| [00:30] <idnar> http://blog.launchpad.net/ppa/failed-to-fetch-errors-for-ppas contains what is, as far as I can determine, a really confusing typo | |
| [00:31] <idnar> it looks like "All of these refer to different components within the same PPA. You only need the first four PPAs" should read "for" instead of "four" | |
| [00:31] <wgrant> idnar: Hah, indeed. Fixing. | |
| [00:33] <wgrant> Fixed. | |
| [00:34] <wgrant> idnar: I thought it was right when I reviewed it. | |
| [00:34] <wgrant> Someone actually corrected it to 'four' afterwards... | |
| [00:34] <idnar> heh | |
| [00:37] <lifeless> I would say 'PPAs only need the first component' | |
| [00:37] <lifeless> and avoid the ambiguity | |
| [00:38] <wgrant> Only need, or only have? | |
| [00:42] <lifeless> have would be better | |
| [00:42] <lifeless> I was rephrasing the existing sentence | |
| [00:43] <wgrant> Right, was just proposing further alterations. | |
| [00:43] <wgrant> Is done. | |
| [01:22] <kalikiana> Is there a way to mark all Committed bugs as Released? | |
| [01:31] <wgrant> kalikiana: There's no way to do it directly in Launchpad, but you can do it easily through the API, and there are several existing launchpadlib scripts that do that. | |
| [01:34] <wgrant> lp:~mgiuca/+junk/launchpad-tools is one I've used. | |
| [01:34] <wgrant> But there are many around that are slightly different. Some close all Fix Committed bugs in a milestone, others all in a project, some add comments, some don't... | |
| [01:39] <kalikiana> wgrant: that looks perfectly like what I need, thanks a lot. I have only used the UI so far | |
| [02:00] <lostogre> who here was working on bind 9.8.0rc1? | |
| [02:20] <kalikiana> wgrant: I can't seem to use the script, it eats all my memory + swap and I need to kill it before it does anything :-( | |
| [02:20] <kalikiana> Would you have any idea why that happens? | |
| [02:22] <wgrant> kalikiana: What's the traceback if you kill it while it is eating? | |
| [02:24] <kalikiana> it seems to be in File "/usr/lib/python2.7/site-packages/launchpadlib/credentials.py", line 242, in __call__ | |
| [02:24] <kalikiana> webbrowser.open(self.authorization_url) | |
| [02:24] <kalikiana> it doesn't really make sense that opening a browser would need much memory | |
| [02:26] <wgrant> One would think not... | |
| [02:26] <kalikiana> wgrant: would you happen to know what url I need to open? | |
| [02:27] <wgrant> It varies, because it contains the token URL. | |
| [02:27] <kalikiana> hmm | |
| [02:27] <wgrant> What happens if you strace it? | |
| [02:28] <kalikiana> oh, very good idea, it seems to choke on xdg-open | |
| [02:28] <wgrant> I've never seen that before. Does xdg-open work if you invoke it manually? | |
| [02:34] <kalikiana> looks like it loops. and changing $BROWSER to not be xdg-open fixes it. which is odd because I didn't touch that variables in months | |
| [02:35] <kalikiana> but not an lp issue in any case then | |
| [02:40] <kalikiana> wgrant: works now, thanks again | |
| === ferrousw1eel is now known as ferroushweel | |
| === ferroushweel is now known as ferrouswheel | |
| [08:11] <micahg> uh, I just had a bugzilla.gnome status just go to unknown | |
| [08:29] <lifeless> bug ? | |
| [08:30] <micahg> bug 657227 | |
| [08:30] <ubot5> Launchpad bug 657227 in gnome-shell (Ubuntu) "gnome-shell doesn't respect user font settings" [Undecided,New] https://launchpad.net/bugs/657227 | |
| [08:34] <lifeless> its marked resolved duplicate in the bug watch portlet | |
| [08:36] <lifeless> and the duplication was just now | |
| [08:36] <micahg> lifeless: ok, but still, should it go to unknown | |
| [08:37] <micahg> should I set the new upstream or do you need this for testing? | |
| [08:37] <lifeless> micahg: https://bugzilla.gnome.org/show_bug.cgi?id=636868 | |
| [08:37] <lifeless> unconfirmed | |
| [08:37] <micahg> ok, but unconfirmed = New | |
| [08:37] <lifeless> micahg: I don't know quite what the expected behaviour is here | |
| [08:37] <lifeless> I would file a question about this and leave it be | |
| [08:37] <lifeless> we should be able to do better | |
| [08:37] <micahg> question as opposed to bug | |
| [08:38] <lifeless> yes | |
| [08:38] <lifeless> bugs are for defects | |
| [08:38] <lifeless> questions are for, well, questions | |
| [08:38] <lifeless> 'whats meant to happen' is not a bug :) | |
| [08:40] <micahg> https://answers.launchpad.net/launchpad/+question/146088 | |
