[00:42] <Zaery> How do i download a build such as https://launchpad.net/~ubuntu-wine/+archive/ppa/+buildjob/2127712
[00:42] <Zaery> ?
[00:54] <wgrant> Zaery: There are usually links to files in the "Built files" section, but we delete PPA packages a week after they are superseded.
[00:54] <wgrant> So old ones like that are no longer available.
[00:55] <Zaery> oh well, then i'm off to testing some newer versions
[00:55] <wgrant> It's unfortunate, but to keep them all forever would require a *lot* of disk space.
[00:55] <Zaery> yeah, i understand
[00:56] <jfi> Zaery: if you want the last build and stay uptodate, you can simply add the ppa
[00:58] <jfi> Zaery, "sudo apt-add-repository ppa:ubuntu-wine/ppa" will add the repository to your system, then you can get the packages with the software center
[02:04] <micahg> I have a build faillure w/out a build record on armel, do you need the logs for this or can I just retry?
[02:04] <wgrant> micahg: openjdk, or something less evil?
[02:05] <micahg> wgrant: xulrunner-1.9.2
[02:05] <wgrant> micahg: Do you have a link?
[02:05] <wgrant> openjdk took out a few builders last night.
[02:05] <micahg> https://launchpad.net/ubuntu/+source/xulrunner-1.9.2/1.9.2.14+build3+nobinonly-0ubuntu2/+buildjob/2299489
[02:06] <wgrant> micahg: Retry it.
[02:06] <micahg> wgrant: thanks
[02:06] <wgrant> Doesn't look like it should cause another builder massacre.
[02:06] <wgrant> And if it does, I can hopefully resolve it.
[02:07]  * micahg actually needs to upgrade a new version, but let's see if the rebuild works
[02:07] <micahg> *upload
[06:12] <foxbuntu> Any LP admins around to help me out with an issue?
[06:14] <wgrant> foxbuntu: No admins around during the weekend.
[06:14] <wgrant> What's the issue?
[06:14] <wgrant> I have limited superpowers.
[06:15] <foxbuntu> wgrant, well I had some malware on a p.o.s. winbloz machine I fixed as a favor take control of one of my google accounts a break some things and DaveMorris suspended my LP account to prevent it spamming
[06:16] <foxbuntu> just wondering if I can get it unlocked now that I have the security issues resolved
[06:16] <wgrant> Ah, yes, I heard about that.
[06:16] <wgrant> I can suspend accounts, but not unsuspend them. You'll need to ask back here on Monday, or email feedback@launchpad.net.
[06:16] <foxbuntu> yeah...very ugly stuff
[06:16] <wgrant> Glad to hear you got it resolved, though.
[06:17] <foxbuntu> yeah...teach me to help others with their junk PCs...
[06:17] <foxbuntu> ;)
[06:17] <foxbuntu> Thanks wgrant, I will get in contact with someone post weekend
[08:09] <trijntje> Hi all, is there a way to review suggestions done in launchpad offline? For example to get them commented out in a .po file?
[08:10] <trijntje> oops, wrong channel, ill add some background
[08:11] <trijntje> I'm a translator for Ubuntu and I would really like to be able to review suggestions done in launchpad offline, is there a way to do this?
[09:55] <ikt> heya, can anyone tell me why it's importing my code and saying that a team that I'm in is the one changing the code?
[09:55] <ikt> https://code.launchpad.net/~ikt/ium/trunk
[10:00] <jml> ikt: email address matching
[10:01] <jml> ikt: the email address used in the commit is the email address of that team.
[10:01] <ikt> yeah I thought so
[10:01] <ikt> I changed it
[10:01] <ikt> to ts@ikt.id.au
[10:01] <ikt> from ikt@ikt.id.au
[10:02] <ikt> it hasn't updated :s
[10:02] <jml> ikt: I don't think it will.
[10:02] <jml> ikt: it makes the links on-import
[10:02] <jml> ikt: future commits will be different, though.
[10:03] <ikt> yeah I changed the email address, then uploaded a file, imported it, and it still thinks it's the same :s
[10:03] <ikt> delete branch and try again?
[10:03] <jml> ikt: sure. why not?
[10:08] <ikt> jml, nope still says it :/
[10:15] <wgrant> jml: revisionauthor :(
[10:15] <jml> wgrant: it caches?
[10:17] <wgrant> jml: As they are imported, the revision's author field is parsed and it's linked to a RevisionAuthor. Each RevisionAuthor is linked to a Person as one with a matching email address is seen.
[10:17] <wgrant> I don't believe there's any facility to change existing ones when an address moves or is delted.
