[00:00] <daveisadork> why did it only fail with a nested merge?
[00:03] <thumper> daveisadork: a commit was done after a merge
[00:04] <thumper> daveisadork: and bzr 2.2 added a hard constraint that an identity is needed
[00:04] <thumper> daveisadork: and one wasn't provided
[00:04] <thumper> daveisadork: bzr 2.2 is in maverick
[00:04] <daveisadork> i see
[00:06] <daveisadork> well thanks... the recipes and daily build setup is awesome
[01:04] <micahg> can launchpad do bzr code imports?
[01:04] <wgrant> micahg: As in mirroring a remote bzr branch?
[01:04] <micahg> wgrant: yes
[01:04] <poolie> yes
[01:04] <micahg> awesome :)
[01:04] <wgrant> micahg: Yes. Go to the project page, click 'Register a branch', and select 'Mirrored'
[01:05]  * micahg is working with a team on a sourceforge project, but would like to leverage bzr build recipies/PPAs
[03:05] <hyperair> hi. is there a way i can prevent other members in the team from receiving build failure messages for packages i upload to the team PPA?
[03:13] <StevenK> hyperair: Sadly, not that I'm aware of.
[03:14] <hyperair> hmm that sucks.
[03:14] <hyperair> shouldn't it only alert the uploader about build failure messages?
[03:15] <StevenK> That point is debatable
[03:15] <hyperair> why so?
[03:15] <hyperair> you don't see all ubuntu developers getting emailed for all build failures that any ubuntu developer uploads to the archive.
[03:15] <StevenK> Because other people on the team may be able to, or better placed to fix the build failures
[03:16] <StevenK> It's useful in some cases -- but I do agree it should be configurable
[03:16] <hyperair> i think those cases are rather remote.
[03:17] <StevenK> Why so?
[03:18] <hyperair> because the uploaded package is generally already on the uploader's computer, and the person who knows best about how to fix it, or who to ask for help to fix it, is the uploader him/herself.
[03:19] <rockstar> hyperair, you could set the contact email address of the team to be your email address, or a mailing list that people can opt into.
[03:19] <hyperair> i'd ask, why don't you spam all build failures in ubuntu to all ubuntu developers then?
[03:20] <StevenK> Because Ubuntu isn't hosted in a PPA
[03:20] <ajmitch> PPA = personal, whether than be a single person or a team
[03:20] <ajmitch> rather than a whole distribution
[03:20] <hyperair> ah
[03:21] <ajmitch> I think the intent of teams having PPAs is that it's (generally) a fairly small set of people that can upload to them
[03:22] <hyperair> in the case of banshee-team, the team's also subscribed to bug reports.
[03:38] <mwhudson> any problem in launchpad can be solved with another layer of team memberships
[05:52] <actionx> Hi! the Launchpad accepts project with Git?
[05:54] <spiv> Launchpad can import git branches into bzr for you.
[05:54] <spiv> It can't host git repositories directly.
[05:55] <actionx> ok
[06:01] <actionx> spiv: Can the bzr save the project in a DB or just create directories?
[06:01] <actionx> for example, in Mysql
[06:03] <SpamapS> actionx: er, uh.. why would one want to do that?
[06:03] <SpamapS> actionx: bazaar is, itself, a database, much more suited to revision control than mysql
[06:05] <actionx> SpamapS: correctly
[06:06] <actionx> now I know
[06:06] <actionx> tks
[08:26] <thopiekar> moin
[08:26] <thopiekar> what have I done wrong here? http://launchpadlibrarian.net/53580554/buildlog.txt.gz
[08:29] <yofel> thopiekar: I don't know much about it, but I think that should be 'nest packaging' instead of 'merge packaging'
[08:30] <thopiekar> hmm ok
[08:30] <yofel> thopiekar: at least that's what we use here https://code.edge.launchpad.net/~neon/+recipe/project-neon-recipe and it works
[08:30] <thopiekar> you are right I'm new here.. what about a branch with a debian folder in it and replacing it with another debian folder?
[08:31] <wgrant> thopiekar: Why replace, rather than merge?
[08:31] <wgrant> bzr does the whole merging thing fairly well...
[08:32] <thopiekar> ahh ok so I need to use the merge method when replacing and nest when adding something?
