[00:06] infinity: unless lp confuses me again, now would be a good time to give back happy alex on lpia ;) [00:07] happy alex? [00:07] Nafallo: see backlog [00:07] Nafallo: (iow, the joke was already made) [00:08] sistpoty: ah. I /lastlogged and couldn't see what you meant first. then I saw the "and" ;-) === lutin is now known as Lutin [00:09] or maybe lamont? (since infinity seems to be away) [00:11] dpkg-buildpackage: source only upload: Debian-native package [00:11] what did I do wrong? [00:12] TomJaeger: wrong name of orig.tar.gz, I assume (btw.: for such packing question I assume that #ubuntu-motu is more suited) [00:12] okay, thanks === elmo changed the topic of #ubuntu-devel to: Launchpad is going down from 00:55 UTC until 01:15 UTC for a code update || Archive: Feature Freeze, DocumentationStringFreeze | Ubuntu 8.04 LTS Beta released | 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 ht [00:42] * slangasek blinks [00:45] slangasek: Well, be thankful the limit isn't what it used to be in IRCnet. :-) [00:46] oh, no, I'm just blinking at the downtime === elmo changed the topic of #ubuntu-devel to: Archive: Feature Freeze, DocumentationStringFreeze | Ubuntu 8.04 LTS Beta released | 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 [01:16] Okay, so this is the first time (since I got a shiny new laptop with decent 3D) that I've run compiz instead of immediately switching back to metacity... [01:17] Can anyone tell me if I can "fix" the fact that the fancy 3D Alt-Tab switcher doesn't understand Shift-Alt-Tab to cycle backwards? [01:18] infinity: I'd lie, if I say I could in case you'd give back alex and happy on lpia :P [01:19] sistpoty: I'd love to, if LP wasn't down. [01:19] infinity: it still is? [01:19] sistpoty: Oh, it's back. :) [01:19] :) [01:20] sistpoty: So, how do I fix the compiz task switcher? :P [01:21] infinity: there are various plugins that provide alt-tab like switching, so it may be that one of them can do it. [01:21] infinity: good question, I use KDE though, and I lied :P [01:21] although the one that I have doesn't understand it [01:22] james_w: Yeah, the default one seems to not know how to go backwards, which makes me grumpy. [01:22] james_w: Given that every window manager (heck, every OS) I've used in the last decade could do so. :P [01:22] infinity: run ccsm [01:22] infinity: you may have to install it first [01:22] simple-ccsm? [01:22] select "Application switcher" [01:23] you need the real thing I think [01:23] the advanced button in simple-ccsm spawns it [01:23] on the "Bindings" tab there is a "Prev window" setting, and it is unbound by default [01:24] Genius. [01:24] I've never looked at any of this before. [01:25] Of course, binding that should be the default, IMO, but I don't know how deeply I care about filing a bug to fix it this late. [01:26] * infinity squeals with glee. [01:26] james_w: You're my hero for the next 30 seconds. [01:26] james_w: If you ever needed a favor from me, ASK FAST. [01:26] umm, umm... [01:28] too slow [01:31] james_w: You had a chance to ask infinity's hand in marriage, and you blew it? [01:32] dammit [01:37] asac: I've found the problem you were having, apt-get source only responds to a source package name without a version. I think I can work around it by picking the first binary package out of debian/control. I think it's unlikely that the first package will have been added in that revision. [02:10] what's the status of launchpad? [02:10] is it rebooted and done it's maintenance? thanks :) [02:10] nepbabu: there is a #launchpad [02:11] looks like it's all over though. [02:11] james_w: tia! but the SSL cert is invalid says ff [02:12] nepbabu: are you on firefox 3? [02:12] james_w: yes [02:12] ff3b4 james_w [02:12] I don't know then, it works here [02:12] hmm.. [02:13] any buildd admin still around who could give back hscolour and haddock on lpia? [02:18] * sistpoty tries with highlighting the indirect members: elmo, Spads, lamont, Ng for the give-backs of hscolour and haddock on lpia :) [02:43] Can somebody help me with bug #195953 ? [02:43] Launchpad bug 195953 in wacom-tools "Tablet input resolution tied to display resolution" [Medium,Triaged] https://launchpad.net/bugs/195953 [02:45] There's the same problem with plain mice. I've been saying it should be fixed. :-) [02:53] I'm just trying to get a new upstream version uploaded. How would I go about getting sponsorship for the new version and do I need a freeze exception? [02:56] TomJaeger: take a look at https://wiki.ubuntu.com/SponsorshipProcess and https://wiki.ubuntu.com/FreezeExceptionProcess [02:58] Wow, that sounds complicated [02:58] TomJaeger: What package? [02:58] wacom-tools [02:58] Is it bug fixes or new features? [02:58] I don't know why nobody feels responsible for this [02:59] bug fixes [02:59] TomJaeger: 20,000 packages and ~100 developers. You do the math. [02:59] Is this the right channel to ask for somebody to apply a fix for a bug? https://bugs.launchpad.net/ubuntu/+source/qt4-x11/+bug/208556 [02:59] Launchpad bug 208556 in qt4-x11 "qdbuscpp2xml uses moc-qt3 instead of moc-qt4" [Undecided,New] [02:59] ScottK2: 100 developers, most of whom are inactive. [03:00] That too [03:03] TomJaeger: Since it's a Main package it's a rather simpler process. [03:04] TomJaeger: Is the release in Debian already? [03:04] no [03:04] andresj: You will probably have more luck with that package on #kubuntu-devel [03:06] ScottK2: s [03:06] "it's maths, jeremy" [03:06] The thing that bugs me is that all these procedures will take just a few minutes for someone who knows what they're doing, but will take me forever to figure out [03:06] TomJaeger: but once you've figured it out [03:07] TomJaeger: I can understand that, but the first thing it takes is motivation. [03:07] it's not like the people who can do it are sitting around scratching their backsides, they're busy on other things that they feel are more important [03:07] TomJaeger: I've got about a dozen things on my list to get done before release. If I get to 4, I'll be lucky. [03:08] ok, ScottK2 [03:10] I'm more than happy doing things like tracking down bugs, providing patches and talking to upstream. But trying to package that stupid package has been a complete nightmare so far. I feel like my productivity there is close to zero. [03:11] OK. [03:12] I'd suggest focus on writing a bug that makes the case for why it's important and why it's safe. [03:12] I've done that [03:13] TomJaeger: What bug? [03:13] bug #195953 [03:13] Launchpad bug 195953 in wacom-tools "Tablet input resolution tied to display resolution" [Medium,Triaged] https://launchpad.net/bugs/195953 [03:14] If it doesn't get fixed, Tablet PCs are almost useless with hardy [03:14] Right. [03:15] slangasek: Would you please look at ^^^^ bug and give an opinion on if you'd accept a new wacom-tools version to fix it at this point? [03:16] Do I know why the debdiff is so large? It's a complete mystery. The diff between the two directories is 600k, and that's mainly autoconf stuff [03:16] TomJaeger: He's the Release Manager. [03:23] ScottK2: I would potentially accept *a* new wacom-tools version, but to know whether I'd accept *this* wacom-tools version it would have to go through the FFe process; and I probably wouldn't be the one to sponsor it myself, I think it'd be better to have it done by someone with more direct knowledge of the issues [03:23] slangasek: I understand. I'm just looking for a rough idea of if it's worth pursuing. Who'd be the core-dev best equipped to look into this? [03:25] ScottK2: bryce, maybe? [03:25] i.e., Mr. Inkscape [03:25] Right. [03:25] slangasek: Thanks. [03:26] TomJaeger: Now you have an idea that this might be doable and who you want to discuss it with rather than worry overmuch about packaging it yourself. [03:27] ScottK, thanks [04:27] hey slangasek I have something for you :-) [04:27] slangasek: http://people.ubuntu.com/~bryce/bisect/ [04:28] slangasek: these are pre-git bisected -intel debs for testing. [04:30] (the page is still a work in progress) [04:32] bryce: how are you building them? [04:32] james_w: bunch of scripts [04:33] james_w: I'd gotten stuck down a rathole trying to get them to build in PPA, but PPA has turned out to be worthless for this [04:33] I was just wondering if you be able to basically run "git bisect run debuild" [04:33] james_w: well, ironically 'git bisect' never gets used [04:34] It obviously wouldn't work just like that, but it's an interesting idea. [04:34] bryce: :-) [04:35] yeah basically I pull a bunch of info out of git log, sort the commit id's, then do a bunch of git reset's and such to cycle through them, plunk in debian/ dirs, debuild, pbuilder, rsync, and generate the web page [04:38] bryce: TomJaeger was here earlier looking at bug #195953 and hoping for someone to champion getting it uploaded. slangasek pointed vaguely in your direction. [04:38] Launchpad bug 195953 in wacom-tools "Tablet input resolution tied to display resolution" [Medium,Triaged] https://launchpad.net/bugs/195953 [04:41] ScottK2: whoa that's an insanely huge debdiff [04:42] rats, it crashed my ff3 [04:42] :-D [04:42] bryce, the debdiff b/w 0.7.7.7 and 0.7.9.3 is more than 5M [04:42] A nice little 3.5 MiB debdiff. [04:45] I made a little mistake in the "original" tarball there. the doc directory should be one