[00:02] <micahg> gilir: xubuntu will be uploading an indicator gtk2 stack if lubuntu is interested in it
[00:03] <gilir> micahg, that would be great :-)
[00:03] <micahg> gilir: so, there should be no reason to unseed it actually as the binaries should still be in the archive
[00:04] <gilir> micahg, yes, but the ISO build is failing for now because of this
[00:04] <micahg> orly?  that's not good
[00:05] <micahg> which part is failing
[00:05] <gilir> micahg, it's a depend on our seed, but it's normal :-)
[00:05] <gilir> I'll add it back when the binaries will be available
[00:06] <gilir> micahg, any help needed for the packaging of the gtk2 stack ?
[00:07] <micahg> gilir: not sure, mr_pouit is doing it, I'm guessing he could probably use some help (should just be reuploading the old version as a new source with the gtk3 binaries removed or something like that)
[00:08] <gilir> micahg, ok, I'll ping him tomorrow to see if he needs any help
[00:11] <micahg> gilir: thanks!
[03:14] <TheDrums> micahg: For what reasons would backportpackage fail?
[03:19] <TheDrums> http://paste.openstack.org/show/20454  Granted, I have no idea what I'm doing.
[03:21] <micahg> TheDrums: you don't have a key available to sign the upload, you might need -k and your keyid
[03:22] <TheDrums> Figured it was that, had hoped it wasn't.
[03:22] <micahg> TheDrums: I'm happy to stage uploads in my backports PPA if you'll test, just tell me which releases (or file the bugs)
[03:22] <micahg> *through requestbackport (12.04+) of course :)
[03:23] <TheDrums> Well, I figured the one in debian stable or testing was as good as any.
[03:23] <TheDrums> (And had the needed feature that I think is neededa)
[03:23] <micahg> it's in quantal :)
[03:24] <TheDrums> That'd work fine too.
[03:25] <TheDrums> I'd be good with testing, would you just pull it over from Q?
[03:26] <micahg> TheDrums: yeah, and build it on the release in question
[03:34] <TheDrums> SSO is killing Firefox, funtimes.
[03:37] <TheDrums> micahg: Bus Error (core dumped), looks like I won't be doing this right now either. :/  I take it you do need that request?
[03:38] <micahg> TheDrums: tell me which releases and I'll file the request and subscribe you
[03:38] <micahg> TheDrums: you still have to do the testing though:)
[03:38] <TheDrums> No problem!  I was going to try filing another day if so, but precise.
[03:38] <TheDrums> Thank you very much!
[03:39] <micahg> you're welcome
[03:39] <micahg> TheDrums: please fill in the reason on the bug after I subscribe you
[03:40] <micahg> TheDrums: bug 1041010
[03:40] <TheDrums> micahg: "Added feature to pass commands to ssh such as keyfile and port"  Seem understandable?  If so, I'll fill.
[03:41] <micahg> TheDrums: uploaded, there's a slight PPA backlog, should be good by Sunday hopefully
[03:50] <TheDrums> Finally was able to edit it, confirm button was missing.
[04:58] <shadeslayer> so, I'm looking at bug 1040828, just to be clear, is it just a matter of dropping gambas2-gb-db-sqlite2 and making gambas2-gb-db-sqlite3 provide gambas2-gb-db-sqlite2 ?
[04:59] <shadeslayer> I guess I could also drop libsqlite0-dev
[04:59] <shadeslayer> I guess it'll need a transitional package
[07:10] <dholbach> good morning
[11:25] <Laney> yay, we might have some buildd time for haskell this weekend
[11:30] <geser> how far are we with the rebuilds?
[11:33] <Laney> maybe 2/3
[12:58] <gch-scilifelab> hi! A question: I've packaged a software that has a Makefile. In this makefile of corse there is the install rule
[12:58] <gch-scilifelab> the thing is that the package is built without any problems, but... when installing the resulting .deb nothing is installed
[12:58] <gch-scilifelab> the files that should go to /usr/local/bin as the Makefile says are missing
[12:59] <gch-scilifelab> any help please?
[12:59] <Zhenech> does the makefile honor DESTDIR?
[12:59] <gch-scilifelab> yes it does
[12:59] <gch-scilifelab> wait I'll paste it to you
[12:59] <Zhenech> :)
[13:00] <directhe`> gch-scilifelab, so in your dirty build tree, you have a debian/mypackagename/usr/local/lib/stuff folder?
