=== jamesh_ is now known as jamesh [02:28] hello all [03:00] New bug: #220082 in launchpad ""Top contributor" is whack" [Undecided,New] https://launchpad.net/bugs/220082 [03:01] * RAOF guesses that's been filed by jml :) [03:05] hee hee [03:17] heh [03:18] * RAOF scores! [03:24] How do you all test your package installations? Is there a way to actually _install_ the package with pbuilder? [03:24] awmcclain: Yes. 'pbuilder login' will log you in to the pbuilder environment. [03:28] RAOF: And then what, apt-get or dpkg install it? Doesn't everything get lost when you do pbuilder build? [03:29] awmcclain: Yes, it does. And yes, you then dpkg --install it. [03:29] RAOF: so; login to the pbuilder environment, build the package, then install it? [03:30] No. Build the package. Login. Copy the built package in (using bind mount, or whatever), then install it. [03:35] ROAF: But wait, if you tried to install the package using dpkg inside pbuilder, woudln't you have to manually download all the dependencies for running the package? [03:36] Hobbsee: hi are you here? [03:36] awmcclain: Yes. That being one of the points. [03:36] awmcclain: You'd dpkg --install it, then run apt-get -f install to install the broken dependencies. [03:38] RAOF: Ahhh! I'd been asking about that second step in other channels for 2 months! I didn't think it'd work. Is bind mount documented in a pbuilder doc or man page somewhere? [03:40] awmcclain: Probably. I'd be somewhat careful with bind-mount, obviously. I've accidentally blown away ~ with a not-careful-enough rm. [03:40] RAOF: Wonderful. Exactly what I want to hear. [03:40] ;) [03:42] awmcclain: It's possible to do the bind-mounting automatically and safely with pbuilder, I know. I've just never bothered to learn (particularly since I use sbuild) [03:42] RAOF: Ok, i'll poke around. [03:43] BINDMOUNTS=/var/cache/pbuilder/result [03:43] ;) [06:02] does anyone know witch language pack to channel source to translation? [06:36] how do I clear a bzr push lock? [06:36] bzr break-lock [06:38] no workie [06:39] You may need to retry it several times if you'd tried several pushes while it was locked. [06:39] (to work through a backlog of processes on the server waiting to take the lock) [06:39] ah [06:39] Alternatively, they'll time out eventually. [06:40] how long is eventually? [06:40] (They ought to time out immediately, I'm not sure why they don't. There's a bug about this somewhere...) [06:41] Several minutes, roughly. I'm not sure exactly how long, because I don't understand exactly why they don't give up immediately like they should. [06:41] someone in ubuntu-motu answered my q [06:42] the ssh process is still around [06:43] Oh? Which ssh process? [06:44] hmm - actually that didn't solve it [06:44] there were some ssh connection processes hanging around on my machine [06:44] but === asac_ is now known as asac [06:45] Interesting! [06:45] * jetsaredim =>> #bzr [08:00] Goooooooooooooooooood morning Launchpadders! === jetsaredim is now known as jetsaredim_ [08:15] morning [08:23] hello [08:23] Can I request that a vcs-import be stopped on a lp project in here? [08:23] clsk: you can, although the relevant people aren't necessarily around [08:24] clsk: so filing a question at https://answers.launchpad.net/launchpad-bazaar is probably more reliable [08:24] ok [08:24] thanks [08:25] hi all [08:31] jtv: here? [08:31] wow! [08:31] jml's change for easier branch linking landed [08:31] that is just _so_ nice [08:32] mdke: yes, here [08:32] jtv: thanks for your emails. I just wanted to check whether the po files I'd edited are now fixed... [08:33] mdke: want me to re-upload them? [08:33] jtv: can you tell by eye from the diff whether the changes I made are appropriate? [08:33] or do you need to upload them to be sure? [08:34] mdke: pulling in the upload as we speak. [08:34] mdke: uploading is an effective way to find out! [08:34] maybe someone could tell me the way to see all new translations, translated in launchpad ? I only see "changed in launchpad" filter, but I need to see translations, added in launchpad [08:34] jtv: alright. But I didn't want you to go to the trouble if I'd edited them the wrong way or something [08:35] That's "new in launchpad." [08:35] Those