Couple of bzr-svn related issues...

Jelmer Vernooij jelmer at
Mon Mar 9 13:36:35 GMT 2009

Hi John,

On Mon, 2009-03-09 at 09:03 -0400, John Szakmeister wrote:
> I've been trying to use bzr-svn more (and it keeps getting better,
> BTW), but I've run into a couple of issues.  I reported the issue
> about the target project not being set, and you kindly pointed out
> that it was fixed on trunk.  I pulled in the change that fixed it (had
> to hand type it since it's painful for me to import this stuff... I'm
> not connected to the internet).  Not it doesn't backtrace, but it
> created the branch in the wrong location.  I told it to put the branch
> in branches/my-test-branch, but it ended up using the branch nickname
> instead.  As a user, I'm a little confused by that behavior.  You may
> have fixed this in a subsequent revision, and when I get back to my
> machine at home, I'll try out trunk just to be sure.
This is a bug that needs fixing in both bzr and bzr-svn; the bzr-svn
side of it has been fixed, the bzr bit is in progress
( As a workaround,
you can use the svn-push command.

> The other issue I ran into is that bzr-svn backtraces if I try to bzr
> push into the top-level of the repo (a sibling to trunk).  I found
> this out because I was just trying to get a feel for how things would
> look pushing an entirely new branch into Subversion that wasn't
> originally branched from there.  So I created an empty test repo...
> and you can see where that went.
Please file a bug report about this and how to reproduce it.

> Finally, I'm curious if it's possible that in a normal workflow (of
> branching trunk, making changes and pushing to the branches area of
> Subversion) if it's possible to get that extra copy step of trunk to
> branches with no other changes.
I'm not sure I understand what you mean? What operation of bzr-svn in
this situation is problematic here?


Jelmer Vernooij <jelmer at> -
Jabber: jelmer at

More information about the bazaar mailing list