[00:26] Would anyone happen to know why mdadm installs udev rules which create /dev/md/* symlinks ? [00:26] Why the extra names, instead of just using /dev/md* ? === dendrobates is now known as dendro-afk [02:08] SpamapS, pitti: Hi. "pkgsel" package for Lucid-SRU(Bug #797985) is still in unapproved queue. Could you take a look to push lucid-proposed? Thanks in advance. [02:08] Launchpad bug 797985 in pkgsel (Ubuntu Lucid) "Please export Japanese translation from Launchpad" [Medium,In progress] https://launchpad.net/bugs/797985 === dendro-afk is now known as dendrobates === _LibertyZero is now known as LibertyZero === hunger_ is now known as hunger === hunger is now known as hunger__ === hunger__ is now known as hunger [08:18] Daviey: your cloud thing is losing it's mind again and spamming us with failure mails === ogasawara_ is now known as ogasawara [08:52] nobuto: we'll get to it; I'll probably ask our latest SRU team member RAOF to review it :) === Adri2000 is now known as Guest89352 === Guest89352 is now known as Adri2000 === fenris is now known as Guest83722 === ogra is now known as Guest4143 === Guest4143 is now known as ogra_ === ogra is now known as Guest69691 === Guest69691 is now known as ogra_ [11:06] I have a package which has only static files which shall be installed, and I've put the list of the files into debian/install. However, I need to pass --sourcedir to dh_install. How do I do that when I'm using the default %: dh $@ rules? [11:07] sveinse: You can write an override_dh_install: target that will be called instead of dh_install. That target would call dh_install --sourcedir=whatever. [11:08] Are there any variables available for the root of the sources? [11:09] The root of the sources? debian/rules is guaranteed (by dh_test) to be called from the root of the sources (or at least the directory containing debian/), so CURDIR will be what you want. [11:10] Excellent. Thanks [11:23] pitti: What is the SRU policy WRT translations? The package for Bug #797985 appears to contain a full export of launchpad translations, so has a bunch of translation changes unrelated to the specific bug, but we probably want them all. [11:23] Launchpad bug 797985 in pkgsel (Ubuntu Lucid) "Please export Japanese translation from Launchpad" [Medium,In progress] https://launchpad.net/bugs/797985 [11:50] RAOF: I can cut it down if you like, but there didn't seem a strong reason to do so [11:52] cjwatson: I agree. As the newest SRU team member I'm just checking that everyone *else* agrees, too :) [11:56] https://launchpad.net/~linaro-validation/+archive/ppa === chrisccoulson is now known as chrisccoulson_ === chrisccoulson_ is now known as chrisccoulson [12:42] Is it possible to rename the installed file when using the listing in debian/install? [12:45] sveinse: no, sorry, that's a documented limitation (in dh_install(1)) [12:46] sveinse: if you need that, create the containing directory in debian/dirs, and then have an override_dh_install rule that does 'install -m0755 source-file-name debian/packagename/usr/bin/target-file-name' or whatever [12:46] yes, thanks [12:48] Is it considered bad to do 'install -D ...'? (instead of using debian/dirs I mean) [12:51] cjwatson: ^ === ogra is now known as ogra_ [12:52] sveinse: doesn't matter much [12:59] mvo: hi! vte 2.90 is in unstable now, I think next gdebi upload could be targeted unstable as well. [13:03] DktrKranz: awsome news! === arand is now known as mew === mew is now known as mewerner [14:20] cjwatson, kees: just sent a tb@ mail about moving the meeting today [14:24] pitti: are you available? [14:24] mdeslaur: yes, I am; I can come over to the security room [14:24] ok? [14:24] pitti: sure, cool [14:24] your chairs are a lot better than our's :) [14:25] pitti: yes, they are :P [14:25] pitti: some of our chairs were stolen though [14:25] kees: Dude! Don't tell him, he won't come [14:34] What are the options in the section line in debian/control ? I thought I could put universe here, but lintian refuses to accept it [14:34] ha at the security team being a victim of chair theft :-) [14:42] sveinse: The universe keyword is added by the archive software, and does not get put into actual package control files. [14:48] Laney: bad security I'd say :) [14:48] * Laney expects a USN