[00:20] hello [00:22] how would I go about the following issue?: a cross-compiler depends on pre-compiled runtime libraries. with them, I build the compiler. then I can build the runtime libraries again. So the runtime libraries depend on themselves. [00:23] That sounds like a typical bootstrapping problem; I'd guess off the top of my head that gcc does that, and I'm fairly sure mono does, too. [00:25] it is possible to build them in many steps, but then I'd have to bundle the runtime source with gcc and have a somewhat complex debian/rules. [00:26] You're bootstrapping; you're going to have a somewhat complex debian/rules *regardless*. [00:27] heh, makes sense. [00:27] You can bundle binaries in the source package as long as they get rebuilt during the build - ie: patching some source and then rebuilding should Do The Right Thing™ [01:46] poor chris. [01:51] I wonder what's bouncing me. [01:51] RAOF: "Excess Flood" [01:52] That seems odd, given I wasn't actually doing anything at all. [01:52] Indeed [01:52] Perhaps a bug against smuxi is in order. [01:55] RAOF: there was a huge netsplit a little while ago [01:56] Ah. Maybe that triggers it? Was that when I started bouncing? [01:57] RAOF: yes [02:12] what's the program that's similar to 'apt-get source package' but instead pulls the latest code from launchpad? [02:13] stochastic: pull-lp-source [02:13] thanks micahg [02:13] hi all, can anyone look here: https://bugs.launchpad.net/ubuntu/+source/cln/+bug/508995 and tell me why the patch I provided doesn't work? [02:13] Ubuntu bug 508995 in cln "Please merge cln (1.3.1-2) from Debian Testing" [Undecided,New] [02:42] wgrant: O hai -- how long does it take qa.uw.com/ftbfs to notice something does now build? [02:43] StevenK: It makes a *lot* of requests to Launchpad, so it only runs every six hours. [02:43] That frequency is arbitrary. [02:45] wgrant: Yeah, sounds reasonable [02:56] wgrant: so the next step is to get some batching happening in the API? [03:12] Oh, poot. I misspelled “Requires.private” as “Requires.Private”. [03:15] Hey folks! Anyone up to advocate/review qt-shutdown-p? It is a small program to shutdown the computer. http://revu.ubuntuwire.com/p/qt-shutdown-p [03:50] anyone know if there is a way to tell pbuilder-dist to save changes? I can't seem to get the pbuilder --save-after-login flag to work with it [03:52] depends on the type(s) of changes, but generally, --save-after-exec --override-config [03:52] crimsun: I just need to add my ppa to my apt sources [03:53] normally I'd login and do --save-after-login, add the ppa exit and all would be good [03:53] but with pbuilder-dist it doesn't seem to honor that flag (or I'm doing something silly) [03:55] It does (or should) honour that flag, but it also overrides a bunch of your config with each invocation. Say, for example, that you don't want mirror.kernel.org as your mirror? Currently, you're out of luck with pbuilder-dist. [03:56] That might also cause it to kill your added entries in sources.list; I'm unsure. [03:59] RAOF: well I used pbuilder and pointed at the resulting .tzr file and it seems to have saved it, but oddly enough when I logged in with pbuilder-dist it didn't show me the saved data [03:59] I'll go ahead and build and see if it works [03:59] I need my PPA to match a dependency that isn't pushed yet [04:03] crimsun: hi there, can you have a look here: https://bugs.launchpad.net/ubuntu/+source/cln/+bug/508995 if you have a minute or so [04:03] Ubuntu bug 508995 in cln "Please merge cln (1.3.1-2) from Debian Testing" [Undecided,New] [04:05] crimsun: why is the patch not applying? I did it using bzr [04:14] if something is marked won't fix, if it's released with a proper changelog entry, will the status change to Fix Released? [04:15] RAOF: well, I went in with pbuilder added them, updated, everthing is ok. Then I run pbuilder-dist and it doesn't seem to obey :( [04:15] guess I'll have to go straight up pbuilder for now [04:16] micahg: You mean: there's a changelog entry * foo, bar: fixes (LP: #123456), where bug 123456 