On Dec 7, 2007 10:04 AM, David Glasser <glasser@davidglasser.net> wrote:
> On Dec 7, 2007 8:27 AM, Miller, Eric <Eric.Miller@amd.com> wrote:
> > > Oh, and just to make sure: if we're doing an update at a depth that
> > > should affect a directory A but ordinarily would not affect its
> > > children (or its non-file children), and we get a delete for A, we
> > > really should be doing a recursive delete anyway, right?
> > >
> > > (ie, the problem here is that the test cases fail, not that they fail
> > > with an unintelligible error, right?)
> >
> > For me the biggest problem is that it results in a working copy that is
> > unrecoverable using svn commands.
>
> Oh, sure, that goes without saying :)
>
> Thanks for the bug report, by the way! While in the future it would
> be nice to clarify the version you're using, and to refrain from
> opening a bug report without a buddy, this definitely is an issue we
> should fix. (While it's not a regression from pre-1.5, that's only
> because -N *never* really worked pre-1.5).
This bug isn't limited to depthiness; it also means that "svn cleanup"
doesn't really work for pending deep deletions. You can see this if
you insert an abort(); before the svn_wc__run_log call in
update_editor.c(do_entry_deletion), do an update which deletes some
directory, and try "svn cleanup": it fails.
--dave
--
David Glasser | glasser_at_davidglasser.net | http://www.davidglasser.net/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Dec 11 00:27:17 2007