[01:58] <pygi> hello
[01:58] <pygi> trying to configure bzr smart server
[01:58] <pygi> I get this when I try to pull
[01:58] <pygi> branch that is
[01:58] <pygi> bzr: ERROR: Unknown branch format: "error\x01Generic bzr smart protocol error: bad request 'GET /libisofs-ng/.bzr/branch-format HTTP/1.1\\r'"
[01:59] <jelmer> pygi: Looks like you are using http to access it??
[02:00] <pygi> jelmer, yup
[02:00] <pygi> jelmer, why wouldnt I be able to do that?
[02:01] <jelmer> pygi: The smart server talks a custom protocol
[02:01] <jelmer> use the bzr:// branching scheme
[02:01] <pygi> jelmer, and no way I could make it use http? :P
[02:03] <jelmer> pygi: No, you can use any HTTP server if you want http
[02:04] <pygi> if I had access, then yes :)
[02:07] <jelmer> pygi: ?
[02:07] <pygi> jelmer, to apache I meant :)
[02:07] <pygi> dont worry, I can just use bzr :)
[02:08] <pygi> bzr protocol I meant :p
[02:09] <pygi> ok, one more question
[02:09] <pygi> I run bzr serve on a custom protocol
[02:09] <pygi> how do I do branching/etc from that port?
[02:13] <Odd_Bloke> pygi: 'bzr branch bzr://<host>:<port>/<path, I would expect.
[02:13] <Odd_Bloke> I haven't used bzr serve in months though. :p
[02:17] <pygi> hmmmm, connection refused
[02:17] <pygi> I'll have to look into it
[02:17] <pygi> thanks Odd_Bloke
[04:54] <abentley> spiv: ping
[07:51] <ubotu> New bug: #186285 in bzr "commit fails if the working tree contains an unknown file with undecodable name" [Undecided,New] https://launchpad.net/bugs/186285
[08:47] <lifeless> hi
[09:03] <i386> lifeless: hi
[11:54] <snod> morning, are there bzr 1.1 packages for ubuntu gutsy available? the repository listed on the site still has 1.0
[11:54] <dato> snod, try https://launchpad.net/~bzr/+archive/
[11:56] <snod> ah thanks alot :)
[15:35] <ubotu> New bug: #181773 in bzr-svn "Incorrect revision numbers in bzr:revision-id:v3-trunk1" [Undecided,New] https://launchpad.net/bugs/181773
[15:53] <jelmer> hmm, does bzr have problems with hardlinks?
[15:53] <jelmer> what sort of problems could a vcs have with hardlinks?
[16:14] <radix> jelmer: multiple hardlinks to the same file within one branch?
[16:14] <jelmer> radix: Why is that a problem?
[16:15] <radix> jelmer: I'm just asking if that's what you're doing.
[16:15] <jelmer> I somehow have the feeling I'm missing something obvious :-)
[16:15] <radix> It sounds like it could definitely lead to some weird behavior, but I don't know, it depends what you're expecting.
[16:15] <radix> jelmer: "hard links are scary" is the obvious thing :)
[16:15] <jelmer> (-:
[16:15] <dato> jelmer: do you mean hardlinks outside from .bzr only?
[16:15] <dato> outside of*
[16:15] <jelmer> dato: I'm adding support for bzr to etckeeper
[16:16] <jelmer> dato: and it complains for git and hg when it sees hardlinks
[16:16] <jelmer> I was wondering if it should do that for bzr as well
[16:17] <radix> Imagine two files linked to the same resource. Merge from a branch that changes each of them in distinct ways.
[16:17] <radix> hard links are generally a bad idea. :-)
[16:19] <snod> i just wonder: has anybody tried to use bzr over sshfs?
[16:20] <radix> snod: no, but I wouldn't want to, because sshfs is really unreliable (for me at least)
[16:20] <radix> snod: I assume you're talking about the FUSE-using filesystem
[16:20] <radix> I wonder why you'd want to use it instead of just pushing to bzr+ssh URLs
[16:22] <snod> i don't want to commit every tiny change
[16:22] <radix> tiny change?
[16:22] <radix> why would you need to do that?