is currently set to “won't fix”? I think that'll set it to fix released. [04:16] Launchpad bug 123456 in xine-lib "podcast crashes amarok" [Undecided,Fix released] https://launchpad.net/bugs/123456 [04:16] RAOF: that's my question :) [04:17] I'm fairly sure it'll set it to fixed released. Do you not want it to? [04:17] RAOF: no, I want it to so I don't have to make more bug noise ;) [04:18] micahg: Although I have to wonder why something that's marked wontfix is getting fixed. [04:19] ScottK: source package change [04:19] Generally we're pretty conservative about putting that on a bug. [04:19] firefox 3.6 will return to the firefox source package which was formerly firefox 2 [04:20] * micahg put it there in a lot of cases [04:21] it was correct at the time [04:33] IF it was correct at the time, why isn't it still correct? "Won't Fix" should be reserved for stuff we really aren't expecting to fix. Notabug and the like. === micahg1 is now known as micahg [04:38] persia: it was firefox 2 which was won't fix, now it's firefox 3.6 which will be fix released [04:38] we intended to stick with versioned source packages until upstream decided to do rapid releases [04:38] I still think it oughtn't have been won'tfix. That should only be used where we specifically intend not to fix something. [04:39] Getting it fixed later, even by accident, makes us look indecisive. [04:40] persia: so, you're saying I should comment anyways and explain? [04:40] Leaving a comment in the bug explaining why circumstances have changed would mak sense, and changing the bug from "wontfix" to "triaged" or something with the comment. [04:41] Then you go from "Triaged" to "Fix Released" with the upload. [04:41] persia: ok, I'll do that, it's only 3 bugs in this case [04:41] But try to avoid these :) [04:41] yep, it's unfortunate, but necessary [05:26] persia: Sometimes circumstances change and so wontfix should change too. [05:27] ScottK: Agreed. I just think that 1) it deserves explanation when it happens, and 2) we should try not to do that. [05:27] Certainly. [05:28] I'd rather look indecisive than make present virtue out of past necessity. [05:30] I'd rather avoid both, but given the choice I have to agree. [07:11] good morning [07:12] morning === ApOgEE__ is now known as ApOgEE [09:42] BlackZ: so does autotrash use a setup.py based installation? [09:42] dholbach, yes [09:43] you could try just using /usr/share/doc/debhelper/examples/rules.tiny (with debhelper 7) [09:43] DktrKranz: do you have a good example for an python app using dh7? [09:44] dholbach: gaupol :) [09:44] hi there, i submited an new package version to an bug report https://bugs.launchpad.net/ubuntu/+source/python-pysnmp4/+bug/509495 what does i nedd to do now, to get it replaced in repositry [09:44] Ubuntu bug 509495 in python-pysnmp4 "pysnmp version switch broken" [Undecided,New] [09:45] thanks POX [09:45] BlackZ: apt-get source --diff-only gaupol; less gaupol*.diff.gz [09:45] BlackZ: that'll give you an idea how to do it [09:46] dholbach, thx [09:46] BlackZ: and if you get stuck, just ask in here, somebody will be able to help you :) [09:55] dholbach: zless, I guess (and gaupol has to be >= 0.14-1) [09:55] POX: less works too [09:58] not here (I don't have LESSOPEN defined) [10:02] POX: ah ok [10:04] johe|work: you need to provide a debdiff (modification to the *source* package), and see https://wiki.ubuntu.com/SponsorshipProcess [10:04] randomaction, thx [10:04] johe|work: see also https://wiki.ubuntu.com/Bugs/HowToFix (it tells you how to produce a debdiff) [10:27] Could someone please review package ipwatchd ? http://revu.ubuntuwire.com/p/ipwatchd === dholbach_ is now known as dholbach [12:30] dholbach: ping. regarding the whole MC losing quorum thing, how should i go about applying as a MOTU, if i can't make the next meeting? (it's 3AM local time) [12:33] hyperair: You could try an email application, but otherwise, you'd do best to wait until the 2nd February DMB meeting, at which we can hope some direction is provided. [12:33] Note that we haven't