Getting a grip on the UUID

The DT manual says that the UUID “is unique to each item and more than one item can’t have the same UUID. When email messages are imported, they are given such an identifier based on metadata from the actual email. This means you can’t import the same message into a database more than once, as that would result in more than one file having the same UUID.”

Is the UUID also derived from metadata for pdf files and for text capture from web pages?

I ask because through some unfortunate user error I wound up with a lot of documents duplicated across several databases. As I try to clean this up I’m realizing that if I drag 100 selected documents to a new database, the 25 that remain selected, don’t move, and get reported as “UUID already in database” I can safely delete those. Or can’t I?

Thanks for any help!


Is the UUID also derived from metadata for pdf files and for text capture from web pages?


We have had rare reports of this but haven’t been able to reproduce it at will.

  • Are the files from DEVONthink or DEVONthink To Go?
  • Are they in databases syncing with DEVONthink To Go?

No, DT2G can’t be implicated because I haven’t gotten around to DEVONthink to go.

I’m guessing it’s an error I introduced with profligate dragging of tags from one database to another, including an earlier copy of one of the databases—I’m certainly dealing with far more replicants than I’ve ever seen before. But I needed to make sure I had all of the captured research, and so I chose to go for duplication that I could clean up later. This began in DT2 and while I was in the middle of this project along came DT3. Once I realized how much cleaner and more powerful 3 is, I accelerated the process.

Of course, the cleaning has gotten tedious, so was hoping there might some tricks I could use to assist. In the end, I’ll know what’s really in my library, so it’s all good.