W3C home > Mailing lists > Public > ietf-dav-versioning@w3.org > July to September 2001

RE: List of areas in the specification where properties are empty /not set

From: Clemm, Geoff <gclemm@rational.com>
Date: Wed, 15 Aug 2001 16:45:05 -0400
Message-ID: <3906C56A7BD1F54593344C05BD1374B103F8AB46@SUS-MA1IT01>
To: ietf-dav-versioning@w3.org
   From: Peter Raymond [mailto:Peter.Raymond@merant.com]

   Section 3.9 in the additional postconditions (DAV:auto-checkout)
   for PUT, the spec says "the DAV:checked-in property MUST be
   empty". I propose it should instead read "the DAV:checked-in
   property MUST be removed".

Done.

   Section 4.4 in the additional postconditions
   (DAV:cancel-checked-out) for UNCHECKOUT, the spec says "and the
   DAV:checked-out property no longer is set".  I propose it should
   read "and the DAV:checked-out property is removed".

Done.

   Section 11.2 the description of the MERGE method says "a
   version-controlled resource with a non-empty DAV:merge-set". I
   propose it should read "a version-controlled resource with a
   DAV:merge-set property".
   Section 11.6 in the additional preconditions
   (DAV:merge-must-be-complete) for CHECKIN, the spec says "The
   DAV:merge-set and DAV:auto-merge-set of the checked-out resource
   MUST be empty".  I propose it should read "The checked-out resource
   MUST not have the DAV:merge-set or DAV:auto-merge-set properties".
   Somewhere in the spec perhaps we should state that when a property
   is removed from a resource it is no longer returned by PROPFIND or
   in any <DAV:prop> response from any method.

A client might resolve one merge at a time, which means it
will be successively PROPPATCH'ing shorter strings into the
DAV:merge-set property.  I'd like to allow it to PROPPATCH
the empty string into the DAV:merge-set, and have it mean
the same thing as removing the DAV:merge-set.

Cheers,
Geoff
Received on Wednesday, 15 August 2001 16:36:10 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 13:57:42 GMT