[10:28] <ikt> I don't understand why it linked it in the first place
[10:29] <ikt> my name is the same on github as it is on lp, and I'm the one who setup the link and the series
[10:34] <wgrant> ikt: It uses the email address that you committed with.
[10:36] <ikt> gah :(
[10:37] <wgrant> Hm? That's what it should do.
[10:38] <wgrant> But it should also know to remove the cached link when you remove the email address from a team.
[10:46] <ikt> wgrant, indeed, going to open a bug
[10:47] <ikt> what does it say if the email address that it is committed with isn't on lp?
[10:47] <wgrant> ikt: It displays the name, without a link.
[11:01] <wgrant> jml: Still around?
[11:05] <jml> wgrant: yeah
[11:05] <jml> wgrant: what's up?
[11:06] <wgrant> jml: Any chance you could time a query on staging for me?
[11:06] <jml> wgrant: yeah sure.
[11:06] <jml> wgrant: EXPLAIN ANALYZE?
[11:06] <wgrant> EXPLAIN ANALYZE SELECT COUNT(revisionauthor) FROM revisionauthor LEFT JOIN emailaddress ON emailaddress.email = revisionauthor.email WHERE revisionauthor.person != emailaddress.person;
[11:06] <jml> got it.
[11:06] <jml> wgrant: http://paste.ubuntu.com/576399/
[11:07]  * jml makes a note to set up a pipe-driven staging db convenience thingummy
[11:11] <wgrant> jml: Thanks.
[11:11] <wgrant> That's quick enough to run in garbo, so this is an easy fix!
[11:16] <wgrant> Evening lifeless.
[11:17] <lifeless> hiya
[15:47] <aroman> how long does it take for users to get upgrades visible when changing a package in a PPA?
[15:52] <hangfire> Launchpad slow for anyone else?
[16:03] <DaneAU> hangfire, definitely is for me
[16:19] <aroman> Please try againSorry, there was a problem connecting to the Launchpad server.
[16:36] <hangfire> Well, got through eventually :)
[17:36] <ti4mi> hello,
[17:36] <ti4mi> how can I have two versions of the same package in my PPA?
[17:36] <ti4mi> I would like to have a stable and a development package...
[18:56] <jelmer> ti4mi: You can't have two versions of the same package in a PPA - create two PPA's
[19:05] <magcius> where's the bug for slow rendering on firefox+nvidia?
[19:45] <arand> What are the p-series and o-series I see in the "copy packages" for PPAs?
[19:50] <jelmer> arand: it's a bug they're showing up there
[19:50] <arand> Ah, right :)
[19:50] <jelmer> arand, you can't actually copy to those series yet. The names o-series and p-series are used because the codenames for Ubuntu O and Ubuntu P haven't been decided yet
[19:50] <jelmer> we have already created those distro releases for ubuntu so bugs can be targetted to them
[20:16] <dotwaffle> something up with launchpad right now? (nothing in the topic, i notice)
[20:18] <lifeless> no
[20:18] <lifeless> bah
[20:22] <Ampelbein> lp seems a bit slow, that is all.
[20:23] <lifeless> any page in particular ?
[20:24] <Ampelbein> https://bugs.launchpad.net/ubuntu/+source/compiz-plugins-main for example
[20:24] <Ampelbein> it has only 2 open bugs but takes a good 10 seconds (maybe more) to load
[20:25] <lifeless> rendered for me in 1.06 seconds
[20:25] <lifeless> checking latency now
[20:25] <lifeless> 1.4 seconds to transmit - so we've got some in-dc queueing
[20:26] <lifeless> (total of 1.4 seconds I mean)
[20:26] <lifeless> Ampelbein: this is probably the overload we're already aware of; I'm working on further config changes to handle this now
[20:27] <Ampelbein> lifeless: ok, i just tested with wget: http://paste.ubuntu.com/576614/ 43 seconds
[20:28] <Ampelbein> most of the time spent at Waiting for Response
[20:28] <lifeless> Ampelbein: are you logged into launhcpad ?
[20:29] <Ampelbein> lifeless: yes
[20:29] <lifeless> so, I see the massive latency with wget
[20:29] <lifeless> not with chromium
[20:30] <lifeless> and its fast in wget now
[20:31] <Ampelbein> lifeless: yes, now its 0.275 seconds ;-)
[20:31] <lifeless> adding no-cache -> slow
[20:32] <Ampelbein> lifeless: wasn't the rendering time reported on the page when there still was edge.lp?
