Review Request: VCS::checkout()/import() Replace the VcsMapping argument with a pair of arguments VcsLocation and KUrl

[Matthew Woehlke]

Andreas Pakulat wrote:
> On 14.05.09 18:30:12, Matthew Woehlke wrote:
>> Fabian Wiesel wrote:
>>> Does a non-recursive import make sense? I simply adopted the feature from VcsMapping.
>> How was this addressed?
> 
> Use the source Matthew ;)

Hmm... the source contains (minor) errors in the documentation :-). I'd 
fix it, but I feel like the wording on the sourceRepository parameter 
should be improved at the same time, and nothing comes to mind 
immediately how to do that. There is still 'therefor' (technically a 
word but likely incorrect*) and mention of 'checkout()' (and the 
trailing space isn't really needed, either :-) ).

(* ...and I'm not sure if it's best just changed to 'therefore' or 
replaced with a different word, e.g. 'so' or 'thus', or perhaps 
something else entirely. Thoughts?)

Okay, so createWorkingCopy still has a recursion parameter.

Oh, and there are references to "a mapping" that could probably be altered.

>> While it's not typical, AFAIK svn supports  non-recursive checkouts
>> (or at least updates).
> 
> Yes, checkouts and updates, but not import. Import means taking a local
> directory in putting it into the VCS - the first time. Yes, svn has a
> non-recursive (or in newer versions a --depth argument).

Ah. It's been too long.

> IMHO a non-recursive import is too much a corner case to warrant an
> entry in the interfaces, if a VCS supports that and wants to offer it it
> can provide its own vcs-specific interface.

Makes sense.

-- 
Matthew
Please do not quote my e-mail address unobfuscated in message bodies.
-- 
"Please remain calm... I may be mad, but I am a professional."
   -- Mad Scientist (actual author unknown)