[00:00] <wgrant> Or that.
[00:00] <epkugelmass> is it possible to create a branch that only certain people can access?
[00:00] <wgrant> But it's nothing now, so it's less obviously a template.
[00:00] <persia> epkugelmass: Last I heard that was only available as a commercial option, after asking a question against launchpad.
[00:00] <epkugelmass> persia, thanks.
[00:00] <wgrant> https://edge.launchpad.net/launchpad/+faq/208 may be handy.
[00:01] <wgrant> ... for those who wait.
[00:01]  * wgrant disappears to uni.
[00:02] <persia> Oh, the pricing and contact stuff has all been implemented.  Cool.
[00:37] <sbeattie> Hrm, despite the topic, I'm unable to browse the branch at https://code.launchpad.net/~kees/+junk/cpu-checker or bzr branch it.
[01:25] <sbeattie> And for the record, according to kees, the branch that I can't browse (https://code.launchpad.net/~kees/+junk/cpu-checker) was last pushed to around 8 hours ago.
[01:28]  * thumper looks at sbeattie's branch
[01:42] <thumper> sbeattie: yes the mirrored copy looks somewhat screwed, trying to work out how to fix
[02:14] <sbeattie> thumper: thanks, appreciated!
[03:21] <Peng> Anyone got advice for not accidentally revealing details about a private bug? They are marked, but it's not super-obvious, and I might forget... :-\
[03:21] <Peng> s/forget/forget to check.
[03:22] <Peng> It's probably not a big deal -- most bugs aren't that secret, and if they are, it's obvious from the subject matter -- but I dunno, I feel uncomfortable.
[04:05] <Andre_Gondim> Is there any way to change the behavior mail list, to when anyone no needs reply to all, only reply for the list?
[04:11] <lifeless> Andre_Gondim: thats a mail client setting
[04:11] <lifeless> ctrl-L in evolution
[04:11] <Andre_Gondim> lifeless, but in maillist I can set this option
[04:12] <Andre_Gondim> if reply to the sender or to the list, I just want to set reply to the list, not to sender
[04:12] <lifeless> oh, reply-to-mangling
[04:12] <lifeless> not on launchpad lists, thank $deity
[06:43] <poolie> https://edge.launchpad.net/bzr is timing out repeatedly :/
[06:44] <wgrant> poolie: Known. i think the fix is committed.
[06:44]  * wgrant checks.
[06:46] <wgrant> Ah, no, not fixed yet.
[06:46] <wgrant> Bug #534462
[09:30] <donri> if i move away from github i'm going to miss the wiki and web host service for documentation; is there a good place to get that for open source projects? sorry if it's off-topic.
[09:58] <soren> donri: I know of a few projects that use wikia.com. I don't know if it's any good, though.
[09:59] <idnar> http://pbworks.com/ is what I seem to be using for most of my projects
[09:59] <donri> isn't wikia filled with advertisement, and only wiki?
[10:00] <idnar> I think it's only free for non-commercial use, though
[10:04] <soren> donri: Sorry, I though you were asking for a wiki.
[10:05] <idnar> soren: he wanted wiki + document hosting, I think
[10:05] <soren> idnar: It's hardly uncommon for wiki's to hold documents :)
[10:06] <donri> by documentation i mean static html, properly served as such (not "download file via some interface")
[10:09] <donri> and deployable via command line (ftp, scp, rsync, some vcs)
[10:18] <donri> like, i could use packages.python.org for that if the project is python-related, but i'm looking for something more general.
[10:18] <donri> (wiki is a secondary need.)
[11:29] <donri> how does blueprints differ from bugs?
[11:40] <jelmer> donri: Blueprints are used for creating specification, where there's design work that needs to happen.
[11:40] <jelmer> *specifications
[11:41] <donri> still not sure how that differs from bugs
[11:50] <jelmer> donri: for bugs it's generally very clear what the problem is and what needs to be fixed, for features it's usually less clear. What does the UI need to look like, what use cases need to be handled, etc.
[11:51] <idnar> ugh
[11:51] <idnar> how would I get all of the bugs targetted at a milestone via the API?
[11:52] <donri> i don't really get how that differs practically
[11:54] <idnar> ah, looks like searchTasks might be what I want
[11:55] <idnar> donri: I'd say that blueprints are for things that are larger in scope than will comfortably fit in a single bug
[11:55] <idnar> there might be multiple bugs associated with a blueprint, or there might just be a lot of design / specification work that doesn't comfortably fit into a bug
[11:56] <idnar> the split is kinda awkward, though
[11:56] <idnar> especially since blueprints can't actually be stored on launchpad
[11:56] <donri> huh
[11:57] <jelmer> donri: the actual blueprints are usually on a wiki page, but there is an entry in Launchpad to make tracking their status easier
[11:57] <donri> i'm only talking about the launchpad feature though
[11:58] <donri> unrelated to ubuntu specs
[11:58] <idnar> donri: I was also only talking about the launchpad feature
[11:59] <wgrant> Note that the Launchpad feature was initially very very heavily modelled on the Ubuntu spec process, and has remained largely unchanged since.
