Is it possible to write a plugin that redacts portions of a pdf based upon user permissions stored in a centralized repository?

[chrisf1386106]

Hello,

I want to write an enterprise digital rights management software for my organization; it's fairly simple and I'm trying to do a cost-benefit-analysis of writing one vs. buying one such as the LifeCycle Enterprise Suite.  I want to define what sections within a pdf are readable for groups within an organizations.  All users within the organization that open the pdf will only be able to see what they are authorized to see with the remaining redacted (and encrypted).  It's not even clear to me something like the LifeCycle Enterprise Suite can do this; does the sdk theoretically provide this capability?

Thanks!

[Test Screen Name]

I am guessing you want to redact at a central location (server) prior to delivery to the end user. The SDK cannot be used in this case because (a) the SDK automates Acrobat (b) Acrobat is not for server use.

The Adobe PDF Library might be appropriate, but I am not sure that it includes redaction functionality. (On no account try to create one with editing tools, this is doomed to catastrophic disclosures!)

If Adobe have nothing, there exist third party redaction enterprise servers.

[chrisf1386106]

Thanks for the response - what about an approach where the whole document is encrypted, and upon opening the document the plugin checks a central repository for users' rights and unencrypts appropriately.  This is very similar to how I understand Adobe's silver partner's FileOpen product suite (Cloud-based Document Rights Management | Encryption & Authentication | FileOpen)  can be set up, except they only allows for encryption at the page level, making me wonder if there was some fundamental limitation within the SDK that prevents this capability at a more fine grained level (paragraph, sentence, etc.)?

Chris

[Bernd Alheit]

This is not a limitation of the SDK. May be a limitation of the PDF Specification.

[Test Screen Name]

The lowest level encryption could apply is a stream. Which might represent everything on the page, or not. But it is not context free: the positions etc. depend on other resources which might not be yet decrypted. Can't see that working, though it's an appealing idea.