[01:41] <RenatoSilva> do you use bzr-email for notification on branch changes?
[01:41] <RenatoSilva> if not, what do you use instead?
[01:42] <wgrant> RenatoSilva: They use their own thing. Not a bzr plugin.
[01:45] <RenatoSilva> ok
[02:09] <RenatoSilva> it would be nice if you could upload a zipped branch in Launchpad
[02:10] <thumper> RenatoSilva: why?
[02:10] <thumper> RenatoSilva: the email out of LP was originally copied from bzr-email
[02:12] <RenatoSilva> thumper: I've heard of another option with mail template, do you know any?
[02:13] <thumper> RenatoSilva: I'm not quite sure I understand what you are asking
[02:13] <RenatoSilva> about zip: in case of you can't push
[02:14] <RenatoSilva> a better option to bzr-email
[02:14] <RenatoSilva> also, a suggestion: push through zip upload
[02:14] <thumper> RenatoSilva: we have some work around emailing merge directives to launchpad
[02:14] <thumper> RenatoSilva: right now we can make a new branch from a merge directive
[02:14] <thumper> RenatoSilva: but this also proposes for merging
[02:15] <thumper> RenatoSilva: it pulls in the changes from the directive
[02:15] <RenatoSilva> I don't get you
[02:16] <thumper> a merge directive is what you go when you say `bzr send`
[02:16] <thumper> a bundle is an text representation of some revisions
[02:16] <thumper> a merge directive contains the revision bundle by default
[02:17] <thumper> you can email a new branch to LP and propose it for merging with a single email
[02:17] <RenatoSilva> oh
[02:19] <RenatoSilva> so I send an email with the patch and it creates a branch?
[02:19] <RenatoSilva> or merges
[02:20] <thumper> RenatoSilva: right now it needs to be a merge directive, not a patch
[02:20] <thumper> RenatoSilva: and it has some error edge cases right now that aren't handled particularly well
[02:22] <RenatoSilva> ok it's better wait
[02:23] <RenatoSilva> but can't I upload a directive?
[02:23] <RenatoSilva> by form
[02:23] <thumper> no, just email
[02:27] <RenatoSilva> ok
[02:27] <RenatoSilva> is bazaar.conf fetched when I bzr branch lp:*?
[02:27] <RenatoSilva> I want to use bzr-email at work, bring branch home, stop using bzr-email, work, commit, and push
[02:29] <thumper> not sure
[02:41] <RenatoSilva> #bzr confirmed: no
[03:06] <RenatoSilva> I would like to try sending a merge proposal to launchpad, how can I do it?
[03:06] <RenatoSilva> How should I save the bzr send? I put a .patch extension
[03:06] <spiv> RenatoSilva: https://help.launchpad.net/Code/Review#Email%20interface
[03:14] <RenatoSilva> the branch is amnually created?
[03:14] <RenatoSilva> if approved
[03:14] <RenatoSilva> I thought it was automated
[06:18] <DBO> does LP update merge diffs if the merge gets new revisions pushed to it after the merge is proposed?
[06:25] <mwhudson> no
[06:25] <mwhudson> not yet, anyway
[06:25] <DBO> thank you
[06:32] <RenatoSilva> bzr branch lp:bzr-email: FATAL ERROR: Disconnected: No supported authentication methods available
[06:33] <RenatoSilva> can't I branch as anonymous after a lp-login?
[06:39] <jmarsden> RenatoSilva: I think you can't "undo" the lp-login easily, you have to edit or rm a file or two under .bzr/ before it will use http to access the repository.  Someone was discussing this eitehr here on on #ubuntu-motu a few days ago I think.
[06:39] <RenatoSilva> jmarsden: thanks for the info at least
[06:40] <RenatoSilva> jmarsden: something like bzr branch --annonymous, and bzr lp-logout would be nice
[06:43] <jmarsden> RenatoSilva: Yes.  I think just bzr lp-logout would be sufficient.
