[00:54] <vila> pl0sh: branches will automatically find a repo above them in the file system hierarchy, if the repo is on a server, readers/writers client need to be able to access it on the server
[00:56] <vila> pl0sh: this is valid for all protocols between the client and server. bzr+ssh is the most efficient implementation these days and as such the preferred one if bzr is available on the server
[00:57] <vila> pl0sh: which is true for freebsd
[13:46] <masklinn> using fast-export/fast-import to convert a repo from hg to bzr seems to flip around a number of merge commits, resulting from a "mainline" (log withou -n0) composed of the commits of the merged branches, rather than the commits from trunk/default/whatever. Is this a known issue? Is there a way to fix it?
[13:47] <masklinn> the repository looks correct on the hg side, with the "trunk" being the straight mainline and the rest branching from/merging into it.
[15:47] <maxb> masklinn: That seems pretty weird. You should probably begin by finding out if the flipping has occurred in the export or the import - i.e. is the fastimport data stream correct
[16:01] <masklinn> maxb yep, checked further (by trying to trip through git as well), the issue is at the export from hg (I'm guessing bzr's hg export script and the hg-fast-export one for git floating around have the same origin, they yield the same flipped merge commits)
[16:02] <masklinn> So I tripped through git via hg-git before importing back into bazaar, and that worked
[18:00] <lode> Hi
[18:00] <lode> Anyone here knowing where to suggest a new translation / report a translatopn bug in the bzr explorer diff tool?
[20:22] <poolie> hi lode, hi vila
[21:37] <jam> hi poolie, good early morning to you
[21:49] <lifeless> hi jam
[22:26] <mathrick> hiya
[22:26] <mathrick> mathrick@hatsumi:[...]/BMB/tools/lib-overrides$ bzr join cl-gtk2/
[22:26] <mathrick> bzr: ERROR: File id {TREE_ROOT} already exists in inventory as InventoryDirectory('TREE_ROOT', u'cl-cairo2', parent_id='toolsliboverrides-20110525170907-q1x2hf7rrxb0i21i-1', revision=None)
[22:26] <poolie> hi there
[22:26] <poolie> jam, do you want to have a chat now
[22:26] <mathrick> I get that when I try to join a repo after having previously joined another one
[22:27] <poolie> mathrick: hi, this is a bug
[22:27] <mathrick> is there anything I can do to work around this?
[22:27] <poolie> i recognize it; I don't know a good workaround
[22:27] <poolie> do you, jam?
[22:27] <poolie> hm
[22:27] <mathrick> poolie: ah, pity. Is it known and recorded somewhere?
[22:27] <mathrick> somewhere being lpad of course :)
[22:29] <poolie> https://bugs.launchpad.net/bzr/+bug/370710
[22:29] <mathrick> thanks
[22:29] <poolie> so there is a workaround there, though not an ideal one
[22:32] <mathrick> poolie: hmm, the repo is pulled from github, is it expected then that it'll have a {TREE_ROOT} id?
[22:32] <poolie> through bzr-git/
[22:32] <mathrick> yes
[22:33] <jam> hey poolie. I'm around a bit, but getting close to midnight here. So if it isn't too involved, certainly
[22:33] <poolie> is it imported by launchpad, or by you?
[22:33] <poolie> oh, hi
[22:33] <poolie> i was confused
[22:33] <poolie> i think i momentarily forgot you weren't still in Chicago
[22:33] <jam> :)
[22:33] <poolie> can we set a time for this week, maybe your Tuesday morning?
[22:34] <jam> poolie: whatever time works for you. Tues AM is good I think
[22:34] <mathrick> poolie: me, I did bzr clone git://...
[22:35] <poolie> is 8 too early?
[22:36] <poolie> mathrick: hm i would have thought if you did that into a format 2a repository it would get a unique root id
[22:36] <poolie> is it a very old import?
[22:37] <mathrick> poolie: yes and no, I first imported my old github branch which was made using bzr-git quite some time ago
[22:37] <poolie> hm actually that'd be 9am
[22:37] <mathrick> but since I don't need that (I can grab the upstream directly and not pull on top of my own github branch), I'm trying that now
[22:40] <mathrick> poolie: hmm, no, I still get it
[22:40] <mathrick> mathrick@hatsumi:[...]/BMB/tools/lib-overrides$ bzr info cl-gtk2
[22:40] <mathrick> Standalone tree (format: 2a)
[22:47] <jam> poolie: 9am should work, I usually get back around then after taking K to school. I'm off to bed for now, have a good day
[23:00] <mathrick> poolie: any idea how I could look into the tree root not getting a proper id?
[23:29] <poolie> mathrick: so normally the initial add will give it a unique id
[23:29] <poolie> in bzr 2a formats and onwards
[23:29] <mathrick> poolie: initial add?
[23:30] <poolie> the first 'bzr add' after 'bzr init'
[23:30] <poolie> i don't know if bzr-git is perhaps overriding that
[23:30] <mathrick> hmm
[23:30] <poolie> or if it's just that you first made this branch earlier
[23:30] <mathrick> has jelmer disappeared? I haven't seen him in a long time
[23:30] <poolie> hm, it may well be overriding it to get deterministic imports
[23:30] <mathrick> poolie: I retried on a fresh clone and the same happens
[23:30] <poolie> he was on holiday last week and today (monday)
[23:31] <poolie> and the week before that we were sprintig
[23:31] <poolie> he would be the person to ask
[23:42] <mathrick> poolie: right, but I didn't know about the holiday (though I kinda suspected that)