are messages that weren't translated in the imported files, but are translated in Launchpad. [08:36] mdke: and yup, now that I diff, I see a problem with line endings. [08:37] mdke: So the \n is gone but the bad line ending that probably caused the problem is still there. [08:37] jtv: damn. in each case I removed the \n on the following line and added a " on the end of the preceding line... [08:38] mdke: Hmm... you're not editing on a Mac by any chance? I see two quoted-string lines concatenated with a ^M between them. [08:38] jtv: with gedit on Ubuntu [08:41] mdke: Where the problem happened, there used to be a ^M. Those ^Ms now seem to have replaced the newlines. [08:42] jtv: maybe this is a problem with gedit. I don't really understand anything about editors or line endings [08:42] mdke: when it comes to line endings and platform interoperability, I don't think there's a single right thing to do. It's just a mess. [08:43] jtv: is there a simple reliable problem I can open these po files with? Perhaps if I use poedit or something? Will that sanitise them? [08:43] problem/program [08:43] How Freudian [08:43] * mdke didn't say anything about his mother [08:43] mdke: I'm trying now [08:44] "Now, how did your mozzer come into zis..?" :) [08:44] ;) [08:44] jtv: thanks. I have to go to work now, perhaps we can continue by email. Seems we are a way off sorting this out. But there is no deadline for this [08:45] mdke: try vim. It shows the ^M as a weird character. [08:46] mdke: bzr diff -r date:2008-04-10 | grep '-.*\\n' should also reveal the problems (amid a bit of extra output, unfortunately) [08:46] mdke: sure, back to email! [08:47] jtv: thanks [09:04] jtv: I don't see filter "new in launchpad" :( look for example https://translations.edge.launchpad.net/ubuntu/hardy/+source/jockey/+pots/jockey/lt/+translate [09:10] mantiena: you're right, sorry. There's only "changed in launchpad." [09:13] jtv: so, maybe there is another way, which doesn't exist in UI ? Maybe I can add some parameter in URL ? [09:13] mantiena: nope, sorry! [09:15] mantiena: otoh I believe "changed in Launchpad" will show you these messages (in addition to the ones that were already translated but have a different translation in Launchpad) [09:15] jtv: no, "changed in launchpad" filter displays only changed messages :( [09:16] at least in edge.launchpad.net [09:19] mantiena: looks like it. Not much time to change that at the moment, unfortunately, unless it's urgent. :-/ [09:40] Fujitsu: around? [09:40] bigjools: I am. [09:41] Fujitsu: hey there - quick question for you, we're mooting the idea of allowing anyone who can upload to a component retry its builds as well. How does that sound to you? [09:41] That sounds very sane. [09:42] I've been wondering for quite some time why that wasn't already how it worked. [09:42] right - they can already tie up buildds anyway ... [09:42] Seems easier than adding extra celebrities, although you *might* still want that for rescoring. [09:42] indeed [09:42] It's how it works for PPA, and yes, we can just upload rebuilds anyway. [09:50] New bug: #220156 in malone/1.2 "Deal with persons not having an e-mail address when importing bug comments" [Undecided,Confirmed] https://launchpad.net/bugs/220156 [10:22] New bug: #202301 in dell "Firefox not translated/localized (dup-of: 195013)" [High,Confirmed] https://launchpad.net/bugs/202301 [10:30] New bug: #220170 in soyuz "build priority of language packs too high" [Undecided,New] https://launchpad.net/bugs/220170 [10:31] Yay! [10:33] <\sh> woot? [10:36] \sh: It'd be great to see it fixed. [10:39] <\sh> Fujitsu, rebuilds? [10:41] \sh: Langpacks should be scored lower than everything else. [10:41] <\sh> Fujitsu, ah this one [11:02] Does anyone know the difference between "Packaging information" and "Show distribution packages" on a project's page? [11:04] mpt: One's on the homepage, badly designed and ordered incorrectly. The other isn't. [11:04] ah, they're exactly the same info, just with different columns [11:04] vs. [11:04] * mpt giggles [11:05] Oh, I didn't know there was a +distributions. [11:05] I think you need to wave your wand over it. [11:05] or my Delete key. [11:05] Perhaps so. [11:06] New bug: #218848 in