[06:44] <RenatoSilva>  bug 349143
[06:47] <jmarsden> You found it :)  Loosk like you can rm ~/.bazaar/authentication.conf and edit your launchpad username out of ~/.bazaar/bazaar.conf to get the desired effect.  But that's hard to remember :)
[06:56] <RenatoSilva> :)
[07:20] <RenatoSilva> what's Status Whiteboard for branches bug linking?
[07:21] <kiko> RenatoSilva, just somewhere to put status information
[07:22] <RenatoSilva> ok
[07:22] <kiko> you can use it as you see fit -- it's a whiteboard, so no parsing is done of it
[07:24] <RenatoSilva> ok
[07:34] <persia> Hello.  I've been told that some people get reject messages for unsigned uploads to a PPA and some people don't.  I'd expect nobody to get any error messages because Soyuz wouldn't be able to figure out if it was a forged upload.  Does anyone know which is the correct behaviour?
[07:38] <robin_> bzr: ERROR: Invalid url supplied to transport: "lp:~scut-tang/mysql-server/mysql-6.0-infoschema": The remote branch at ~scut-tang/mysql-server/mysql-6.0-infoschema has no URL specified.
[07:40] <robin_> what is wrong? The bug 382795 has not be fixed yet?
[07:40] <robin_> But I can't push my branch still?
[07:43] <spiv> robin_: it's apparently a remote branch, not a hosted one.
[07:43] <spiv> robin_: but weirdly, it's a remote branch with no remote URL set.
[07:43] <spiv> jml: ^ any idea?
[07:44] <jml> spiv, plenty of idea :)
[07:44] <jml> robin_, hi
[07:44] <robin_> spiv: ?with no remote url set?
[07:44] <robin_> jml: ah?
[07:44] <spiv> robin_: my guess is that what you want to do is delete that branch (via the web UI) and then push it, assuming you want that branch hosted on launchpad.
[07:45] <jml> robin_, I'm in the middle of a few things right now. Please give me a moment.
[07:45] <robin_> jml: ok
[07:45] <kiko> robin_, what spiv said, though it's weird that you ended up in that situation -- maybe rename it and we can look at it later
[07:45] <robin_> ok, I try to delete it in the web UI, and push it again.
[07:47] <jml> so
[07:47] <spiv> robin_: did you create that branch via the "Register a branch" web page?
[07:47] <jml> spiv, kiko: this was the branch that caused the memory crisis last week
[07:47] <robin_> spiv: no, I push it directly.
[07:47] <jml> we manually hacked it to be remote in order to make sure it didn't take down our servers.
[07:48] <kiko> ah.
[07:48] <kiko> ah yes
[07:48] <kiko> robin_, how long does it take to push that branch?
[07:48] <robin_> kiko: long time-_-!
[07:48] <spiv> jml: ah!
[07:49] <robin_> kiko:  I delete the branch, and push it now. It works and is transferring.
[07:49] <kiko> robin_, you know, you might want to wait a few minutes until we decide if we are going to roll an update to codehosting in the next hour
[07:49] <kiko> because we'll have to do a bounce
[07:49] <kiko> if spm agrees to it :)
[07:49] <robin_> kiko: I am pushing it.  Is it ok?
[07:51] <kiko> robin_, you may have to abort and restart if we manage to convince spm to re-roll, just that
[07:52] <robin_> kiko: ok. I stop it. When I can push it again? I wait for it for 2 weeks~~
[07:52] <kiko> robin_, hang on there
[07:54] <robin_> kiko: ok
[07:55] <kiko> robin_, we're not rolling out now. how much time is "long time" when you said that above?
[07:55] <kiko> days?
[07:56] <robin_> kiko: I am not sure. May be several hours.
[07:57] <robin_> kiko: I run the push command and left computer.
[07:57] <kiko> robin_, soon we'll have you testing bzr 2.0 format which should be a big difference?
[07:58] <robin_> kiko: ok. Tell me how to do it is fine.
[07:59] <robin_> kiko: But I hope it can work as soon as possible.
[09:11] <Saviq> danilos: you around?
[09:21] <Peng_> Ohh, branch descriptions are back? Yay!
[10:11] <Peng_> Is it possible to trigger a full remirror? Without having to learn the API? It's just a silly, 1 MB test branch, but I realized that lp:~mnordhoff/+junk/pytz-2008f.dev6rr was badly packed.
