[00:03] <johnl> hi. how can I import a particular git branch into launchpad (as opposed to importing master, which seems to be default)?
[00:04] <mwhudson> johnl: you can't yet :/
[00:05] <mwhudson> jelmer will fix this sooner or later i guess :)
[00:05] <johnl> I did suspect that, ta.
[00:06] <johnl> I've pulled from the specific git branch into a bzr branch I created locally, but I don't quite understand how to push this to a branch on launchpad (on my project).
[00:06] <johnl> my project is: https://code.launchpad.net/ceph
[00:07] <johnl> if I register a new branch, it's on my personal launchpad account, not on the ceph project (like the imported git one)
[00:07] <johnl> any ideas what I'm missing?
[00:07] <johnl> I'm rather the bzr newb, sorry :/
[00:09] <thumper> johnl: bzr push lp:~you/ceph/branchname
[00:09] <thumper> johnl: you don't need to create the branch on LP before you push
[00:09] <thumper> johnl: one will be created
[00:16] <johnl> hrm, that looks better than my last attempt (which had +junk in it). thanks.
[00:17] <johnl> but it's still lp:~me/ceph..., wheras the trunk is lp:ceph
[00:17] <johnl> is that normal?
[00:20] <mwhudson> yes
[00:24] <maxb> johnl: In summary, *every* LP branch is fundamentally associated with a person|team. Ones which are the focus branch of a series, and more so the focus branch of a default series, get additional short aliases
[00:24] <maxb> lp:~person/project/branchname
[00:24] <maxb> lp:project/series
[00:24] <maxb> lp:project (default series)
[00:25] <maxb> except for source package branches, which are different again :-)
[00:25] <maxb> lp:~person/distro/distroseries/sourcepackage/branchname
[00:26] <maxb> lp:distro/distroseries-or-distroseries-pocket/sourcepackage
[00:26] <maxb> lp:distro/sourcepackage
[00:27] <maxb> It may look complex, but fundamentally, it's always a ~person or a project-or-distro at the start, with reasonable hierarchy after that
[01:03] <falktx> hey guys
[01:03] <falktx> I got a package upload stuck...
[01:03] <falktx> https://launchpad.net/~kxstudio-team/+archive/ppa/+build/2095304
[01:04] <falktx> it finished building 6 hours ago
[01:04] <falktx> can you help?
[01:04] <maxb> wgrant: looks like we got another one.
[01:05] <maxb> falktx: It would seem to be an intermittent bug with Launchpad. You can generally work around it by reuploading the package for another try
[01:06] <falktx> maxb: an, ok, thanks
[01:06] <falktx> maxb: why does keep happening?
[01:07] <maxb> unknown
[01:07] <maxb> I'm just filing a bug
[01:08] <falktx> k
[02:51] <yofel> any reason why I get mailed about *successfull* recipe auto builds?
[05:46] <dnivra> hello. i'm trying to login into ubuntuforums using my launchpad id. i'm permanently signed in at launchpad website-checked by logging into launchpad.net. but it still redirects to login.launchpad.net when i click 'sign in using launchpad id'. why isn't my login detected?
[09:52] <tvst4000> Hey all, is this the appropriate place to get help uploading to a PPA?
[09:52] <tvst4000> I got the following error: "Rejected: Unable to find source package MY_PACKAGE_NAME_GOES_HERE in lucid"
[09:52] <bigjools> tvst4000: yes, and it seems like an obvious problem
[09:53] <tvst4000> Good! But not obvious to me, though
[09:54] <tvst4000> I'm uploading the package the same way that always worked for my other packages. The source compiles fine, but then gets rejected with that error.
[09:55] <bigjools> MY_PACKAGE_NAME_GOES_HERE is not a real source name is it :)
[09:55] <bigjools> it's probably wrong in debian/control
[09:56] <tvst4000> I changed it for the IRC :)
[09:56] <tvst4000> The actual message was
[09:56] <tvst4000> Rejected:
[09:56] <tvst4000> Unable to find source package cardapio-awn/0.9.162-ubuntu2-lucid1 in lucid
[09:56] <tvst4000> Sorry for the confusion.
[09:56] <bigjools> ok
[09:57] <tvst4000> I'm uploading cardapio-awn to my PPA, with two versions: one for lucid and one for maverick.
[09:57] <tvst4000> The maverick version worked fine, as  you can see here: https://launchpad.net/~cardapio-team/+archive/unstable/+packages
[09:58] <tvst4000> The lucid version gave me that error -- but for no apparent reason, since the procedure was the same for both!
[09:58] <bigjools> are you uploading binaries?
[09:58] <tvst4000> Nope, sources
[09:59] <bigjools> that error is only emitted when it's looking for the source for a binary upload.  OTOH it could be a bug, let me check the code some more.
