[00:52] <poolie> oh ok thanks
[00:52] <poolie> jelmer, of course we don't need to verify every bug, just make sure it's not broken
[01:07] <meth> can i do conditional commits ?
[01:07] <meth> or a patch commit you might call it
[03:18] <poolie> meth, what do you mean?
[03:21] <meth> git add -p
[03:21] <meth> would like that
[03:23] <poolie> try bzr-interactive
[03:27] <poolie> https://launchpad.net/bzr-interactive
[03:31] <meth> cool
[04:11] <BlindWolf8> anyone alive in here?
[04:14] <spiv> Often!
[04:16] <BlindWolf8> hi spiv!
[04:16] <BlindWolf8> your name looks familiar :-)
[04:16] <BlindWolf8> do you have time to field a bazaar question for me?
[04:19] <spiv> BlindWolf8: depends on how hard it is ;)
[04:20] <spiv> BlindWolf8: in general it's best to just ask your question, lots of folks are usually quiet but check the scrollback periodically and might answer
[04:20] <BlindWolf8> spiv: well, I always try and be polite
[04:21] <BlindWolf8> people giving fre support and all
[04:21] <spiv> This channel is here to talk about bzr :)
[04:21] <BlindWolf8> *free
[04:21] <BlindWolf8> OK, so...here's the deal...
[04:21] <spiv> You're quite welcome to just pop in and say "hi, I'm having X problem..." :)
[04:21] <BlindWolf8> I have a server...running Ubuntu 11.10
[04:21] <BlindWolf8> bzr is installed on it via the PPA
[04:21] <BlindWolf8> bzr 2.4.2 is on all the clients and the server
[04:22] <BlindWolf8> putty 0.61 is installed on all clients
[04:22] <BlindWolf8> clients are all windows boxes
[04:22] <BlindWolf8> clients gave me public keys to pop on the server and use pagent to authenticate
[04:23] <BlindWolf8> I have the command for bzr_ssh_path_limiter listed in the auithorized_keys file
[04:24] <BlindWolf8> on the first line of the keys file, it goes <command for bzr_ssh_path_limiter> <ssh key>
[04:24] <BlindWolf8> all other lines are simply public keys
[04:24] <BlindWolf8> first, I am assuming that just the first line is supposed to have the bzr_ssh_path_limiter command?
[04:27] <BlindWolf8> I have done various tests on my end and I am able to pull files from the server no problem, but none of the other people I work with can
[04:27] <spiv> If you only want the first key to use the bzr_ssh_path_limiter
[04:27] <spiv> Each line is independent of the others, read the authorized_keys man page that comes with OpenSSH for the exact details.
[04:28] <BlindWolf8> ah........
[04:29] <BlindWolf8> I thought it was executed on the user's profile, not per key....
[04:29] <spiv> No, it's per key
[04:30] <BlindWolf8> that solves that problem, thanks!
[04:30] <BlindWolf8> I have another slight issue though
[04:30] <spiv> So some people use it as a relatively cheap way to have per-user repos hosted by a single user account.
[04:31] <BlindWolf8> some clients get a temp file error that deals with pagent...but I'm using 0.61 with no isseus, even though the issue is supposed to be with 0.61
[04:31] <spiv> I don't know much about pageant.
[04:31] <BlindWolf8> relavent bug report: https://bugs.launchpad.net/bzr/+bug/820805
[04:38] <poolie> o/ spiv
[04:38] <spiv> Hey poolie
[09:03] <mgz> morning all.
[09:26] <jelmer> hi mgz
[10:27] <mgz> what's the way of finding out what versions of packages PQM has installed?
[10:30] <jelmer> mgz: checking with a losa, probably
[10:31] <mgz> okay, what's the way of changing the PQM code so it displays the versions it has installed?
[10:32] <jelmer> mgz: you'd have to modify the script that is used to verify bzr submissions, I'm not sure if that's publicly available somewhere
[10:34] <lifeless> mgz: change Makefile
[10:34] <lifeless> mgz: in the bzr tree
[10:35] <lifeless> mgz: PQM is -extremely- simple. Thats why everyone gets so confused by it.