[10:13] <kiko> Peng_, I'm not sure what you're asking -- is this a remote branch we are mirroring?
[10:23] <Peng_> kiko: Yeah. I'd like it to be remirrored from scratch, since the current one isn't very optimal. Anyway, it doesn't matter. I'm abandoning that branch.
[10:40] <Peng_> If LP is having problems mirroring a branch, where should I bring it up? Here? Answers?
[10:55] <intellectronica> Peng_: best is to file a question
[10:55] <Peng_> intellectronica: ok
[11:56] <wgrant> Why does the bugtask table not replace the stock product/distro icons with the customised branding, if any?
[11:56] <intellectronica> interesting question
[11:57] <wgrant> It would be nice to just be able to glance at the table and seek immediately to the right task.
[11:57] <intellectronica> i can give it a try, since i'm anyway working on redoing it, but i somehow doubt it will look good or be helpful
[11:57] <wgrant> Hm, OK.
[11:57] <intellectronica> yeah, it's definitely worth a try
[11:57] <wgrant> What are you doing to it? AJAXing?
[11:57] <intellectronica> yup
[11:58] <wgrant> Nice.
[11:58] <wgrant> Except for the whole JS thing.
[11:58] <intellectronica> what do you mean?
[11:58] <wgrant> Coding JS == pain
[11:58] <intellectronica> i mean, it's a nightmare to develop using JS, but is it a problem for you as a user?
[11:58] <intellectronica> oh, you bet!
[11:59] <wgrant> No, no.
[11:59] <wgrant> As long as it works, which Launchpad's seems to in general.
[12:46] <Scurz> hey
[12:48] <Scurz> in a team, for a mailing list, the team's admin can't subscribe himself all team's members ?
[13:07] <rodrigo_> hi
[13:08] <rodrigo_> I'm trying to build a new version of json-glib in my PPA, but want to copy the source package for karmic, so where do I copy it from? (as it's said in https://help.launchpad.net/Packaging/PPA#Copying%20packages)
[13:08] <rodrigo_> 'Copy packages' link on my PPA just allows me to copy from my PPA to somewhere else
[13:09] <wgrant> rodrigo_: You can select the same PPA on that page.
[13:09] <bigjools-afk> rodrigo_: you need to re-upload it if you want to change the package, otherwise there's no need to copy from Ubuntu itself
[13:09] <wgrant> Oh, I see. From real Karmic.
[13:09] <rodrigo_> right, but I want to have the latest for karmic, and just update that
[13:10] <rodrigo_> (or should I just use apt-get source?
[13:10] <bigjools-afk> rodrigo_: if you're changing the source package, yes, apt-get source, modify, upload
[13:10] <rodrigo_> ok
[13:10] <bigjools-afk> copying just re-publishes the same stuff that's already published elsewhere
[13:11] <rodrigo_> ah
[13:13] <aboudreault> If a package doesn't require any modification between hardy,intrepid,jaunty.
[13:14] <aboudreault> You upload it in hardy. and copy binaries from hardy to intrepid and hardy to jaunty. ?
[13:14] <aboudreault> and not the contrary ? (jaunty to intrepid ...)
[13:16] <wgrant> aboudreault: It's more likely to work going from Hardy.
[13:16] <wgrant> Much more likely, in fact.
[13:16] <bigjools-afk> yep, in that respect it's similar to a new Ubuntu series
[13:17] <aboudreault> yeah, that what i've noticed, if I build a package from jaunty...some dependency will have a higher version that needed.
[13:17] <aboudreault> all right. just wanted to be sure :)
[13:30] <wgrant> sinzui: What did you change the series display name to?
[13:30] <sinzui> Mozilla Firefox trunk series
[13:30] <wgrant> Excellent.
[13:51] <ondrej> hi
[13:51] <sanxiyn> Hello, I am trying to set my location,
[13:51] <ondrej> i would like to ask two things
[13:51] <sanxiyn> and when I click Korea and "Update", location is set to Turkey.
[13:51] <sanxiyn> WTF?
