On 07/22/2010 03:01 PM, Daniel Shahaf wrote:
> C. Michael Pilato wrote on Thu, Jul 22, 2010 at 14:53:51 -0400:
>> On 07/22/2010 02:12 PM, C. Michael Pilato wrote:
>>> Hrm... 'svn relocate' just calls svn_client_relocate() as well. I don't
>>> see any calls to svn_wc_is_wc_rootX() or anything. Does that allow partial
>>> relocations, too?
>>
>> To answer my own question: Yes. 'svn relocate', when run on a working copy
>> subtree, will only relocate that subtree.
>>
>> Am I wrong to think that this is ... odd?
>>
>
> The way it is is consistent with every other subcommand.
Hey, nobody in this community is more vocal proponent of consistency than I
am. :-) But I do wonder about the utility of this particular sort of
consistency. This isn't just a normal switch. This is a relocation. In
the context of Subversion 1.7-era functionality, what does it even mean to
have a single working copy whose citizens have different repository root
URLs? I know there's a vision of supporting that situation someday (much as
CVS did with its working copy integration feature), but I'm not aware of
anyone specifically coding for it today.
--
C. Michael Pilato <cmpilato_at_collab.net>
CollabNet <> www.collab.net <> Distributed Development On Demand
Received on 2010-07-22 21:38:58 CEST