[05:57] <cdbs> is bzr broken in Natty?
[05:58] <cdbs> okay, got the bug
[05:58] <lifeless> https is
[06:02] <cdbs> lifeless: so is there a workaround?
[06:02] <cdbs> I tried to branch using the bzr:// protocol
[06:06] <lifeless> cdbs: I'm not sure; perhaps running with python2.6 ?
[06:06] <lifeless> e..g python2.6 `which bzr` branch bzr://....
[08:30] <thumper> lifeless: ping
[08:32] <lifeless> hi?
[08:32] <thumper> how do I create a transport from an existing transport?
[08:32] <thumper> I want a directory under the existing transport
[08:32] <thumper> (which is a shared repo)
[08:32] <thumper> so I have repo.user_transport
[08:32] <thumper> and I want a directory under that to create a branch in
[08:34] <lifeless> t.clone(relpath)
[08:34] <thumper> ta
[08:35] <xanalogica> I'm looking for a way to move a working tree to a specific revision but cannot find it;   basically I want:    wt.update(revision_id=27)
[08:43] <xanalogica> hmm, it looks like it might be clone(wt_rootpath, revision_id='........'), where you basically clone a specific rev *on top of yourself*.  Interesting.
[08:50] <lifeless> if you want a specific rev
[08:50] <lifeless> revert() is generally a sufficient answer
[08:53] <xanalogica> I don't see that wt.revert() accepts a rev id or version though?
[10:12] <lifeless> xanalogica: wt.revert(old_tree=wt.branch.repository.revision_tree(...))
[11:00] <xanalogica> lifeless, thanks I have a working solution now, using revert().
[23:05] <knighthawk> I'm trying to do a checkout of a branch from a central repo into a new branch on my local drive. when I type 'bzr checkout sftp://path/to/repos' I get error Not a branch. I'm guessing that has something to do with my local setup. but I'm not sure what.
[23:10] <maxb> knighthawk: If you really are giving a path to a repository, then a repository is not a branch
[23:10] <maxb> Try 'bzr info URL'
[23:19] <knighthawk> its hanging
[23:20] <knighthawk> bzr: ERROR: No repository present:
[23:21] <knighthawk> that's when I point it at the branch. when I point it at the repos it hangs.
[23:22] <maxb> hrm
[23:23] <maxb> Could you give the URLs (un-obfuscated) of the branch and repo, so I have a better idea what's going on here?
[23:24] <maxb> Also, running 'find -ls' in the repository and pastebinning the result would be helpful to understand what you have
[23:24] <knighthawk> the branch is called grws its in a repos call bzr-repos sftp://repos.grdev.com/opt/bzr-repos/grws
[23:27] <maxb> OK, can I see 'find /opt/bzr-repos/.bzr -ls' and 'find /opt/bzr-repos/grws/.bzr -ls' ?
[23:28] <knighthawk> just a sec.
[23:29] <knighthawk> http://fpaste.org/w5To/
[23:30] <maxb> erm. That doesn't look so good. It's just an empty bzrdir, there's no repository there.
[23:31] <knighthawk> http://fpaste.org/cjzg/
[23:31] <maxb> That's a branch+tree alright, but it would seem that the repository that is supposed to be present at /opt/bzr-repos/ is missing
[23:32] <knighthawk>  history | grep 'bzr init-repo'
[23:32] <knighthawk>    21  bzr init-repo bzr-repos
[23:32] <maxb> OK..... so it was there once, but is very much not there any more
[23:33] <maxb> There should be an /opt/bzr-repos/.bzr/repository/ but there is not
[23:33] <knighthawk> okay so is there a way for me to make /opt/bzr-repos a repo again without blowing away the branches inside of it?
[23:33] <maxb> If the repository that was containing the data for the branches is lost, the branches are mostly useless now
[23:34] <knighthawk> lovely.
[23:36] <knighthawk> so I have a branch on my local machine.  will that help me restore anything?
[23:37] <maxb> cd /opt/bzr-repos
[23:37] <maxb> mv .bzr broken.bzr
[23:37] <maxb> bzr init-repo .
[23:38] <AfC> and then re-pull the content into those branches
[23:38] <maxb> Then, locally, 'bzr push url://to/opt/bzr-repos/some-name-for-branch'
[23:39] <maxb> The branches are primarily a pointer to the tip revision, plus a dictionary of tags
[23:39] <maxb> If, by means of other push/pull operations, you cause the relevant revisions to be present in the recreated repository, the old branches will work again
[23:45] <knighthawk> thanks guys I *never* would have figured that out.