Remove a branch

[John Arbash Meinel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Sebastjan Trepca wrote:
| Thanks guys and +1 for being a bug :)
|
| Sebastjan
|
| On 4/29/08, Martin Pool <mbp at canonical.com> wrote:
|> On Tue, Apr 29, 2008 at 7:31 AM, John Arbash Meinel
|>  <john at arbash-meinel.com> wrote:
|>  >  Well, you can 'rm -rf .bzr/branch' in that directory. I guess it isn't
|>  > something
|>  >  we really worked out. Usually a branch is in it own directory, so you can
|>  > just
|>  >  get rid of the whole containing dir. But if it is at the root of a shared
|>  > repo,
|>  >  you don't want to get rid of the repository.
|>
|>
|> Maybe it's a bug that we allow push to create a branch in a control
|>  directory that previously contained just a shared repository?  I
|>  realize it's technically quite feasible but it seems like an unusual
|>  use, and probably a mistake.
|>
|>
|>  --
|>  Martin <http://launchpad.net/~mbp/>

I personally feel like it should be supported. I may be the only one, but I sort
~ of like the nested namespaces. Where http://host/project is the mainline of
'project' and http://host/project/branch is a custom branch of it. It even has
the property that "bzr co http://host/project" gives you a directory named
"project" instead of the generic "trunk".


I may be the only one that feels that way, though. Other people seem to have a
harder time handling a nested namespace. So we could explicitly forbid it. If we
did, then I would suggest doing it at all levels. (So
http://host/branch/subbranch would also be forbidden.) Though that starts to get
into trouble if you want to have nested branches (aka nested trees, but no
working trees right now.)

John
=:->
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkgXHIsACgkQJdeBCYSNAAM7qgCgvCPH56yd2jfEIo08UR3cnu43
lgoAmwQyso4A2XeXp1dLUpv9yqDy83s5
=x1p4
-----END PGP SIGNATURE-----