[11:59] <donri> so it basically assumes that you complement it with actual specing elsewhere?
[12:05] <jelmer> donri: yeah, it mainly provides you with an easy way to track specs
[12:05] <donri> aye, thanks
[12:12] <persia> For some value of "easy" :)  Take care not to get more than ~200 specs, or you'll get lost.
[12:14] <idnar> heh
[12:14] <persia> Also, take care not to close specs by accident: they become very hard to find afterwards.
[14:00] <idnar> lazr.restfulclient.errors.HTTPError: HTTP Error 412: Precondition Failed
[14:01] <idnar> why am I getting that from launchpadlib?
[14:07] <thekorn> idnar, there are a few bugreports about this issue, which have a few workarounds, let me try to find one
[14:08] <idnar> I'm reading a few of those now
[14:08] <idnar> I tried throwing in an lp_refresh(), which didn't seem to help
[14:09] <idnar> ah, I think I found the problem; I was setitng .status on a bug task directly, instead of calling transitionToStatus
[14:10] <thekorn> idnar, yes, transitionTo* will work, because it is using POST methods, seems like PATCH is affected by this bug
[14:10] <idnar> hmm, I spoke too soon, that worked a few times but then gave me the same error again
[14:12] <idnar> so, not really sure how to work around this. my code is here: http://paste.ubuntu.com/391783/
[14:13] <thekorn> idnar, you don't need lp_save() when using task.transitionToStatus()
[14:13] <geser> I'm not sure, but do you need to call .lp_save() when using .transitionToStatus?
[14:14] <Breaking_Pitt> One question
[14:14] <Breaking_Pitt> I need to install launchpad
[14:14] <idnar> thekorn: oh, I see
[14:14] <Breaking_Pitt> but i have created the ssh keys to be sent to launchpad in one mchine
[14:14] <Breaking_Pitt> but i have to isntall it in other machine
[14:15] <Breaking_Pitt> i have to generate new ssh keys again?
[14:15] <idnar> I just read bug 534066
[14:15] <idnar> okay, yeah, got rid of lp_save() and now it works fine
[14:15] <thekorn> idnar, and you should also use   milestone.searchTasks(status=["Fix Committed",])    to not iterate over all (open) tasks for this milestone
[14:16] <idnar> thekorn: excellent point
[14:16] <deryck> intellectronica, did we get james_w's fix for the above bug deployed again?
[14:17] <intellectronica> deryck: nope. it bounced because of an unrelated test failure and i just got it finally landed.
[14:17] <intellectronica> deryck: i will arrange a cp for both this fix and the heat display one
[14:18] <deryck> intellectronica, excellent.  thanks!
[14:18] <thekorn> great! - good news
[14:28] <persia> jml: After watching traffic in both #launchpad and #launchpad-devel for a week or so, I think you *do* have sufficient traffic (on discrete topics) to benefit from two channels.
[14:28] <jml> persia, ok. :)
[14:29] <persia> (in reply to your comment last week, which I then didn't have enough data to answer)
[14:29] <jml> persia, I have this feeling that experienced Launchpad developers watch this channel less.
[14:30] <jml> persia, but I'm glad to have an opinion from a seasoned external observer :)
[14:30] <persia> jml: I suspect that the seasoned developers *do* watch it less, but I suspect that there would be less peer-support between users was all the -devel traffic here.
[14:30] <persia> (and I believe that this peer-support is critical to community development)
[14:31] <jml> persia, I agree.
[14:37] <jml> btw, I just posted http://code.mumak.net/2010/03/get-started-with-launchpadlib.html
[14:38] <jml> some people might find it useful
[14:44] <jelmer> Breaking_Pitt: hi
[14:44] <jelmer> Breaking_Pitt: You should be able to copy the ssh keys between machines (they live in ~/.ssh/).
[14:44] <Breaking_Pitt> hello jelmer
[14:44] <Breaking_Pitt> ok
[14:44] <Breaking_Pitt> thanks
[14:44] <Breaking_Pitt> with this answer
[14:44] <Breaking_Pitt> I can continue
[14:44] <Breaking_Pitt> thank you
[14:46] <jelmer> You're welcome
[15:01] <Breaking_Pitt> jelmer, I have copied the ssh keys from the machine in where I've setup the first install of Launchpad and don't seem to work now
[15:01] <Breaking_Pitt> some advice please?
