=== mark06 is now known as mark06-away [01:44] SamB_MacG5: sorry, I can't test at the moment [01:44] oh well :-( [01:44] SamB_MacG5: libsvn-dev conflicts with heimdal so I can't build subvertpy [01:46] * SamB_MacG5 installs a thing [01:46] SamB_MacG5: ask me again next week when I don't need this Kerberos stuff as badly :) [01:47] this is NT's fault, isn't it? [01:48] it's libpq-dev's fault for depending on krb5-dev [01:49] * SamB_MacG5 doesn't get why libapr depends on so much stuff either ... [01:50] * SamB_MacG5 *poof* === mark06-away is now known as mark06 [02:16] anyone knows which bug is the one mentioned above about adding email change to bzr-rewrite? I have searched for email and mail but can't find any http://bit.ly/PTiVqw === mark06 is now known as mark06-away === mark06-away is now known as mark06 === mark06 is now known as mark06-away === mark06-away is now known as mark06 === mark06 is now known as mark06-away === mark06-away is now known as mark06 [03:25] wouldn't sed work over .bzr for replacing committer field if we ever get a way to uncompress the relevant files? [03:27] not really [03:31] mark06: in principle yes, but for referential integrity you need to change the revids too [03:33] referential integrity like in relational databases? [03:43] or are revids some sort of hash out of commit data? [03:45] they are guids [08:04] morning! === jam1 is now known as jam === mmrazik is now known as mmrazik|lunch === mmrazik|lunch is now known as mmrazik [11:23] Hi all. Whom can I ping to review my little merge-request? :) It has "pending" status for 2 months already. [11:25] here is good. [11:25] okay, it's https://code.launchpad.net/~torkvemada/bzr/commit_hooks/+merge/115348 [11:25] I think yours is the one I pass... right [11:26] I skipped that in my last big review-all-pending-branches sweep [11:27] the main reason being there's been a bunch of discussion about where to put commit hooks before that I wanted to dig up, and your proposal didn't include examples of how you intended to use these new ones to quickly compare about [11:28] but, as concrete things you can do to get this landing, adding some tests would be the main thing [11:29] mgz: I've already pointed in this very conference, what I added it for (it was 2 months ago, yes :) ) [11:29] just a little link: http://bazaar.launchpad.net/~torkvemada/+junk/bzr-cimage/view/head:/__init__.py [11:29] referencing that stuff in the merge proposal is the kind of thing that helps the reviewer [11:30] ok, I'll add the reference [11:30] could you point out the place in sources, where I should add the tests? [11:31] I'm not familiar with bzr sources [11:32] only seeing https://lists.ubuntu.com/archives/bazaar/2012q2/074812.html on cursory mailing list search for similar questions... [11:34] trkv: I'll try and find some earlier hook addition proposals for you for reference [11:38] trkv: see the following mps for some hints [11:38] [11:38] [11:46] ok, thanks, I'll look at it [12:01] Hello, I'm new to bzr, but I usually use git, there's an equivalent command to "git remote add ..." ? [12:02] Well, since there's not an equivalent to git remote's... 8-} [12:03] What larger task are you trying to accomplish? [12:05] but I can have different origins in bzr? [12:05] I want to pull from a public repositroy, make changes and commit them to a private one [12:05] then pull from public again... [12:06] Mmph, I have a feeling terminological differences are going to bite us here... [12:06] :) [12:06] But pull and push (and merge, which you probably want in the mix here) use different saved locations. [12:09] what do you mean with saved locations? [12:09] Where 'bzr push' pushes to if you don't give it a URL on the command line. [12:09] (ditto pull pulling from and merge merging from) [12:10] ah ok, it is the last one used no? [12:10] First one used, actually. If there's no saved location, the first run saves what you set. [12:11] Future explicit locations aren't saved unless you use --remember. [12:11] you know where does bzr save it? [12:11] It's in the branch config. You can see the list in 'info'. [12:11] And some other config-related command that vila will remind me of at some point. [12:13] fullermd: bare 'bzr config' [12:13] Not in public! [12:13] thanks, will try to understand better the concepts now [12:14] tictacbum: The main concept to keep in mind is the branch-centric as opposed