[20:33] <lifeless> Ampelbein: the render time is in the body of the page
[20:33] <lifeless> Ampelbein: developers that are logged in get it in the top right of the page near their username link
[20:35] <Ampelbein> lifeless: as a non-developer I have to look at the soure: 'At least 91 queries/external actions issued in 1.73 seconds'
[20:47] <lifeless> Ampelbein: yes
[20:48] <lifeless> Ampelbein: eventually we may let non devs turn this on voluntarily, but its a bit geeky to have the in main UI for everyone
[20:48] <lifeless> Ampelbein: this is roughly what I see - http://people.ubuntu.com/~lifeless/showtime.png
[20:48] <lifeless> (thats an older font/style - but sample place)
[20:48] <Ampelbein> lifeless: I can understand that. If I really wanted I could make a greasemonkey script to add the info.
[20:49] <Ampelbein> but for a non-developer it's mostly unimportant information
[21:42] <lvh> Hello!
[21:43] <lvh> I have a problem :-(
[21:43] <lvh> People want me to use Github.
[21:43] <lvh> This is most unfortunate.
[21:43] <lifeless> hi
[21:43] <lvh> But I have found a potential solution, and it seems quite obvious so perhaps someone else has tried something similar already.
[21:43] <lifeless> is this for your project, or their project?
[21:44] <lvh> lifeless: Mine.
[21:44] <lifeless> its not really up to them then is it ? :)
[21:44] <lvh> First of all, trunk commits (lp:someproject) get mirrored to Github. People contribute code preferably on Github in the form of pull requests.
[21:44] <lvh> lifeless: Well there's two problems there.
[21:44] <lvh> lifeless: 1) I have about 6 to 12 months to live so bus numbers are nice things to have high values for.
[21:44] <lifeless> lvh: oh :(
[21:44] <lvh> lifeless: 2) A recent discussion on Convore has shown that some people really really really dislike Launchpad.
[21:45] <lvh> Up to the point that they will quite happily ignore a project that does what they want.
[21:45] <lvh> Because it's not on Github.
[21:45] <lifeless> lvh: I'm curious what things they really dislike about LP
[21:45] <lvh> Now, of course, that's stupid, because Github doesn't have a workable issue tracker.
[21:45] <lvh> lifeless: I keep hearing the same things and I am compiling them into a blog post.
[21:45] <lvh> It's mostly UI.
[21:46] <lvh> And people disliking loggerhead.
[21:46] <lifeless> I'm glad you're writing it up.
[21:47] <lifeless> we're (obviously) working on improving things
[21:47] <lvh> So anyway, the idea is that I use Github's API to detect pull requests, and then I create a branch on lp from trunk plus that diff
[21:47] <lvh> And initiate a merge proposal.
[21:47] <lifeless> sounds nice
[21:48] <lvh> I know, right!
[21:48] <lvh> the best part is that the github users don't bother me
[21:48] <lvh> Something on which I've also ranted
[21:48] <lvh> I'm sick and tired of people who have only ever used git their entire life telling me how much everything that isn't git sucks
[21:48] <lifeless> hmm, I've done everything I planned to do today and its only 11am :(
[21:49] <lvh> (also because most of the time I know a git a helluva lot better than them)
[21:49] <lifeless> lvh: amen
[21:49] <lvh> lifeless: Clearly you need a better issue tracker!
[21:49] <lvh> lifeless: I met pivot tracker today
[21:49] <lvh> I could see it working for small teams working consistently on a project
[21:49] <aroman> why did my package that builds perfectly on lucid, fail to build when I changed it to maverick in debian/control? http://launchpadlibrarian.net/65759906/buildlog_ubuntu-maverick-i386.elementary-wallpapers_0.2.1_FAILEDTOBUILD.txt.gz
[21:49] <lvh> Open source typically has more contributors and more randomness in contribution patterns though
[21:49] <lifeless> lvh: interesting
[21:50] <lifeless> aroman: Found files in /usr/lib/python2.6/site-packages (must be in dist-packages for python2.6).
[21:50] <lifeless> debian/elementary-wallpapers/usr/lib/python2.6/site-packages
[21:50] <lifeless> aroman: from that log
[21:50] <lifeless> debian/elementary-wallpapers/usr/lib/python2.6/site-packages/elementary_wallpapers-0.2.1-py2.6.egg-info
[21:50] <aroman> lifeless: what exactly does that mean?
[21:51] <lvh> lifeless: http://b.lvh.cc/why-do-people-hate-launchpad-so-much
[21:51] <lifeless> aroman: just what it says
[21:51] <lvh> lifeless: If you want I'll ping you when I finish the followup article.