[08:34] <spiv> Well, you use merge to merge another version of the branch.
[08:34] <spiv> You use nest to put a copy of another branch somewhere inside the checkout.
[08:35] <yofel> hm, https://code.edge.launchpad.net/~neon/kdepimlibs/trunk fails constantly with 'sqlite3.OperationalError: database is locked'    bug?
[08:42] <spm> yofel: you're using firefox? try a different browser and/or exit/restart firefox?
[08:43] <yofel> spm: er, sorry, it's not the page that times out but the vcs-import
[08:44] <yofel> http://launchpadlibrarian.net/53585298/neon-kdepimlibs-trunk.log
[08:44] <spm> yofel: haha; right. ok. that's a different problem then.
[08:55] <sobczyk> how to edit attachements in launchpad bug report?
[09:00] <maxb_> yofel: It means that there's another vcs-import from the kde repository running on the same import host
[09:01] <yofel> ah, that would explain why this happens to kdelibs too http://launchpadlibrarian.net/53617052/vcs-imports-kdelibs-kde4.log
[09:01] <yofel> shouldn't they block each other instead of crashing?
[09:01] <maxb> ideally
[10:02] <thopiekar> I'm waiting now more than an hour for this import: https://code.edge.launchpad.net/~canola/canola/e-dep
[10:02] <thopiekar> is that "normal"?
[10:10] <thopiekar> and another thing.. the import I am talking about includes different software (http://svn.enlightenment.org/svn/e/) when making a recipe for python-ethumb for example I want to use lp:~canola/canola/e-dep/trunk/BINDINGS/python/python-ethumb/ but after editing the recipe it changes the branch location back to lp:~canola/canola/e-dep.. is it because the branch isn't imported yet or isn't it possible to make a recipe like that?
[10:14] <maxb> thopiekar: it's normal in the sense it's a manual process
[10:14] <maxb> The purpose of the process is to catch errors in import URLs which will tie up resources doing the wrong thing
[10:15] <maxb> thopiekar: For example, in this case, the URL is pointing to a repository root, containing trunk,tags,branches subdirectories. This will not do the right thing
[10:16] <maxb> If that import proceeded unmodified, you would end up with a bzr branch containing a trunk,tags,branches directory structure, which would be of no use
[10:16] <maxb> thopiekar: I assume you wanted trunk imported?
[10:16] <thopiekar> yes
[10:16] <thopiekar> so I should change the url?
[10:17] <maxb> yes, well, ask a vcs-import operator (e.g. me) to do it
[10:17] <maxb> However, in this case, the branch you want is already registered for import at lp:~vcs-imports/enlightenment/trunk
[10:18] <maxb> However, it is failing to import: Uses characters invalid in Bazaar paths (\)
[10:18] <maxb> Unfortunately you won't be able to import this at present
[10:18] <thopiekar> :/
[10:20] <thopiekar> maxb: what about http://svn.enlightenment.org/svn/e/trunk/ethumb/? would importing this location work?
[10:21] <maxb> thopiekar: I can't easily tell for certain, but if you'd like to request it (under the enlightenment launchpad project, not canola), we can give it a try
[10:22] <thopiekar> when making such imports don't I need to be member of enlightenment's team?
[10:23] <maxb> No - anyone can create a branch of a public project
[10:23] <maxb> Part of the purpose of a project record in Launchpad is to collate all activity around a particular codebase
[10:24] <maxb> both official and unofficial
[10:24] <thopiekar> k, however I joind now E17 Users
[10:26] <thopiekar> hmm it doesn't seem that it's possible to add a import, here https://code.edge.launchpad.net/enlightenment
[10:27] <maxb> thopiekar: ah, sorry, Launchpad sucks at actually providing a link to the create form
[10:27] <maxb> let me find it
[10:28] <thopiekar> thanks
[10:28] <maxb> https://code.launchpad.net/enlightenment/+new-import
[11:01] <thopiekar> maxb: works! https://code.launchpad.net/~canola/enlightenment/python-ethumb
[11:01] <thopiekar> https://code.launchpad.net/~canola/enlightenment/ethumb
[11:02] <thopiekar> just waiting for emotion now ;)
[11:02] <maxb> what about them?
