[00:13] <wgz> okay, that only took me all evening
[00:14] <wgz> but have a build of pyqt that works with trunk python here so can be used with trunk bzr
[00:25] <wgz> where is the syntax of BZR_PLUGINS_AT actually documented?
[00:26] <wgz> http://doc.bazaar.canonical.com/beta/en/user-reference/configuration-help.html#bzr-plugins-at
[00:31] <wgz> but... bzr-explorer still doesn't want to run inplace for some reason
[00:35] <wgz> meh, I need the dir named 'explorer' not 'bzr-explorer'... what's the point of the @ then? just take the last dir component.
[00:36] <wgz> and now it's bed time again.
[00:40] <wgz> http://paste.ubuntu.com/843793/
[06:30] <robert_ancell> how do I delete a wrong tag in LP bzr?
[06:36] <mgrandi> is there a program all you guys use to look at diff files?
[07:05] <jvargas> hi.
[07:05] <mgrandi> hi
[09:26] <jml> sorry, I've forgotten
[09:26] <jml> how do I fetch a new remote branch into my colo repo?
[09:39] <mgz> morning
[09:42] <jml> mgz: hi
[09:42] <jml> I need some help with colo
[09:42] <jml> I've forgotten how to fetch a new remote branch into a colo
[09:42] <jml> And now I'm getting conflicts sometimes when I switch branches
[09:43] <jml> that doesn't make any sense to me.
[09:45] <mgz> hey jml
[09:46] <mgz> it's possible to get conflicts when switching if, for instance, a directory didn't exist but has other contents in your tree
[09:47] <mgz> or is it something else?
[09:47] <jml> mgz: hmm. maybe it is that.
[09:48]  * jml looks
[09:48] <mgz> I can't remember if there was a fancy way to branch into a new colo dir, but `bzr branch remote file:,branch=newlocal` should work
[09:48] <jml> mgz: no, it's a text conflict in NEWS and a couple of other files
[09:49] <jml> mgz: in this particular branch, I'm switching from testtools-0.9.12 to testtools-0.9.13. Am really perplexed by the conflicts.
[09:50] <mgz> jml: and the tree was in a clean state when doing it? that would be odd.
[09:55] <jml> mgz: yeah, no text changes etc.
[09:56] <jml> mgz: http://paste.ubuntu.com/844150/
[09:57] <jml> mgz: easy enough to reproduce the problem.
[10:00] <jml> mgz: further, I have no idea what to do when I get a conflict like that.
[10:01] <mgz> well `bzr revert . && bzr resolve` I guess
[10:02] <mgz> the basic transform between .13 and .12 works fine here
[10:03] <mgz> so, I guess there must be a more specific bug, if you would do the honour of filing that jml?
[10:05] <jml> mgz: sure thing.
[10:07] <jml> mgz: https://bugs.launchpad.net/bzr/+bug/933362
[10:08] <mgz> ...well, I've just got myself in a weird state trying to replicate your setup
[10:09] <mgz> I wanted a colo branch, but somehow have "Unshared repository with trees and colocated branches (format: 2a)"
[10:09] <mgz> and now can't do any operations in that dir, as it's not a branch.
[10:17] <jml> mgz: huh
[10:17] <jml> mgz: the pastebin starts from scratch, fwiw.
[10:18] <jml> incidentally, I wish I knew what public API python-subunit had so I could clean up some of these unnecessary imports.
[10:19] <jml> hmm, now, to figure out how these tests manage to pass
[10:20] <mgz> yeah, I was trying to be clever by creating the branches in a slightly different way to see if it was swtich -b at fault
[10:20] <mgz> but screwed up the first step
[10:21] <mgz> `bzr branch testtools file:colotesttools,branch=trunk` does the weirdness
[10:23] <mgz> jml: well, either everything without an underscore, or I guess having a look at what using subunit and being pragmatic
[10:23] <jml> in general, I think I quite like colo, but one thing it makes slightly harder is opening up the trunk version of a branched file for side-by-side comparison.
