OK so we can reproduce this for y'all now.
We have a tiny svn repo (1.6.3) that could be attached to a bug-report
and a single merge command that will show it immediately.
Is that OK, or would the python bug reproductions scripts be better ?
Regards,
- Paul
On Sep 17, 2009, at 6:28 PM, Paul Hammant wrote:
> Hey folks,
>
> Question: Should svn merge ever fail within the first few seconds
> of what should be two days of three-way merge arbitration ?
>
> How we discovered this :
>
> We used a self-built svn from branches/1.6.x today (@39407). We
> double checked via ldd that it is statically linked.
>
> > cd branch
> > path/to/svn merge -r10182:11149 https://svn/repo/trunk/foo .
> svn: Attempt to add tree conflict that already exists at 'foo/src/
> something'
> svn: Error reading spooled REPORT request response
> > path/to/svn st
> C foo/src/something
> > local edit, incoming delete upon merge
>
> It is odd to us that Subversion does not go past this merge-
> conflict. Marking it in some way, sure we can work with conflicts
> later. Exit of the executable is something that's leaving us
> incredibly nervous.
>
> Svn 1.6.4 is the same, FYI.
>
> Thoughts?
>
> - Paul
>
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=1065&dsMessageId=2396240
To unsubscribe from this discussion, e-mail: [users-unsubscribe_at_subversion.tigris.org].
Received on 2009-09-18 02:19:12 CEST