Library.

There's an attribute in the ETD xml called <nonce> with what looks like an 
encrypted value - that is presumably a key for the server, not the PDF password.

Here's their advice (with typical linux fob-off):

http://www.bl.uk/reshelp/atyourdesk/docsupply/productsservices/sed/sedfaq/index.html