[11:02] <maxb> oh, yes
[11:45] <jelmer> 'morning maxb
[12:07] <maxb> jelmer: Hi
[12:11] <jelmer> maxb: to comment on my closing the space-in-url bug as fixreleased, I think the issue you're actually hitting now is similar to the problem that didn't allow you to import the svn test repo from cvs2svn
[12:13] <maxb> hmm
[12:13] <maxb> ok, I will investigate
[13:12] <thopiekar> what is wrong here? http://launchpadlibrarian.net/53633167/buildlog.txt.gz
[13:15] <wgrant> thopiekar: Your recipe or branches are wrong: there's no debian/changelog present.
[13:19] <thopiekar> ahh I need lp:~canola/canola/canola-packaging/canolad/debian but it always changes it to  lp:~canola/canola/canola-packaging/ why?
[13:19] <thopiekar> and how can I fix that? , wgrant
[13:22] <thopiekar> I mean it my recipe (https://code.edge.launchpad.net/~thopiekar/+recipe/canola-daemon-git) when I "nest" the debian/* files to the cloned branch
[13:23] <wgrant> thopiekar: You can't nest a subtree of a branch.
[13:24] <thopiekar> hmm.. k, so I'll need to make for every package a new branch with the debian/* files?
[13:25] <thopiekar> or maybe mirrored branches?
[13:25] <wgrant> A better solution might be to have a packaging branch branched off the trunk of each project.
[13:25] <wgrant> So you branch trunk, add in the debian/ directory, and then you can merge it in your recipe.
[13:26] <thopiekar> k, thanks
[13:48] <nhandler> If translations for a project are entirely maintained on LP (and the po files are never touched directly from outside of LP Translations) is there any reason not to have them synced back to the same branch LP is checking for .pot files (trunk in this case) ?
[13:51] <dpm> nhandler, I think danilos, henninge or jtv can answer that one best, but afaik, there is no reason for not using the same branch, it should only make your life easier by letting LP (and translators!) do all the translations work for you. You might want to watch bug 490668 as well
[13:51] <danilos> nhandler, well, if they are entirely maintained in LP, you should probably not import them from the branch, or you'd hit a bug with continuous imports (i.e. translations getting imported, then exported with just the Exported-Date changed, and so forth)
[13:52] <danilos> nhandler, dpm linked a bug that I was referring to, thanks dpm :)
[13:52] <dpm> :)
[13:52] <nhandler> dpm, danilos: That is a good point.
[13:52] <nhandler> Thanks
[13:55] <danilos> nhandler, note that you completely work-around the bug by setting your translation imports to import only templates, and the only thing you lose is the "green bar" (i.e. translations will stay marked as "done in launchpad")
[13:56] <nhandler> danilos: I just made that change :)
[13:57] <danilos> cool :)
[15:06] <smoser> hey all, i was under the impression that I could do something like:
[15:06] <smoser> https://bugs.launchpad.net/ubuntu/+filebug?field.tags=ec2-images
[15:06] <smoser> and that a bug filed by following that link would then have 'tags' prepopulated with 'ec2-images'.
[15:06] <smoser> is that not the case ?
[15:06] <smoser> er... i meant to say, "that is not the case"
[15:06] <smoser> is there a way to do that ?
[15:25] <jml> smoser, Good question. I don't know.
[15:25] <jml> deryck might
[15:27] <deryck> jml, smoser -- it won't work because of the ajaxy dupe finder stuff we do.  You can only set the initial summary that way.
[15:28] <deryck> I can't think of a way to do this.
[15:28] <deryck> gmb, can you?  See ^^
[15:29]  * gmb reads backscroll
[15:30] <gmb> jml, deryck, smoser: That's not something you can do at the moment. What we should do is pass on the field values from +filebug to the filebug form when we create it. I'll file a bug.
[15:31] <smoser> so its likely that this did work at somepoint in the past?
[15:31] <smoser> i'm just wondering if I'm imagining things
[15:31] <deryck> yes
[15:31] <smoser> ok thanks
[15:32] <gmb> smoser, It worked before we introduced the AJAX dupefinder.
[15:32] <gmb> Because it was an ordinary Zope form.
[15:32] <jml> gmb, thanks.
[15:34] <gmb> jml, deryck, smoser: https://bugs.edge.launchpad.net/malone/+bug/617416
[15:34] <deryck> gmb, thanks for filing the bug!
