[OAI-implementers] resumptionToken and metadataPrefix

Tim Brody tim@tim.brody.btinternet.co.uk
Fri, 24 May 2002 16:57:56 +0100


ResumptionToken is *exclusive* - an OAI-compliant harvester must not supply
a metadataPrefix with a resumptionToken.

(As has been mentioned in the past by Herbert, if a prefix could be
supplied, the query could be changed, therefore the result set would be

If a repository requires a metadataPrefix it isn't OAI-compliant, so either
contact the repository administrator to fix it, or you'll need to code an
explicit "work-around" for broken implementations.

All the best,

----- Original Message -----
From: "David Smith" <dasmith@perseus.tufts.edu>
To: <oai-implementers@oaisrv.nsdl.cornell.edu>
Sent: Friday, May 24, 2002 4:42 PM
Subject: [OAI-implementers] resumptionToken and metadataPrefix

> I apologize if this has been addressed before; chapter and verse in the
> archives would be welcome.
> I am confused by the interaction of resumptionToken and metadataPrefix
> on a ListRecords request. Some data providers will not respond properly
> to the resumptionToken unless a metadataPrefix is supplied. Others
> appear to object to including metadataPrefix.
> I cannot find anything about this in either version of the protocol,
> but, again, I could be overlooking something obvious. Although a prudent
> harvester would try both options, I would be interested in what seems
> best practice for data providers.
> Thanks
> David
> _______________________________________________
> OAI-implementers mailing list
> OAI-implementers@oaisrv.nsdl.cornell.edu
> http://oaisrv.nsdl.cornell.edu/mailman/listinfo/oai-implementers