[21:51] <lifeless> lvh: I'd love that
[21:51] <aroman> lifeless: :/ what should I do to rectify the situation?
[21:52] <lifeless> aroman: change your package to install into the dist-packages dir not site-packages
[21:52] <lifeless> aroman: how you do that will depend on how you're doing the packaging
[21:52] <aroman> hmm okay
[21:52] <lvh> lifeless: Unfortunately not all feedback is useful
[21:52] <aroman> lifeless: why would that change lucid>maverick, do you think?
[21:52] <lvh> lifeless: It's mostly like RAAAARRR THEIR UI IS FREAKIN' TERRIBLE
[21:53] <lifeless> aroman: python 2.5 -> 2.6
[21:53] <lvh> without any mention of what it does poorly in particular
[21:57] <lifeless> lvh: http://b.lvh.cc/designing-a-continuous-deployment-system-caut is interesting, nicely put
[21:57] <lvh> lifeless: I'm talking about it at Pycon.
[21:57] <lvh> lifeless: http://paste.pocoo.org/show/349175/
[21:58] <lvh> lifeless: chatham house rule applies etc
[21:58] <lifeless> lvh: we have safeguards that don't transition a server if it doesn't come up properly - so we may wedge during the deploy, but we don't generally end up totally down
[21:58] <lvh> lifeless: Interesting
[21:59] <lvh> lifeless: I've seen many tricks to implement those
[21:59] <lvh> They're mostly app-speciifc and I'm hoping to come up with useful generalizations for them
[21:59] <lvh> lifeless: In that paste the first two paragraphs are useful but the last one doesn't add anything
[22:00] <lvh> It's just "I had a really bad experience with LP" but no details what where when etc so no ide awhat actually happened and how we could fix it
[22:00] <lifeless> lvh: thats a great paste; I'd like to show this stuff to our product team
[22:00] <aroman> lifeless: fascinatingly, the maverick version of the package that one that failed to build was based off of (ubuntu-wallpapers) has absolutely no changes that might change what you described in it's debian folder
[22:01] <aroman> so i really dont know what else could be different
[22:01] <lifeless> aroman: different dependencies / setup.py invocation ?
[22:01] <lvh> lifeless: Sure, go ahead
[22:01] <lvh> lifeless: I made it clear chatham house rule applied to anything sent to me in this context
[22:02] <aroman> lifeless: nope, absolutely no difference that would cause a change
[22:02] <lifeless> lvh: cool, thanks
[22:15] <lvh> lifeless: Also a very common complaint: Nobody can figure out how to (easily) access code. "View branch content" ought to be called "View code", or something.
[22:16] <lvh> Apparently it's only active lp users that have figured it out once that figure *that* out
[22:16] <lvh> people prefer Github's approach of shoving code in your face
[22:16] <lvh> (also people don't particularly like loggerhead so perhaps the the two things could be put together)
[22:17] <lifeless> yeah
[22:18] <lifeless> we need to make loggerhead more deeply integrated
[22:22] <lvh> I'm starting with features that sound like easy fixes
[22:22] <lvh> Like "remove this button it is pointless" or "move this box to this page"
[22:23] <lvh> fixing loggerhead is obviously a requirement since github's browser is more than a bit nicer but that's many orders of magntiude more of an investment than moving a box
[22:29] <lifeless> yeah
[22:29] <lifeless> you've read about our new team structure?
[22:30] <lvh> I haven't
[22:31] <lifeless> http://blog.launchpad.net/general/changing-how-we-track-launchpads-bugs-questions-and-blueprints and http://blog.launchpad.net/general/announcing-launchpad-squads
[22:36] <lvh> lifeless: Oh, a point on loggerhead that I actually completely agree with
[22:37] <lvh> Apart from the fact that I don't use Loggerhead so I don't care
[22:37] <lvh> The date format is hard to parse. Github and Bitbucket use relative, human-readable timedeltas, except for dates sufficiently far away (where they use a plain date). You could still make the real date accessable on mouseover if people wnat that feature.
[22:37] <lifeless> that should be fairly easy to do - just file a bug
[22:38] <lvh> Right, there's some bugs people linked me to too
[23:34] <lvh> lifeless: Your interest made me get off my lazy bum backside and do things
[23:34] <lvh> lifeless: http://b.lvh.cc/a-compiled-list-of-launchpads-perceived-flaws
[23:37] <lifeless> cool!
[23:49] <lifeless> lvh: we're going to fold bugs and blueprints together
[23:49] <lifeless> lvh: a more capable bugtracker and no separate spec tracker - https://dev.launchpad.net/IssueTracker