Re: svn copy onto itself is possible

Lübbe Onken <L.Onken@rac.de> writes:

> Maybe I'm misunderstanding something here, but I think SVNs behaviour is
> dangerous:

How is it dangerous? It behaves the same as unix 'cp'. There are no
shocking surprises here.

cp path/A path/B --> if path/B doesn't exist, creates path/B
                      if path/B does exist, creates path/B/A

> 1) First Copy
> svn copy http://myserver/myrepos/trunk/ http://myserver/myrepos/tags/Tag_01
> creates http://myserver/myrepos/Tag_01, when http://myserver/myrepos/Tag_01
> does not already exist.
> This is what I expect to happen.
>
> 2) Second Copy
> svn copy http://myserver/myrepos/trunk/ http://myserver/myrepos/tags/Tag_01
> creates http://myserver/myrepos/Tag_01/trunk, when
> http://myserver/myrepos/Tag_01 already exists.
> This is where I'd expect a warning, that the destination already exist,
> telling you where the copy would really end up and allowing you to cancel or
> continue the operation.

So you're saying that Unix 'cp' is equally dangerous and misleading?

That if I run 'cp path/A path/B', and it would create 'path/B/A', Unix
should be giving me warnings?

I don't agree.

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Wed Oct 1 15:41:27 2003