[05:48] <AfC> Yet another bad day trying to use Git.
[05:48] <bob2> it's not so complicated nowadays
[05:49] <AfC> bob2: I just discovered Git is incapable of tracking empty directories. #lame
[05:49] <bob2> yeah
[05:49] <bob2> it tracks files and symlinks only
[06:04] <felipec> AfC: at least it's capable of tracking more than one branch
[06:07] <bob2> all useful vc systems can do that though
[06:07] <felipec> bob2: and by branch I don't mean a repository
[06:08] <bob2> yes I know what a branch is
[06:08] <bob2> are you whinging about bzr-colo not existing or something?
[06:09] <felipec> bob2: I don't know what bzr-colo is, but I meant that bzr branch != git branch, more like git clone
[06:10] <bob2> ok!
[06:10] <bob2> anyway, bzr-colo or just use git
[06:11] <felipec> I'm looking at bzr-colo, and it looks like it's more like mercurial branches... still not the same
[06:11] <bob2> it's not
[06:11] <felipec> origin/master <- that I like
[06:12] <lifeless> felipec: bzr-colo is nearly identical to git's model of branches
[06:12] <lifeless> its not at all like mercurial
[06:13] <bob2> I think that's a complaint about branch identifiers not the branch model
[06:13] <felipec> lifeless: so, I can have master, devel, feature-a, origin/master, origin/next, bob/master, alice/feature-b, etc.?
[06:14] <lifeless> felipec: yes, they sit in a subdirectory of off the root
[06:14] <lifeless> and AIUI can have arbitrary names
[06:15] <felipec> lifeless: it's not about the name, it's about tracking a remote repository
[06:17] <felipec> it looks like there's support for a single remote repository, or workspace... I guess that's better than nothing
[06:17] <lifeless> felipec: each branch tracks a remote branch
[06:17] <lifeless> felipec: and there is a (not optimised) plugin to track entire repos
[06:18] <lifeless> gluing the two together should be fairly easy, if you want to do that
[06:21] <felipec> all right, my git-remote-bzr is looking good :) http://pastie.org/5179942
[08:19] <AfC> Yet another bad day trying to use Git.
[08:20] <AfC> I just discovered `git status` writes a file somewhere
[08:21] <AfC> Unreal
[08:22] <bob2> are you running it over fuse
[11:29] <AfC> bob2: no. It's .git/index.lock that gets {..., CLOSE_WRITE, CLOSE}.
[16:55] <Wiz_KeeD> hey guys
[16:55] <Wiz_KeeD> if i made a commit and i would like to change the title of the commit...how do i do that?
[17:04] <jelmer> Wiz_KeeD: bzr uncommit && bzr commit
[17:07] <Wiz_KeeD> aww i have to uncommit :(
[17:07] <Wiz_KeeD> i have to revert to that revission
[17:07] <Wiz_KeeD> uncommig
[17:07] <Wiz_KeeD> commit
[17:07] <Wiz_KeeD> then head back
[17:24] <jelmer> Wiz_KeeD: yep - you can't really edit existing revisions, just create new ones
[17:25] <Wiz_KeeD> heh
[17:25] <Wiz_KeeD> thanks man
[17:25] <Wiz_KeeD> i'll be more careful then
[17:28] <maxb> It would be quite nice to have a 'bzr commit --amend' that implicitly based its parent revision on the parent-of-the-parent and preloaded the old commit message in the editor
[17:30] <jelmer> indeed - there is a bug report open about that I think
[18:21] <SamB_MacG5> okay, what the heck ... I have to deal with stale locks on *launchpad* now?
[18:48] <jelmer> SamB_MacG5: only if connections actually got interrupted, etc?
[20:58] <SamB_MacG5> jelmer: I don't know how it came to exist, really ...
[20:58] <SamB_MacG5> but it was like a hundred days old
[20:59]  * SamB_MacG5 doesn't check his cron mail very often ...
[22:41]  * SamB_MacG5 thinks the SVN protocols ought to add a FAST-EXPORT <path> type of command ...
[22:43] <fullermd> If we just replace all SVN usage with some DVCS or other, we can bypass that whole problem   8-}
[22:44] <SamB_MacG5> yeah, but while people are still using various DVCSes with SVN repositories ...
[23:03]  * SamB_MacG5 discovers debianlp:; doesn't remember seeing that in the bzr docs ...
[23:47] <felipec> jelmer: I was able to import/export to/from bzr without bzr-fastimport: http://pastie.org/5186077
[23:50] <felipec> still missing a few features, but not bad for a couple days of work :)