[10:24] <jml> mgz: ISTR subunit deliberately exporting 'content' and 'content_type' from testtools. ICBW though.
[14:04] <alansaul> Hey guys, I want to merge some changes I accidently made in my trunk branch (but havent yet commited) into a new branch
[14:04] <alansaul> so ive made the branch, but now I want to commit to that branch, not to my current one, how do I do that?
[14:12] <mgz> the easy way is just to do `bzr branch trunk new_feature` then `bzr pull --overwrite -r-4 trunk trunk`
[14:12] <mgz> the starting state of trunk is what you want in your feature branch
[14:12] <mgz> and then you can just chop off the last N revisions from trunk
[14:13] <mgz> ...the second command may be wrong, but there is a way of doing that (or you can just uncommit several times)
[14:29] <leo2007> why can't there be two bzr command running on the same repo simultaneously?
[14:30] <jelmer> leo2007: there can be multiple commands running on the same repository
[14:30] <jelmer> leo2007: you can't write to the working tree state file with multiple processes concurrently though
[16:11] <abentley> jelmer: What do you think of using the colocated branch name as default nick?  I've got an mp up for that, but poolie wanted to see what you thought.  https://code.launchpad.net/~abentley/bzr/name-nick/+merge/93319
[16:15] <jelmer> abentley: I thought I'd replied, but something seems to have eaten my email
[16:16] <jelmer> abentley: I think that's a good idea,  I'll review again
[16:16] <abentley> jelmer: great.  Thanks.
[17:38] <abentley> jelmer: landing.  At last.
[17:39] <abentley> jelmer: poolie suggested putting my "nc:" plugin in core, perhaps as "x-nc:" to indicate experimental nature.  Any thoughts?
[17:41] <jelmer> abentley: I haven't looked at the code yet.I think it would be reasonable to add it as nc: or x-nc:
[17:41] <jelmer> abentley: Of course the goal is to eventually just support passing colocated branch names to commands
[17:42] <jelmer> abentley: but in the mean time it would be very useful; plus, I don't think there's a clear disadvantage in supporting it
[17:43] <jelmer> I'm not sure if "nc" is the best prefix, though I don't really have any brilliant ideas on a better prefix
[17:43] <mgz> no contest.
[17:43] <abentley> jelmer: I'd use colo if the plugin didn't use it already.
[17:44] <jelmer> hmm, yeah
[17:46] <abentley> jelmer: "ncolo:"?
[17:47] <abentley> jelmer: "jcolo" ? :-)
[17:50] <jelmer> abentley: perhaps just "co:" ?
[17:50] <mgz> kolo would be cooler, and it's jelmer, so, should use
[17:50] <mgz> jk:
[17:51] <abentley> jelmer: wfm.
[17:52] <jelmer> mgz: I didn't realize using a k makes something cooler, let me make a note of that :)
[20:48] <fullermd> jelmer: It doesn't.  It makes it kooler.
[22:02] <jono> hey all
[22:02] <jono> hey all
[22:02] <jono>  I am trying to branch and keep getting:
[22:02] <jono>  ConnectionReset reading response for 'BzrDir.open_2.1', retrying
[22:02] <jono>  Permission denied (publickey).
[22:02] <jono>  bzr: ERROR: Connection closed: Unexpected end of message. Please check connectivity and permissions, and report a bug if problems persist.
[22:02] <jono>  I imported my public key after I did bzr launchpad-login
[22:05] <wgz> if you do `ssh yourusername@bazaar.launchpad.net` what do you get?
[22:05] <wgz> if you still have a key error, that's what needs fixing, if it says no shells, you've done everything right
[22:11] <wgz> jono: any luck?
[22:11] <jono> wgz, still no luck
[22:11] <jono> wgz, we are discussing this in #launchpad
[22:12] <jono> I think this is a key error
[23:09] <spiv> "Permission denied (publickey)." strongly suggests a key problem, yse.
[23:09] <spiv> *yes