[16:23] <snod> as far as i understand a push just pushed commited revisions?
[16:23] <radix> yes.
[16:23] <snod> *pushes
[16:23] <radix> (you can also 'bind', which creates a checkout, which means committed revisions are automatically pushed)
[16:24] <snod> so if i edit a file locally and have to try/execute/whatever it on a remote place i would have to do a commit for every change i do in the src file?
[16:24] <radix> ok, so you want to try/execute/whatever it on a remote place. that wasn't clear.
[16:25] <snod> :)
[16:25] <asabil> you will need a bzr update
[16:25] <asabil> on the remote side to update the tree
[16:25] <radix> asabil: well, he doesn't even want to commit in the first place, so update isn't going to help either :)
[16:26] <snod> i just ried bzr on an sshfs-mounted dir
[16:26] <snod> gives me an error: bzr: ERROR: exceptions.AttributeError: 'NoneType' object has no attribute 'abort'
[16:26] <radix> snod: well, like I said, I would avoid using sshfs because I've found it pretty unreliable, even for simple read-only stuff.
[16:26] <snod> i will keep that in mind :)
[16:27] <snod> (the error occurs on commit, init and add worked)
[16:32] <asabil> snod: why don't you want to commit ?
[16:32] <jelmer> wow, that was easy
[16:32] <jelmer> etckeeper support for bzr in < 30 minutes
[16:34] <snod> asabil: the changes are tiny like forgotten brackets or a misspelled word
[16:34] <asabil> snod: those deserve commits as well
[16:35] <asabil> snod: what you are trying to do will bring troubles upon you
[16:35] <asabil> and it is exactly what makes Visual Source Safe so ... Unsafe
[17:45] <ubotu> New bug: #159143 in bzr-svn "Handling of svn:author" [Wishlist,Triaged] https://launchpad.net/bugs/159143
[18:37] <vila> radix: I use bzr over sshfs for months without nay problems (using -o workaround=rename solved the only problem I ever encountered)
[18:43] <jelmer> hey vila
[18:43] <jelmer> vila: do you happen to know how to retrieve a configuration variable for a specific section?
[19:08] <vila> jelmer: from memory, it should be possible to query a ConfigObj object for a specific section and get a dict in return
[19:17] <jelmer> vila: Ah, yep
[19:17] <jelmer> Just found it
[19:21] <jelmer> hmm, is it possible to change the tree from a pre-commit hook?
[19:22] <jelmer> crap, apparentlynot :-(
[19:25]  * jelmer files a bug
[19:30] <ubotu> New bug: #186422 in bzr "Ability to modify the tree from a pre-commit hook" [Undecided,New] https://launchpad.net/bugs/186422
[19:38] <bronson> I have a legacy svn tree...   I'd like to check a bzr project into a branch in that tree.
[19:38] <bronson> Is there a good way of doing this?
[19:38] <bronson> Normally I want to manipulate svn trees in bzr...   This time, though, I want the opposite.
[19:45] <jelmer> bronson: I think "bzr svn-push" is what you're looking for
[19:46] <bronson> jelmer: hm, probably.
[19:46] <bronson> I'll have to read up on it.
[19:53] <bronson> jelmer: "bzr: ERROR: These branches have diverged. Use the merge command to reconcile them."
[19:53] <jelmer> bronson: You can't push to an existing branch
[19:53] <bronson> But that doesn't really make sense since this branch has never existed in svn.
[19:53] <bronson> The branch in svn is empty.
[19:53] <jelmer> bronson: Empty is an existing branch
[19:54] <bronson> Hm, I'm not understanding...  How am I supposed to set up the svn-push?
[19:55] <jelmer> bronson: from inside of the bzr branch, run something like:
[19:56] <jelmer> bzr svn-push svn://host/repository-path/branches/mynewbranch
[19:56] <jelmer> where svn://host/repository-path/branches exists but "newbranch" doesnt yet
[19:57] <bronson> Ah.  And then I populate mynewbranch with another bzr branch, then push it all back up to the host repo.
[21:44] <lifeless> moin
[21:51] <brink_> Does bzr split work?
