=== mw is now known as mw|out [03:07] What's the environment variable that controls which editor bzr uses to write a commit message? [03:11] EDITOR [03:19] looks like BZR_EDITOR [03:28] both [03:28] BZR_EDITOR will override EDITOR [06:06] Is there some way to mark a branch as dead with bzr? [06:07] I know I can rm -rf the branch but not everyone has access to the server so I was wondering if there's a way to do that from within bzr. [07:35] New bug: #216157 in bzr "push to empty repository causes error" [Undecided,New] https://launchpad.net/bugs/216157 [08:39] abentley: any idea about this: [08:39] File "./start-bundlebuggy", line 26, in [08:39] from bundlebuggy.controllers import Root [08:39] File "/home/robertc/bundlebuggy/bundlebuggy/controllers.py", line 41, in [08:39] File "/home/robertc/bundlebuggy/bundlebuggy/model.py", line 163, in [08:39] File "/usr/local/lib/python2.5/site-packages/Elixir-0.5.1-py2.5.egg/elixir/entity.py", line 651, in __init__ [08:39] File "/usr/local/lib/python2.5/site-packages/Elixir-0.5.1-py2.5.egg/elixir/statements.py", line 48, in process_mutators [08:39] File "/usr/local/lib/python2.5/site-packages/Elixir-0.5.1-py2.5.egg/elixir/statements.py", line 35, in process [08:39] File "/usr/local/lib/python2.5/site-packages/Elixir-0.5.1-py2.5.egg/elixir/relationships.py", line 916, in handler [08:39] File "/usr/local/lib/python2.5/site-packages/Elixir-0.5.1-py2.5.egg/elixir/relationships.py", line 366, in attach [08:39] File "/usr/local/lib/python2.5/site-packages/Elixir-0.5.1-py2.5.egg/elixir/properties.py", line 119, in attach [08:39] AttributeError: merge [08:46] it seems to be this: [08:46] http://elixir.ematia.de/trac/browser/elixir/trunk/elixir/properties.py [08:47] line 119 does delattr [08:47] -> rel.attach(entity, name) [08:47] /usr/local/lib/python2.5/site-packages/Elixir-0.5.1-py2.5.egg/elixir/relationships.py(366)attach() [08:47] -> super(Relationship, self).attach(entity, name) [08:47] > /usr/local/lib/python2.5/site-packages/Elixir-0.5.1-py2.5.egg/elixir/properties.py(119)attach() [08:47] -> delattr(entity, name) [08:47] (Pdb) entity [08:47] [08:47] (Pdb) entity.merge [08:47] [08:47] (Pdb) delattr(entity, 'merge') [08:47] *** AttributeError: merge [08:47] but entity here is a class not an instance [09:58] hi, anybody for core guys around? when bzr.dev will start 1.5 development? [10:05] bialix: should be started already as far as I know: it always starts after the rc1 release... [10:06] dato: I can't find dedicated branch for 1.4 that's why I'm asking. Probably poolie did not finish or publish it yet\ [10:06] possibly [10:09] I don't wanna start 1.5 myself because I have not enough information [10:11] err, I hope nobody kill me for my bad english [11:10] hey folks! [11:17] howdy [13:23] hi all. a question about the announce list. the domain i view my options on (and i access the list via) is lists.ubuntu.com, but my subscription is for lists.canonical.com. its also seems to be on a different mailman server then the ubuntu lists. is this intentional? [13:25] Bambi_BOFH: afaik lists.ubuntu.com and lists.canonical.com are the same server [13:28] dato: that explains a bit i guess. [13:37] interestingly, bzr is listed in https://lists.ubuntu.com/mailman/admin , which i woudl have thought only had one mailmans instance listed [14:52] If I want to maintain a continuous mirror of an upstream CVS repository of non-trivial size/complexity (pkgsrc, freebsd, that kinda stuff), would anyone say there is a suitable software for doing this in a reliable manner at this time? The purpose is to keep a mirror that is branchable in a distributed fashion. [14:53] cvs2svn/cvs2git + fastimport seems to be recommended, but it seems mostly for one-time conversions. [14:53] I know about tailor but have had bad luck with it generally. is there a rock-solid solution that people are using? [14:57] the one used by launchpad isn't *rock* solid, but they use it to do incremental imports of some pretty large stuff. [14:58] I can't remember the name though unfortunately, #launchpad might be able to help [14:59] Thanks! [15:06] New