David Blaikie wrote:
Hi, I'm doing a much less ambitious "skim" (i.e., checking out code
from VSS and adding directly to SVN) migration of our VSS repository.
I think that in order to migrate active branches and labels from VSS,
I should check out each branch separately and then add it directly to
the branches directory in SVN.
My approach would be this:
1. create a new project directory with branches/ tags/ and trunk/
directories
2. check out latest version of code from VSS
3. add checked-out tree into trunk/ in SVN
4. individually check out branches from VSS
5. add each as a separate branch in branches/ directory in SVN
6. individually check out labels from VSS
7. add each as a separate tag in tags/ directory in SVN
does this sound correct? Would this allow developers to merge
branches or tags with the trunk at a future point in time?
Hello David,
Unfortunately this approach will lead to a lot of wasted/duplicated
space in your repository. Ordinarily Subversion creates branches and
tags using its "cheap copies" so that only changes in a branch take up
space. But if you manually check these files in several times as
suggested, you won't be creating cheap copies but rather full duplicate
copies of all files multiple times.
Whether that option is still desirable would be a judgment call on your
part; if your source tree isn't very large it may still be helpful but
you won't be able to perform the ordinary commands available with copied
files such as diff against the original and merging.
So I'd probably recommend against that option and instead finish out any
existing branches using VSS then create new ones in SVN after you migrate.
toby
_______________________________________________
vss2svn-users mailing list
Project homepage:
http://www.pumacode.org/projects/vss2svn/
Subscribe/Unsubscribe/Admin:
http://lists.pumacode.org/mailman/listinfo/vss2svn-users-lists.pumacode.org
Mailing list web interface (with searchable archives):
http://dir.gmane.org/gmane.comp.version-control.subversion.vss2svn.user