[22:13] <jelmer> hey lifeless
[22:41] <igc> morning
[22:43] <abentley> lifeless: morning
[22:44] <lifeless> hey abentley
[22:44] <lifeless> public holiday here, so I'm not really around :)
[22:44] <abentley> When'd you get back?
[22:46] <abentley> brink_: yes
[22:48] <igc> hi lifeless (who isn't actually around) :-)
[22:48] <igc> hi abentley
[22:48] <abentley> Hi igc
[22:49] <igc> abentley: still in Oz?
[22:49] <abentley> Nope, got home yesterday.
[22:49] <igc> cool. Hope you had a good trip
[22:50] <abentley> It was good, and I'm glad I went.  I'm missing the warm weather.
[22:50] <igc> :-)
[22:50] <abentley> Went from a t-shirt to a fleece shirt and down jacket.
[22:55] <brink_> abentley:  I can use split to remove a directory, but the directory turns into a totally messed up branch.   I've tried every combination of upgrades from pack-0.92 I could think of.   None of them seem to work.  I'm using 1.1o on OS X.
[22:55] <abentley> In what way is it messed up?
[22:57] <brink_> It's messed up like this: bzr 1.1.0 on python 2.5.1.final.0 (darwin)
[22:57] <brink_> arguments: ['/opt/local/bin/bzr', 'commit', '-msplit']
[22:57] <brink_> encoding: 'US-ASCII', fsenc: 'utf-8', lang: None
[22:57] <brink_> plugins:
[22:57] <brink_>   difftools            /Users/joe/.bazaar/plugins/difftools [unknown]
[22:57] <brink_>   extmerge             /Users/joe/.bazaar/plugins/extmerge [unknown]
[22:57] <brink_>   launchpad            /opt/local/lib/python2.5/site-packages/bzrlib/plugins/launchpad [unknown]
[22:57] <brink_>   lessdiff             /Users/joe/.bazaar/plugins/lessdiff [unknown]
[22:57] <brink_>   multiparent          /opt/local/lib/python2.5/site-packages/bzrlib/plugins/multiparent.pyc [unknown]
[22:58] <brink_> *** Bazaar has encountered an internal error.
[22:58] <brink_>     Please report a bug at https://bugs.launchpad.net/bzr/+filebug
[22:58] <brink_>     including this traceback, and a description of what you
[22:58] <brink_>     were doing when the error occurred.
[22:58] <abentley> Please use a pastebin for that.
[22:58] <brink_> Sorry.  What's pastebin?
[22:58] <abentley> ubotu: paste
[22:58] <ubotu> pastebin is a service to post large texts so you don't flood the channel. The Ubuntu pastebin is at http://paste.ubuntu-nl.org (make sure you give us the URL for your paste - see also the #ubuntu channel topic)
[23:00] <brink_> Here is the error http://paste.ubuntu-nl.org/53801/
[23:02] <brink_> Basically, I did a bzr split <dir>, went into <dir> and did a bzr commit -m"split"
[23:03] <brink_> It seems like bazaar decided all the stuff from the root needed to be renamed into the subdirectory.   When I do a commit on it then it chokes.
[23:03] <abentley> brink_: I believe a bug is already registered for that.
[23:03] <abentley> I'm looking it up
[23:04] <brink_> Is there a work around?
[23:04] <abentley> bug 185211
[23:04] <ubotu> Launchpad bug 185211 in bzr "bzr commit fails on imported branch when renaming a directory with non-ASCII-name" [Critical,Confirmed] https://launchpad.net/bugs/185211
[23:05] <igc> bbiab
[23:06] <abentley> I'm not aware of one.
[23:06] <brink_> I do not believe that any of my characters are non-ascii.
[23:06] <abentley> The traceback is the same, isn't it?
[23:07] <brink_> That may be, but if so the title of the bug is incorrect.  My characters are all ASCII.
[23:10] <abentley> Well, I would say these look like two symptoms of the same underlying problem.
[23:22] <abentley> brink_: I can't reproduce the bug here, btw.
[23:22] <abentley> Not with split, I mean.
[23:43] <brink_> Not sure I can make a smaller example.  I'll give it a try.