post haste [14:51] * micahg sends USN-CHAI-R [14:53] TheMuso, what can be put into the section keyword? [14:54] anyone know which package contains GtkFileChooserDialog in Oneiric? [14:59] sveinse: http://www.debian.org/doc/debian-policy/ch-archive.html#s-subsections [15:00] ok, so the debian sections also applies for ubuntu. Thanks [15:00] brendand: that'd be probably the gtk package [15:01] brendand: https://launchpad.net/gtk [15:02] evfool - not possible to assign a bug to it though. i think that's a project page anyway, not necessarily a package [15:03] i guess it may be https://launchpad.net/ubuntu/+source/gtk+3.0 ? [15:04] brendand:yep, that's it [15:20] Laney: you mentioned it were interested in the TB meeting? we're moving it 2 hours earlier, fyi [15:21] kees: righto. I'm just interested in hearing the outcome to my mail really :-) [15:21] thanks for letting me know [15:21] okay :) [15:36] james_w: do you know whether status.u.c. publishes the db files anywhere? I need the current one for debugging [15:36] james_w: I can't access viburnum.canonical.com [15:37] pitti, I didn't change that code, so it does [15:37] http://status.ubuntu.com/ubuntu-oneiric/oneiric.db doesn't exist [15:37] http://status.ubuntu.com/ubuntu-oneiric/ubuntu-oneiric.db [15:37] oh, ubuntu-oneiric.db [15:37] *headdesk*, sorry [15:37] thanks! [15:37] np [15:38] james_w: we retargeted a whole bunch of oneiric specs to p, and the tables updated properly, but not the charts [15:38] hmm [15:38] james_w: I suppose the charts still count the dropped WIs somehow [15:38] not sure what would cause that [15:39] pitti, you did it by changing the series on the blueprints? [15:39] yes [15:39] but I suppose that didn't clean the specs from the db [15:39] hmm [15:40] I'm still not quite sure what happened [15:40] just that we dropped some 50 WIs, and there's no visible dent at all on http://status.ubuntu.com/ubuntu-oneiric/canonical-desktop-team.html [15:43] hm, on second look I actually can't find any inconsistency in the numbers [15:45] I'll just do an experiment [15:47] james_w: do you know when the WI tracker cron fires? [15:48] pitti, *:33 [15:48] james_w: thanks [15:51] tseliot: if [ "$1" = install -o "$1" = upgrade ] && [ -z "$2" ] [15:55] mvo: any availability to talk about the stuff we didn't get around to talking about? [16:11] cnd: can x11proto-input 2.0.2-2 be merged? [16:12] @conference/canonical <--- All-Hands? a sprint? [16:12] Error: "conference/canonical" is not a valid command. [16:12] udevbot_: sit down you pile of persnickety Python. [16:12] Error: "sit" is not a valid command. [16:13] mneptok: platform rally [16:14] (nee sprint) [16:14] Don't forget LP [16:14] mdeslaur: I will come over at 17:00 or tomorow morning, is that ok? [16:14] oh yes [16:14] mvo: sounds good, thanks [16:15] that's some coincidence. i just bought "Platform Rally" for the PS3. [16:15] Indeed, we're doing a bunch of good work. Unless you're infinity. [16:15] That came out wrong. :-( [16:15] StevenK: welcome to my world. [16:16] mneptok: You should see a doctor for that. [16:16] StevenK: i tried. the doctor said solving that issue is like putting a Band-Aid on an amputee. :/ [16:20] slangasek, I'm not sure what you're asking? [16:21] I'm sure it could be merged, it looks like a bug fix according to the version numbers [16:21] there could be conflicts, but I'd be surprised if there were [16:26] cnd: I'm asking if it's safe to merge the new upstream version; you're listed on https://merges.ubuntu.com/main.html as the last person to touch it, I haven't looked at the delta at all and don't know if from your perspective there's a risk of regressions caused by pulling the new upstream version [16:36] slangasek, let me take a look at debian's changes, but I can't imagine it would be a problem [16:36] my patch to x11proto-input is pretty large though :) [16:36] it includes the prototype xinput 2.1 support [16:37] but that should be rather self-contained away from the rest of xinput 2.0 and earlier stuff [16:41] slangasek, looks like it should be fine save for one trivial conflict [16:41] I pushed a small patch upstream that landed in 2.0.2 [16:41] that change was rolled up as part of my 1_xi2.1.patch, instead of being a separate patch [16:42] I should have made it a