bug: #216330 in bzr "Bzr crashes when executing log command over smart server protocol" [Undecided,New] https://launchpad.net/bugs/216330 [15:20] <[1]ian> Hi all. I am using Bazaar 1.3 for windows and got the error message bzr: ERROR: Generic path error: 'd62egvgtpqqy0o38y0r4.fetch': Failure: unable to [15:20] <[1]ian> rename to '../packs/7f3c802bce73af94369ea3e20c4755dd.pack') [15:20] <[1]ian> when I tried to push to a server. Does anyone know what may have caused this? [15:21] what protocol are you using to push? [15:21] <[1]ian> ssh, in the form of sftp [15:25] odd - it's trying to do a simple file rename on the remote server but failing [15:25] what os is the remote server and what ssh server are you running? [15:33] <[1]ian> server is running redhat linux with openssh and my client laptop is running parmiko ssh to get authication.conf to work [15:40] how i can send package of slackware for download in bazaar website? [15:40] i build the package for slackware [15:43] where do i find the code for serve ? [15:43] it's not anywhere in ~/lib/python/bzrlib/ [15:44] Pengwn: start by cmd_serve() in builtins.py [15:44] someone? [15:45] someone here is administrator of bazaar? [15:46] data thanks. [15:46] sorry data=dato [15:47] happens all the time ;) [15:47] herson: Just link it from the download page on the bazaar-vcs.org homepage [15:47] herson: it's a wiki [15:48] jelmer: i can put in there? [15:48] herson: yep [15:48] jelmer: tanks [15:51] wow I need to learn python sytax :-) [15:51] what's a def run . [15:52] is that part of class cmd_serve ? [15:54] I don't think ctags support python ? [15:55] it does [15:56] oh ok. [16:09] how to a go to a defn is it not [ in vim [16:09] or is ctags -R not enought to recg python? [16:11] Pengwn: try ctrl-5 [16:11] hey when did this feature out :-) [16:12] or is it only for python? [16:13] no, it's just vim [16:13] ok actually ctrl-] also working . [16:13] thanks Dato. will remember that.:) [16:15] wow damn limited support for python. ok think will have to do it the hard way or will try eclipse . [16:17] jelmer: i can put my packagens in bazaar launchpad? [16:18] herson: I think so [16:19] jelmer: how? you know? [16:19] If you're a member of the bzr team you should be able to upload [16:19] if you're not yet a member, probably best to email Martin about it [16:26] jelmer: but,  Which team works to keep a package of a system? [16:27] herson: how do you mean? [16:29] jelmer: I have to create a new slackware team for developers to send and maintain the packages? [16:30] poolie: hey mean, you be here? [18:21] Hello there! [18:22] Just a quick question - does bzr support nested repos? I'd like to create a 'projects repository with separate repos within it. Is it going to work reasonably? [18:23] * beuno points matid at LarstiQ [18:23] I'm trying to achieve something like creating an svn repository with svn:externals pointing to other repos. [18:28] LarstiQ: Hello, are you there? [18:34] matid: yes [18:35] ah [18:35] * LarstiQ promptly hides again [18:35] *g* [18:35] :D [18:35] matid: give me a couple of minutes to eat a bit and such [18:35] LarstiQ: Sure, no problem. [18:45] New bug: #216424 in bzr "bzr-svn crashes with conflict error" [Undecided,New] https://launchpad.net/bugs/216424 [18:55] matid: so, nested trees are not done, but what is available is at https://code.launchpad.net/~larstiq/bzr/nested-trees [18:57] LarstiQ: Cool, thanks. [18:57] LarstiQ: Do you possibly have any experience working with bzr-svn? [18:59] matid: yes, I do. [18:59] LarstiQ: Which version of bzr-svn would you recommend? I got 0.4.9 working with bzr 1.3.1 but it failed to push changes back via svn+ssh [19:00] Now I'm trying to do the same with trunk version of bzr-svn and bzr 1.4rc1 but it gives me segmentation faults on push. [19:01] matid: Personally I'm now using 0.4.9 with 1.3 and 1.4rc1 [19:01] matid: but the method of access is http with apache2, not svn+ssh [19:04] LarstiQ: Will try this way. [19:06] matid: are you intending to use bzr-svn with the repository you just mentioned, or are the questions