[15:35] <smoser> gmb, thanks.
[16:33] <pc_magas> How can I get a back up of my gpg keys?
[16:35] <pc_magas> How can I get a back up of my gpg keys?
[17:28] <paultag> sinzui: prod
[17:28] <sinzui> hi paultag
[17:29] <paultag> sinzui: Hey, you just deactivated a project of mine ( not that big of a deal, I thought it could have been sketchy )
[17:29] <paultag> sinzui: I was trying to get some of my git stuff in bzr via LP. Can I do that in a +junk branch?
[17:29] <sinzui> +junk is fine with git branches.
[17:29] <sinzui> you have an unlimited number of branches
[17:30] <paultag> sinzui: Righto. Thanks. Sorry for clouding up LP.
[17:30] <sinzui> paultag, you can also push some of your branches to projects that are already registered. they remain yours, but it is clear your branches are based on existing works
[17:31] <sinzui> paultag, np, project registration does a poor job of explaining what a project is...a shared area for communities to work on code and bugs
[17:31] <paultag> sinzui: yeah, but most of them don't have a lp project, and I figured one minor project would be a bit easier on the system over 5 or so very minor projects
[17:31] <paultag> sinzui: aye
[17:32] <paultag> sinzui: anywho, thanks
[17:33] <sinzui> paultag, not so in this case. you will get terrible bzr push/pull performance because each branch will be assumed to be the branch based on the project's focus of development series. since there is nothing in common, you will often see 100% change sets
[17:33] <paultag> Ahhha
[17:33] <paultag> right, that makes sense
[17:47] <Muscovy> Is there a way to make a PPA with the same name as a deleted one?
[17:51] <lfaraone> Hi, I've been getting an OOPS all day when trying to access bug 320915.
[17:51] <lfaraone> (most recently, OOPS-1686EB2333)
[17:53] <Muscovy> I get OOPS-1686D1310 when I try to view it.
[18:52] <Muscovy> Are there any dangers/problems related changing your username, other than needing to remove your PPAs?
[18:56] <Muscovy> My username was generated based on my email when I first joined Launchpad, and I never bothered to change it before.
[18:56] <Muscovy> But I'm worried it might cause problems with memberships, karma, etc.
[18:56] <beuno> Muscovy, it will affect code as well
[18:57] <beuno> it won't for membership or karma
[18:57] <yofel> jelmer: anything that can be done about those? http://launchpadlibrarian.net/53640313/vcs-imports-kdelibs-kde4.log happens when more than one kde import is scheduled on the same import machine
[18:58] <jelmer> yofel: python-tdb can be installed on the launchpad machines; that's not really my area though, you'd want to talk to one of the launchpad code folks (abentley, rockstar or thumper)
[18:59] <yofel> k, thx
[18:59] <rockstar> jelmer, we should chat about that.
[18:59] <Muscovy> beeno: what sort of issues?
[19:00] <Muscovy> beuno*
[19:06] <maxb> Installing python-tdb would be nasty - the importds would recache millions of revisions
[19:06] <maxb> We'd probably get banned from svn.apache.org again :-)
[19:06] <maxb> Also, having been using python-tdb locally on a large repository, I have to say, it's performance utterly sucks if the whole thing isn't in cache
[19:07] <beuno> Muscovy, the URLs for all the branches change
[19:08] <beuno> all your branches, that is
[19:11] <Muscovy> Oh, that's fine in my case, all my code is in group branches.
[19:11] <Muscovy> Thanks. :D
[19:11] <jelmer> maxb: Is that comparing it with python-sqlite ? Because I'm quite sure python-tdb's performance is better in that regard.
[19:12] <maxb> I'll admit I haven't used sqlite in a while. I should try it to recompare
[19:12] <maxb> However, I've had to resort to catting tdb files to /dev/null at times
[19:13] <jelmer> maxb: running "svn log -v" on the apache repository won't get us banned; I've done it myself a couple of times. We got banned because we opened lots (and I mean lots) of connections through cscvs.
[19:13] <jelmer> maxb: Catting tdb files to /dev/null? Why?
[19:13] <maxb> It makes bzr-svn go faster
[19:14] <maxb> It seems that the access pattern of tdb accessing its own data is a lot less efficient than getting the whole thing read into cache up front
[19:15] <jelmer> maxb: I'm not sure I follow, tdb *is* the cache.