[09:59] <tvst4000> Ok.
[10:00] <tvst4000> It could also be that my understanding of the PPA-building process is flawed. But the little I know has worked without problems so far :)
[10:01] <bigjools> tvst4000: what time did you upload the one that was rejected?
[10:02] <tvst4000> Some 10 minutes ago. A few seconds before the maverick version, which was accepted.
[10:03] <tvst4000> Actually, more like 50 minutes ago.
[10:03] <bigjools> you did a lucid one earlier today it seems
[10:04] <tvst4000> Yes, there was another lucid and maverick earlier today, some 8 hours ago. But those had a reason to fail :)
[10:04] <bigjools> it was accepted as far as I can see
[10:04] <bigjools> are you getting a build failure notification?
[10:05] <tvst4000> The build seems to work, as I get an "Accepted" message. But then,
[10:05] <tvst4000> I also get a "Ubuntu Installer" message that says
[10:05] <tvst4000> Rejected:
[10:05] <tvst4000> Unable to find source package cardapio-awn/0.9.162-ubuntu2-lucid1 in lucid
[10:05] <tvst4000> (by message I mean email)
[10:05] <bigjools> I see your problem
[10:05] <bigjools> it's a build failure
[10:06] <bigjools> you uploaded 0.9.162-ubuntu2-maverick1 to lucid
[10:06] <tvst4000> Oh, did I?
[10:06] <bigjools> and your binary is built for the 0.9.162-ubuntu2-lucid1 source
[10:06] <tvst4000> Wait, let me check...
[10:07] <tvst4000> oops, sorry, closed the window :)
[10:07] <tvst4000> just a sec.
[10:08] <tvst4000> Oh, I see!
[10:08] <tvst4000> Doh!
[10:09] <tvst4000> Thanks for helping me debug this. My fault entirely
[10:09] <bigjools> no worries
[10:09] <bigjools> the error message could be better to be fair
[10:09] <bigjools> if you want, file a bug
[10:09] <tvst4000> I just may :)
[10:10] <tvst4000> thanks for your help
[10:10] <tvst4000> ttyl
[10:10] <bigjools> enjoy
[10:24] <owenmorris> mrevell: Is there any problem with package uploads to ppa?
[10:24] <owenmorris> I am getting the following error: Unhandled exception processing upload: 'NoneType' object has no attribute 'md5'
[10:25] <bigjools> owenmorris: known bug
[10:25] <bigjools> you can work around it by uploading a new orig tarball
[10:25] <bigjools> it's happening because the old one expired due to a package deletion, and the code that looks for existing files is mishehaving
[10:25] <bigjools> misbehaving, even
[10:26] <owenmorris> bigjools: thanks, my tarball is generated from git-buildpackage
[10:26] <owenmorris> so the name is generated automatically
[10:26] <owenmorris> will I have to add a new changelog entry and rebuild
[13:34] <lamothe> Hi all, I seem to have busted my bzr branch, can someone please tell me where I can put in a request to fix this?
[13:56] <cdbs> lamothe: What is the exact thing you did wrong?
[13:58] <lamothe> cdbs: Hi there!
[14:02] <lamothe> Not sure what I did wrong but ... I can get to my code  at https://code.launchpad.net/~gnome-media-player-development/gnome-media-player/0.2
[14:02] <lamothe> *I can't
[14:03] <beuno> lamothe, and you are part of the ~gnome-media-player-development team?
[14:03] <lamothe> Yes.
[14:04] <beuno> lamothe, what error are you getting?
[14:05] <lamothe> I can get to that page but any attempt to dig deeper results in "Sorry, there was a problem connecting to the Launchpad server."
[14:05] <lamothe> e.g. http://bazaar.launchpad.net/~gnome-media-player-development/gnome-media-player/0.2/revision/110
[14:05] <beuno> lamothe, so, that url is the code browser
[14:05] <beuno> which has been known to fail
[14:06] <beuno> it's not really a problem with your branch
[14:07] <beuno> not sure what the process these days for restartng codebrowse is
[14:07] <lamothe> I think that link used to take me to a source code browser.
[14:07] <beuno> yes
[14:07] <beuno> what I'm saying is that the server is down
[14:07] <beuno> rather than a problem with your branch  :)
[14:07] <beuno> losas, ping?
[14:08] <lamothe> Oh, ok, it's been like that for a few hours.
[14:08] <cdbs> we have been having problems with loggerhead for quite some time
[14:08] <mthaddon> beuno: howdy
[14:08] <beuno> oh hai mthaddon!
[14:08] <beuno> mthaddon, is codebounce maybe down?
