[01:23] sistpoty: I discussed Eric with lex79 on #kubuntu-devel. He's going to look into 4.4.2 from Unstable. [01:23] ScottK: excellent, thanks! === mrcurrington is now known as mrc === mrc is now known as mcurrington [08:23] bug #558946 [08:23] Launchpad bug 558946 in ubuntu "FFe: Sync ipheth 1.0-1 (universe) from Debian sid (main)" [Undecided,Incomplete] https://launchpad.net/bugs/558946 [08:23] How do I create an install log? piuparts? [08:24] I have a sbuild install log which I can attach, but I wasnt sure of the best way to create an install log. Any pointers would be much appreciated... [08:26] pmcenery: dpkg --install the package and attach the log to the bug report. [08:29] iulian: thanks. Is that log just the output to the terminal, or was there some /var/log file you refer to? [08:30] pmcenery: The former. [08:30] iulian: ok. thanks. will do that [08:30] You're welcome. === shadeslayer is now known as evilshadeslayer [13:39] Hey guys, is the package wallpaper-tray going to be in Lucid, there is no publishing data for it. [13:55] no, it got removed from lucid: (From Debian) ROM; reimplemented upstream; Debian bug #553167 [13:55] Debian bug 553167 in ftp.debian.org "RM: wallpaper-tray -- ROM; reimplemented upstream" [Normal,Open] http://bugs.debian.org/553167 [13:57] geser: The website for the project is just a placeholder, doesn't look like it's developed anymore [14:01] probably [14:01] Better off setting a bug in Jaunty as Won't Fix then I guess [14:02] which one? [14:03] bug 237738 [14:04] Launchpad bug 237738 in wallpaper-tray "wallpaper-tray crashes (often) when adding new (large) folders" [Undecided,Incomplete] https://launchpad.net/bugs/237738 === skipper_ is now known as bdrung [15:40] ScottK: Good catch on that empty binary package, fixing now [15:40] (and I had used -v correctly) [15:41] OK [15:41] that uncovered a bug in haskell-devscripts, which is a win [15:42] Nice. [15:55] when cherrry picking upstream VCS commits, is it better to do it as 1-1 patch or 1 patch for everything upstream? [15:55] 1-1 patch. [15:56] use quilt [15:57] hyperair: package already uses quilt :) [15:58] hyperair: for patch naming can I do XXX-reason-I'm-patching-1.patch XXX-reason-I'm-patching-2.patch ? === hannesw__ is now known as hannesw [16:01] micahg: right. [16:01] micahg: wait a sec. [16:02] micahg: this is a series of commits fixing one issue? [16:02] hyperair: well, I'm cherry picking the xulrunner 192 port fixes from upstream vlc [16:02] =\ [16:02] 1-1 would be good then, i think [16:03] especially in the case of git format-patch kind of patches [16:03] hyperair: with the naming convention I proposed or should I do XX1-reason.patch XX2-reason.patch? [16:04] i think i'm not seeing the whole picture here. [16:04] exactly what patches are these? [16:04] like can i know their original names, for example? [16:05] is upstream using git? [16:05] hyperair: well, there are a series of commits that fix different issues with porting to xul192 [16:05] hyperair: yes [16:05] if the patches are git-formatted i'd just use theo riginal names. [16:05] just quilt import them all in [16:05] * micahg checks the original name [16:06] hyperair: they don't have names [16:06] hyperair: IIRC, upstream mozilla uses Hg. [16:06] ... [16:07] okay, i give up [16:07] this is turning my brain into spaghetti [16:07] furst one says git, then another says hg [16:07] hyperair: one example: http://git.videolan.org/?p=vlc.git;a=commitdiff_plain;h=6dbe4986f7c11370c2bc275491d4502f5f4c3c60 [16:07] then they don't have names [16:07] ScottK: this is for vlc :) [16:07] but format-patch names all the patches nicely [16:07] Oh. [16:07] so what. the. **** is going on?! [16:07] Nevermind then. [16:07] * micahg never used format-patch before [16:07] ... [16:08] okay, that explains it [16:08] go use format-patch and take those patches to shove straight into quilt [16:11] hyperair: silly question...where do I get/use format-patch? [16:12] you install git-core, and you run "git format-patch blah blah" [16:12] hyperair: ah, it's a git thing :) [16:12] ScottK: do you know who's spearheading the get Sugar into Lucid effort? [16:13] micahg: IIRC it was lfaraone and highvoltage. [16:13] ScottK: thanks [16:15] * lfaraone waves. [17:31] * abogani waves [17:31] I'm looking for sponsors: https://wiki.ubuntu.com/AlessioIgorBogani/linux-rtPPUApplication [17:31] Thanks! === yofel_ is now