| [10:41] <slangasek> is there a way someone here could kill https://launchpad.net/~vorlon/+archive/multiarch/+buildjob/2274512 for me? | |
| [10:42] <slangasek> it wasn't meant to try to build in that ppa, it was supposed to get binaries copied from a different ppa but I was too hasty and tried to copy before the binaries were fully published | |
| [10:44] <wgrant> slangasek: Hm, it shouldn't have let you copy if there were unpublished builds... | |
| [10:44] <slangasek> oops ;) | |
| [10:46] <wgrant> Unfortunately we can't sensibly kill a building build without poking the builder directly. | |
| [10:46] <wgrant> And even if I was to do that, it's possible the aborted build would block you from copying the i386 binaries in later. | |
| [10:48] <slangasek> wgrant: well, I'm already routing around the need to copy by playing towers of hanoi with another ppa, but y'all might want that builder back sooner than 9 hours from now | |
| [10:49] <wgrant> Heh. Let me try to kill it... | |
| [10:50] <wgrant> Ah, actually, I can't. | |
| [10:50] <wgrant> I can do it for recipe builds, but not binary builds yet. | |
| [10:50] <slangasek> ok :/ | |
| [16:21] <lvh> Hello! | |
| [16:21] <lvh> Is it possible to rename a project, or do I ask for it to be removed and create a new one? | |
| [16:21] <lvh> Removal might actually be better: it's a replacement of a dead project. | |
| [16:49] <micahg> lvh: file a question on answers.launchpad.net/launchpad | |
| === marienz_ is now known as marienz | |
| [16:58] <maxb> lvh: Renaming is possible, if it makes sense to do so | |
| [17:03] <JoshBrown> How do I remove tags from a Bazaar branch on Launchpad? | |
| [17:05] <maxb> JoshBrown: 'bzr tag --delete -d lp:.... tagname' | |
| [17:06] <maxb> However... any merging of pushing from another branch which still has the tag will re-add it | |
| [17:06] <maxb> s/of/or/ | |
| [17:08] <JoshBrown> maxb: Thanks, that's exactly what I'm looking for! | |
| [17:27] <lvh> micahg: Thanks, I just did. | |
| === kalikiana1 is now known as kalikiana | |
| [17:48] <JoshBrown> When you create an original deb package on Launchpad, what should it be called? Am I correct in thinking it should be something along the lines of `name_1.0-0ubuntu1~ppa1_all.deb`? | |
| [17:50] <micahg> JoshBrown: launchpad only accepts source uploads | |
| [17:51] <JoshBrown> micahg: Okay, but nonetheless, what should the package be designated as? Do I include ubuntu1, ~ppa1, etc? | |
| [17:54] <micahg> JoshBrown: I usually do ~series~ppa1 where series is the release targeted, if this is the first upload of that version, you can use UPSTREAMVERSION-0ubuntu1~series~ppa1 | |
| [17:58] <JoshBrown> micahg: I was mainly wondering whether it was `0ubuntu1` or just `0`. Thanks, that answers my question! | |
| [17:59] <micahg> JoshBrown: technically either would be fine since 0 is lower than 0ubuntu1 | |
| [18:56] <maxb> JoshBrown: People exercise varying degrees of care about version numbers in PPAs, and there are several different schools of thought | |
| [18:57] <maxb> For example, I think micahg's example is inverted, in the sense that series is usually the least significant part of your packaging efforts, so goes at the end. | |
| [18:57] <micahg> maxb: the problem with that is you override official backports | |
| [18:58] <maxb> Also, I'll explicitly avoid including 0ubuntu1 in versions unless I'm actually deriving a PPA package from an official ubuntu package version including that | |
| === Lcawte|Away is now known as Lcawte | |
| [18:59] <maxb> hmm. I cynically care little about that on the basis that official backports seldom seem to actually happen :-/ | |
| [19:00] <JoshBrown> maxb: So you'd package as `name_1.0-0~ppa1~maverick`, for example? | |
| [19:01] <maxb> I usually go with ~maverick1, and increment that number if I need to make a series-specific build fix | |
| [19:02] <JoshBrown> maxb: Great, thanks! | |
| [19:02] <maxb> Also, I tend to call things 0ppa1 rather than 0~ppa1 | |
| [19:03] <maxb> Mainly because I believe in using the magical ~ operator only when relevant, rather than as an arbitrary separator | |
| [19:04] <maxb> Though I admit that I'm relying on the happenstance that 'ppa' < 'ubuntu' in doing so | |
| [19:09] <micahg> maxb: actually, now with the extras repo, that isn't necessarily a good idea as it's 0series1 for the versioninig | |
| === menesis1 is now known as menesis | |