[05:36] <Logan_> Can any UDDers please perform this? $ requeue_package.py --full initramfs-tools
[05:36] <bob2> don't you want #launchpad instead?
[05:36] <Logan_> Someone recommended that I go here for this request.
[05:38] <Logan_> jam, maybe?
[05:39] <Logan_> Or lifeless?
[05:39] <jam> Logan_: lifeless is no longer @canonical, but I'll look into it. One moment.
[05:39] <Logan_> Alright, thanks. :)
[05:40] <Logan_> I'm requesting it based on http://package-import.ubuntu.com/status/ and the linked https://bugs.launchpad.net/udd/+bug/714622 , which recommends running that command for packages that encounter that bzr failure. It has worked in the past.
[05:43] <jam> Logan_: I have requeued your package.
[05:44] <Logan_> Awesome, thanks jam. :)
[08:40] <gour> morning
[08:42] <gour> i've a need to use svn to contribute to some open-source project, so wonder how does bzr-svn compare with hg & git's options?
[08:42] <gour> otoh, it's sad to see bzr development stalled :-(
[08:45] <lifeless> bzr-svn is pretty good
[09:09] <mgz> mornin'
[09:10] <jam> morning mgz
[09:10] <gour> lifeless: what's going  to happen with bzr?
[13:31] <mathrick> hiya
[13:32] <mathrick> so what exactly is the difference between development-colo and bzr-colo, and how am I supposed to use the former?
[13:42] <jiku> hello All..
[13:42] <jiku> im pretty new to bazaar and would like to implement it in my workplace..
[13:43] <jiku> bzr-explorer is good and would like to master it so that i can talk to developers on how to use it and create a workflow.
[13:43] <jiku> 1 issue im facing is..
[13:43] <jiku> i have a revision in the server with name, think-blog..
[13:43] <jiku> i checkout in a client (win7) machine with bzr-explorer installed.
[13:43] <jiku> it asks for sftp password when doing checkout which is normal.
[13:44] <jiku> once i download, if im not wrong, the revision with its history is saved in the client..
[13:44] <jiku> but here whenever i open bazaar explorer and open that working tree, it asks for passwords.. when i do a diff, it asks for passwrods..
[13:44] <jiku> why is it still dependent on the server which actually should be doing everything from the local perspective ?
[13:44] <jiku> any ideas ?
[13:45] <jiku> or is it that im doing it wrong ?
[13:49] <jiku> hi,.. anybody ?
[14:03] <mgz> jiku: you don't want a lightweight checkout, you want to make a proper branch
[14:04] <jiku> mgz, oh.. ok...
[14:04] <jiku> i was doing that by default..
[14:04] <jiku> let me try again.
[14:04] <mgz> you then also may want to put the passwords in your authentications.conf file, see `bzr help authentication`
[14:05] <jiku> ok..
[14:05] <jiku> let me try this..
[14:05] <mgz> or, use an agent so you don't type the password each time but use a public key instea
[14:05] <mgz> d
[14:06] <mgz> mathrick: you don't want to use development-colo, but you can use the builtin colo support in 2.6, it's just a little less polished than the bzr-colo plugin
[14:09] <jiku> mgz, hmm..
[14:09] <jiku> looks like, whenever i do a checkout using explorer, it does  with --lightweight enabled.
[14:09] <mgz> you don't want to do checkout
[14:09] <mgz> you want a branch
[14:09] <mgz> (I suspect)
[14:10] <mgz> if you have options over workflows, you really don't want to emulate the centralised server thing of svn etc
[14:10] <jiku> mgz, true.. yes..
[14:11] <jiku> mgz, was confused with git (never used but just read).. there you checkout and branch.. (if im not wrong)
[14:11] <mgz> depends what you think you can get everyone else to use, but having a full branch locally, and working with trunk+merge is how to get the main benefit of dvcs
[14:12] <jiku> yes.. right.. this is what we need actually.. there shouldn't be a centralized locking..
[14:12] <mgz> in git, saying 'checkout' means get me all the history, and branch just... prints the branch name or something
[14:12] <mathrick> mgz: right, but still, how do I use it? I don't see much explaining how to do it
[14:13] <mgz> mathrick: basically, you just create a new standalone branch, do `bzr switch -b trunk` or whatever to name the default branch
[14:13] <mathrick> ah
[14:13] <mgz> then you can do `bzr switch -b feature` to create a feature branch
[14:14] <mgz> `bzr switch trunk` to go back to trunk
[14:14] <mathrick> and what about cloning, will it carry over all branches?
[14:14] <mgz> the main missing thing is a nice addressing scheme for colocated branches outside of a few commands
[14:14] <mathrick> I see an open bug about it, so I guess not?
[14:14] <mgz> right, it doesn't, though there's a workaround of some kind I think
[14:15] <mathrick> what'd that be?
[14:16] <mgz> not sure, but it's reasonably trivial to iterate over local branches and push them all up, there's a plugin that does something like that for shared repos
[14:16] <mathrick> this was what I hoped most for, along with fixing the fact that bzr-colo refers to branches with absolute paths, which breaks alternative approaches of just sharing the same workspace dir amongst many machines
[14:17] <mathrick> mgz: well, yes, but it needs thinking on my part and remembering to do it. As much as I think git's model is stupid, the fact that fetch carries over everything is a boon
[14:17] <mathrick> not just my part, but everyone I talk to
[14:17] <mathrick> which is a bigger problem really
[14:18] <mathrick> mgz: does 2.6 need anything in the way of the branch format to support colocated branches?
[14:18] <mgz> nope, that's all done
[14:18] <mathrick> so just 2a?
[14:18] <mgz> it's just making it all pretty to use that's unfinished
[14:19] <mathrick> mhm
[14:20] <mgz> I'm using it for one project currently as an experiment
[14:20] <mgz> apart from some confusion setting up, and needing to know to type file:,branch=FEATURE in a few circumstances, I've not had any issues
[14:21] <mgz> that said, I've not wanted to push all branches, push has been specifically for updating remote trunk after feature branch merge
[16:41] <Logan_> jam: Can you please do < requeue_package.py --full iscsitarget > when you get the chance? Thanks so much!
[16:41] <Logan_> (Or any other UDDer, for that matter.)
[16:44] <mgz> warum?
[16:46] <mgz> I can do this... well, I could look at the package log as well, but would prefer to not take 15 minutes looking at it first.
[22:51] <nDuff> Is shelve expected to work with bzr-svn? I'm getting "NotImplementedError: <bound method SvnBasisTree.path_content_summary of <SvnBasisTree for '<SvnWorkingTree of /home/cduffy/VC/boundsetter>'>>"
[22:52] <jelmer> nDuff: no, I don't think it works with bzr-svn