known as yofel [17:50] abogani: who sponsored your linux-rt work till now? those person are the best to judge on your application [17:53] geser: Since Gusty until Hardy(included) Ubuntu Kernel Team since Intrepid until Lucid (included) Luke Yelavich. I think that Luke don't sponsor me. [17:55] abogani: have you asked him? [17:56] geser: Yes he said me that he don't trust on my packaging knowledge. [17:56] without any sponsored contributions/uploads it will be hard to get PPU rights for a package and your current sponsor(s) are in the best situation to judge if you are ready or not [18:01] abogani: and you don't trust his assessment of your packaging skills and try nonetheless to apply? I don't want to anticipate the results of your application but if your sponsor doesn't trust you why should the DMB grant your PPU rights? [18:04] geser: No one have ever worked on this package. Neither Ubuntu Kernel Team nor Luke have worked on/changed my (upload) works. So linux-rt package is result of *only* my work. If you think that I don't deserve PPU rights you should know that 1) I push realtime kernel in Ubuntu 2) unilt Feisty I triage bugs I backported upstream fixed 3) I working on packaging when sine Intrepid linux-rt became a standalone universe package. [18:05] geser: In any case I'll accept your choice. === lukjad007 is now known as lukjad86 [19:00] is there any person works or got expirence in backports? [19:02] Several folks: ask a question, and someone will answer :) [19:10] I want to backport trac package to karmic, jaunty, intrepid :) [19:10] for security fix reason [19:10] backports aren't for bugfixes. [19:11] If there are security issues, you might want to drop by #ubuntu-hardened and talk about the best way to get them resolved. [19:13] DktrKranz: I think I recall you doing an empty binary scan over the Debian archive recently. Do you still have the scripts you used to do it? [19:13] (noticed a bug in hlibrary.mk which causes empty binaries to be produced) [19:19] Laney: there's a lintian check for that already working === lukjad007 is now known as lukjad86 [19:19] DktrKranz: Oh, that's cool. I'll do that [19:20] Laney: http://lintian.debian.org/tags/empty-binary-package.html [19:20] but lintian.d.o doesn't pick up the case I'm looking at [19:20] interesting [19:20] some times ago, we ran lintian checks on Ubuntu packages as well, I don't know if it's still running [19:20] It's not. [19:20] I want to fix this in Debian anyway [19:21] Laney: which one? [19:21] I've been meaning to get it running again, but haven't yet found the setting-up-whole-archive-lintian-runs-for-the-unfamiliar guide yet :) [19:21] DktrKranz: e.g. libghc6-hsql-mysql-dev [19:21] when I build it lintian picks it up [19:21] but doesn't appear on that interface [19:21] (maybe it's only run on the souce packages) [19:21] no, can't be [19:22] definitely not [19:22] it searches the binaries [19:23] Laney: which lintian version do you use, 2.3.4? [19:23] I'm running it in a sid chroot, so 2.3.4 indeed [19:23] gah [19:23] let me emphasis, I *do* see the warning locally [19:23] but the website doesn't [19:24] ah [19:24] * DktrKranz logs in to lintian.d.o to see which version it runs [19:26] haskell-hsql-mysql (1.7.1-4) [19:26] that's the same [19:29] Laney: btw, I processed several haskell NEW uploads, and luckily they were OK. [19:31] persia: if you need some config, I can try to see if setup for lintian.d.o is available, and eventually pass it [19:32] DktrKranz: I believe it to be on alioth somewhere, although I haven't found it. If you do find it, it will speed me running lintian over Ubuntu :) [19:34] DktrKranz: "the same"? [19:34] as in empty? [19:35] Laney: I checked if the empty packages in the archive is the same lintian laboratory is know of, and version matches [19:36] the Lintian version is too old then? [19:36] you should also spank whichever ftp team member NEWed that package ;) [19:36] * DktrKranz checks if it was him... :) [19:37] N: Processing source package haskell-hsql-mysql (version 1.7.1-4) ... [19:37] N: Removing haskell-hsql-mysql [19:37] W: haskell-hsql-mysql source: changelog-should-mention-nmu [19:37] W: haskell-hsql-mysql source: source-nmu-has-incorrect-version-number 1.7.1-4 [19:37] mmmh, lintian has the correct version, so why check is failing? [19:38] root@chicken:/# lintian libghc6-hsql-mysql-doc_1.7.1-4_all.deb [19:38] warning: lintian's