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

RE: Versioning collections question

From: Tim Ellison <Tim_Ellison@uk.ibm.com>
Date: Wed, 27 Jun 2001 09:48:40 +0100
To: "Clemm, Geoff" <gclemm@rational.com>
Cc: DeltaV <ietf-dav-versioning@w3.org>
Message-ID: <OF2F1B34D4.538DD323-ON80256A78.002FAFB5@portsmouth.uk.ibm.com>
"Clemm, Geoff" <gclemm@rational.com> wrote:
> A couple of small clarification/corrections:
>
>    From: Tim Ellison [mailto:Tim_Ellison@uk.ibm.com]
>
>    By checking out a collection you make the collection
>    resource (including its internal members == "bindings")
>    mutable.  This means to rename /a/b to /a/c then the
>    collection /a/ must be mutable, but /a/b and /a/c can
>    be immutable versions.
>
> Tim was using "version" in an informal way here.  /a/b and
> /a/c are checked-in (and therefore immutable) version-controlled
> resources, but they aren't versions in the formal DeltaV sense
> of the word.

Oops, that just slipped out, change "immutable vresions" to "checked-in
version-controlled resources".

> >    So if you PROPFIND depth 1 on a version of a collection
> > you will discover the internal members are bound to version
> > history resources.
>
> Actually, a collection version is not itself a collection
> (it's state is defined by its DAV:version-controlled-binding-set).
> We did at one time have the internal members of a collection
> version bindings to version-history resources, but that
> information was moved to the DAV:version-controlled-binding
> set property. So you'd just ask for a PROPFIND;Depth=0 for
> the DAV:version-controlled-binding-set.

Goofed again, thanks for correcting that.  An instance where I should use
my eyes rather than relying on my memory<g>

Tim
Received on Wednesday, 27 June 2001 06:22:16 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 13:57:41 GMT