[13:00] <directhe`> i.e. after you've run dpkg-buildpackage - this is NOT something you should have created yourself
[13:01] <gch-scilifelab> http://pastebin.com/hYX6fzNL
[13:01] <Zhenech> gch-scilifelab, first line should be
[13:01] <Zhenech> DESTDIR?=/usr/local/bin
[13:01] <Zhenech> (do not overwrite the destdir if you pass it on the commandline)
[13:01] <gch-scilifelab> outch
[13:03] <gch-scilifelab> Zhenech, ok, but I think this is working (I think), because before I had the absolute paths (without the DESTDIR env var) and the package was not building
[13:03] <Zhenech> post logs :)
[13:03] <gch-scilifelab> directhe, I dind't create any folder by hand, if this is you're asking
[13:03] <gch-scilifelab> ok
[13:05] <gch-scilifelab> the last log http://pastebin.com/ujesty7y
[13:07] <Zhenech> gch-scilifelab, as the error says: you have files in usr/local/, which is not permited
[13:10] <gch-scilifelab> but.. I don't understand the error message sorry, It says "dh_usrlocal: debian/freebayes/usr/local/bin/freebayes is not a directory", and of course is not a directory, I doesn't even exist
[13:10] <gch-scilifelab> I'm a bit lost :S
[13:10] <directhe`> wait, that's not how destdir works
[13:12] <directhe`> destdir should be the root of a tree - e.g. to install files into /usr/bin/ then you use DESTDIR=/usr and install into ${DESTDIR}/bin
[13:12] <Zhenech> oh, right
[13:13] <directhe`> or, in this case, DESTDIR=debian/freebayes/usr/
[13:13] <gch-scilifelab> ahh
[13:14] <gch-scilifelab> so I should modify my Makefile with DESTDIR=/usr and then in the install rules install -m 0755 bin/freebayes $(DESTDIR)/local/bin??
[13:14] <Zhenech> DESTDIR?=/usr
[13:14] <gch-scilifelab> ops yes
[13:15] <directhe`> ( http://www.gnu.org/prep/standards/html_node/DESTDIR.html )
[13:16] <directhe`> gch-scilifelab, your make file should use DESTDIR?=/usr/local and install to ${DESTDIR}/bin
[13:16] <directhe`> and debian's build will override DESTDIR
[13:16] <directhe`> it is correct for upstream software to assume /usr/local
[13:16] <directhe`> the point of /usr/local is it's for user-installed bullshit which didn't deserve to come from a package. packaged things go into /usr
[13:17] <directhe`> so upstream, assume /usr/local
[13:17] <directhe`> downstream, you build assuming the "install prefix" is /usr - but the "make install" rule needs to respect DESTDIR because during the package build, the files are make installed into a temporary folder
[13:21] <gch-scilifelab> I see, thank you directhe
[13:21] <gch-scilifelab> so in the link you sent me it says: You should not set the value of DESTDIR in your ‘Makefile’ at all
[13:21] <gch-scilifelab> if I don't set it but I prepend the DESTDIR var in the install rules
[13:21] <gch-scilifelab> shold it work?
[13:56] <gch-scilifelab> directhe, doing what you say seems not to work at all. I mean, the package is built, but when I try to do a dpkg --install, this error is rised: dpkg: error processing ../../freebayes_0.9.6-0ubuntu5_all.deb (--install):
[13:56] <gch-scilifelab>  trying to overwrite directory '/bin' in package iputils-ping 3:20101006-1ubuntu1 with nondirectory
[13:56] <gch-scilifelab> is if like it was ignoring the DESTDIR when installing
[14:11] <gch-scilifelab> any idea?
[14:26] <gch-scilifelab> well I've to go, thank you very much for your help:-)
[15:26] <dholbach> I just put up some very crude, but working code at lp:~dholbach/+junk/lintian-todo which might help us put up working lists for https://wiki.ubuntu.com/UbuntuDevelopment/BugFixingInitiative (search for a lintian tag or package, etc. and it will show you which packages are in Debian and which are in Ubuntu and what lintian has to complain about)
[15:26] <dholbach> just in case anyone is interested
[16:10] <micahg> dholbach: you do realize that we have a lintian lab now, right?
[16:11] <dholbach> I'm not I realise
[16:11] <dholbach> I'm not sure I realise
[16:11] <micahg> dholbach: http://lintian.ubuntuwire.org/quantal/
[16:11] <Laney> lintian.ubuntuwire.com
[16:11] <dholbach> yes - it's used as a basis for it
[16:11] <micahg> ok
[16:11] <dholbach> but I didn't find it easy to figure out which of the packages listed were in Debian and which in Ubuntu
[16:13] <dholbach> have a great weekend everyone
[16:14] <micahg> dholbach: we could probably update the lab to show which packages are in Debian
[16:14] <dholbach> if that's possible, that'd be great
[16:14] <dholbach> alright, got to go - see you :)
[16:20] <xnox> micahg: ideally the lab should run twice, once for debian & once for ubuntu and then show the diff or tags introduced in/by ubuntu changes
[16:21] <tumbleweed> xnox: broder is the person you need to talk to
[16:22] <broder> xnox: "patches welcome". alternatively, you can get this information from UDD - it imports both debian and ubuntu lintian tags, so you can do something like create table ubuntu_only_lintian as select * from ubuntu_lintian except select * from lintian;
[16:23] <xnox> broder: wow, I didn't know udd had that =)
[16:23] <xnox> awesome
[16:23] <broder> yeah, it's great as a crutch to demotivate you from actually writing better features :)
[16:25] <tumbleweed> broder: btw, \o/ on another fun CTF (and nicely done, wgrant)
[16:25] <broder> haha, thanks :)
[16:26] <broder> i actually had very little to do with this one. mostly a little bit of playtesting and some sysop work when things started
[16:26] <broder> but i will pass that along
[16:26] <broder> (andy, greg, and sidd did most of the work)
[16:26] <tumbleweed> the python code wasn't the prettiest, but at least it wasn't all ruby :P
[16:33] <micahg> hyperair: did you forget to subscribe ubuntu-release to your banshee FFe or is it not ready yet?
