Moving the discussion to the XMPP working group mailing list since this is RFC-related.
It looks like Google's docs for this are here: http://code.google.com/apis/talk/jep_extensions/jid_domain_change.html Should the client just use this bare JID the next time it logs in? If so, we may need to make a change to 3920bis to make this clear. If we're contemplating making a change in -bis, we should make the correct one, not just loosen up the schema. Of course, in your implementation, there's nothing that says you can't use any schema you like to do validation, since validation is not required and the schemas are non-normative. -- Joe Hildebrand From: [email protected] [mailto:[email protected]] On Behalf Of Mittal Thakkar Sent: Tuesday, June 23, 2009 12:43 AM To: [email protected] Subject: [Standards] Modifying the schema for auth(RFC 3920) Hi, The clients using the libpurple 2.6.x like Adium and Pidgin sends the following stanza for auth : <auth xmlns='urn:ietf:params:xml:ns:xmpp-sasl' mechanism='DIGEST-MD5' xmlns:ga='http://www.google.com/talk/protocol/auth' ga:client-uses-full-bind-result='true'/>. At our server xml parsing fails as in the schema for auth it supports only one attribute ie. mechanism. wrt. Appendix C.4 of RFC 3920. Is it valid if we allow any attribute( of other namespace ) for the <auth/> as RFC is silent about it. The schema we want to use is as follows: <xs:element name='auth'> <xs:complexType> <xs:simpleContent> <xs:extension base='xs:string'> <xs:attribute name='mechanism' type='xs:string' use='optional'/> <xs:anyAttribute namespace='##other' processContents='lax'/> </xs:extension> </xs:simpleContent> </xs:complexType> </xs:element> -- Thanks, Regards, Mittal Thakkar