language-pack-gnome-fi-base "XML error when going to site needing certificate exception when using fi_FI.UTF-8 or de_DE.UTF-8" [Undecided,Fix committed] https://launchpad.net/bugs/218848 [11:06] The question is which one is more useful. [11:07] Both layouts are useful :( [11:10] Okay, how is each useful? [11:10] I mean, how is each useful in a way that the other isn't [11:11] Well, if you ever manage to convince another distro to use LP, or import Debian stuff, the layout of each will diverge significantly. [11:11] I might want to know where I can get version X of this project. [11:11] Or I might want to know what versions distro Y has of this project. [11:12] ALthough the latter usecase is handled by the SP page already. [11:13] So +packages shows information in a new way, while +distributions really just aggregates what's on //+source/ [11:15] so that's just a sort order, really [11:15] No. [11:15] or a grouping [11:15] Right. [11:15] Grouping. [11:15] grouping by distribution vs. grouping by series [11:15] Yep. [11:16] ok [11:16] That's the only non-textual difference between the pages... how very odd. [11:35] New bug: #220183 in launchpad ""Packages by distribution" vs. "Distribution packages"" [Undecided,New] https://launchpad.net/bugs/220183 [11:41] Looking further through the upstream linking stuff, the UI is one of the nastiest I've seen in LP... [11:42] See the `Update' link on https://edge.launchpad.net/ubuntu/feisty/+source/soundconverter, for example. [11:42] It points to a page which contains a link to do the actual updating. [11:42] And one cannot delete the link by providing an empty value for the series. [11:47] actually that link is on the page you were on before anyway [12:45] poolie: yes [13:05] New bug: #220207 in blueprint "I can register a blueprint for a project that "Doesn’t use Blueprints"" [Undecided,New] https://launchpad.net/bugs/220207 === kiko-afk is now known as kiko [13:19] fsking launchpad. [13:19] er, echan. that was supposed to go to -release. [13:20] * Hobbsee wishes launcphad would not break around the time of release! [13:20] Error ID: OOPS-842EC22 [13:21] good afternoon kiko [13:22] hey Hobbsee === mrevell is now known as mrevell-lunch [13:22] Hobbsee, where's that happened? [13:22] kiko: https://edge.launchpad.net/ubuntu/hardy/+queue?queue_state=1&queue_text=&start=20 (timeout) [13:22] it's a timeout [13:22] keep trying [13:22] kiko: i haven't been able to accept from the queue in a good couple of days. [13:23] Hobbsee, non-edge any better? [13:24] kiko: nope. (OOPS-842B1655) [13:24] https://devpad.canonical.com/~jamesh/oops.cgi/842B1655 [13:25] kiko: it hasn't worked since launchpad went down for 13+ hours yesterday, and was having trouble a week or so prior, too. [13:25] then it worked midweek. so go figure. [13:28] oh good, launchpad has logged meetings [13:28] 4am local. no way i'll make that. [13:42] kiko: any other ideas? [13:52] er, make that 14+ hours. [13:53] annoying+ hours, at any rate. [13:54] * Hobbsee grumbles at the current borkage which means that the queue can't be processed without drescher access. [13:54] how's that for annoying, right around release? [13:54] Hobbsee, get bigjools to help, but I wonder why nobody in Ubuntu proper has the same problem. [13:54] bigjools is definitely the person who knows what's going on with that [13:54] Everyone else uses drescher. [13:54] kiko: because all the canonical people use the SSH access, as it's quicker and more convenient, and i'm the only one who uses that section. [13:55] of course [13:55] Hobbsee, we'll get a fix and have it rolled out. let me talk to bi [13:55] * kiko cringes [13:55] julian. [13:55] bigjools: any idea on what i can do, short of sit back and relax? [13:55] heh [13:55] (he's out for lunch) [13:55] ahhh [13:56] Hobbsee, do you often override section and binary priority? [13:56] kiko: no [13:56] okay. [13:56] kiko: components would be the regular one [13:57] Hobbsee, except new packages are already overridden by default to universe, now? [13:57] kiko: sure, but that won't help in the case of multiverse packages. [13:57] kiko: so, fsvo "regular" - each time we get a multiverse package [14:11] Hobbsee: I am taking a look at your timeout problem. It certainly looks like some stuff can be optimised. [14:11] bigjools: right [14:20] Hobbsee, we know what the problem is; it's a consequence of calculating bug subscribers properly and closing bugs from changelogs. [14:20] Hobbsee, intellectronica and I are discussing it. [14:20] kiko: cool, OK, thanks. [14:20] kiko: i thought they were doing that when they first hit unapproved, whether they ended up getting approved or not [14:23] no, only when accepted. === jetsaredim_ is now known as jetsaredim [14:25] kiko: i thought i saw the opposite happen. [14:25] kiko: i went looking for a bug, but it had been marked as fix released, evne while it was still stuck in unapproved. [14:26] Hobbsee, maybe somebody marked it manually -- that's not the same thing. [14:27] kiko: i didn't see mail indicating that, but perhaps. === mrevell-lunch is now known as mrevell [14:43] Hobbsee, argh, this bug is *really* hard to fix [14:43] kiko: heh :) [14:43] kiko: i'm not surprised, as it's not the first time it's broken. [14:44] kiko: if you dont' get it rolled out in the next 24h or so, it can wait till the next cycle, but it definetly needs to be fixed and working for intrepid, due to the stuff that will soon be announced. [14:45] kiko: why was it working midweek then? [14:45] Hobbsee, sheer luck and statistics -- performance is like that. [14:46] Hobbsee: it's going to be directly related to the number of simultaneously accepted items [14:46] bigjools: ahhh. i usually only accept one at a time. [14:46] Hobbsee, it's not more broken now than it was before, IOW -- you just can't notice the cause [14:46] kiko: hehe, yeah [14:46] Hobbsee: right - so try one at a time and see if it works for you as a temporary workaround [14:47] bigjools: got one of the other people in the release team to accept a whole bunch, but will try with the next one [14:47] (or will try with them asleep) [14:47] great, let me know how that works out for you [14:51] Hobbsee, actually, you know what? https://edge.launchpad.net/launchpad/+bug/219346 is to blame :) [14:51] Launchpad bug 219346 in tseries "r-base transition" [Undecided,In progress] - Assigned to Cesare Tirabassi (norsetto) [14:51] Ahaha. [14:52] That's not even a very big bug! [14:52] Fujitsu, heh :) [14:53] kiko: why? [14:54] Hobbsee, because we need to notify implicit subscribers for all those packages, which means lots of queries for calculating [14:54] kiko: twitch. [14:55] Doesn't a load of the bug page itself have to also perform those calculations? Or is the main work in the actual notifying bit? [15:00] Hobbsee, for now the workaround is to ask someone with shell access to approve any upload which fixes a bug with many (3+ tasks) [15:00] Hobbsee, you can probably approve other uploads without any problem [15:01] Hobbsee, and we're looking into this bug for 1.2.4 [15:01] kiko: OK, cool, thanks. [15:01] kiko: i wish you luck on your extra menu agenda item, too [15:02] Hobbsee, which extra menu agenda item? :) [15:02] kiko: the one mpt said that he put on there yesterday. [15:02] That agenda item has nothing to do with menus afaik [15:03] * Hobbsee wonders where she got menu from. [15:03] oh, i meant *meeting* [15:03] bad tab-completion :-) [15:03] mpt: bad brain completion. i've already been told off for not being psychic today, and actually doing what i was told. [15:04] mpt: so i'm thinking the world is somewhat broken - my brain should be allowed to be too! [15:04] kiko: looks like i can accept one upload at a time, yes [15:06] Hobbsee, as long as it doesn't touch that bug, as I told you [15:07] Hobbsee, in fact, I bet you can approve many uploads at a time as long as none of them fix that bug. [15:07] kiko: already tried that. i can't. [15:07] kiko: tried to accept 2 sugar packages, got a timeout, so tried again with 1, which worked. [15:08] Hobbsee, can I see the OOPS ID for that one? === rockstar` is now known as rockstar_ [15:08] kiko: i don't still have the #, sorry - but i wouldn't imagine it would be too hard to find? [15:09] Hobbsee, it's actually pretty hard to find [15:09] well, until tomorrow. [15:09] oh right [15:09] kiko: i'll try dogfood? [15:10] Hobbsee, no difference. [15:10] Hobbsee, just paste in OOPS IDs if you actually do get OOPSes which aren't in actions that try to approve bugs with many tasks. === kiko is now known as kiko-afk [15:25] New bug: #220251 in malone "Calculating subscribers for bugs with multiple tasks is expensive" [High,New] https://launchpad.net/bugs/220251 === Rinchen` is now known as Rinchen === Rinchen is now known as rinchen [15:33] hi, how do i get a sponsor for my updated package? [15:33] spacepluk: Perhaps by waiting about two weeks, and asking in #ubuntu-motu. === rinchen is now known as Rinchen [15:34] Fujitsu: thanks, I've asked on #ubuntu-motu but got no response. I guess they are very busy with the release. [15:35] spacepluk: There's a bit more activity now, but you won't have a chance of getting an update in for another couple of weeks unless it fixes a very critical bug. [15:36] Fujitsu: It doesn't :P [15:41] New bug: #220257 in launchpad "Shipit permissions broken" [High,New] https://launchpad.net/bugs/220257 [15:41] New bug: #220258 in soyuz "Accepting queue items causes some very slow SQL queries" [High,Confirmed] https://launchpad.net/bugs/220258 [16:21] hi all [16:21] i just need help on ubuntu mirroring [16:21] we have a mirror and it seems uptodate [16:21] https://launchpad.net/ubuntu/+mirror/ubuntu.fastbull.org-release [16:22] but it doesn't appear in launchpad list [16:22] and also in ubuntu list [16:44] lamont, do you have any sage advice for bostik ^^ [16:45] lamont: please help me ;) [16:47] it was disabled (out-of-date?), was probed an hour ago... I expect that LP just hasn't gotten around to caputring all the now-current archives back into the cd archives list [16:51] ok [16:51] i'm setting up a new second fastbull mirror then i will add to lauchpad [17:22] Can launchpad.net be used to translate a project hosted elsewhere? VCS system is Subversion. [17:22] I've read through the launchpad.net Help docs a bit... it seems like it may be possible but I just wanted to confirm, if possible. [17:22] I've registered it: https://launchpad.net/mifos/ [17:23] and was just looking into "registering a Series"; prompted by https://help.launchpad.net/VcsImports [17:26] ah, nevermind, I think I found where to set the upstream source ( https://launchpad.net/mifos/trunk/+source ). Nice! Launchpad is pretty sweet. === mrevell is now known as mrevell-dinner [18:05] New bug: #220305 in soyuz "The queue page does not update the result count after accepting/rejecting" [Undecided,New] https://launchpad.net/bugs/220305 === TeTeT_ is now known as TeTeT [18:50] New bug: #220316 in soyuz "subscribe to all uploads to a ppa" [Undecided,New] https://launchpad.net/bugs/220316 === stdin_ is now known as stdin [20:27] Is it possible to rename a project in launchpad? [20:31] Seveas: yes if you open question :) [20:33] emgent, nice [20:35] that should probably be on help.launchpad.net somewhere... [20:35] nah [20:36] https://answers.edge.launchpad.net/launchpad/+question/28529 [20:37] yeah, saw quite a few of those types of requests: http://tinyurl.com/4j5nnm [21:43] hi [21:44] how i should upload a orig file with DSC? [21:45] i just started using ppa, and want to build a package which original tarball is not uploaded [21:45] so i got this error: Unable to find bicon_0.2.0.orig.tar.gz in upload or distribution. [22:05] hey there [22:05] is there some RDF guru around? [22:06] I'm trying to parse launchpad's RDF export on groups. I currently use this query: SELECT ?nick WHERE (?any, , ?nick ), which works, but includes the name of other groups. [22:07] how to modify that query so that other (recursive) groups do not appear? [22:11] siretart: you may wanna try #swig [22:11] swig as in the language binding generator? [22:11] siretart: no, this one is semantic-web interest group [22:12] heh, neat. I thought that was a snub. :) [22:12] the channel of binding generator is #swig on irc.efnet.org [22:12] ah, I see [23:21] New bug: #220390 in launchpad "Update navigation menu API to support having a title label over the menu" [Undecided,New] https://launchpad.net/bugs/220390 === barry is now known as barry-away [23:52] Launchpad is hiding the email address of a user from me [23:52] can someone please tell me the email address of Marien Zwart ? [23:53] bignose, it's hiding it from you because you're not logged in :)