[13:52] <ondrej> 1) is it possible to delete a project in launchpad?
[13:52] <sanxiyn> ondrej: https://help.launchpad.net/Projects/Closing
[13:53] <sanxiyn> e.g. you need to file a request, and it's not automatic.
[13:53] <kiko> sanxiyn, seriously?
[13:54] <sanxiyn> kiko: You mean location? Yes 100% serious.
[13:54] <sanxiyn> Try it yourself.
[13:54] <kiko> sanxiyn, what URL are you at?
[13:54] <sanxiyn> https://launchpad.net/~sanxiyn
[13:58] <salgado> sinzui, didn't you investigate something similar to what sanxiyn is describing?
[13:59] <sinzui> salgado: apparently geonames believe Norwich UK is on the Kazakistan border
[14:01] <sinzui> salgado: I believe the mismatch in timezones and names relates to the ancient pytz launchpad uses. I suspect the bug that Launchpad does not know Indian cities will also be fixed by updating pytz
[14:02] <salgado> hmmm
[14:03] <ondrej> sanxiyn, thanks
[14:03] <ondrej> and my second question:
[14:04] <ondrej> there is a branch that is imported from cvs and gets updated regularly
[14:04] <ondrej> unfortunately, the devs of the original project don't like this
[14:05] <ondrej> so i would like to remove it, as i have initiated the whole thing
[14:05] <sanxiyn> Yes, I too think Launchpad VCS import is evil.
[14:05] <ondrej> the question is, is there a way, when i keep my local branch, to update it from the original cvs?
[14:06] <ondrej> sanxiyn, i find it useful simply because working with cvs hurts compared to a modern vcs
[14:06] <ondrej> but the project in question is not my call
[14:06] <ondrej> so i need to have it closed and the branch removed
[14:07] <ondrej> i have created the project only because of the import
[14:07] <kiko> sanxiyn, whyy do you think that?
[14:07] <ondrej> development take place elsewhere
[14:07] <kiko> ondrej, to be honest, that's a perfectly valid use of launchpad
[14:07] <kiko> ondrej, we have lots of projects registered that don't use launchpad
[14:07] <kiko> and that are available so that people can track either code or bugs remotely
[14:08] <ondrej> kiko, i know. that's why i did it and found it perfectly fine. legal by gpl and everything
[14:08] <kiko> ondrej, what project is that? I can get somebody to go and contact them.
[14:08] <sanxiyn> kiko: It's one more place to check -- which people may not want.
[14:08] <kiko> sanxiyn, to check.. for what?
[14:08] <sanxiyn> kiko: You know how controversial Launchpad's translations are?
[14:09] <sanxiyn> People translate on Launchpad, which is wasted, since that's not coordinated with upstream translation, etc.
[14:09] <kiko> sanxiyn, I know there is controversy in many things, from wikipedia to gmail to launchpad translations, yes
[14:09] <sanxiyn> Same for VCS import (if it's not intentional.)
[14:09] <kiko> sanxiyn, so, two things
[14:09] <ondrej> sanxiyn, exactly their argument, they want to keep it central. i don't thing a mirror of the repo breaks that, but i am not in a position to argue
[14:09] <ondrej> it's called cp2k
[14:09] <ondrej> http://cp2k.berlios.de
[14:09] <kiko> sanxiyn, VCS imports are completely read-only, so I don't quite understand what the comparison with translations is there
[14:09] <sanxiyn> ondrej: I do think it's their call (not to speak of I agree with them)
[14:10] <sanxiyn> lol VCS import is read-only :)
[14:10] <kiko> sanxiyn, as for translations, that is true for /upstream/ translations, but there's a set of mechanisms to ensure that the right policy is advertised and applied there
[14:10] <ondrej> sanxiyn, i agree. therefore i do not want to argue and want to have it fixed.
[14:10] <kiko> sanxiyn, for distributions like ubuntu, the requirements are a bit different
[14:10] <ondrej> my opinion is besides the point.
[14:11] <sanxiyn> kiko: Yes, I do understand it has improved (right policy advertised etc)
[14:11] <sanxiyn> On the other hand, initial implementation was very distasteful.