[19:15] <maxb> tdb can be very slow at reading if the entire tdb file is not present in kernel disk cache
[19:31] <mathiaz> hi - is there a way to request that all my +junk branches be upgrade to the 2a format?
[19:40] <jelmer> mathiaz: Hi; There is no easy way to mass-upgrade other than talking to a LOSA.
[19:42] <jelmer> mathiaz: Alternatively, it might be possible to script something using bzrlib and launchpadlib.
[19:55] <mathiaz> jelmer: is it possible to upgrade using launchpadlib?
[19:56] <mathiaz> jelmer: or should a specific bzr command be called?
[19:56] <mathiaz> jelmer: is there an upgrade-remote-repository command in bzr?
[19:57] <thopiekar_> hi .. why don't you support adding subdirs to recipes?
[19:57] <jelmer> mathiaz: launchpadlib won't let you upgrade remote branches directly as far as I know, but it should allow you to get a list of all your +junk branch URLs
[19:58] <jelmer> mathiaz: "bzr upgrade" supports remote URLs, although it might do some of its work locally (I'm not sure how much)
[19:58] <mathiaz> jelmer: great
[19:58] <mathiaz> jelmer: seems like a good plan
[19:58] <mathiaz> jelmer: thanks for the help
[20:07] <thopiekar_> or is it possible to copy, move and remove files in recipes?
[20:31] <thopiekar_> :/
[21:25] <flacoste> sinzui: how can I make the green bar full on https://edge.launchpad.net/pylkk?
[21:26] <flacoste> sinzui: is there a way to say, it seems that can only happen if i say the translations is done on LP
[21:26] <sinzui> Not yet
[21:26] <flacoste> sinzui: is that what the new tristate data model intends to solve?
[21:26] <sinzui> brad landed the db 3 days agio
[21:26] <flacoste> got it
[21:27] <flacoste> sinzui: that works nicely though, kudos to you and your team!
[21:27] <sinzui> My blocked card is blocked because I need to use the tristate
[21:27] <sinzui> It looks very code
[21:28] <sinzui> It looks very nice on the page I mean
[22:19] <thopiekar> maxb: ping
[22:19] <maxb> pong
[22:20] <thopiekar> "20100813 maxb Please register a project record for pydispatcher in Launchpad and re-request this import within that project. Marked invalid." What is a project record? and where can I register it? , maxb
[22:20] <maxb> Go to the front page of Launchpad and choose "Add project"
[22:21] <thopiekar> k
[22:27] <thopiekar> maxb: it seems that pydispatcher is highly outdated.. just talking with my project team what to do..
[22:27] <maxb> outdated? OK. Well, you can still get an import set up anyway it you like
[22:30] <thopiekar> so I need to create a project and add it there as code and make a recipe?
[22:30] <thopiekar> maxb: ^
[22:31] <maxb> a recipe? no
[22:31] <maxb> Just create the project and the import
[22:31] <thopiekar> I mean to make a import..
[22:31] <thopiekar> yeah.. I just want to import and make a recipe, you know ;)
[22:32] <thopiekar> thank you for helping me, maxb!
[22:32] <maxb> np
[22:54] <thopiekar> maxb: just about the debian/* files that I need to build the Canola packages.. Is there no way to get the subdirs of this branch? http://bazaar.launchpad.net/~canola/canola/canola-packaging/files
[22:55] <maxb> bzr-builder, as it's currently designed, cannot do that
[22:55] <thopiekar> I can't import every debian/* files for every app because git doesn't support subdir cloning like svn , for example
[22:55] <thopiekar> :/
[22:55] <maxb> Is there a reason you're maintaining multiple packages in a single branch?
[22:56] <thopiekar> yep the reason is the project wanted to keep it like that (in the past)..
[22:57] <thopiekar> but ok I'll contact the man who is maintaining the mainlines..
[23:16] <maxb> thopiekar: Your other option would be to consider enhancing bzr-builder
[23:17] <thopiekar> for sure!
[23:17] <nhandler> Before I file a bug, could someone give this a try. Go to a page like https://translations.edge.launchpad.net/classbot/trunk/+pots/classbot and click the 'Last Edited' column. Does it sort the column chronologically or alphabetically/numerically ?