[14:08] <cdbs> codebrowse
[14:08] <cdbs> not codebounce
[14:09] <mthaddon> beuno: nagios checks look fine - what seems to be the problem?
[14:09] <beuno> cdbs, well, we have a nickname for it, because it bounces up and down so much  :)
[14:09] <beuno> mthaddon, http://bazaar.launchpad.net/~gnome-media-player-development/gnome-media-player/0.2/revision/110
[14:09] <mthaddon> hmm, our check url is http://bazaar.launchpad.net/~vcs-imports/busybox/main/changes and that seems to work fine
[14:10] <beuno> hm
[14:10] <beuno> that branch seems to not work
[14:10] <mthaddon> the one you're looking at, you mean?
[14:10] <beuno> jelmer, you around by any chance?
[14:10] <beuno> mthaddon, yes
[14:11] <beuno> mthaddon, if the server is up, maybe it's a bug then, thanks
[14:11] <mthaddon> yeah, both servers are up (we have two load balanced codebrowse instances these days)
[14:11] <lamothe> I have been screwing around i.e. deleted an old branch named 0.2, renamed another to 0.2, then changed the owner.
[14:12] <beuno> lamothe, could you file a bug>
[14:12] <beuno> ?
[14:12] <lamothe> beuno: Sure, is that for soyuz?
[14:12] <beuno> lamothe, launchpad-code
[14:13] <lamothe> beuno: Ok, ta, will do.
[14:13] <lamothe> Thanks.
[14:13] <beuno> thanks lamothe
[14:23] <jelmer> beuno: sorry, wasn't watching #launchpad - what's up?
[14:30] <beuno> jelmer, any ideas why this would be failing?  http://bazaar.launchpad.net/~gnome-media-player-development/gnome-media-player/0.2/revision/110
[14:51] <geser> bigjools: Hi, do you have an idea why the "xubuntu" package set is not in the lp.packagesets collection (LP API)?
[14:53] <bigjools> geser: I don't know.  Did you talk to Ubuntu folks?
[14:55] <geser> bigjools: who exactly? and how can they change it? using edit-acl.py from ubuntu-archive-tools I can query it, so it seems to exist but when I iterate over lp.packagesets it's not listed there
[14:55] <geser> so I don't even know where the problem is and who can fix it
[14:59] <bigjools> geser: check with cjwatson in the first instance
[14:59] <cjwatson> it's not my problem
[15:00] <cjwatson> as geser says, the package set exists - it's just not showing up in the packagesets collections
[15:00] <cjwatson> *collection
[15:00] <cjwatson> I don't know why
[15:35] <geser> bigjools: should I file a bug about it?
[15:35] <bigjools> geser: yes, sounds like a bug
[15:35] <bigjools> thank you
[16:18] <ScottK> Is it by design that soyuz doesn't notice and take a package out of depwait when the missing build-depend is newly provided only by a virtual package?
[16:18] <ScottK> See any of the libpadre-plugin* packages currently depwait in Natty (the one that FTBFS I retried by hand)
[16:18] <bigjools> ScottK: I will check
[16:19] <ScottK> bigjools: Thanks.
[16:28] <bigjools> ScottK: I can't see anything obvious, there must be another constraint that is not triggered
[16:28] <ScottK> bigjools: OK.  Do I need to file a bug on this or do you have it?
[16:28] <bigjools> ScottK: I am not sure if there's a bug
[16:28] <bigjools> how long were those waiting once the dependencies were satisfied?
[16:29] <ScottK> bigjools: 9 hours so far.
[16:29] <ScottK> I think if they were going to come out of depwait, they would have.
[16:32] <bigjools> ScottK: can you double check that all its deps are satisfied (in the right component and pocket)
[16:32] <bigjools> if you're convinced they are, then please file a bug by all means
[16:33] <bigjools> and put all the relevant dependency info in there
[16:33] <ScottK> bigjools: https://launchpad.net/ubuntu/+source/libpadre-plugin-vi-perl/0.23.0-1/+build/2094813 was like the others until I retried it manually.  it has all package deps met.
[16:33] <ScottK> OK.
[16:36] <bigjools> ScottK: soyuz uses apt_pkg to check the dependencies, it's possible it has a bug.  Are any of the versions a little funky?
[16:37] <ScottK> I don't think so.  I suspect it's a virtual package issue.
[16:37] <ScottK> That may be apt_pkg's fault though.  Dunno.
[16:38] <bigjools> there's nothing in the code that is concerned with that, we call apt_pkg.ParseDependencies(), look them up in LP and then use apt_pkg.VersionCompare()
[16:39] <bigjools> with the usual component/pocket/arch restrictions
[16:40] <ScottK> Both are in Universe and it clealry found it when I did the manual retry.
