Rather than a new method, an ADDMEMBER specification could require support for the Expect mechanism which solves that problem rather neatly for any method. Lisa On Feb 17, 2005, at 1:04 PM, Jim Whitehead wrote: > > A few comments on extending the functionality defined in -addmember-00: > > 1) One problem clients currently encounter with PUT is determining > whether > they have permission to write to a specific location. At present, a > client > must submit a PUT (or ADDMEMBER) with the entire entity body, then > wait for > the response, or the authentication challenge. It would be very nice > to have > a TESTPUT/TESTADDMEMBER method that sees whether a write (via PUT or > ADDMEMBER) would succeed, or if an authentication challenge is needed. > > Yes, a server supporting ACLs could be queries to see if writing were > allowed, but it still doesn't provide an easy way to pre-determine the > authentication information. > > 2) It would be very helpful to allow ADDMEMBER to have atomic > PUT+PROPPATCH > capabilities, to write both the resource and its metadata in one > action. > > - Jim > >Received on Thursday, 17 February 2005 21:44:24 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 12 October 2007 17:53:22 GMT