unmapped_lock

[Joe Orton]

On Fri, Oct 10, 2008 at 05:39:00PM +0100, Alexander Chow wrote:
> So, I have been working on the new litmus 0.12 and, especially, this new 
> unmapped_lock test.  I got things to work for LOCK of an unmapped URL and 
> create an empty resource.  However, when it gets to the subsequent UNLOCK 
> call, it fails because a DELETE is made in between.  If the resource is 
> deleted, how can the lock still exist to be unlocked?  This does not seem 
> to be covered in the RFC.
>
> Not sure if there is a hole in my understanding, the RFC, or litmus.   
> Thanks.

4918 section 6.1 says:

     8.  If a request causes the lock-root of any lock to become an
       unmapped URL, then the lock MUST also be deleted by that request.

so in the LOCK/DELETE/UNLOCK sequence, it is indeed correct that the 
UNLOCK should fail.  I've removed the DELETE call for the next release.  
Thanks for the report!

Regards, Joe