unrelated? [19:06] LarstiQ: You mean nested repos stuff? No. [19:07] LarstiQ: Trying to pull the same repository via svn+https gives me bus error. [19:08] * LarstiQ blinks [19:08] LarstiQ: http://pastie.caboo.se/179726 [19:12] LarstiQ: And that's what I got when I tried to push with svn+ssh: http://pastie.caboo.se/179729 [19:12] hiya [19:12] mathrick: Hello [19:12] matid: :) [19:12] is there a real reason there's a difference between bzr pull and bzr up on a checkout? [19:13] it appears to be rather spurious to me [19:13] ie. I'd expect bzr pull without any parameters to perform an update [19:20] * fullermd wouldn't... [19:20] fullermd: is there a reasonable scenario where you'd want to perform both a pull and up on a checkout and have them do different things? [19:21] Of course. A checkout is just a working tree on a branch. update updates the working tree, pull updates the branch. [19:32] matid: seems like you hit a python-subversion bug [19:33] jelmer: Argh... I'm trying to get it working since morning. [19:34] jelmer: What is the recommended set up as of now? I'm running trunk 1.6 subversion with bzr 1.3.1 and 0.4.9 bzr-svn. [19:34] jelmer: On Mac OS X, fyi, [19:34] matid: I wouldn't recommend 1.6 at the moment but rather 1.5 [19:36] jelmer: http://svn.collab.net/repos/svn/branches/1.5.x/ or http://svn.collab.net/repos/svn/tags/1.5.0-beta1/ ? [19:36] matid: The crash over https is definitely a subversion bug. I'll see if I can reproduce it here and fix it. [19:36] humm, 1.5.x hit a release finally? [19:36] jelmer: I *think* I had the same thing some time ago [19:37] but I can't remember what the problem was [19:37] mathrick: It can't occur with 1.4 since it doesn't even have that function [19:37] ah [19:37] matid: the other bug I'm not sure [19:37] matid: Can you run this in a python interpreter: [19:38] import svn.core; print svn.core.SubversionException(1,"blah").args [19:38] (1, 'blah') [19:39] matid: thanks [19:39] matid: Can you try pushing over svn+ssh again but set the BZR_PDB=1 environment variable? That should put you into a python debugger [19:40] once you're in the debugger can you run "print err" and "print err.args" ? [19:40] jelmer: Sure. [19:41] http://pastie.caboo.se/pastes [19:41] Sh** [19:41] jelmer: print err.args outputs () [19:42] and type(err) ? [19:42] [19:43] err.file and err.line? [19:44] 'subversion/libsvn_ra_svn/internal_auth.c':114 [19:45] and perhaps err.apr_err ? [19:45] 210007 [19:47] jelmer: I can try compiling 1.5.x instead of 1.6 to see if it works. [19:49] matid: Once sec, I've got a patch coming up to deal with this issue for svn+ssh [19:50] jelmer: Cool. [19:55] matid: http://samba.org/~jelmer/bzr/bzr-svn-svn1.5exceptionargs.diff [19:58] jelmer: Applied, testing atm. [20:00] jelmer: http://pastie.caboo.se/pastes [20:00] jelmer: Seems like there's a similar problem in a different file. [20:04] hmm, looks like I need to patch all places where SubversionException is being caught [20:04] it may be more worthwile to fix this upstream since this is a regression from 1.4 [20:05] jelmer: I'll try 1.5 in the meantime, what do you think? [20:06] matid: 1.5 probably has the same issue [20:06] I'm discussing it with the svn folks at the moment [20:07] jelmer: How about https? Should it work with 1.5 and its bindings? [20:08] matid: I think you'll find the segfault is there as well [20:08] jelmer: We'll see, I'm going to try it out. [20:12] jelmer: Yeah, it's the same. push with svn+ssh gives me the same error as in 1.6 and switch to svn+https fails with bus error. [20:22] jelmer: Any news? [20:24] matid: working on two patches in parallel to fix those issues [20:30] jelmer: Cool, thanks in advance. [21:58] jelmer: Are you going to publish a patch or simply add it to trunk of bzr-svn? [21:58] jelmer: I've got to be going soon, that's why I'm asking,. [22:03] matid: I'm working on getting it fixed in svn [22:03] jelmer: Ah, right. [22:03] matid: The patch is just a workaround around svn breaking [22:04] matid: one of the other svn devs made a patch that should fix it properly [22:04] matid: see http://paste.lisp.org/display/59044#1 [22:04] How do I do per-repository post-commit hooks in bzr? [22:04] That is, I don't want my post-commit hook to apply to *all* my repos. [22:05] matid: That patch should fix pushing over svn+ssh for you [22:05] jelmer: Is it 1.6 only? [22:05] matid: both 1.5 and 1.6 [22:05] jelmer: Cool. [22:05] Slant: Usually we add a configuration setting and let the post commit hook look at that [22:06] Slant: this is a python-based hook? [22:06] jelmer: I just want to run a shell script after a commit. [22:15] jelmer: Seems like it really fixed the issue.... [22:16] jelmer: But another one appeared - http://pastie.caboo.se/179770 :) [22:16] matid: can you change that line to (self, message, apr_err) ? [22:17] jelmer: Sure. [22:18] Is there a method in place already for having bzr run a script after a commit? [22:19] jelmer: That's what I got after fixing that line: http://pastie.caboo.se/179771 [22:21] jelmer: What do you think? [22:22] matid: thanks [22:22] * jelmer looks up what 210007 means [22:28] SVN_ERR_RA_SVN_NO_MECHANISMS, [22:28] matid: this is over svn+ssh:// or over svn:// [22:28] ? [22:29] svn+ssh [22:30] hmm, that's kinda strange [22:31] it shouldn't be doing any auth stuff itself if you're tunneling over ssh [22:32] Yeah... [22:33] using svn 1.5 manually with svn+ssh works ok? [22:40] jelmer: Oh, right. [22:40] jelmer: It stopped to work :) [22:40] svn: Commit failed (details follow): [22:40] svn: Cannot negotiate authentication mechanism [22:44] matid: wow, that's odd - this works fine with svn 1.4 ? [22:44] jelmer: Yeah, it worked without any problems. [22:47] http://www.nabble.com/Crazy-svn%2Bssh-commit-regression-in-trunk-and-1.5-td15349239.html looks relevant [22:48] matid: in other words, configuring with --without-sasl may fix it [22:48] jelmer: Yeah, I've seen it. I'm in the middle of applying it as we speak. [22:50] hi there, i will love if anyone could give me a direction on how to upload project code in lauchpad to use bzr [22:51] Pretto, is your project already versioned with bzr? [22:51] beuno, no, well i think no [22:52] Pretto, this might help: https://help.launchpad.net/BzrHowto [22:52] beuno, thank you so much [22:53] Pretto, your welcome, and feel free to drop by with more questions :) [22:53] beuno, ok... i think i will have a lot of them [22:56] jelmer: Cool, works great now. [22:57] jelmer: I'm looking forward to the svn folks fixing svn+https anyway ;) [22:59] jelmer: Thanks for your help and see you! [23:00] matid: see r30564 in subversion trunk [23:23] beuno, i did it.. thank you [23:27] Pretto, congrats! :) [23:32] Where's the changelog for 1.4r1? I'm getting a zr: ERROR: Cannot lock LockDir(http://bamboo.fluther.com/repos/pandasite/trunk/.bzr/branch/lock): Transport operation not possible: http does not support mkdir() [23:32] Or, could that be related to the client being at 1.2? [23:33] Oh, wait, no, in this use can client and server are running 1.4 (since they're on the same machine) [23:33] s/can/case [23:36] awmcclain: I think you get that error if you try to push over http [23:36] Let me get the commnad that failed it... [23:39] ah [23:39] ok [23:39] after i do a lightweight checkout [23:39] over http [23:39] i'm doing a bzr mv [23:39] locally [23:41] Why would a local mv need to lock the remote server? [23:47] awmcclain: And you're committing that move, I assume? [23:47] No [23:47] jelmer: bzr mv /usr/local/src/panda-deploy/releases/20080412223856/panda/* /usr/local/src/panda-deploy/releases/20080412223856 [23:47] Oh, ok. That's weird then. So the "bzr mv" command is actually the one that raises the error? [23:47] bzr: ERROR: Cannot lock LockDir(http://bamboo.fluther.com/repos/pandasite/trunk/.bzr/branch/lock): Transport operation not possible: http does not support mkdir() [23:47] Yes [23:47] Correct [23:48] hmm, that indeed shouldn't happen [23:48] Any chance you can file a bug about this? [23:48] FTR this worked in 1.2 [23:51] New bug: #216533 in bzr "ERROR: Revision ... not present in ... trying to branch libsndfile repository" [Undecided,New] https://launchpad.net/bugs/216533