[23:18] <thopiekar> something like a "sh" command that can execute console commands for copy, move and remove files..
[23:18] <maxb> Launchpad will not offer arbitrary command execution in recipes
[23:18] <maxb> I was thinking of perhaps adding a 'symlink' bzr-builder directive
[23:19] <maxb> That way you could nest your packaging branch, and symlink the appropriate debian subdir
[23:20] <wgrant> nhandler: It works OK in Firefox, but not Chromium.
[23:20] <wgrant> Odd.
[23:22] <nhandler> wgrant: Thanks for catching that. I only tested chromium. Let me file a bug
[23:24] <thopiekar> great idea maxb!
[23:24] <thopiekar> how can I get involved and work on bzr-builder?
[23:26] <maxb> james_w is the main developer, you might like to catch him here on IRC and run ideas past him - then branch, hack, and propose a merge!
[23:27] <maxb> huh
[23:27] <maxb> although the latest revision in its trunk is "Fixes for the nest-part instruction. Thanks Andrew."
[23:27] <wgrant> Hah. Convenient.
[23:27] <maxb> Which sounds like it might work for your use case
[23:28] <thopiekar> maxb: you mean me? :P
[23:28] <maxb> yes
[23:28] <maxb> and now, I shall depart#
[23:28] <thopiekar> so nest-part is another command?
[23:30] <thopiekar> how does nest-part work?
[23:30] <wgrant> nest-part packaging lp:~foo-dev/foo/packaging debian
[23:30] <wgrant> But LP doesn't support it yet.
[23:30] <bdrung> nest-part let's bzr crash
[23:30] <wgrant> bdrung: Hm?
[23:31] <bdrung> bzr: ERROR: exceptions.AttributeError: 'module' object has no attribute 'MergeIntoMerger'
[23:31] <thopiekar> wgrant: you mean: nest-part packaging lp:~foo-dev/foo/packaging/and/some/subdirs debian?
[23:31] <bdrung> http://pastebin.com/ukEqRveK
[23:32] <bdrung> wgrant: ^
[23:32] <wgrant> thopiekar: No. For that, you'd do 'nest-part packaging lp:~foo-dev/foo/packaging and/some/subdirs debian'
[23:32] <wgrant> bdrung: I wonder if it needs a newer version of bzr.
[23:33] <bdrung> wgrant: i am running lucid
[23:33] <thopiekar> wgrant: great then!
[23:33] <thopiekar> when will lp support it?
[23:33] <wgrant> Not sure. The issue that bdrung just ran into might be inconvenient.
[23:34] <bdrung> at least it's reproducible
[23:35] <bdrung> wgrant: what do you think about adding a "safe" removal function to the recipe (that can be enabled on LP)?
[23:35] <bdrung> use case: run rm -rf debian/patches
[23:36] <thopiekar> wgrant: when will be nest-part available on lp?
[23:36] <thopiekar> can't wait to make recipes with it!
[23:36] <bdrung> thopiekar: quote: "Not sure. The issue that bdrung just ran into might be inconvenient."
[23:37] <thopiekar> ;/
[23:38] <thopiekar> I'll go now and take some sleep ;)
[23:38] <thopiekar> good luck making lp better!
[23:51] <njin> someone around
[23:52] <wgrant> njin: Nobody.
[23:53] <njin> m? wgrant is a bot? ;-)
[23:53] <wgrant> Heh.
[23:53] <njin> i need help
[23:53] <wgrant> What's the problem?
[23:54] <njin> sometimes i receive a reply mail but on the bug page don't result
[23:54] <njin>  Da: xsnake <393090@bugs.launchpad.net> Reply-to: Bug 393090 <393090@bugs.launchpad.net> A: marconifabio@hotmail.it Oggetto: [Bug 393090] Re: mmc: Timeout waiting for hardware interupt Data: 11/08/2010 01:21:32
[23:55] <njin> Oops, that is an old mecssage, better i go to bed....
[23:55] <njin> sorry
[23:56] <bdrung> njin: you are the second person saying that... :)
[23:56] <njin> i'm cloned... help
[23:57] <njin> ;-)
[23:57] <bdrung> njin: i can't help you because i need sleep ;)