separate patch back when I added it :( [16:43] removing lines 94-101 (a small diff hunk) of 1_xi2.1.patch should do the trick [16:49] bdmurray: /usr/share/apport/general-hooks/ubuntu.py === beuno is now known as beuno-lunch [17:06] pitti: do you know how often the ddebs.ubuntu.com packages files are rebuilt ? [17:07] 10 3,7,11,15,19,23 [17:07] apw: ^ for oneiric [17:07] pitti: and for older releases ? [17:08] 10 2,10,18 [17:08] pitti: thanks :) === beuno-lunch is now known as beuno [17:48] What is the difference between dh_install and dh_auto_install ? [17:49] have you read the man pages? they have quite good descriptions [17:52] Yes, but admit skimming them. Let me rephrase: dh_install handles the debian/install files while, dh_auto_install takes care of calling make install, right? [18:01] I have a rule here where I override install. How should I pass $DESTDIR to the target makefile? override_dh_auto_install: \n make INSTALL_ROOT=$(DESTDIR) install does not work, INSTALL_ROOT is empty === ogra is now known as Guest32535 === Guest32535 is now known as ogra_ [18:16] sveinse: dh_install/dh_auto_install> right, that's a good summary [18:19] It seems DESTDIR environment variable is not set when going into the override_dh_auto_install rule [18:29] Hmm. I'm clueless and google doesn't tell me much. [18:29] How do you guys handle install if the Makefile wants to install to INSTALL_ROOT instead of DESTDIR? [18:30] Can't patch Makefile, as it's auto generated [18:31] sveinse: patch the non-auto-generated source file [18:32] This is a Qt application. That would mean to patch upstream qmake in another package... [18:32] sveinse: look at how other qt apps have been done maybe? [18:33] "apt-cache search qt" may give you a few pointers of other qt-based applications [18:34] Doing a quick search of qt4-x11 sources, it seems qmake is not patched for chaning INSTALL_ROOT to DESTDIR at least. Checking apps... [18:35] But override_dh_auto_install of qt4-x11 reveals it: $(MAKE) install INSTALL_ROOT=$(CURDIR)/debian/tmp/ [18:36] Not what I expected thou [18:39] sveinse: oh, sorry - INSTALL_ROOT and DESTDIR are equivalent except for the name? Yeah, no need to patch that, the described override is pretty much what one would expect [18:39] (provided that dh_auto_install doesn't actually know how to detect qmake) [18:41] Hmm. Not quite. If I remove my override_dh_auto_install, I see dh running "make -j1 install DESTDIR=/home/user/packages/lmas/debian/lmas", while the above override uses debian/tmp [18:47] sveinse: that depends on the dh compat level; with a modern package, $(CURDIR)/debian/lmas is correct [18:47] it also differs for multi-binary packages [18:48] qt4-x11 is unlikely to be a good example to refer to for Qt *application* packaging [19:03] cjwatson: This should be a single-binary package. Are there any other places than debian/control that control if it's a multi or single binary package? [19:05] sveinse: nope [19:06] Given a newer dh compat level and if $(CURDIR)/debian/lmas is correct, then I need to find my own package name unless I want to hardcode the latter "lmas" part of the path [19:07] $(shell dh_listpackages) [19:07] (assuming single-binary) [19:09] excellent, now things look better. Thanks === dendrobates is now known as dendro-afk [19:57] Hi! [20:05] I'm not familiar with the development process of Ubuntu. I read the wiki, but I'm still a bit unsure about the significance of DebianImportFreeze. [20:07] The flashrom package in Debian unstable is a bit outdated and the release of a new upstream flashrom version is expected to happen in the next few weeks. Is there a chance to get that new (mostly bugfixes) flashrom version included in Oneiric, and if so, can it be made a copy of the Debian flashrom package which is hopefully updated directly after the flashrom release? [20:12] carldani: DebianImportFreeze only means that automatic copies of packages Debian -> Ubuntu will stop, once the updated flashrom package is in Debian someone has to request a sync from Debian (assuming the upload to Debian happens between DebianImportFreeze and FeatureFreeze) [20:18] geser: Thanks for the info! === dendro-afk is now known as dendrobates === yofel_ is now known as yofel === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates