- From: Julian Reschke <julian.reschke@gmx.de>
- Date: Tue, 4 Mar 2003 16:42:02 +0100
- To: "Clemm, Geoff" <gclemm@rational.com>, "WebDAV" <w3c-dist-auth@w3.org>
> From: w3c-dist-auth-request@w3.org > [mailto:w3c-dist-auth-request@w3.org]On Behalf Of Clemm, Geoff > Sent: Tuesday, March 04, 2003 4:11 PM > To: WebDAV > Subject: RE: Move and Delete (was: bind draft issues) > > > > I believe it is simpler for a client if the DELETE model stays > consistent (i.e. doesn't change when another binding is added > to a member of the collection). Well, just because it's simpler doesn't necessarily make it an acceptable model (acceptable to everybody, that is). > Also note that it is actually quite complex to state under what > conditions it is OK to partially delete nested bindings; namely, > "when a collection is deleted, a nested binding can be deleted > from the collection only if there is exactly one binding to any > resource on the path from the collection to that nested binding". I don't think we want to say that. Some servers will want to implement DELETE as simply removing the binding, others will want to ensure that they can also cleanup all members (and fail if they can't). I think both models make sense, and that the presence of support for an explicit BIND call doesn't really indicate that the former model always applies. It certainly doesn't for the Unix file system (rmdir() is only allowed for empty directories) and our system. Julian -- <green/>bytes GmbH -- http://www.greenbytes.de -- tel:+492512807760
Received on Tuesday, 4 March 2003 10:42:37 UTC