[10:35] <mgz> that only works if it's running a test I think lifeless
[10:36] <lifeless> mgz: yes... that was the context of your question right ?
[10:36] <lifeless> '23:31 < mgz> okay, what's the way of changing the PQM code so it displays the versions it has installed?
[10:36] <lifeless> '
[10:36] <mgz> but lp:pqm doesn't look too hard to poke, though actually deploying a change is probably impossible
[10:36] <lifeless> pretty easy - get the losa to pull in the branch
[10:37] <lifeless> and yes, lp:pqm should be fairly straight forward, though I'm horribly confused about what you're trying to do.
[10:37] <mgz> okay, I will poke this.
[10:37] <mgz> lifeless: I want to look at pqm.bazaar-vcs.org and know what version of testtools is being used
[10:37] <lifeless> mgz: I don't think that question makes sense unless a test is runnig.
[10:38] <mgz> meh, maybe not.
[10:39] <mgz> there's not a sensible way of getting ther versions of python packages unless they're being used
[10:40] <mgz> debian packages would kind of work, but I imagine that would also be a harder sell.
[10:41] <lifeless> there is that, and that the place to check is in a chroot, of which there can be multiple versions at once (especially when bzr is transitioning)
[10:41] <lifeless> having the web UI go off and probe arbitrary chroots which may or may not be active or even usable at the time ... seems odd
[10:41] <lifeless> vs having Makefile report it for you.
[10:42] <mgz> adding more output to the Makefile doesn't actually solve my problem though
[10:42] <lifeless> what is your problem ?
[10:42] <mgz> you'd still have to push a branch to see any relevent information
[10:43] <mgz> ^as stated above
[10:43] <lifeless> mgz: I don't follow :)
[10:44] <mgz> I, right now, want to know something, and right now, the page just says "0 scripts" and the time
[10:45] <lifeless> ah.
[10:45] <lifeless> so in general, to know this, I ask a losa
[10:45] <mgz> there's no change I can do in lp:bzr to remedy that, and it seems the same goes for lp:pqm
[10:46] <lifeless> mgz: you could send a patch through that will print out the versions and return 1
[10:46] <mgz> so my ping saving idea is no good.
[10:46] <lifeless> mgz: making the test suite fail and thus not landing anything.
[10:46] <lifeless> mgz: you could even just have that sitting around ready to go at a moments notice
[10:48] <mgz> that only helps me, which is much the same as just asking an admin.
[10:49] <lifeless> I must be really tied
[10:49] <mgz> but it's okay, sometimes there's not an easy way of adding code to save work,
[10:49] <lifeless> *tired*
[10:49] <lifeless> because I thought you said you wanted to know
[10:50] <mgz> I do, and when the answer was "ask someone" I wondered if I could avoid that by fixing the web display.
[10:50] <mgz> The answer to the second question is "no"
[10:50] <lifeless> its more yes-but-its-not-as-simple-as-all-that :(
[10:51] <mgz> So, I will go back to the initial issue and forget the generalisation.
[10:51] <lifeless> :)
[12:54] <mgz> ah, vila is still suffering
[12:54] <mgz> all I really have left today is the impressive cough
[13:21] <mgz> dammit, last change somehow hasn't fixed the random failure on babune natty
[13:22] <mgz> what were the chances of even hitting it again right after...
[16:20] <mgz> heh, bialix and I having the same idea in these bugs at the same time
[17:20] <mgz> I'm sure Gordan or someone fixed an issue related to unquoting paths from config
[17:20] <mgz> but I now can't find it
[17:54] <Amis> Hello o/
[17:54] <Amis> I'm sure it's somewhere in the documentation but I can't find how exactly create a new branch in a shared repository. Any help?
[17:57] <mgz> Amis: just `bzr init` as normal.
[17:58] <Amis> bzr init "remote shared rep location" project?
[17:58] <Amis> oh wait
[17:58] <mgz> eg, `bzr init-repo repo; bzr init repo/branch`
[17:59] <mgz> you can also branch into the shared repo and it will do the right thing
[18:05] <Amis> yeah, right, thanks