[15:02] <jelmer> Breaking_Pitt: Does ssh-add -l list your keys correctly?
[15:02] <Breaking_Pitt> seems that not
[15:02] <Breaking_Pitt> Could not open a connection to your authentication agent.
[15:23] <RobOakes> Hi, does anyone know how to continue a bzr push request?  I was making an initial commit to launchpad with a big SVN repository (23990 revisions).  The bzr client stalled halfway through and I *really* don't want to start over from scratch.
[15:24] <RobOakes> The error message says something about finish_writing / finish_reading ...
[15:24] <RobOakes> The initial error was: The medium 'SmartSSHClientMedium(connected=True, username=u'robertsoakes', host='bazaar.launchpad.net', port=None)' has reached its concurrent request limit. Be sure to finish_writing and finish_reading on the currently open request.
[15:26] <mrand> https://code.launchpad.net/~mrand/mythplugins/fix-repeated-file-ext-511653/+merge/20969/+preview-diff/+files/preview.diff is timing out... gives "Please try again" constantly.  All other pages work fine, both before and after repeatedly trying that.
[15:26] <pmatulis> will a PPA always be updated by apt when the same software/package appears in -updates?
[15:27] <persia> pmatulis: Not at all.
[15:27] <persia> pmatulis: A PPA is updated if/when someone uploads to it.  This is unrelated to the local system.
[15:28] <pmatulis> persia: so a PPA is really like its own package then, according to how it is named?
[15:28] <noodles775> What he said :), Note, the ppa index will display "There is a newer version of this package available", but the users system will always download the latest that it finds via apt's sources.
[15:30] <persia> pmatulis: A PPA is it's own *archive* containing any number of packages.
[15:31] <persia> noodles775: As a refinement: any apt instance will grab the newest it finds, even if it's not associated with a specific host (e.g. someone playing with quinn-diff)
[15:32] <noodles775> Aha.
[15:33] <pmatulis> to be clear, packages in a PPA remain independent of non-PPA packages?
[15:33] <maxb> RobOakes: I don't _think_ bzr push is resumable, but you might want to ask #bzr before you take my word for it
[15:34] <persia> pmatulis: Entirely.
[15:34] <pmatulis> persia: thx
[15:34] <RobOakes> Thanks, I'll try them.  Do you know if there is a way to do a shallow push?  All I really need is the most recent revision.  The repository has a 15 year version history, and that's just overkill for what I'm doing.
[15:34] <persia> pmatulis: Note that this relationship is complicated when instantiated in any given apt-capable environment, as those tend to pull from several sources.
[15:37] <pmatulis> persia: can you give me a concrete example of such a complication?
[15:40] <persia> pmatulis: Two archives are enabled.  One has foo_1.1 the other has foo_1.1+ppa1.  foo_1.1+ppa1 is the install candidate.  The first archive gets updated with foo_1.2.  foo_1.2 is now the install candidate, and any changes in foo_1.1+ppa1 are no longer present, unless they were merged into foo_1.2
[15:44] <pmatulis> persia: alright.  now such a scenario would only ever arise after a Ubuntu release upgrade right?  since Ubuntu doesn't upgrade major versions within a release
[15:46] <persia> pmatulis: Same rule applies for 1.2.3-4ubuntu5 getting increased when someone has 1.2.3-4ubuntu5+ppa1 : either the critical bugfix or the PPA changes won't be applied (which depends on how apt compares versions, which you can test with dpkg --compare-versions)
[15:49] <pmatulis> persia: basically i don't want a PPA package updated.  so the safe way is to compare versions prior to any upgrade
[15:49] <pmatulis> (updated by a package in 'main' or 'universe' for example)
[15:49] <persia> pmatulis: You installed a PPA package, and you want it never overridden?
[15:50] <pmatulis> persia: i will want to do this, yes
[15:50] <persia> pmatulis: OK.  Just don't upgrade until the PPA has merged each change.
[15:50] <persia> (or don't upgrade that package)
[15:50] <persia> You can't force this on users without doing things that can break their systems.
[15:51] <persia> If you're in a controlled environment (large deployment), having a custom repository containing only approved changes is the usual model.  This is also sometimes used for commercial deployments (e.g. preinstalls)
[15:52] <pmatulis> persia: right, thanks
[18:54] <Laney> lamont: please kill https://launchpad.net/ubuntu/+source/ghc6/6.12.1-12/+build/1526050 - it's obviously hung
[18:55] <Laney> is it reasonable for soyuz to grow per package/arch timeout support as wanna-build has?
[20:53] <nailora1> i cannot find the bug report about bug watches for bugzilla.gnome.org being disabled / causing problems. do you know where it is?
[22:29] <stani> is something wrong with launchpad translations? I have requested to download the po and mo files more than 10 hours ago and still did not receive anything.