[14:11] <ondrej> still, i like to work with bzr much more than cvs
[14:11] <kiko> ondrej, I'm going to talk to somebody about contacting cp2k; ping kfogel when he is around about this and him and I will see what to do
[14:11] <ondrej> and would therefore like to keep the ability to see a read-only bzr version of the original cvs repo
[14:12] <ondrej> kiko, i would really like to avoid trying to convince them
[14:12] <ondrej> i have already talked to them
[14:12] <ondrej> and been told it was discussed internally
[14:14] <kiko> ondrej, I guess my point is that they can't actually ask for it to be removed IYKWIM
[14:14] <ondrej> although the license of the code allows me to do what i have done and much more, i would like to avoid anything that the devs find problematic, even if it is for reasons i don't agree with
[14:14] <kiko> ondrej, if you find it useful, then it's hard to argue for getting it removed
[14:14] <kiko> and the process in this situation is for us to contact them
[14:14] <ondrej> kiko, i totally agree. formally, it is sound.
[14:17] <ondrej> kiko, if you really want to get into this, then the right place is the mailing list at google groups: http://groups.google.com/group/cp2k
[14:17] <kiko> yeah
[14:17] <kiko> thanks!
[14:17] <ondrej> but please note that i really advice against this
[14:18] <ondrej> kiko, see also my original announcement: http://groups.google.com/group/cp2k/browse_thread/thread/ba88294ae132314e#
[14:18] <ondrej> i have also talked privately with fawzi about this
[14:20] <ondrej> now to get back to the technical point: is there a way to update the imported branch locally?
[14:20] <kiko> ondrej, you can probably run cscvs locally
[14:20] <kiko> I have never tried this, but it's open source: launchpad.net/launchpad-cscvs
[14:24] <wgrant> kiko: I made the mistake of trying to run cscvs once...
[14:24] <wgrant> kiko: It's not fun.
[14:25] <kiko> you can say that again
[14:25] <ondrej> wgrant, exactly my experience
[14:25] <ondrej> that's the reason i ask :-)
[14:25] <ondrej> i know it should be possible *in principle*
[14:25] <wgrant> For Subversion or Git it's easy, as there are good bzr plugins for those.
[14:26] <ondrej> i would appreciate a howto or a pointer to what command actually runs at the servers upon import
[14:26] <ondrej> wgrant, unfortunately, it's cvs in this case
[14:27] <kiko> ondrej, at the server? cscvs works completely remotely from the server
[14:28] <ondrej> kiko, i mean launchpad servers
[14:28] <ondrej> i assume you need cscvs, a local branch and a remote cvs repo
[14:29] <kiko> ondrej, ah. the best person to ask would be mwhudson to be honest
[14:29] <ondrej> kiko, ok, thanks. he did the original import, as there were some trouble. i'll try to ask him.
[14:30] <kiko> thanks
[14:30] <kiko> I'll talk to karl as soon as I'm back
[14:30] <kiko> laters!
[14:30] <ondrej> thanks, bye...
[14:42] <ondrej> mwhudson, are you here?
[14:43] <wgrant> ondrej: It's 2am where he is.
[14:43] <wgrant> So I hope not.
[14:44] <abentley> Peng_: We mirror on a logical level, not a per-file level.  The amount of packing in the source isn't relevant.
[14:44] <ondrej> wgrant, ah, thanks for the sanity check, missed that :-)
[15:25] <slayton> I'm trying to build a package on launchpad but it keeps failing, here is the log http://tinyurl.com/me2m2c here are the relavant lines: http://paste2.org/p/269016  what I don't get is when I build the package on my local machine using the debs hosted in the repository the package builds fine...
[15:26] <slayton> it appears that the build machine on launchpad isn't recognizing that dbus-c++ of the packages is installed, but dbus-c++ contains the correct pkconfig file and I have verified that it gets installed in the correct place on my local machine
[15:26] <slayton> the only way i can replicate the bug on my local machine is to uninstall dbus-c++
[15:26] <slayton> but looking at the build log dbus-c++ is clearly being installed during the build process
[16:38] <geser> slayton: are are the dependencies of the .pc file available?
