W3C home > Mailing lists > Public > ietf-dav-versioning@w3.org > July to September 2002

RE: Marshalling Depth > 0 responses for REPORTs, WAS: Replacing t he Label header with a DAV:labeled-version report

From: Julian Reschke <julian.reschke@greenbytes.de>
Date: Fri, 16 Aug 2002 11:36:43 +0200
To: "Clemm, Geoff" <gclemm@rational.com>, <ietf-dav-versioning@w3.org>
Message-ID: <JIEGINCHMLABHJBIGKBCOEHKFCAA.julian.reschke@greenbytes.de>

Geoff,

another (hopefully, last one) question.

Assume that the collection at the root of my WebDAV namespace happens to be
versioned. How do I use locate-by-history to find it?

Shouldn't this

"The response body for a successful request MUST be a DAV:multistatus XML
element containing every version-controlled resource that is a member of the
collection identified by the request-URL,..."

be

"The response body for a successful request MUST be a DAV:multistatus XML
element containing every version-controlled resource that is  a member of
the collection identified by the request-URL or the collection itself,...."

instead?


> -----Original Message-----
> From: ietf-dav-versioning-request@w3.org
> [mailto:ietf-dav-versioning-request@w3.org]On Behalf Of Clemm, Geoff
> Sent: Friday, August 16, 2002 12:05 AM
> To: ietf-dav-versioning@w3.org
> Subject: RE: Marshalling Depth > 0 responses for REPORTs, WAS: Replacing
> t he Label header with a DAV:labeled-version report
>
>
>
> The DAV:locate-by-history report finds any collection member
> for that history, not just internal members.  In particular,
> the example in 5.4.1 shows a Depth:0 (the default) request which
> finds the /ws/public/x/test.html member of the /ws/public collection.
>
> So a DAV:locate-by-history report only makes sense with the
> default Depth header value of "0".
>
> What report are you thinking of that makes sense with a non-zero
> depth, but which can only be applied to collections?
>
> Cheers,
> Geoff
>
> -----Original Message-----
> From: Julian Reschke [mailto:julian.reschke@greenbytes.de]
> Sent: Thursday, August 15, 2002 10:35 AM
> To: Clemm, Geoff; ietf-dav-versioning@w3.org
> Subject: RE: Marshalling Depth > 0 responses for REPORTs, WAS: Replacing
> t he Label header with a DAV:labeled-version report
>
>
> > From: ietf-dav-versioning-request@w3.org
> > [mailto:ietf-dav-versioning-request@w3.org]On Behalf Of Clemm, Geoff
> > Sent: Thursday, August 15, 2002 3:20 PM
> > To: ietf-dav-versioning@w3.org
> > Subject: RE: Marshalling Depth > 0 responses for REPORTs, WAS: Replacing
> > t he Label header with a DAV:labeled-version report
> >
> >
> >
> > A Depth:infinity DAV:locate-by-history request makes no sense,
> > so having it return something ugly (i.e. a 403 for every non-collection
> > member) is not a problem.
>
> Well.
>
> It does make a lot of sense if you don't know which collection the VCR is
> in -- this is why we need it.
>
> So the use case is:
>
> - I have a VHR
> - I need to lookup VCRs on a server for which the
> DAV:version-history points
> to my VHR
>
> How am I supposed to do that if not using depth infinity?
>
> > On the other hand, a Depth:infinity DAV:labeled-version request does
> > make sense, and it is reasonable for it to return a 403 for every
> > non-version-controlled member of the collection.
> >
> > In general, when a request asks you to apply a report to every member
> > of a collection, it seems reasonable to indicate the result (possibly,
> > an error) for each member of that collection.
>
> Which basically renders all REPORTs that apply to collections (are there
> others?) useless if you depth = 0 or depth = 1 isn't enough.
>
>
>
Received on Friday, 16 August 2002 05:37:15 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 13:57:43 GMT