On Tue, Jan 8, 2013 at 1:14 PM, Stefan Sperling <stsp_at_apache.org> wrote:
> On Tue, Jan 08, 2013 at 01:08:04PM +0100, Johan Corveleyn wrote:
>> -0 for now, but maybe I don't understand. Are you guys talking about:
>> 1) 'svn mv wc1/A wc2' will error out, and user will have to figure
>> out that he needs to execute 'svn copy; svn delete'
>> 2) 'svn mv wc1/A wc2' will do the same as 1.7 did, under the hood.
>> The difference being that option 1 is a regression for an existing use
>> case, and option 2 is actually not a regression (the cross-wc move
>> just doesn't get the "move-tracking" enhancement that is new in 1.8).
> We're talking about option 2.
Ok, phew :-).
Then let's say +0. I'd like it even better if this could be handled
correctly with move-tracking, but I can't help in any meaningful way
myself with that (and I rather agree that we shouldn't postpone
releasing for too long, and keep nice things from our users), so I'll
thankfully accept any working solution you guys (anyone) can implement
in a reasonable timeframe :-).
Received on 2013-01-08 13:23:37 CET