[16:34] <micahg> xnox: go for it :)
[16:36] <micahg> xnox: but, the Ubuntu only stuff should be separate, it's nice to see all tags for Ubuntu as a whole as well in case we want to drive a fixing of a certain tag in Debian as well, like cjwatson did with the dpkg-maintscript-helper stuff
[16:36] <xnox> micahg: true, true.
[16:37] <Laney> the UDD table doesn't have historical data
[16:37] <Laney> so you can't guarantee to be able to find the base version
[16:57] <xnox> Laney: hmm...
[16:57] <xnox> didn't think of that
[16:58] <micahg> base version shouldn't matter much, if the tag is in Ubuntu and not in Debian and the package is, just flag that with the versions in Debian that it's not it
[16:59] <micahg> *not in
[16:59] <micahg> there shouldn't be too many packages where we're totally diverged from Debian (~200-300)
[17:00] <micahg> maybe less
[17:07] <ogra_> micahg, are you still the spare-time-chromium maintainer ?
[17:07] <ogra_> it segfaults on arm currtently (havent collected data yet but will do so the naxt days)
[17:11] <micahg> ogra_: spare time is the keyword, been sorely lacking, kenvandine has been doing uploads as of late, I hope to get it going again before quantal release, if you have a patch, go ahead and submit a merge proposal or bug/debdiff
[17:14] <ogra_> heh no, i only have a bug atm :)
[17:14] <ogra_> no patches
[17:14] <ogra_> but i switched ac100 to lubuntu ... so chromium gets some arnm testing now
[17:14] <ogra_> *arm
[18:49] <jtaylor> did we ever transition libnetcdf?
[18:49] <jtaylor> I don't see a tracker
[18:50] <jtaylor> nevermind libnetcdfc != libnetcdf
[20:07] <xnox> is there a Kubuntu dev channel?
[20:08] <TheLordOfTime> isnt it #kubuntu-devel?
[20:08] <TheLordOfTime> i think it is :P
[20:35] <commandoline> Hello, I'm trying to improve my packaging skills a bit by packaging a python 3 (and only 3) application. Does anyone know an example application which is also packaged for python 3 only so I can look how it's done there? Most documentation is for porting python 2 apps to python 3 and I can't manage to succesfully build a deb based on that.
[20:39] <tumbleweed> commandoline: python3-chardet jumps to mind: http://anonscm.debian.org/viewvc/python-modules/packages/python3-chardet/trunk/debian/rules?revision=15258&view=markup
[20:40] <commandoline> tumbleweed: thanks, seems just what I needed :)
[20:59] <commandoline> tumbleweed: That package gave away some of the clues I was missing, it works now. Thanks again.
[21:20] <TheLordOfTime> is there a method to request removal of a package, and have that package instead point to a newer package name?
[21:20] <TheLordOfTime> (curious)
[21:21] <tumbleweed> http://wiki.debian.org/Renaming_a_Package
[21:22] <TheLordOfTime> tumbleweed:  specifically in Ubuntu, though.,
[21:23] <TheLordOfTime> or are the guidelines the same?
[21:23] <tumbleweed> the same
[22:42] <wgrant> tumbleweed: Thanks
[23:44] <xnox> can anybody give me a crash course in haskel transitions?
[23:44] <xnox> http://people.canonical.com/~ubuntu-archive/transitions/ghc.html
[23:44] <xnox> rebuild - if it builds upload?
[23:44] <xnox> or
[23:44] <xnox> rebuild - if it builds & becomes installable upload?
[23:46] <micahg> xnox: things have to be built in order to be installable, so, you just make sure if you're uploading something, it doesn't depend on something else uninstallable
[23:46] <xnox> ok
[23:50] <xnox> micahg: and the fact that we have dependency levels 1 & 14-51 means that we started a new cycle without fully finishing a previous one?
[23:50] <xnox> or circular deps?