[16:43] <slayton> geser, it appears that they are... the only thing required by dbus-c++ is dbus-1 which appears to be installed during the build process
[16:44] <geser> but not libdbus-1-dev which contains the .pc file referenced in your .pc file
[16:46] <slayton> ah.... ok I'll add that thanks
[16:47] <geser> slayton: and I'm not sure if pkg-config will find dbus-c++ if the file is named dbus-c++-1.pc (you can try if "pkg-config --libs dbus-c++" works)
[16:50] <slayton> geser, so I  guess I need to rename the package dbus-c++-1
[16:52] <geser> slayton: I don't have much experience with pkg-config so better check yourself how the file should be named (and what else is needed) that pkg-config will find your lib and your build moves further
[16:52] <slayton> geser, ok great. I really appreciate your help
[17:18] <slayton> Hmm.... for some reason my amd64 package doesn't build but the lpia and i386 packages build just fine... the build process is failing on dh_shlibdps, here is the build log http://tinyurl.com/nvg2s6  and here are the relavent build lines: http://paste2.org/p/269107   why would I get this error on amd64 and not on the other 2 build systems
[17:38] <Sam-I-Am> slayton: is there a ppa dependency thats missing its amd64 build?
[17:41] <bigjools> I'm no expert on this, but it looks like the package needs another build dependency for amd64
[17:43] <Scurz> in a team, for a mailing list, the team's admin can't subscribe himself all team's members ?
[17:45] <bigjools> barry: ^
[17:53] <bigjools> or abentley? --^ (since you're on call)
[17:57] <barry> Scurz: no.  team members must subscribe themselves
[17:57] <Scurz> okay
[18:19] <slayton> Sam-I-Am, no all of the packages are present under amd64
[18:30] <abentley> rockstar: skype?
[18:30] <rockstar> abentley, sure.
[18:52] <slayton> bigjools, what do you mean by " it looks like the package needs another build dependency for amd64"
[19:25] <slayton> can anybody explain to me why a build might fail on amd64 but not on lpia or i386?  all the dependencies are either in the ubuntu repos or the ppa... and all of the other amd64 .debs built fine
[21:59] <kklimonda> hmm.. is it a feature that when I push a packaging branch to LP it links to all bugs it closes?
[23:17] <Peng_> abentley: This is a dev6rr branch. If the source branch is not optimally compressed (I forgot to pack after upgrading), doesn't bzr copy that verbatim? If it does in fact recompress it, the LP branch is twice as large as it should be for some other reason. :\
[23:51] <thumper> kklimonda: yes
[23:51] <kklimonda> thumper: is it helpful? packaging branch doesn't actually have a fix for a bug itself.
[23:52] <thumper> well...
[23:53] <thumper> it should only link to branches where there is a mainline revision that says --fixes lp:12345
[23:53] <thumper> it may be that we have to change our linking criteria
[23:53] <kklimonda> when you use debcommit to commit a change to package it uses --fixes lp:12345
[23:54] <thumper> ah
[23:54] <thumper> kklimonda: if it fixes a bug?
[23:54] <kklimonda> yes
[23:54] <thumper> well, if it isn't useful, then we need to work it out :)
[23:56] <kklimonda> maybe i'm using it wrong (it being bzr for packaging the way I use it, lp for publishing changes or something else ;) )? that's my branch: https://code.edge.launchpad.net/~kklimonda/transmission/ubuntu
[23:56]  * thumper takes a look
[23:57] <Peng_> If my question gets turned into a bug, should I mark the question as solved or leave it open until the bug is fixed?
[23:57] <thumper> Peng: no idea
[23:57] <Peng_> thumper: :D
[23:57] <thumper> kklimonda: perhaps we need to change the linking so if the bug is fix commtted or fix released, we don't link ?
[23:57] <Peng_> I should check my old answermail to see what other people do.
[23:58] <thumper> kklimonda: file a bug about the behaviour, what you are doing, and what you'd like to happen :)
[23:59] <kklimonda> thumper: thanks
[23:59] <kklimonda> i'll do it