[00:02] strycore, it looks like it has been fixed upstream but has not been packaged in debian yet [00:04] that is a new version upstream so the fix won't make it into karmic. Your best bet, given it is a regression with a simple fix, would be trying to request a SRU i think [00:04] https://wiki.ubuntu.com/StableReleaseUpdates [00:04] im kind of new to this though so I might be wrong [00:04] yes this bug is fixed in convertall 0.4.3, I took just one line from it. But I know that were not going to have it before Lucid so it might be nice to patch convertall 0.4.2 so it actually works [00:05] and it's the smallest patch ever ! [00:07] so then get the fix in karmic: https://wiki.ubuntu.com/StableReleaseUpdates =) [00:15] ok i've subscribed the ubuntu-sru team and nominated the fix for karmic, but I'm not a motu, there are things I can't do === asac_ is now known as asac === nxvl_ is now known as foobar23 [03:57] what is the equivalent pbuilder/pdebuild command for 'dpkg-buildpackage -sa -S' ? [04:13] pmatulis: pdebuild -S -sa # should work fine, as should pbuilder --build --debuildopts "-S -sa" foo.dsc === yofel_ is now known as yofel [11:14] havp === ripps is now known as ripps|sleep === AlanBell_ is now known as AlanBell === thekorn_ is now known as thekorn [14:06] dtchen: Can you hit me up if you come around? [14:41] what is the equivalent pbuilder/pdebuild command for 'dpkg-buildpackage -sa -S' ? [14:43] --debbuildopts -sa [14:43] pmatulis: why would you need that? for pbuilder you need to have a .dsc file already [14:44] I have to use -sa occasionally to ensure the .orig.tar.gz gets included in the .changes file [14:45] yes, but you don't use pbuilder to build your source package, do you? [14:46] geser: my goal is to have PPA's for any release i choose [14:47] geser: and pbuilder provides that in general. needless to say i'm quite new at this [14:48] pbuilder takes your source package and builds the binary debs from it [14:48] geser: No not usually, good point :) [14:49] you don't need a lucid pbuilder to build a source package for lucid (in most cases) [14:51] geser: so if my host is jaunty how do i get LP to build karmic PPA's? [14:51] you target karmic in your changelog entry [14:52] that lets LP know that it should build it for karmic [14:53] geser: alright, easier than i thought then [15:23] dtchen: OK, I just uploaded libsdl 1.2.14 to experimental if you get bored and want to test :) [15:26] heya bddebian :) [15:27] Hi sebner [16:17] fyi uploads to revu isn't working [16:18] it says "wrong password"! [16:20] revu is down [16:38] i'd like to package this: http://www.pre-emptive.net/doco/w3cvalidator-command-line-html-validator [16:38] ScottK: do you know if your changes to some r-cran-* packages are still needed? (Don't build with xvfb on Sparc to work around LP 410711) [16:38] Launchpad bug 410711 in xorg-server "Xvfb failed to start on sparc" [Undecided,Fix released] https://launchpad.net/bugs/410711 [16:38] geser: That should be fixed now. [16:38] so sync those packages again? [16:38] at work we tend to produce native packages, but I know that's frowned upon [16:38] geser: I have three R packages that should be able to sync now that the new r-base is in. [16:39] geser: Yes. If you want to file the syncs, I'm fine with that. [16:39] (just let me know) [16:39] I'd like to keep the original source code under version control [16:39] I guess the ones I looked at to get some FTBFS of the list [16:39] what's the best way to do that and comply with debian / ubuntu guidelines? [16:41] seems like keeping original source under version control might make it awkward to comply with requirement to work from "original tarball" [16:41] ScottK: I'll testbuild them and file sync request then [16:41] geser: Thanks. [16:41] jjlee: Keep it all in your VCS, but exclude the debian dir when you make the tarball for release. [16:41] are there tools to generate patches (and even generate "original tarball") from a version control repository? [16:41] Yes, there are, but it depends a lot on what you use for a VCS. [16:42] git by preference, but bzr if it integrates better with LP (which I assume it does?) [16:43] It does, but it really only matters if the project is hosted on LP. [16:43] ScottK: how do I generate "original tarball" from repo? does it have to be bit-for-bit the same as the real original tarball? Does tar include timestamps in archives? [16:43] I do want to use LP's PPAs [16:43] don't care especially about the rest of LP's features [16:44] jjlee: You want to look at pristine-tar as the tarball has to have a consistent md5sum. [16:44] though I guess it's nice for issue tracking to be in the official place [16:45] ScottK: thanks -- this binary delta that pristine-tar talks about contains the timestamp (&c.?) stuff extracted from original tarball, presumably? [16:46] Something like that. I don't know the details. [16:46] ScottK: thanks anyway, looks promising! [16:50] is there other LP stuff I should do other than just uploading my source package to PPA (other than getting it into {multi,uni}verse)? [16:50] stuff that makes other maintainers / regular users lives easier? [16:50] like using standard issue tracker? [17:26] ScottK: sync requested for r-cran-time{date,series}: bug #489632 and bug #489636 [17:26] Launchpad bug 489632 in r-cran-timedate "Sync r-cran-timedate 2100.86-2 (universe) from Debian unstable (main)" [Wishlist,Confirmed] https://launchpad.net/bugs/489632 [17:26] Launchpad bug 489636 in r-cran-timeseries "Sync r-cran-timeseries 2100.84-2 (universe) from Debian unstable (main)" [Wishlist,Confirmed] https://launchpad.net/bugs/489636 [17:27] Good. Thanks. === azeem_ is now known as azeem [17:35] i have been working my way down the universe list on M-o-M looking for and filing sync requests. I have not been asking or notifying the previous uploader as I just noticed geser doing. Should I not just go down the list filing syncs (when appropriate)? [17:39] mannyv: it's good to at least let the previous uploader know it to avoid double work [17:42] geser, ok will do. That'll be a chore, iv'e filed about 20 sync requests [17:43] mannyv: If you look at the top of that page, it's the first thing that's suggested. [17:47] ScottK, technically it says it before doing a merge. I am not trying to be pedantic but syncs (at least the ones i have been touching) are not much work and so not likely to be sitting in someone work log half done. But I take your point and will ask first [17:48] mannyv: The other thing is that it isn't rare for the previous uploader to know something relevant about the package and if it can be synced, for example, I knew it was now OK to sync the packages that geser asked about due to following the bug I uploaded a workaround for and knowing that r-base had recently been synced. It'd have been a lot tougher for someone who hadn't been following the packages to know. [17:52] ScottK, ive been skipping over the packages with more complicated. I have focused on obvious easy changes like changing a Depends or renaming getline to get_line [17:52] That's good, but still please ask. [17:54] i will [18:27] anybody know how to explicitly specify gpg key to use when signing a file? [18:27] jjlee: -k [18:28] hyperair: man gpg says that's the same as --list-public-keys [18:29] jjlee: hyperair is correct. [18:29] jjlee: sorry, i was talking about debsign. were you talking about something else? [18:29] * ScottK was too [18:30] hyperair: sorry, I was referring to gpg itself [18:31] jjlee: look for default-key in the manpage [18:31] jjlee: see --sign in the manpage [18:31] hyperair: thanks, that led me to --local-user [18:32] ah === siretart changed the topic of #ubuntu-motu to: Ubuntu 9.10 released! | Want to get involved with the MOTU? https://wiki.ubuntu.com/MOTU/Contributing | Sponsor queue: http://is.gd/2y76G | http://qa.ubuntuwire.com/ftbfs | http://people.canonical.com/~ubuntu-archive/NBS/ | http://qa.ubuntuwire.com/debcheck | latest rebuild failures: http://people.ubuntuwire.org/~wgrant/rebuild-ftbfs-test/test-rebuild-20090909-karmic.html | REVU back [19:10] yay [19:48] is this a suitable name & version for my ppa package (from release 1.0.1 of http://www.pre-emptive.net/doco/w3cvalidator-command-line-html-validator)? w3c-markup-validator-commandline_1.0.1ubuntu1~ppa1 [19:48] jjlee, that's still a native package version [19:50] ah, thanks. So 1.0.1-1ubuntu1~ppa1? Do I need the ubuntu1? [19:51] 1.0.1-1ubuntu1~ppa1 means: [19:51] upstream version 1.0.1 [19:51] debian version 1 [19:51] ubuntu modification 1 of debian version [19:51] So 1.0.1-1~ppa1 I guess [19:52] ppa package 1 based upon the above, but versioned to be "lower" than the non-ppa version [19:52] (should one exist) [19:52] right, I want it to be lower [19:52] (though none exists yet) [19:52] 1.0.1-0ubuntu1~ppa1 [19:52] it's not a modification of a big-D Debian package [19:52] unless you plan on targeting debian rather than ubuntu with a "real" package [19:53] use 0ubuntu1 for packages not based on debian versions [19:53] this is a new package [19:53] So 1.0.1-0ubuntu1~ppa1 ? [19:53] jjlee: you're still using 0ubuntu1 if you're going to upload to Ubuntu not Debian. [19:54] I'm going to upload to ubuntu, not Debian, yes [19:54] thanks [20:19] This package has a perl module in it. Not sure I want to take the obvious debian package name for that w3c-markup-validator-perl. [20:19] Seems reasonable just to keep it all in a single "binary" package? [20:19] it's not in CPAN [20:24] Can anybody suggest an example perl debian package that's midway in the level of debian/rules auto-magic-ness between libwww-perl (only a couple of lines in debian/rules) and libuuid-perl (big ugly makefile)? [20:30] DktrKranz: thanks for getting ubuntu-dev-tools into Debian :) [20:31] DktrKranz: So are we going to continue to maintain u-d-t the same way and you will just "sync" it in Debian ? [20:35] RainCT: hehe, np :) [20:36] nhandler: I think it's safe, I only diverge in a dep (which it will be applied in Lucid too, as soon as dependency stack will be synced), and Maintainer (to avoid ML bouncing) [20:37] now that we sync from testing, it's not worthy to wait 10 days [20:38] the only issue I've encountered is some unicode issues in buildd script [20:39] DktrKranz: btw, have you seen http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=543315? [20:39] Debian bug 543315 in sbuild "mk-sbuild-lv does not export http_proxy to schroot environment" [Normal,Open] [20:40] looks like they want mk-sbuild-lv in the sbuild package [20:40] kees: ^ [20:41] Even better. [20:41] The less we have to maintain, it's good. [20:43] RainCT: I agree some tools should be included in other packages (i.e. devscripts), and leave here only Ubuntu-specific ones === ripps|sleep is now known as ripps [21:56] DktrKranz: what unicode issues exist in the buildd script? [22:13] hi [22:13] im just strart [22:14] and i like so much that version of ubuntu [22:14] can some one hellp me [22:15] i want install ubuntu on my usb [22:15] is that keep me eny change i have made on setings [22:15] mihailikos: #ubuntu is the place for support [22:15] mihailikos: This is a development channel; your question would be better asked in #ubuntu. [22:15] aha [22:15] ok [22:15] sorry [22:49] should the .changes file I upload to launchpad end in _source.changes, as this page suggests? https://help.launchpad.net/Packaging/PPA/Uploading [22:49] mine doesn't [22:50] though the .changes file does point to an .orig.tar.gz [22:50] (it also points to a .dsc, and a .diff.gz, and a .deb) [22:58] jjlee: Launchpad only accepts source uploads. You need to pass '-S' to debuild or dpkg-buildpackage. [23:00] I'm using git-buildpackage, didn't see any similar-looking options [23:01] oh, it passes options along [23:01] thanks