[00:49] hello. can anyone help me with compiz [00:50] this is not a help channel; see #ubuntu for help with Ubuntu 7.10, or #ubuntu+1 for hardy [00:50] or #compiz-fusion for compiz support. [00:51] thank you === Onyxyte is now known as Quaver2 [02:05] is ubuntu doing the summer of code? [02:06] hunh, I am unable to branch the casper bzr repo... [02:08] So where's a good place to discuss hardy installation failures? [02:09] ubuntu+1? [02:10] So I can go there and see if someone'll sponsor a bugfix? or #ubuntu-bugs? [02:14] Caesar, You'd file a bug and attach your bug fix and subscribe either ubuntu-universe-sponsors or ubuntu-main-sponsors depending on what part of the archive the package resides in. [02:20] keescook: I've documented it, yes, but it needs to be approved by $work. === cody-somerville is now known as somerville32 [02:33] crimsun_: ah! okay, thanks. :) [02:54] keescook: What error do you get? === slangasek changed the topic of #ubuntu-devel to: Archive: Feature freeze | soft freeze for alpha 6 | Development of Ubuntu (not support, not application development on Ubuntu) | #ubuntu for support and general discussion for dapper/edgy/feisty/gutsy, #ubuntu+1 for hardy | #ubuntu-motu for getting involved in development | http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://wiki.ubuntu.com/HelpingWithBugs [03:12] hm. hope there was nothing in main i wanted to add. [03:14] so [03:14] why are the alphas just called alpha? [03:16] because "why are the alphas called alphas?" is an easier question to answer than "what's a 'tribe'?" [03:19] <[reed]> slangasek: "bearing in mind that the Feature Freeze is also now in effect[3]" <-- except you don't have a [3] at the bottom of your mail ;) [03:19] \o/ [03:19] it's always something, isn't it [03:19] :) [03:19] <[reed]> rly [03:19] <[reed]> ;) [03:19] I could send a followup entitled '[3]' [03:22] Caesar: ah, bug #198110 was your installer bug? [03:22] Launchpad bug 198110 in ubuntu-keyring "ubuntu-keyring doesn't install" [High,Triaged] https://launchpad.net/bugs/198110 [03:26] Anyone else's icons all disappear today? I'm getting "Cannot open pixbuf loader module file '/usr/local/etc/gtk-2.0/gdk-pixbuf.loaders'" when I try to run most GTK programs too [03:27] no [03:27] hmmm [03:31] Caesar: upload sponsored, cheers [04:00] cmpc, eh? [04:00] ogra: does that mean i get sane installtion methods in Hardy? ;) [04:43] slangasek: thanks [05:22] 5~/c [05:22] hi TheMuso ;-) [05:22] Hey LaserJock. [05:42] slangasek: Apparently soyuz has grown the ability to make sure that Main backports only get built from Main sources. So my Postfix backports for Dapper/Edgy are depwait. [05:42] slangasek: Any idea if there's a solution? [05:43] Note for the record that the existing Postfix 2.4.5 backport in Dapper couldn't be there if this had always been the case. [05:46] ScottK: hmm, we could move the postfix backport to the universe section? :) [05:47] slangasek: I've really no idea. I'm sure this is another "Feature". [05:47] Because we wouldn't want to mix the unsupported backports from Main with unsupported backports from Universe. [05:48] is there any reason /not/ to just move the postfix backport to universe? [05:48] slangasek: can different pockets have packages in different components? [05:49] lamont: sure [05:49] what are they dep-wait on, I wonder? [05:49] slangasek: I guess not. [05:49] * lamont is tired after a workout [05:49] tinycbd [05:49] * lamont pukes [05:49] wasn't that main in dapper/edgy? [05:49] Nope [05:50] slangasek: If you can do that, please do. [05:50] yeah. make it universe [05:50] * ScottK has quit thinking about these sorts of things as problems in launchpad. They're just encouragement to work harder on NM. [05:51] well, it looks like change-override.py, at least, is not accomodating [06:11] slangasek: alternatively, we could pull the semi-decent tinycdb back to dapper-backports/main... :=) [06:11] * lamont sleeps before he gets shot. [06:13] lamont: If we did that we could probably avoid the FTBFS on Dapper AMD64, so that wouldn't be bad. [06:15] * milli wonders if lamont will clean up the puke [06:16] ScottK: rails fix should make it good [06:31] hello [06:31] I'm looking for the latest version of the "screens and graphics" program [06:32] how would I get ahold of that? [06:33] milli: I'll fix up rails tomorrow. I need to get to bed. [06:41] Good morning [06:42] Good morning pitti. [06:46] hi pitti [06:47] ahoy LaserJock [06:48] hi highvoltage [06:49] It's a LaserJock! [06:55] hiya Fujitsu [06:56] hi [06:56] I want to get involved with ubuntu [06:56] don't really know where to start though [06:58] GhotiPhud: depends on what you're interested in. Have you read http://www.ubuntu.com/community/participate ? [06:58] yeah [06:58] I've just read that and a bunch of the links [06:59] I was just trying to get the "Screens and Graphics" program code [06:59] so I could see if I could get my tv-out to work with it [07:00] I've got it working with the cli [07:00] just can't make it work with that program [07:01] so I figured I'd take a look [07:35] moin [07:44] /part [07:44] :) === \sh_away is now known as \sh [08:48] seb128: bonjour [08:49] seb128: good job with figuring out the brightness issue! I saw the patch on the upstream ML [08:49] seb128: I'll apply it to bzr head soon and upload a new version to my PPA [08:49] pitti: hello, thanks, credits goes to sjoerd who helped me too ;-) [08:49] pitti: thanks [08:49] seb128: ~ppa2 reverted some FDI changes which I thought weren't necessary any more with gvfs [08:49] pitti: we should have gpm mostly back to normal once we have this change [08:50] but now my internal partitions are back on the desktop [08:50] so I think gvfs didn't do any fixes at all, it was still hal [08:50] * pitti hugs sjoerd [08:51] pitti: what change was that? [08:52] seb128: setting volume.ignore for non-/media partitions [08:54] pitti: weird [08:54] pitti: http://paste.ubuntu.com/5250/ [08:55] pitti: it should only display mounts under the user directory or media [08:55] hm [08:55] ah, right [08:56] pitti: ? [08:56] /dev/hdc4 on /mm type xfs (rw) [08:56] I get an icon for this one [08:56] your partitions are under media? [08:56] hum [08:56] seb128: yes, the other two are under /media, they are (invalidly) automounted === asac_ is now known as asac [08:56] seb128: seems that gvfs/nautilus doesn't honor storage.automount_enabled_hint ? [08:57] since /etc/hal/fdi/policy/preferences.fdi disables automounting for those [08:57] seb128: /mm is in my fstab and mounted at boot; it's an internal partition [09:00] seb128: so, if it's supposed to work like that in gvfs, I'd rather fix it there than working around it in hal, WDYT? [09:02] seb128: is there an ubuntu bug # for the brightness issue? [09:04] pitti: I didn't look to the flood of new gpm bugs since we did the battery change, let me look [09:04] pitti: yes, agreed that was should rather fix gvfs than workaround [09:04] seb128: ok, nevermind, I can look myself [09:04] seb128: brightness patch works perfectly for me; I reported back to upstream, will commit it now [09:04] pitti: bug #191725 [09:04] Launchpad bug 191725 in gnome-power-manager "[Hardy] LCD Brightness Applet doesn't work properly" [Undecided,Confirmed] https://launchpad.net/bugs/191725 [09:05] hum, maybe not [09:06] pitti: thanks, if you upload to your ppa I'll give it a try too [09:06] bug mangled [09:09] seb128: uploaded to ppa [09:10] * seb128 hugs pitti, you rock ;-) [09:10] * pitti hugs seb128 [09:10] * Fujitsu was going to file a bug on that, and is glad he doesn't have to. [09:11] DOH [09:11] did ctrl alt backspace again by error [09:12] I blame ctrl-alt also being for workspace switching ;-) [09:12] That always used to hit me too, several times a day, because I didn't release Ctrl+Alt quickly enough after workspace switching. DontZap works well. [09:15] mvo: hm, compiz doesn't save my session, instead I get a crash report on next login :( [09:15] heh, winkey+number works better for workspace switching [09:15] pitti: were does it crash? inside the session plugin? [09:16] mvo: just ?? [09:16] mvo: I'll report it and let the retracers figure it out, ok? [09:17] * Amaranth hides [09:17] pitti: yeah, that is cool. thanks [09:17] * Fujitsu shoots Amaranth for something Compizy. [09:17] mvo: cool, the border issue is fixed in your new compiz as expected ;-) [09:17] * mvo spots Amaranth hidding in a corner [09:17] seb128: great [09:21] seb128: bug 198295, I'd appreciate if you could put your feedback there, too [09:21] Launchpad bug 198295 in hal "FF exception request: update hal to git head" [Undecided,New] https://launchpad.net/bugs/198295 [09:21] anyone else, if you fancy trying out my new hal version and giving feedback here ^, I'll appreciate [09:22] pitti: Is it ~ppa3? [09:22] Fujitsu: yes [09:23] ah, still needs to be published [09:23] pitti: I wgot that a couple of minutes back to test the brightness fix, but is there anything else to look out for? [09:23] No, it was published 3 minutes ago on all archs. [09:23] Fujitsu: do you have a dell laptop? [09:23] Fujitsu: just whether it generally works for you and shows any regression [09:23] seb128: Yes. [09:23] ok [09:23] because the change is dell specific [09:23] Fujitsu: wifi, suspend, removable drives, etc. [09:29] mvo: ah, seems mine was just a dup of bug 131679 [09:29] Launchpad bug 131679 in compiz "compiz.real crashed with SIGSEGV when attempting to unlock screen" [High,Confirmed] https://launchpad.net/bugs/131679 [09:30] pitti: can you reproduce this crash or did this just happend out-of-the-blue? [09:30] pitti: its a nasty one, but we have currently no idea were it might come from and I was not able to reproduce it yet [09:30] brb, testing changes [09:32] Yay, brightness works. [09:33] Wireless power, however, still does not :( [09:46] Fujitsu: heh [09:46] next version [09:46] mvo: yes, I got it twice, I think [09:47] pitti: dellWirelessCtl seems to not like my wireless :( [09:49] Fujitsu: the button doesn't turn the wireless on/off? [09:50] tjaalton: The BIOS-provided hotkey does it, but hal complains every five seconds that dellWirelessCtl failed. [09:51] Fujitsu: oh, ok [09:51] I presume it tries to switch the radio off or on depending on if I've disabled it in NM. [10:28] pitti: [10:28] seb128: just a hunch, is there any way to disable what you do in ubuntu to have the process info just readable by root? maybe policykit is still not prepared to work fine with that. As far as I traced, the first time everything policykit works fine, for subsequent requests it doesn't [10:28] pitti: any idea about that? [10:28] seb128: we can disable it as a last resort, but maybe it'd be more helpful to fix this bug? [10:29] seb128: no idea off-hand, no; I guess something tries to read argv[0] twice, the second time after disabling ptrace() [10:29] and the result should be cached [10:29] but that's just a random blue sky idea [10:29] seb128: what's the recipe for reproducing? [10:30] pitti: the question was rather how to disable this process info thing [10:30] ah [10:30] pitti: the bug is that gst tools work only once [10:31] when you run them again they don't do any change [10:31] you can try users-admin, add an user, close [10:31] run it again and try to remove the user for example [10:31] removing debian/patches/02_noptrace.patch [10:31] it'll do nothing [10:31] hm, then it's not that patch [10:32] seb128: is there a bug # for it for assinging to me? [10:32] I'd like to do it a little later [10:34] pitti: don't bother for now, garnacho said he will have a look [10:34] ok :) [10:34] pitti: bug #188349 is about the issue [10:34] Launchpad bug 188349 in gnome-system-tools "[Hardy] Unable to save manual network configurations using network-admin" [High,Confirmed] https://launchpad.net/bugs/188349 [10:45] phew, pitti's compiz crash was not in the session plugin :) [10:45] of course i have no idea where it is actually happening... [10:45] Amaranth: no, indeed [10:45] but I got it first with the new compiz [10:45] and session saving doesn't work still, so I thought that at first [10:46] when it dies in the eventLoop that means something most likely went wrong some time ago and just got hit [10:46] PITA to debug [11:06] pitti: ugh ofcourse davidz is contemplating a release a few days after doing the git snapshot package :/ [11:09] sjoerd: just read it, yes :) [11:10] sjoerd: but it won't cause a lot of additional work AFAICS [11:10] and who knows how long it'll take still [11:10] neh, we did all the work already [11:10] hopefully the dell laptop fix gets in [11:11] I tested that on mine, works great [11:11] I applied it to the git snapshot and uploaded to PPA [11:11] <\sh> I wonder what is the difference while compiling code...wine compiled on the host works like a charm, but building via sbuild doesn't [11:11] * Fujitsu has no problems with pitti's ~ppa3. [11:14] Huh. I wonder how the ‘web’ got in there. deb http://archive.ubuntu.com/ubuntu hardy main restricted universe multiverse web [11:22] Could we get rsync 3.0.0 to hardy? (sid has it already) [11:22] * cjwatson mentions this thing called feature freeze [11:22] * \sh needs a buildd admin :) [11:25] cjwatson: To rephrase my question, does anyone think it’s worth a freeze exception? :-) [11:25] <\sh> we have really a big problem right now with wine... [11:25] \sh: Does normal sbuild give a working binary? [11:25] <\sh> Fujitsu: that's my problem ... [11:25] Is it just LP sbuild, or any sbuild at all? [11:25] <\sh> Fujitsu: sbuild doesn't pbuilder does (according to scott ritchie) [11:26] <\sh> Fujitsu: building wine on the host system..gives a running wine...and sbuild doesn't [11:26] <\sh> Fujitsu: my sbuild gives no running wine package, as ubuntus sbuild, too [11:26] That's not a buildd admin problem, then. [11:27] <\sh> Fujitsu: well, we need to find the differences of a local source build and a build via sbuild [11:27] <\sh> Fujitsu: what I'm doing now, I'm using a snapshot sbuild chroot and interactive build wine...let's see what's the outcome is.. [11:28] <\sh> Fujitsu: so now, any sbuild gives a crashing wine... [11:29] \sh: yeah, blame poor sbuild. [11:30] <\sh> Mithrandir: I don't blame sbuild...we need to find out what's the difference ... what does sbuild do what a normal source build doesn't do ;) [11:31] <\sh> Mithrandir: and you agree with me, that pbuilder -> works , sbuild -> doesn't work is really a strange thing, yes? :) [11:31] it's a bug in the package which picks up something from the build environment, then. [11:32] <\sh> Mithrandir: and this we need to find out :) [11:39] yay [11:39] * ogra dances ... edubuntu addon works fine on non networked installs :) [11:39] <\sh> ogra: congrats :) [11:40] nice :) [11:42] bah, i suck at kturtle :) [11:48] pitti: could you give back kdeartwork-kde4, kdenetwork-kde4, kdeutils-kde4 ? [11:49] Riddell: done === doko_ is now known as doko [11:50] thanks [12:06] <\sh> Mithrandir: FYI, manually build wine in snapshot schroot (the very same chroot which is used by sbuild) gives working binaries of wine... === Onyxyte is now known as Quaver2 [12:11] pitti: brightness issue? [12:11] Hobbsee: changing screen brightness on dell laptops [12:12] pitti: oh, every few mins, back to the set brightness? yeah, i was meaning to report that. [12:12] pitti: did they manage to fix the "brightness thing comes up every few minutes, and sometimes corrupts the display, and won't go away again" too? [12:12] Hobbsee: no, actually it was about a broken g-p-m status icon when trying to change it [12:13] oh. damn. [12:13] Hobbsee: but maybe your problem is fixed now as well; please try my PPA version [12:13] pitti: where is it? [12:13] it has literally ~ 100 bug fixes, I don't remember all of them [12:13] https://edge.launchpad.net/~pitti/+archive [12:15] Hobbsee: It fixed the randomly corrupt brightness indicator for me too. [12:17] Fujitsu: oh goody [12:17] Fujitsu: do i have to do anything else after it installs and restarts hal? [12:17] pitti: thanks [12:17] Hobbsee: Nope. [12:17] good [12:18] Just press the brightness keys and watch it work magically. [12:19] oh wow. it even shows the correct things now [12:19] Yep. [12:19] way cool. earth is supposedly going to blow up. [12:20] Is there any reason that it doesn't use the nice translucent window that the volume control does? [12:21] noone coded it [12:21] I would have thought that they would have used the same method to display it... [12:26] asac: gnash