authors do not recommend running it with root privileges! [19:38] W: libghc6-hsql-mysql-doc: empty-binary-package [19:38] works here [19:38] what changesfile is it running over? [19:40] it shouldn't process changesfile directly, but packages unpacked [19:40] (at least I see that on lintian.d.o) [19:41] a manual run on that package reveal W: libghc6-hsql-mysql-dev: empty-binary-package for me too [19:46] xkill [19:46] In debian/install, "src/* usr/share/my-package/" was recursive. How can I do the same in setup.py? [19:50] Should I be using MANIFEST.in for that? [19:53] If nobody gives you a good answer after a while, try #ubuntu-app-devel which tends to focus more on upstream best-practices, etc. [19:53] Thank you, will do [20:37] 17:12 < micahg> ScottK: do you know who's spearheading the get Sugar into Lucid effort? [20:38] 17:13 < ScottK> micahg: IIRC it was lfaraone and highvoltage. [20:38] michas: ^^^ that would be maverick, not lucid :) [20:38] highvoltage: is that final then for Lucid, no sugar? [20:41] micahg: lucid is in final freeze real soon, I think pretty much everyone (including me) is doing some scrambling atm to get their things working as good as possible in lucid [20:41] wasn't the plan to have sugar 0.88 in lucid? [20:41] micahg: so I guess it's pretty much final [20:42] lfaraone: ^^^ [20:42] bdrung: parts of sugar works with older and newer versions of it, parts of it is 0.88 and parts of it is 0.86 [20:43] there is still enough time to push the missing bits into universe [20:44] bdrung: that is, if someone picks it up and runs with it :) [21:05] micahg, highvoltage, I really am rather unfamiliar with pyxpcom, and Sugar's useless without it unless I decide to package a bunch of new packages. [21:10] lfaraone: well, my question was is sugar going into lucid...it seems like we're in a circular logic loop [21:11] If it won't be on an ISO we have a couple of weeks to get it sorted. [21:14] micahg: I have exams this week, but feel free to poke me after that if you need any help with Sugar related stuff. [21:15] RainCT: thanks, we're trying to figure out about getting it into lucid and what the requirements are for xpcom integration [21:19] micahg: By the way, have you talked to David Farning? He was working on a Ubuntu Sugar Remix, but I haven't heard of him for weeks. [21:21] RainCT: no, not yet [22:30] how do you have a blank line in the package description in debian/control? debuild keeps complaining about This is probably a duplicate of bug #534743. Can you try booting with the nosplash break nodmraid options, then when you hit the busybox prompt, run dmraid -ay then exit. If the system boots up normally at that point then it's that bug and I will mark this as a duplicate. [22:30] Launchpad bug 534743 in dmraid "dmraid causes udev event feedback loop in Lucid" [High,Triaged] https://launchpad.net/bugs/534743 [22:31] * psusi kicks terminal for not copying that automatically like xterm [22:31] make that complaining about: dpkg-source: error: syntax error in e2defrag/debian/control at line 15: continued value line not in field [22:37] lol [22:38] hey jdong... I'm trying to get https://launchpad.net/e2defrag/ into shape ;) [22:38] was gonna ask about that yesterday :) [22:38] but then realized how much actual work (tm) I had :-/ [22:38] trying to figure out if I can get a build under the project instead of my ppa [22:39] hehe [22:39] well, it seems to work well for me now without uninit_bg and extents [22:39] tackling extents is going to be a challenge [22:39] uninit_bg should be easier [22:39] extents being pretty important [22:39] then after that I have to try and get it to relocate inodes [22:39] yea [22:40] uninit_bg seems like it is going to be pretty important too, especially with lazy_itable_init, which is currently defaulting to off... when I turned that on installing to a 1.5 tb drive went from taking like 15 minutes sitting at 5% complete formatting to like 15 seconds [22:44] * psusi wonders where the hell that dput went [22:44] generally its a good idea to know where you're sending them... [22:46] lp seems to have a way to link to an ubuntu package, but I can't figure out how to upload a new package from the current bzr branch [22:46] in the normal way [22:47] bzr bd -S ... dput [22:47] maco: I sent it to ppa:e2defrag/0.75, but I don't know where it WENT ;) [22:47] bzr bd unknown command === skipper_ is now