RE: Issues/questions regarding sections 3, 4 and 5...

Good.

I already forbid deleting the only version, though I didn't know the
spec actually said I had too.

That didn't prevent me from having someone try to delete the only
version and log a bug because they couldn't. :)


> -----Original Message-----
> From: ietf-dav-versioning-request@w3.org 
> [mailto:ietf-dav-versioning-request@w3.org] On Behalf Of Clemm, Geoff
> Sent: Wednesday, August 22, 2001 12:13 PM
> To: ietf-dav-versioning@w3.org
> Subject: RE: Issues/questions regarding sections 3, 4 and 5...
> 
> 
> Ah, the wonderful ambiguities of English ... (:-).
> 
> This thread was about deleting the "last" version in the 
> sense of "the only version left in the version history".  The 
> result of this would be a version history with no versions in 
> it, which is not allowed by the spec.
> 
> I believe you are referring to the "last" version in the 
> sense of "the one that was created most recently in a linear 
> history". That is totally legal to delete (assuming it is not 
> the only version left in the history), and having the VCR 
> revert to the immediately preceding version is very 
> reasonable (although not required by the protocol).
> 
> Cheers,
> Geoff
> 
> -----Original Message-----
> From: John Hall [mailto:johnhall@evergo.net]
> Sent: Wednesday, August 22, 2001 1:58 PM
> To: 'Clemm, Geoff'; ietf-dav-versioning@w3.org
> Subject: RE: Issues/questions regarding sections 3, 4 and 5...
> 
> 
> 
> 
> >    Reading section 5.6 it took us quite a while to decide how to
> >    delete the last version from a version history.  I think 
> the answer
> >    is "you don't" you must delete the version history 
> itself in order
> >    to delete the last version.  Did we interpret this correctly?  Do
> >    you think we should clarify this in the spec?
> 
> 
> As currently implemented, my server will allow you to delete 
> the last version.  That version goes away and the VCR reverts 
> to the next to last version.
> 
> That solution works because I don't fork.
> 
> But if the consensus is to prohibit this I can prohibit it 
> (except as a side effect of UNCHECKOUT, though).
> 
>

Received on Tuesday, 4 September 2001 12:28:28 UTC