James wrote:
> Andrew Reedick wrote, without attribution:
> > In theory, not that I've tried it recently, and with no guarantee of
> > success, reliability, or data integrity, you could:
> > rename the dir to dir.oops
> > svn co -N svn://server/repos/blah/blah/dir
> > move dir.oops/* dir/.
> > run 'svn status' to see if it really worked. Maybe followed by
> > an 'svn update'
>
> I tried this and it didn't seem to work (the .svn directory was missing
> entries for my files so it still complained about obstructing paths).
This statement confuses me. Checkout to a new working copy; there will be
no existing files to obstruct the checkout. Now copy the files and
directories (but NOT the oops .svn directories) from dir.oops and you should
have everything good to go.
This is essentially what I've done in the past and never had any trouble.
The way I avoid the .svn directories is
$ cp -r dir.oops dir.oops.safe; find dir.oops -type d -name .svn -print |
xargs rm -rf; cp -r dir.oops dir
In Windows, copy and paste dir.oops to a safe place of your choice. Run the
search on dir.oops, looking for .svn of type Folder When the search is
complete, select all the found .svn and delete them. Now enter dir.oops,
Ctl-A to select all, Ctl-C to copy, navigate to dir and Ctl-V to paste
everything in again. (Yes to all will come in handy at this time.)
HTH,
Donald.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: users-help_at_subversion.tigris.org
Received on 2008-01-17 19:59:08 CET