[01:08] <maxb> james_w: Hi. If you're around, I'm staring at the complex comparison logic that the UDD importer uses to decide which order to do imports in, and am wondering why it is the way it is
[01:09] <maxb> I think I can blame the insane branching structure that results when unblocking a long-broken UDD import on this ordering
[01:24] <RenatoSilva> how to diff two branches? I remember using a command for checking if local copy was up-to-date with LP
[01:24] <RenatoSilva> is bzr pull enough?
[01:25] <spiv> RenatoSilva: perhaps you are looking for "bzr missing"?
[01:26] <RenatoSilva> spiv: that's it, thanks!
[02:04] <AfC> spiv saves the day
[03:53] <wallyworld> spiv: can you answer a question about branch revisions?
[03:53] <wallyworld> can you tell me, given a BranchRevision, the correct way to figure out if the revision was from merging in another branch and what that other branch is? Do you look for some attribute of any revision parents? i'm currently looking at the first parent and use getBranch() on that. is that the right thing to do?
[03:59] <spiv> wallyworld: Hmm
[04:00] <spiv> wallyworld: I'm not especially familiar with Launchpad's BranchRevision...
[04:00] <spiv> wallyworld: all bzr records is the DAG of revisions
[04:00] <wallyworld> the BranchRevision joins the branch to its Revisions
[04:01] <wallyworld> in bzr terms, what would you do?
[04:01] <spiv> There's no way in general to go from a revision to a branch that created it.  For example imagine there are two identical copies of a branch pushed to different locations on Launchpad.
[04:02] <spiv> If the tip revision is merged into another branch, which copy would you want to attribute as the merge-from branch?
[04:02] <wallyworld> hmmm. i see what you mean
[04:02] <wallyworld> i'm working on this bug 334336
[04:02] <spiv> In many cases in Launchpad I could imagine that there *might* be one most likely branch.
[04:03] <wallyworld> to answer your question, perhaps the original branch of the tip, not the one merged into
[04:03] <spiv> Which branch is the "original", though?
[04:03] <wallyworld> on seconds thoughts, maybe not :-/
[04:04] <wallyworld> is the bug reasonable then? or should we mark it as invalid?
[04:04] <spiv> I think it is reasonable in intent, but a bit vague.
[04:05] <wallyworld> yeah. i've got something working based on what i explained above. i'll run it by thumper next week and see what he says
[04:05] <spiv> Launchpad does know more facts than just the DAG, like perhaps that merge proposal existed for a branch with that revid
[04:05] <wallyworld> thanks for the input
[04:06] <spiv> I would say that *if* there is one branch on Launchpad whose tip is the revision you are interested in, then you have a pretty good answer.
[04:06] <spiv> And that will probably cover quite a few cases.
[04:07] <spiv> Otherwise maybe something like "the first time Launchpad encounters a revision as a mainline revision, Launchpad should take note of that branch and treat it as the 'original' branch for these purposes"
[04:08] <spiv> There are going to be lots of edge cases no matter what you do, I think.
[04:08] <spiv> (e.g. what if someone does "bzr push --overwrite" or "bzr uncommit" to partially rewind a branch?)
[04:09] <spiv> wallyworld: if you'd like a concrete use-case to guide your work
[04:10] <spiv> wallyworld: what I'd love to have is say I'm looking at the webpage for lp:bzr
[04:10] <spiv> wallyworld: I'd like the list of recent revisions there to link to the branches and even merge proposals corresponding to those revisions
[04:11] <spiv> wallyworld: (in bzr's case, almost all our mainline revisions on lp:bzr and our release branches do have corresponding merge proposals)
[04:12] <wallyworld> spiv: here's my prototype so far : http://people.canonical.com/~ianb/br.png
[04:12] <spiv> wallyworld: that would be nice because it gives me a single page that not only shows the recent changes to the branch, but also provides links to the discussions (in merge proposals, and bugs linked to branches) about those changes
[04:13] <spiv> wallyworld: yes, that sort of thing :)
[04:13] <wallyworld> does that look suitable? i could add in the mp too
[04:13] <wallyworld> so it's just figuring out how to get that "Merged From" branch
[04:13] <spiv> The MP might be good.
[04:14] <spiv> Although it's not too hard to navigate from a branch to a merge proposal, IIRC.
[04:14] <wallyworld> the mp is easy :-)
[04:14] <james_w> maxb, you mean comparing distros, releases, versions etc?
[07:53] <vila> hi all !
[12:15] <mthaddon> vila: do you know who filed RT#42981 (Some committers lack access to some bzr pqm branches)?
[12:33] <vila> mthaddon: yes, me
[12:34] <vila> anything wrong with it ?
[12:34] <mthaddon> vila: np - just couldn't work out who it was from - have replied
[12:37] <vila> mthaddon: oh yeah, indeed sounds sensible...thanks for posting the definitions, we could discuss it further but we will probably keep it this way
[12:37] <mthaddon> ok
[16:58] <dnivra> hello. i am trying to make bzr display a message but I can't display an email address-i get the following error http://paste.ubuntu.com/544937/. how can i fix it?
[17:19] <ovnicraft> hi guys i want to create a branch at revno explicit can pass some like bzr branch lp:proyect -r myrevno ?
[17:23] <vila> yes
[17:24] <vila> see 'bzr help branch' and bzr help revisionspec'
[17:25] <vila> see 'bzr help branch' and 'bzr help revisionspec'
[17:39] <ovnicraft> vila, yes but can help with an example
[17:39] <ovnicraft> ?
[17:39] <ovnicraft> i am trying bzr branch lp:proyect revno:num but create a folder with revno:num
[17:40] <vila> bzr branch -r 42 lp:bzr my-local-branch-at-revno-42
[23:38] <mkanat> So, I just filed bug 691756. Any ideas for a fix or workaround?