You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@subversion.apache.org by Ryan Schmidt <su...@ryandesign.com> on 2006/06/01 10:11:59 UTC

Re: dump | dumpfilter | load -- always renumbers?

On May 31, 2006, at 19:42, Vineet Kumar wrote:

> I'm moving some projects around from one big massive repository into
> multiple smaller repositories, but I want users with checkouts from  
> the
> original repository to not have to do fresh checkouts, and I don't  
> want
> to break existing revision numbers (as referenced in bug reports,  
> etc.)

I don't know the answer to your main question, but I'm pretty sure  
that users will have to check out new working copies; they won't be  
able to use their existing working copies. This is because each  
repository has a unique ID, and the working copy knows it. Even if  
you try to "switch --relocate" the working copy to the new  
repository's URL, the working copy will notice that the ID of the new  
repository doesn't match the ID of the old repository and will bail.

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org

Re: dump | dumpfilter | load -- always renumbers?

Posted by Vineet Kumar <vi...@doorstop.net>.
* Ryan Schmidt (subversion-2006q2@ryandesign.com) [060601 03:12]:
> 
> On May 31, 2006, at 19:42, Vineet Kumar wrote:
> 
> >I'm moving some projects around from one big massive repository into
> >multiple smaller repositories, but I want users with checkouts from  
> >the
> >original repository to not have to do fresh checkouts, and I don't  
> >want
> >to break existing revision numbers (as referenced in bug reports,  
> >etc.)
> 
> I don't know the answer to your main question, but I'm pretty sure  
> that users will have to check out new working copies; they won't be  
> able to use their existing working copies. This is because each  
> repository has a unique ID, and the working copy knows it. Even if  
> you try to "switch --relocate" the working copy to the new  
> repository's URL, the working copy will notice that the ID of the new  
> repository doesn't match the ID of the old repository and will bail.

Sure, what I meant to say was that checkouts of projects that are
remaining in the original repository won't have to be checked out fresh.
I understand that working copies corresponding to projects that are
moved to new repositories will require fresh checkouts regardless, but
in my case, that's a much smaller set of users and so will be less
disruptive.

good times,
Vineet
-- 
http://www.doorstop.net/
-- 
"Computer Science is no more about computers
than astronomy is about telescopes."  -- E.W. Dijkstra