[16:40] <ScottK> Bug #690240 is filed.
[16:42] <bigjools> Thanks - I'll set the Grantinator on that
[16:43] <soren> bigjools: I'm curious. When do you call that?
[16:43] <soren> bigjools: (apt_pkg.ParseDependencies)
[16:43] <bigjools> ParseDepends, my bad
[16:43] <soren> bigjools: Regardless :) When does that get called?
[16:44] <bigjools> when we check the build record's dependencies that were set at the end of the previous build
[16:45] <soren> I don't understand.
[16:45] <soren> This is the code that determines whether a build that's stuck in depwait can now get built.
[16:45] <soren> Right?
[16:47] <bigjools> yes
[16:48] <bigjools> when the build finishes the log contains the dependencies that need to be satisfied, they are parsed out and put on the build record
[16:48] <soren> Ok.
[16:48] <soren> And every time another build finishes, this is checked?
[16:49] <bigjools> then when the cron job fires that checks to see if they're satisfied, it uses ParseDepends() on that list
[16:49] <bigjools> no
[16:49] <soren> Ah, ok.
[16:49] <bigjools> every hour
[16:49] <soren> At intervals.
[16:49] <soren> Gotcha.
[16:49] <bigjools> it would be nice if we had a queueing system eh :)
[16:50] <soren> No comment.
[16:50] <soren> :)
[17:41] <fta> danilos, hi, do you think it would be possible to have (at some point) a way to check strings before they are accepted by lp?
[17:41] <danilos> fta, hi, I've just seen your email
[17:42] <danilos> fta, it would definitely be possible, but I am not sure when that would be because we're definitely going to lack the time to implement it
[17:42] <danilos> fta, however, my suggestion would be to use one of existing formats instead of what you are using, eg. one of http://www.gnu.org/software/hello/manual/gettext/Translators-for-other-Languages.html
[17:42] <danilos> fta, that way, you get at least some checking
[17:43] <fta> danilos, i was thinking about something generic, that would suit anyone, like a simple API. check(string, translation) => boolean. and a project admin can upload a script implementing check()
[17:44] <danilos> fta, that has a lot of problems, the biggest being that we'd have to run that on a sand-boxed server because we can't let everyone run their own scripts on our servers which have access to the full DB or something
[17:44] <fta> sure
[17:45] <danilos> fta, so, that's probably never going to come... the thing that we can do is allow project owners to mark translations as broken, though we don't have an easy way to display that to users yet
[17:45] <danilos> fta (we do have a way to store an error in the DB already)
[17:46] <danilos> fta, basically, at the moment, you can go directly to related messages and mark them as needing review
[17:47] <fta> danilos, is it possible to do with the api? i can't spend my life watching these, but i can sure script it if it's possible
[17:48] <danilos> fta, no, unfortunately
[17:49] <danilos> fta, however, do note that you can use something like sh-format and get basic checking for ${VARIABLE} names
[17:49] <fta> danilos, the search is not good enough either, so i sometimes show the wrong string (or several instead of the one i want to point to)
[17:50] <danilos> fta, as far as that goes, since you also have the POT file, you can use the sequence number instead (i.e. message number "10" in the POT file will be at yourtemplate/LANG/10/+translate)
[17:51] <fta> danilos, but it changes everyday, as i have template updates almost daily
[17:52] <fta> danilos, my only fixed keys are my numeric id (in #:) and the untranslated string itself
[18:04] <danilos> fta, I am surprised the search doesn't work though
[18:05] <danilos> fta, if you search for the full string it should be fine (it will though find messages which contain that string)
[18:06] <fta> danilos, well, when i search for "foo", it finds "foo" but also "foo bar", "foo baz", ...
[18:06] <fta> i'd like to search for ^foo$
[18:11] <danilos> fta, yeah, I know (that's what I said above: "messages which contain that string"), but we won't really be adding that
[18:12] <danilos> anyway, gotta go
[19:56] <ahasenack> hi guys, I'm writing instructions telling someone how to get the sources.list line to access a private PPA. I need a generic launchpad URL for that, instead of something like https://launchpad.net/~ahasenack/+archivesubscriptions
[19:56] <ahasenack> is there a "+me" or something I can use in that url?
[19:59] <matsubara> ahasenack, people/+me/
[20:00] <matsubara>  https://launchpad.net/people/+me/+archivesubscriptions
[20:00] <ahasenack> matsubara: excellent! thanks
[20:00] <matsubara> np
[20:04] <Bogo> ahasenack got a tutorial you're following on how to add applications to your own ppa?! :-/
[20:04] <ahasenack> hmm?
[20:05] <Bogo> crap...accidently closed xchat :P
[20:12]  * Bogo is away: I'm busy...shu...