=== thumper is now known as thumper-afk === thumper-afk is now known as thumper [08:05] hello [08:06] is ploblematic to use non-ascii in commiter name? [08:10] I'm experiencing the issue here: [08:10] https://github.com/felipec/git/issues/38 === mmrazik is now known as mmrazik|lunch === mmrazik|lunch is now known as mmrazik [12:24] jave_: nafaik [12:25] jave_: I'd suspect it is git-remote-bzr where the problem is [12:27] LarstiQ: ok, I delved some more and there seems to be some unicode/ascii conversion issue with git-remote-bzr [12:34] * LarstiQ nods [12:34] jave_: that sounds plausible === wedgwood_away is now known as wedgwood === mmrazik is now known as mmrazik|afk === iBasic is now known as BasicOSX [22:46] is this the official irc channel for bzr? [22:47] i pulled from a different branch, and now everytime i 'bzr update', i get 'bzr: ERROR: [Errno 1] Operation not permitted:...' for some (arbitrary, afaict) file. I rm the file, update again, and I get the same error for another file! [22:47] what could be causing this? [22:48] if I don't rm the file, and update again, I get the same error for the same last file [22:48] it's a huge repository, so if this goes on, it could take a while [22:52] reiserfs [22:54] errno 1 is (on my system anyway, but that low is probably common) EPERM, which would fit. But I can't think of any way on its own bzr could have put things in an EPERMish way. [22:54] The files are writable by $YOU? [22:56] yes. i'm rm'ing them. i can revert them too [22:56] rm'ing wouldn't prove it, since you don't need write on a file (or any perms, for that matter) to delete it. [22:56] And revert would be the same, since it doesn't strictly speaking edit the file; it rm's it too. [22:56] i don't own them, but i belong to the files' group, and they have write access for group [22:57] But doing an update would, so that's the salient difference... [22:57] ok.. trying to edit a file [22:57] (rm'ing a file requires you have write perms to the directory the file is in; the files perms are irrelevant. Hence why /tmp requires the sticky bit) [22:58] i can edit them [22:59] it seems all the problematic files have their permissions changed (-x) [22:59] Mmm. Well, bzr should be able to edit them on your behalf, then... [22:59] it's having difficulty chmod'ing them, it seems [22:59] That could be. Can't chmod a file you don't own. [23:00] ah [23:02] right. that seems to be it [23:03] i'm trying to manage the repository as a different user than the one using it [23:03] apparently it's not working [23:04] Yeah. Always tricky. [23:04] chown.. update completes without errors :) [23:04] With BSD fs semantics (or SysV and use of g+s), you can usually manage to pull it off, but dark corners like that require manual work... [23:11] how do i make sure i have a certain revision id checked out in my working tree? [23:12] revno was giving the revision no. although the update to that rev was not complete [23:12] revno tells you the revno of the head of the branch; revno --tree tells you what the working tree considers itself at. [23:13] and do i have to use another command to resolve the revno to revision id? [23:15] There's a revision-info command that shows both (and it has a --tree as well) [23:16] nice. does it detect a dirty tree? [23:17] Don't think it says anything about that. version-info can. === wedgwood is now known as wedgwood_away