- From: Julian Reschke <julian.reschke@gmx.de>
- Date: Thu, 16 Dec 2004 16:25:26 +0100
- To: Bernard Desruisseaux <bernard.desruisseaux@oracle.com>
- CC: www-webdav-dasl@w3.org, ietf-dav-versioning@w3.org, w3c-dist-auth@w3.org
Bernard Desruisseaux wrote: > [ For those not subscribed to www-webdav-dasl, I've copied > the beginning of this thread at the end of this message. ] > > Julian Reschke wrote: > >> Bernard Desruisseaux wrote: >> >>> The declaration of the DAV:multistatus element in RFC 2518 >>> (section 12.9) specifies that at least one DAV:response >>> element must appear in a DAV:multistatus element. >>> >>> Is that an issue? >> >> >> >> It's an issue with the original definition, which has (or should have >> been) updated in RFC3253 and RFC2518bis. >> >> >> Julian >> > > It's "should have been". :-) > > RFC3253 simply makes reference to RFC 2518, Section 12.9 > for the definition of multistatus. Indeed. This probably is OK because all REPORTs defined in RFC3253 indeed return at a minimum of one result. The new REPORTs on RFC3744 (ACL) may return empty sets and explicitly say so. > RFC2518bis (-06) still declare DAV:multistatus as follow: > > <!ELEMENT multistatus (response+, responsedescription?) > > > Do you want to open an issue for each document, and perhaps > clarify in the search draft (although RFC2518bis should end > up being published before search) ? The DTD fragments aren't normative anyway (thus RFC3744 doesn't attempt to modify them). It's arguable whether RFC2518 needs to be fixed. In a perfect world, RFC2518bis would be done soon and include the base definition for REPORT, but somehow I fear it won't happen in the foreseeable future. Julian -- <green/>bytes GmbH -- http://www.greenbytes.de -- tel:+492512807760
Received on Thursday, 16 December 2004 15:26:00 UTC