plugin is not detected on hardy. The problem is there is no link in /usr/lib/firefox-addons/plugins to /etc/alternatives/firefox-flashplugin. But I don't think this is gnash specific. [12:27] slytherin, that should rather be xulrunner-addons, no ? [12:27] pitti: wow, it really does seem to fix it [12:28] ogra: I think so. But for some reason the link present /usr/lib/firefox/plugins is not working [12:29] Fujitsu: yeah, it simply needs copying code from gnome-settings-daemon into gnome-power-manager [12:29] ogra: And /usr/lib/firefox-3.0b3/plugins is a symlink to /usr/lib/firefox-addons/plugins [12:29] ogra@ceron:~$ ls /usr/lib/xulrunner-addons/plugins/flashplugin-alternative.so [12:29] /usr/lib/xulrunner-addons/plugins/flashplugin-alternative.so [12:29] its there for me and works fine [12:30] and it properly points to /etc/alternatives/xulrunner-addons-flashplugin [12:30] zdzichuBG: Ah, right, forgot they were generated by different apps. [12:30] which in turn points to /usr/lib/flashplugin-nonfree/libflashplayer.so [12:31] ogra: Ok. So there is no /usr/lib/xulrunner-addons/plugins/flashplugin-alternative.so in case of gnash which is causing problem. [12:33] slytherin, aha [12:34] ogra: I will file a bug against gnash [12:34] slytherin, that should be set by the postinst of gnash [12:35] i bet there are already update-alternative lines for the old location [12:36] ogra: A bug is there already it also includes a patch (but no debdiff). bug #194379 [12:36] Launchpad bug 194379 in gnash "enable xulrunner 1.9/firefox 3.0" [High,Confirmed] https://launchpad.net/bugs/194379 [12:37] slytherin: yes flash upload will happen within a few days [12:37] asac: Ok. SO will you also fix the bug I just mentioned in 0.8.2 packaging? [12:41] yes [12:41] * asac lunch time [12:42] asac: ok, thanks [12:51] pitti: libmiracle0.2.4, libmlt0.2.4 and libvalerie0.2.4 only look to depend amongst themselves, so can all be NBS'd. [12:53] StevenK: great, thank you! removed [12:54] * Hobbsee removes yada, and kmos, too [12:54] * ScottK cheers [12:54] pitti: Please tell me if I should log a bug for this. I am using b43 driver with appropriate firmware on my ibook. It works fine but jockey shows that the driver is 'Not Enabled' but it says it is 'In Use'. [12:55] slytherin: please do file a bug, yes; do 'jockey-gtk --list --debug 2>&1 | tee /tmp/jockey-debug.txt' and attach the log [12:56] pitti: I will file a bug right now and will provide additional info tomorrow. [12:56] slytherin: there is already a bug about it, btw [12:57] slytherin: ah, nevermind, that was a different case (enabled, but not in use) [12:58] pitti: now that we have libxcb-enabled libx11 again should we default to sloppy locks like it was for a brief time during feisty? old java's fail currently (bug 87947) [12:58] Launchpad bug 87947 in libx11 "xcb_xlib.c:50: xcb_xlib_unlock: Assertion `c->xlib.lock' failed." [Unknown,Fix released] https://launchpad.net/bugs/87947 [12:59] not really my field of expertise, I'm afraid [12:59] tjaalton, bryce: ^ ? [12:59] yep, it's me ;) [12:59] * pitti hugs tjaalton [13:00] pitti: I'll ask bryce about it later, it should be safe to use sloppy locks by default [13:05] * tjaalton hugs pitti back for adding cmdline-options to jockey [13:06] :) [13:06] I'm currently at adding --check-composite [13:06] and next upload will enable it for fglrx, too [13:06] I can finally fix bug 54335 :) [13:06] Launchpad bug 54335 in linux-restricted-modules-2.6.24 "fglrx should contain install instructions" [Low,Confirmed] https://launchpad.net/bugs/54335 [13:18] slangasek: Apparently republishing the postfix backport for Dapper into Universe wasn't enough to convince soyuz that tinycdb exists. [13:24] ScottK: does the log show it fetching universe in the apt-get update? [13:25] * ScottK looks [13:26] lamont: It shows that it does not. [13:27] then it didn't really go into universe. :( [13:27] Launchpad says it is, but we all know Launchpad lies. [13:28] ScottK: you could always just re-upload postfix to dapper-backports with tinycdb-crack turned off.. [13:28] 05:51 well, it looks like change-override.py, at