to repo-centric worldview. You'll confuse yourself into knots if you try thinking the wrong way (about either system) [12:15] (and the somewhat different usages of 'repo' and so forth, but that's second tier) [12:15] fullermd: thanks :) [12:17] tictacbum: http://wiki.bazaar.canonical.com/MatthewFuller/SpotDocs/PiecesInBrief may be helpful. With a fair amount of existing DVCS experience, it may seem a little elementary, but it's still worth a careful read to catch differences in the worldview. [12:17] Also the PiecesInLength sibling has bits that may help, especially some of the stuff at the end comparing terminology among different DVCSen. [12:18] Which is probably not completely accurate, but the author's always a bit sloppy about stuff. It's mostly right, anyway. [12:19] the author is sloppy about kisses? [12:19] Hey, some people are into that sort of thing! [12:19] (obviously not MANY, or he wouldn't have time to write about version control, but still) === Lasall42 is now known as Lasall [12:40] fullermd: great article, thanks to the author if you see [12:44] Well, I try to avoid him, but it's probably unavoidable to run into him sooner or later. [12:47] after merge: how do i see what commit deleted a file? [12:47] need to know whom to kick [12:47] Through the UI, you're pretty much limited to doing a log -v and /-ing through it. [13:16] back to my qüestion, can a project have more than one repository? [13:18] I tried to push and pull with repo after the command, and ended with a diverged branch [13:51] Crazy request: bzr shelve; bzr unshelve-on-another-machine ssh://... [14:29] LeoNerd: bzr commit; bzr push; bzr uncommit * 2 [14:29] *twitch* [14:34] LeoNerd: do you prefer 'scp .bzr/checkout/shelf/ another-machine/..../.bzr/checkout/shelf/ [14:34] I'd prefer theabove :P [14:34] But it doesn't exist.. so never mind [14:35] hi, is there a UI for bzr like gitk? [14:35] i want to see branch ancestry, see where merges have happened, where branches diverge, which one is missing a commit, etc [14:36] cmars232: qlog from the qbzr plugin [14:36] vila: thx, checking it out [14:45] vila: qlog is exactly what I needed, thanks [14:45] cmars232: make sure to not miss the ability to run it with multiple branches [16:26] Hi there, I'm getting an error very similar to https://bugs.launchpad.net/bzr/+bug/819604 but that's fix released for awhile now and I'm using 2.5.1. I'm running tarmac and while it runs the pre-commit hook, bzr keeps the ssh connection open to the branch but when the tests are over and tarmac resumes, I get the broken pipe error. Any way to workaround the issue? [16:26] Ubuntu bug 819604 in Bazaar 2.4 "when an idle ssh transport is interrupted, bzrlib errors; should reconnect instead" [High,In progress] [16:28] matsubara: jam has some branches up aimed at fixing this, see the latest mp linked from that bug [16:28] mgz, the one superseded? [16:34] mgz, and is there any known workaround before this is released? [16:35] bug 1047325 is probably the other one you're interested in [16:35] Launchpad bug 1047325 in Bazaar "not properly interpreting EPIPE as connection reset" [Critical,Fix released] https://launchpad.net/bugs/1047325 [16:36] basically we only got really good testing of the reconnect stuff when launchpad's copy of bzr was upgraded, so some follow on fixes were needed [16:38] so, you can run a trunk bzr, or cherrypick that fix, or persude tarmac to discard the transport [16:38] I'm not sure which you'll find easiest [16:38] hmm [16:39] I guess running from trunk should do [16:39] I'll try it and come back if it doesn't work [16:39] or lp:bzr/2.5 trunk [16:39] thanks mgz ! [16:39] rather than dev [16:39] hey mgz, matsubara [16:39] we'll try and get a release done when the maas stuff is less pressing [16:40] hi jelmer [16:40] (which we want to bug you about doing some testing for tomorrow) [16:40] jelmer! how's the rain? [16:40] mgz, absent! [16:40] mgz, cool. let me know [16:40] alas. [16:41] mgz: how's things here? [16:41] jelmer: bouncing along okay (and sunny) [16:42] found any samba bugs yet? [16:43] (I know, I know, shocking to even consider the possibility) [16:43] * SamB_MacG5 reconfigures LimeChat to stop highlighting SamB when it occurs as part of a word [16:43] plenty (-: [16:45] SamB: heh. === matsubara is now known as matsubara-lunch === matsubara-lunch is now known as matsubara === yofel_ is now known as yofel