known as bdrung [22:48] you need to install bzr-builddeb [22:48] ohh, I just ran debuild -S [23:09] That ought work just fine as well. [23:12] psusi: Um, ppa:e2defrag/0.75 isn't a PPA. [23:12] The '0.75' PPA for ~e2defrag does not exist. [23:15] psusi: Also, I'm confused. Has e2defrag moved to Launchpad for development? Where was it before? [23:15] Are you the upstream maintainer? [23:26] wgrant, I am now, I took it over and rescued it after it was removed from the archive 2 years ago for not having an upstream [23:26] Ah. [23:26] So, projects don't have PPAs -- people and teams do. [23:26] that's what I thought [23:26] is there a project archive instead of a personal one? [23:27] No. [23:27] For reasons that I cannot fathom, there is not even a way to link a project to PPAs. [23:27] I was wondering about that [23:27] I mean the project links to the branches, why not archive? [23:28] there's a button to link it to a package in the ubuntu archive, but it isn't there [23:28] Well, an archive can contain multiple projects. [23:28] Right. [23:28] But that is stupid and doesn't handle PPAs. [23:28] Even though it was redesigned a couple of months ago. [23:29] so why can you link it to the ubuntu project's archive, but no other project's archive? I thought lp was supposed to be more general than ubuntu? [23:29] so as a separate project, you would think it would have its own archive [23:30] Ubuntu is a distribution, not a PPA. [23:30] ubuntu is setup as a distro not a plain old project in lp [23:30] yes, it has a distinction [23:30] ahh [23:31] It is more general, but a distribution is a special class of object. [23:31] maco: well, we *could* link a project to other distributions which *are* hosted in LP [23:31] IHasDistribution. [23:31] lfaraone: huh what what [23:31] did i say you couldnt? [23:31] maco: you can't, currently. [23:31] i just said that ubuntu gets build-capabilities because its a distro [23:31] (afaict) [23:31] well, I guess for now then I'll have to use ~e2defrag-maintainters ppa and eventually get it uploaded to universe [23:32] The non-Ubuntu linking functionality was somewhat unlinked a couple of weeks ago. [23:32] It was not being used. [23:37] oh by the way, how are you supposed to have blank lines in the description in debian/control? dpkg-source errors thinking the text following is a new field [23:38] There is a special syntax to have blank lines: http://www.debian.org/doc/debian-policy/ch-controlfields.html#s-f-Description [23:38] by using a . [23:39] a single space followed by a single full stop character, is what policy says is the only way [23:39] but persia is quicker than me at getting the link :) [23:39] wow, that's what my guess was ;) [23:41] hrm... bzr bd -S fails since it seems to blindly look for an orig tarball when it shouldn't... seems like a bug [23:43] I suspect there's special syntax you need to get bzr-builddeb to know it's native. [23:43] Why is there no orig tarball? [23:43] because it's a native package [23:44] It doesn't seem like it should be. [23:44] why not? I AM upstream [23:44] psusi: Because you're a well-meaning upstream who wants to support every potential distribution that will ever be invented :) [23:44] What if we need to patch it? [23:44] persia, I am? :) [23:45] wgrant, then we patch it [23:45] psusi: Of course you are :) [23:45] psusi: We can't, sanely, because it's native and we don't have commit access upstream. [23:45] wgrant, but there is no upstream [23:46] There will be if the project gets anywhere. [23:46] psusi: But there could be. Project hosting sites abound. [23:47] I mean I guess you could create a -ubuntu1 package for changes specific to ubuntu, but right now there are none... [23:47] persia, I'm using lp for hosting [23:47] Then there is an upstream. The primary upstream contributor happens to be a distro-maintainer, but that's a coincidence. [23:47] actually I"m not [23:49] I suppose that it could be some day that ubuntu specific changes will be made, but right now it's an initial upstream release, packaged with zero modification [23:58] unless it's something made only for a specific distro (e.g. ubuntu-dev-tools), then it's best to not have it as a native package [23:59] And truth be told, most of the stuff in ubuntu-dev-tools really ought be pushed into other places [23:59] That is true. I would have done so, were it alright to have python scripts in devscripts.