least, is not accomodating [13:28] which sounds like there was a problem [13:28] yeha [13:28] yeah, even [13:29] lamont: If there isn't a soyuz solution I'll do that, but I'd like to get this worked out. Stuff moved from Universe to Main all the time and if this new 'feature' stays with us it will be a substantial complication for backports. [13:51] Ok, my two new packages got rejected with "None" [13:51] how am I supposed to know why if no one puts in a decent comment? [13:52] "If you don't understand why your files were rejected..." [13:52] I think that would be the case for everyone if that is the most common rejection reason ("None") [13:54] Riddell: ping [13:56] Riddell: is the XubuntuY convention supposed to be used for packages that are only in Ubuntu as well? [13:56] BenC: yes unless it's certain that it won't ever be in debian (e.g. the linux packaging) [13:57] hmm, never had that enforced for new packages before...ok [13:57] but then you'd package native anyway [13:57] ogra: it has an upstream, it's just not a Debian sync [13:57] ah [13:57] Riddell: ok, is there anything else wrong? I don't want to reupload just to hit another reject [13:57] BenC: everything else is fine [13:58] Riddell: ok, 2 minutes and they will be back in the queue [14:02] Riddell: upload completed for both with 0ubuntu1 revisions [14:05] hello everybody [14:06] BenC: groovy, accepted [14:06] Riddell: thanks! [14:07] calc: new openoffice binary packages, main or universe? [14:11] I'm trying to package tomcat6, is there a wnpp list for ubuntu? or i should apply to debian's one? [14:14] miguillo: For Ubuntu you file a bug against Ubuntu and tag it 'needs-packaging' and then assign it to yourself. Filing a Debian wnpp bug too is also good. [14:15] Are the issues with the hardy installer and software RAID known? [14:23] Bashtoni: --verbose [14:25] cjwatson: Gives me an error about being unable to remove files from the target [14:26] which CD exactly? [14:26] "hardy installer" is very vague [14:26] cjwatson: Netinstall (PXE) [14:26] Bashtoni: bug 198106 [14:26] Launchpad bug 198106 in partman-target "Configure partitions for RAID1, received: The installer needs to remove operating system files from the install target, but was unable to do so." [Critical,Fix released] https://launchpad.net/bugs/198106 [14:27] Thanks :) [14:27] Bashtoni: I'd appreciate verification of that fix once it hits your mirror [14:28] cjwatson: No problem, will do [14:46] we are going to be doing a merge from debian that has a couple of bugfixes for samba do we need a ffe? [15:18] Riddell: is there a guide to moving to KDE from KDE3 on Hardy? [15:18] KDE4, sorry [15:18] sabdfl: in gutsy or hardy? [15:18] hardy [15:19] i have a KDE3 desktop, want to try out KDE4 again [15:19] sabdfl: just install kde4-core [15:19] ok, thanks [15:19] does that drop all the kde3 stuff? [15:19] or kde4 for unadulterated full thing [15:19] sabdfl: no, they install alongside [15:19] just log out and select kde4 from the display manager [15:19] how do i make sure i get kde4 desktop environment on login? [15:19] ok [15:20] sabdfl get's pretty good tech-support ;) [15:21] yeah. i wonder why. === sourcode_ is now known as sourocde === kylem_ is now known as kylem === \sh is now known as \sh_away === \sh_away is now known as \sh === \sh is now known as \sh_away === \sh_away is now known as \sh === \sh is now known as \sh_away [16:43] Should dpkg-reconfigure setup DPKG_MAINTSCRIPT_PACKAGE (so that dpkg-trigger calls work correctly)? [16:44] hmm [16:44] interesting question [16:46] * cjwatson asks the dpkg-trigger author elsewhere [16:56] propagandist: we think it should but it'll be a bit fiddly to do. Could you file a bug on debconf about it, please, to make sure it doesn't get forgotten? [16:58] glatzor: I'm currently going through kdeguidance bugs and I see in Bug #93749 you said you had some displayconfig-gtk changes you were going to move to the backend. I was wondering if you had done that (I don't have an easy way to test for this bug) or could point me at the right place to check for it in the backend? [16:58] Launchpad bug 93749 