had any successful email applications in a while, and getting it complete in the next six days would be tricky. [12:35] Althernatively, suck it up and get up in the middle of the night ... [12:36] * persia expects some MC members to be attending in the middle of the (local) night. [13:20] dholbach: I finished my job journey and I'm back town, so I can regularly take my session tonight :) [13:47] Hello [13:48] Quintasan: hi [13:48] slytherin: \o [13:48] Why do we have two links pointing to different FTBFS pages in topic? [13:52] Because one if packages that failed to build in the archive and the other is ones that failed to build in a rebuild test. [13:53] ScottK: So is the second url relevant all the time? [13:54] slytherin: Yes. lucas has been doing regular rebuild tests for us this cycle. [13:54] Then it is fine. === dholbach_ is now known as dholbach === highvolt1ge is now known as highvoltage [15:33] I'm using reprepro to manage a repository. When I included a new version of a package, the old one was deleted because it was unreferenced. How do I make it so that all packages always remain in the repository, and it's the user's job to select which one he wants? (and by default he gets the latest version) [15:55] maco, around? [15:55] ramiro: You'll need to either use less smart repository software, or very custom solution. [15:57] ugh, a less smart repository software means I'll have to be less dumb =). which one do you suggest? [15:57] I'm not the right person to suggest any: I don't manage repositories (but we don't much generally here). apt-ftparchive is probably the least fancy of the options. [16:06] ramiro: dak maybe :) [16:09] freeflying: Doesn't dak autoexpire superceded stuff? Anyway, I can't imagine it being useful for less than 10,000 packages. [16:10] persia: Depends on the pain to package ratio you find acceptable. [16:10] heh. I suppose :) [16:11] persia: mini-dak then :) [16:16] ajmitch_: if you have a bit of time - do you think you could help out with reviewing the zope* packages on REVU? === yofel_ is now known as yofel [16:18] Considering there's an active Zope team in Debian, wouldn't they better go there? [16:22] The moreso because they describe themselves as the "Debian/Ubuntu Zope Team" [16:28] hi, if I'd like to have a package merged from debian - what can I do? -https://bugs.launchpad.net/ubuntu/+source/tiemu/+bug/221332 [16:28] Ubuntu bug 221332 in tiemu "The tiemu package is heavily outdated" [Wishlist,Confirmed] [16:31] ? [16:31] benste: First step would be to look at the changes Ubuntu has from Debian in the package and see if they are still needed/relevant. [16:31] benste: There are outstanding Ubuntu changes that need to be investigated before the package is synched. [16:32] benste: From a quick glance, I believe it does need a merge, to deal with iceweasel/firefox differences between Debian and Ubuntu. [16:32] why should tiemu use FF ? [16:32] - where does i have to look for it ? [16:33] benste: In Debian, in tiemu 3.02-1, there is a dependency declared on "iceweasel | www-browser". [16:33] and whom does I have to propose to include the new version even if it's ok - so already build a ubuntu version [16:33] k [16:33] This dependency needs to be modified to be "firefox | www-browser" [16:34] persia: only changing depency is simple isn't it - simply download source, change config and rebuild deb right ? [16:34] Err, sorry. Needs to be "firefox | abrowser | www-browser" [16:34] what about the already build Ubuntu package [16:34] benste: Don't even bother rebuilding. We work exclusively with source packages. [16:34] inlcuded in the PPA of one poster [16:35] persia: Why? [16:35] persia: changing depency is all - if so I'll do right away [16:35] Download the Debian and Ubuntu package sources. Merge the changelogs. modify debian/control to have the right dependency and use "Ubuntu Developers " as the Maintainer. [16:35] I think the odds of anyone having no package that provides www-browser installed is effectively nil [16:35] Put the previous maintainer in XSBC-Maintainer [16:36] Update the changelog, and attach a debdiff to the bug. [16:36] persia: would be kind of you to explain me how to do this - or give some simple resources which will guide me through this process [16:36] ScottK: Because that's what the mozilla team declared should be the dependency. See bug #272772 [16:37] Launchpad bug 272772 in zekr "packages that Depend/Recommend/Suggest firefox (meta-package) must alternatively Depend/Recommend/Suggest abrowser" [High,Fix released] https://launchpad.net/bugs/272772 [16:37] !merge [16:37] https://wiki.ubuntu.com/UbuntuDevelopment/Merging [16:37] benste: That URL should give some guidance. Ask here if you get stuck. [16:37] persia: since it currently Depends/Recommends neither, it's not applicable [16:37] I'll [16:37] fo [16:38] ScottK: Hrm? Are you looking at the same tiemu I'm seeing? [16:38] Depends: libatk1.0-0 (>= 1.20.0), libc6 (>= 2.7), libcairo2 (>= 1.2.4), libfontconfig1 (>= 2.4.0), libfreetype6 (>= 2.3.5), libglade2-0 (>= 1:2.6.1), libglib2.0-0 (>= 2.16.0), libgtk2.0-0 (>= 2.14.1), libpango1.0-0 (>= 1.21.6), libticables3 (>= 3.9.6-1), libticalcs4 (>= 4.6.1-1), libtifiles0 (>= 0.6.6-1), libxml2 (>= 2.6.27), zlib1g (>= 1:1.1.4), firefox | abrowser | www-browser [16:38] persia: I'm looking at you having said it depends iceweasel | www-browser [16:39] My sid apt-cache says it does. Am I out of date? [16:39] No. I'm saying I think that's sufficient. [16:39] I thought that one of the changes we maintained was all the s/iceweasel/firefox/ changes in package relationships. [16:39] Do we have a solution for that now? [16:39] Changing iceweasel | www-browser to firefox | abrowser | www-browser has zero practical effect on user experience [16:40] The only time that would matter is if the user had no package installed that provided www-browser and that would be hard to accomplish [16:40] Well, that removes a whole heap of diffs. [16:40] I'm just saying I think it's not worth the effort to maintain that diff. [16:40] It does. [16:41] Consider it. [16:41] I can see the argument, I just thought that there was some intent behind that. [16:41] I realize it's a change in procedure, but I think it makes sense. [16:41] If there's some other diff, go ahead and do firefox | abrowser | www-browser too, but if that's it, I'd sync it. [16:41] I don't disagree, but I also think that procedure is best changed in combination with interested parties, such as TIL or mozillateam. [16:42] fta: Do you have any opinion about tiemu and iceweasel/firefox ? [16:43] ScottK: If there wasn't some reason to fiddle the names, the simplest solution to the entire mess would be to have firefox Provide: iceweasel. [16:43] persia: I agree with that. [16:43] So, because we didn't do it the easy way, I presume there's some reason. [16:43] That would solve an even large set of problems. [16:44] In theory, changing to firefox | abrowser | www-browser is more correct, but I think it's not worth the trouble. [16:44] I don't pretend to understand that reason, but I'll go along as long as others feel it's better to do it the hard way. [16:45] I don't know that it's been seriously examined in some time. [16:45] That's why I asked TIL :) [17:08] persia: what if I can't find my package in https://merges.ubuntu.com/universe but in https://merges.ubuntu.com/t/tiemu/ [17:08] ? [17:09] benste: then that would mean the wiki page was out of date :) [17:09] lol [17:09] persia: the wiki(how to merge) or the list of merges ? [17:09] The wiki of how to merge. [17:09] so you're using a different site today ? [17:10] /topic says http://people.ubuntuwire.com/~lucas/merges.html , but this may be temporary. [17:10] We've been having a lot of discussion about merges this cycle, and I'm not sure we have a final solution yet. [17:15] Heya gang [17:16] persia: thx for this temp info found the last uploader and Scott replied to the bug too [17:19] persia: His package is in Multiverse. That's why he couldn't find it in Universe. [17:19] :-9 [17:20] Interesting. Per architecture support time: "add case were a