Copying documents

[Florent Guillaume <fg@nuxeo.com>]

Hi all,

As David Caruana pointed out on Twitter, there is no API to copy a  
document.
I'm sure it was considered before, so could people knowledgeable  
explain the reason why it was not felt that this primitive was  
necessary?

I understand that in theory copying can be done by the client by  
reading data and then writing it back into a new document, but this is  
horribly inefficient in many cases and as soon as you have big content  
streams.

What are the problems with introducing a copy primitive? I'm sure very  
few repositories don't have it natively, and furthermore if a client  
can do it then the repository's CMIS bindings themselves can do it as  
well, so why put that weight on the client and the network?

BTW copying folders (and their subtree of documents) would be needed  
to IMHO.

Thanks,
Florent

-- 
Florent Guillaume, Head of R&D, Nuxeo
Open Source, Java EE based, Enterprise Content Management (ECM)
http://www.nuxeo.com   http://www.nuxeo.org   +33 1 40 33 79 87