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

RE: locate-by-history report vs. Depth header

From: Clemm, Geoff <gclemm@rational.com>
Date: Fri, 8 Mar 2002 08:06:57 -0500
Message-ID: <3906C56A7BD1F54593344C05BD1374B1060EE625@SUS-MA1IT01>
To: "Ietf-Dav-Versioning@W3. Org" <ietf-dav-versioning@w3.org>
   From: Julian Reschke [mailto:julian.reschke@greenbytes.de]

   So: depth = 0: the REPORT affects just the request URI, but as the
   report itself is defined to have "colllection member scope", this
   will return all direct members of the collection at the request URI
   having matching version histories?

No, the REPORT affects (is applied to) the configuration rooted at a URL
which includes *all* members of the collection identified
by that URL not just "direct" (what RFC 2518 calls "internal") members.

   depth = 1: also includes collections that are members of the
   collection at the request URI.

Yes, if by "includes", you mean "repeats the REPORT on".  The Depth:0
report "includes" (in the sense of "looks at") all members of the
collection, but it only runs the report once.

Cheers,
Geoff

   > From: ietf-dav-versioning-request@w3.org
   > 
   > The locate-by-history report logically runs against the entire
   > configuration rooted at the request-URL.  So in that sense, it is
   > a "depth:infinity" operation.  But that isn't how REPORT is
   > defined.  REPORT is defined as "being run separately against
   > every resource that satisfies the depth parameter".  Depth:1 says
   > to "run the request against the resource identified by the
   > request-URL, and then again on each of the internal members of
   > the request-URL."  But that is never what you want, because you
   > only want the REPORT run once against the configuration
   > identified by the request-URL.
   > 
   > And that one case where I said that Depth:1 might make sense was
   > just wrong.  You still run Depth:0 (i.e. just run the REPORT
   > once), and the REPORT will find all VCR's that match that version
   > history in any of the workspaces.

   > From: Julian Reschke [mailto:julian.reschke@greenbytes.de]
   > 
   > > From: ietf-dav-versioning-request@w3.org
   > > [mailto:ietf-dav-versioning-request@w3.org]On Behalf Of Clemm, Geoff
   > > 
   > > Actually, the report clearly makes the most sense for depth=0
   > > (perhaps that's what you meant to type?).
   > 
   > You managed to confuse me :-) It the request URI is the parent
   > collection, and the scope is the member of this collection, only
   > depth = 1 seems to make sense. Depth 0's scope would be just the
   > parent collection, correct?
Received on Friday, 8 March 2002 08:07:29 GMT

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