in kde-guidance "[apport] GUI tools don't handle a missing xorg.conf file correctly" [Medium,In progress] https://launchpad.net/bugs/93749 [17:00] cjwatson: sure thing :o) already halfway there [17:01] hi, I'm wondering what criterium is used to decide which packages are shown in gnome-app-install, and which aren't ? [17:01] Xteven: the availability of a desktop file (menu entry) [17:01] hmm [17:02] is that information encoded in the Packages file of the repository somehow ? [17:02] ScottK: one moment [17:02] glatzor: Thanks. [17:03] Xteven: what do you want to do? [17:03] Xteven: /usr/share/app-install [17:03] glatzor: I'm trying to get one of my packages to show up in the gnome-app-install interface :) [17:06] Xteven: do you have got a custom distribution? [17:07] Xteven: otherwise contact Canonical in the case of a proprietary software or try to get it into Universe if it is free [17:07] no, the package I'm building contains a postinst script that configures a bunch of office printers in the building [17:08] Riddell: main [17:08] contacting canonical might be a bit overkill [17:09] Xteven: take a look at the packages: app-install-data und app-install-data-partner [17:09] ok, will do [17:09] thx :) [17:09] Xteven: basically you have to drop a desktop file to /usr/share/app-install and then run update-app-install in the postinst [17:10] Xteven: but this discussion belongs to ubuntu-users [17:11] * Chipzz waves at Xteven :) [17:12] hi Chipzz [17:13] glatzor: running postinst is done at package installation, it doesn't put the package in the app-install listing before it is installed [17:13] kindof odd :) [17:13] Xteven: you need a seperate package with the .desktop file in it [17:14] which I guess creates a circular loop :P [17:14] if I understand correctly, the X-AppInstall-Package tag in a desktop file is somehow encoded into the repository information [17:14] Xteven: afaui, it's not [17:14] Xteven: gnome-app-install just looks at the files on the local system, nothing to do with a repository [17:15] Xteven: dpkg -L app-install-data [17:15] so the list with installable applications (from gnome-app-install) is fixed ? [17:15] not exactly [17:15] it's just not determined from the repo, but from local information [17:16] do dpkg -L app-install-data [17:16] that has a lot of the desktop files that show up in gnome-app-install [17:16] hm [17:16] Xteven: in your concrete case, I would make a kul-app-install package with some .desktop files in it [17:17] so there is no way to get my package in there, unless I hijack app-install-data or app-install-data-commercial [17:17] ok [17:17] like I said, they don't have to be in *that* package [17:19] Xteven: cfr /query ;) [17:19] Xteven: doing system configuring by using packages is not the best approach at all [17:19] what is the best approach ? [17:20] Xteven: depends on your environment [17:20] 100 computers spread out across the planet [17:20] glatzor: pretty big university (> 20.000 students, probably > 30.000 by now) [17:21] Xteven: I don't suppose you can force your users to install using a modified CD? [17:21] no [17:21] cjwatson: well, the output from change-override.py made me fearful that I'd kicked postfix to universe in dapper itself; I tried to set things back and now postfix in dapper-backports shows as being in universe, so I guess that's progress [17:21] glatzor: so custom cd is not really a solution [17:21] no custom cd, no global policy stuff [17:22] thinking more about this, I have another idea [17:22] but I doubt if that's really feasable at all [17:22] Xteven: have you already thought about landscape? [17:22] nope, never even heard of it :) [17:23] I see [17:23] a central config system [17:23] I was thinking more along the lines of ubuntu providing some infrastructure for extra packages which only show up on certain mirrors [17:23] but that would be opening up a whole new can of worms [17:24] I don't want to maintain other people's desktops, I just want to make it easier for them to install 15 printers on their system [17:24] anyway, I have to go to class [17:24] thx for the help and cya later [17:24]