architecture is supported (armel) but not for the full LTS time" in https://blueprints.edge.launchpad.net/ubuntu/+spec/foundations-lucid-support-timeframe-information [17:20] ScottK: Is MoM generally up again, or are we still unsure of a final solution? [17:22] ScottK: Separately, that sense of "supported" probably means "labeled as supported", rather than anything else. Same as hardy lpia in that sense. [17:23] persia: MoM is unmaintained. [17:23] Oh, I didn't realize that obtained for Hardy lpia. [17:23] That was my understanding. As a result, I believe the wiki page on merging to be out of date. [17:23] OTOH, the alternative via bzr is not really mature enough for general use. [17:24] I'm not sure it does for all flavours, but I know that because of the way some of the mobile stuff was done on lpia, lpia desktop didn't work reliably anyway because of library issues. [17:24] So when the (non-LTS) mobile flavour ended support (18 months), I *think* lpia kinda lost support. [17:25] I have no idea about lpia server. [17:25] On merging: right. I don't think we have a good replacement for the merging documentation right now. [17:27] There are some useful hint in the UDD wiki pages. [17:27] hint/hints [17:27] Yes. They haven't worked for me yet, but I'm sure that's only a matter of time. [17:28] Also in the replies to my recent thread on the topic on the UDD mail list. [17:28] It would be useful for someone to work on a page that was the union of those two resources. [19:04] lutin, about bug 510693 [19:04] Launchpad bug 510693 in ecore "Sync ecore 0.9.9.063-3 (universe) from Debian unstable (main)" [Wishlist,New] https://launchpad.net/bugs/510693 [19:05] why do you need an ack from u-u-s? IMHO you should be able to push that by yourself [19:06] (package is still in universe, it seems) [19:06] fabrice_sp: heh, thought it had been promoted to main already. [19:06] sorry [19:07] np: I'll unsubscribe u-u-s then ;-) [19:08] may I subscribe archive admin also? Or you will take care of it? [19:09] fabrice_sp: I'll take care of it, thanks [19:09] ok === gnomefreak76 is now known as gnomefreak [19:42] I have a few packaging problems I was hoping someone could help me with. I want to include a udev-file in my package (which is built with cdbs). but I can't get debhelper to add my udev rule to the package. Supposedly, specifying DEB_INSTALLUDEV_ARGS in debian/rules should take care of it. but it doesn't seen to work. it's just ignored. [19:42] persia, ScottK: wrt tiemu, feel free to do whatever you want with it, i just touched it once as part of a massive firefox meta package update [19:43] fta: More generally, do you know the reason that we fuss with changing all the package dependencies rather than just having firefox and abrowser Provide: iceweasel? [19:43] pucko-: Is this a new package, or a change to an existing package? [19:44] a new package [19:44] dh_installudev is actually run, but I can't get it to take the correct arguments. [19:45] persia, a lot of debian users are using our firefox packages (because they are fresher), as they also have iceweasel, it's best to keep things separate [19:46] fta: So it's worth changing the dependencies of all these packages every cycle? I just thought that by using Provides: we could sync more packages from Debian. [19:46] persia, but you may want to talk to asac about that, those days, i'm mostly focusing on chromium [19:46] OK. I just picked on you because you touched tiemu last :) [19:47] pucko-: You need DEB_DH_INSTALLUDEV_ARGS instead of DEB_INSTALLUDEV_ARGS, but figuring this out is frustrating and tricky. Might I recommend using /usr/share/doc/debhelper/examples/rules.tiny as the base of your debian/rules? [19:47] oh. thanks! [19:49] um.. that file is just 3 lines :-/ [19:50] Yep. Saves lots of messing about. [19:52] You can pass arguments to dh_installudev either by passing them to the dh call (but this passes the same arguments (likely usually ignored) to dh_*), or you can add an override_dh_installudev: rule, and put the complete dh_installudev call in that. [19:52] (Well, there's lots of other ways to do it too, but those might be easiest) [19:52] sounds easier to just use the env variable. now that I know it :-) [19:53] What's your environment variable setting? [19:53] I mean that DEB_DH... thing [19:53] Right. What value are you using? [19:54] oh it worked. DEB_DH_INSTALLDEV_ARGS = --priority=85 === mrpouit is now known as mr_pouit [19:55] pucko-: Here's the two different rules.tiny debian/rules files that do that: http://paste.ubuntu.com/360227/ [19:55] (and no hunting around for special secret variable names) [19:55] now I only have this postinst-has-useless-call-to-ldconfig warning. it is added by debhelper for some reason. can I tell it to ignore it? [19:55] Are you installing any shared libraries? [19:56] well. sort of. it's a smart card reader driver. one .so-file that should go into /usr/lib/pcsc/ somewhere. [19:57] so tha path shouldn't be added to ld.so.conf [19:57] the [19:57] You need to pass arguments to dh_makeshlibs to ignore that. [19:57] ok. === ajmitch_ is now known as ajmitch === arand__ is now known as arand [20:16] hi, is there packages for gegl 0.1 somewhere ? [20:18] feuloren: looks like no, Debian has an open request for it [20:35] hey Daviey, i just reviewed your django-openid-auth branch. got one minor request for you, and i'll be happy to merge it [20:50] hi people, do you know how make ubuntu able to show volume levels for all sound peripherals ( usb also) ? [21:03] fabrice_sp, just seen you've sponsored bug 464422. Thank you ! [21:03] Launchpad bug 464422 in msttcorefonts "package ttf-mscorefonts-installer 3.0 failed to install/upgrade" [Critical,Fix released] https://launchpad.net/bugs/464422 [21:04] jibel, you're welcome: you did all the job ;-) [21:09] jibel: well done, this was a bug with an enormous amount of duplicates [21:10] randomaction, thanks [21:11] two such bugs, actually :) [21:12] if anyone knows how to reach the debian maintainer please ping him. It would be great to fix it upstream too. [21:13] jibel, I saw you reported it there, but no answer :-/ [21:28] could someone review/sponsor my zope.* packages on REVU? http://revu.ubuntuwire.com/u/menesis [21:32] can anyone tell me why there is both a libusb-0.1-4 and libusb-1.0-0? very confusing.. [21:33] in ubuntu karmic [21:36] pucko-: it's commonplace to have different versions of libraries in distribution (or installed on the same machine), as different applications may require different versions [21:37] hm.. [21:40] it seems very specific. why not just libusb0 and libusb1? [21:44] because the files are libusb-0.1.so.4 and libusb-1.0.so.0 [21:45] * Quintasan goes to bed, school etc. [22:05] how long does it take to the package to show up in the ppa after upload? [22:06] jariq: ask in #launchpad. [22:06] thx [22:32] about 5 minutes, anyway [23:49] RainCT, could you ack ~ubuntu-dev-without-bugmail into ~mythbuntu? [23:50] superm1: sure [23:50] thanks [23:52] superm1: Uhm, what has changed since the original proposal when we created that team? [23:52] RainCT, it looks like there is a gateway to catch all the extraneous email now isn't there? [23:53] or was there still some other deficiency? [23:53] superm1: I'm not sure, I'm reading the old mails right now [23:56] superm1: There was no agreement that ubuntu-dev needed access to such branches: https://lists.ubuntu.com/archives/ubuntu-devel/2008-August/026207.html [23:57] superm1: Also, problems with the ~ubuntu-dev-without-bugmail approach were that e-mails other than bugmail were still send to everyone (for example, notifications of new mailing list creations) and that all developers had the Mythbuntu logo on their profile [23:57] Oh right [23:57] well then i guess not worth the troubles still [23:58] once more ~mythbuntu folks are more packaging proficient on their own should just have them apply for per package upload rights and move those branches to ~mythbuntu-dev at that time [23:58] for now it's convenient as it stands [23:58] sorry for the churn, i should have dug that mail up myself and reviewed it