level further down. Not that it changes anything about the size of the debdiff. [04:45] hrm, well looks like it needs some additional packaging work before it can be put in [04:45] TomJaeger: wait, why are you diffing 0.7.7.7 and 0.7.9.3 when hardy currently has 0.7.9.3? [04:46] bryce: heh, my architecture is amd64, not i386... [04:46] slangasek, just for comparison [04:46] bryce: I think a git bisecting howto would've served me better :-) [04:46] slangasek: rats!!! [04:46] slangasek: back to the drawing board [04:47] TomJaeger: erm, comparison of what, exactly? I don't see how 0.7.7.7 can be relevant here [04:48] git bisect is teh awesome. :-) [04:48] slangasek, people are complaining about the size of the debdiff. I'm just saying the last debdiff was even bigger. [04:48] oh, I see [04:48] it's not the size of the debdiff that matters.... [04:48] the issue seems to be that there is prebuilt stuff in the original tarball [04:48] it's the color of the pill that makes the debdiff easier to work with. [05:11] bryce: Yeah. The packaging needs some work by someone who knows what they are doing (e.g. not me). === elkbuntu_ is now known as elkbuntu [05:57] how long does it take for a bzr commit to show up in LP? [05:57] less than 30 minutes typically for the HTTP branch to be updated [05:57] I did a commit 35m ago and it still isn't showing [05:57] oh ok so it should be soon [05:58] the SFTP/SSH developer one is instant of course [05:58] i did a ssh and its not showing up [05:58] yeah, typically for me it hasn't been more than 30 min [05:58] hmm i'll check it out somewhere and verify the commit made it [06:02] wow this is taking a very long time to checkout [06:02] calc: bzr-lp is particularly slow now; they're working on it. [06:03] It'll be like this for another couple eof days, thought. [06:03] Fujitsu: ah ok [06:03] If you use bzr+ssh, it should be quicker to update, but SFTP has to be checked to ensure you haven't uploaded evil stuff. [06:04] yea i am using bzr+ssh [06:04] ok the commit is there, it just haven't shown up on the webpage [06:04] Fujitsu: you mean it's *not* for my video collections? [06:04] jdong: Devastating, I know. [06:09] I think I figured out all the different ways the debian maintainer messed with the original tarball. Things look much better now and the debdiff's down to 600k. === andresj__ is now known as andresj [06:38] Good morning === asac_ is now known as asac === doko_ is now known as doko === hunger_t is now known as hunger === \sh_away is now known as \sh [09:42] yep, we really need the newer wacom-tools, but I never figured out how to compile it and pushed it forward. For some reason the Debian maintainer hasn't updated it, even though he promised to do it two months ago === elkbuntu_ is now known as elkbuntu [11:08] james_w: ah, good to know. thanks! [11:22] tkamppeter_: shouldn't, but in the interest of not accidentally catching unrelated devices I'd transform the list as the existing script does [11:23] Chipzz: gnu tar switch> right, I see; would break compatibility [11:24] <\sh> moins pitti :) [11:25] pitti: and probably in quite a visible and especially *unwanted* way :S [11:28] pitti: as an unrelated data-point, I did man tar on my OSX system, to see which options would be different. And apparently OSX doesn't even ship BSD tar, it ships GNU tar instead. Now if even a BSD based system ships the GNU version, I very much doubt we'll want to ship the BSD version... ;) [11:31] Chipzz: heh, indeed [11:32] <\sh> oh wow...it's a sunny day... === evalles is now known as effie_jayx [11:34] pitti: also, doesn't dpkg use tar internally? I'd hate to see the breakage that would cause.. [11:45] uh, so new upstream versions don't end up in quarantine [11:46] I uploaded a new wacom-tools and thought it would end up in the queue [11:48] Because we aren't in a freeze [11:49] tjaalton: no, it's a trust system. [11:49] tjaalton: and you tend to get the Long Pointy Stick of DOOM!!!!!!!!!!!!!!!™ coming in your directino if you violate it badly. [11:49] well, trust me, this was needed :) [11:49] the old wacom-tools was broken with xserver 1.4 [11:50] q. before i found a channel with the ubuntu people which take care about all the server/mirror to store the .deb of ubuntu. i dont remember the name tho... anybody got suggestion? [11:50] and the new one didn't build until Tom Jaeger fixed the issues [11:54] <\sh> asac: suddenly I see black holes with ff3 while browsing some pages...when I click on the pictures, they are display correctly...but embedded in the page they are black... [12:04] zorglu_: #ubuntu-mirrors [12:05] pochu: thanks [12:11] \sh: dunnow if it's related, but I've had FF beta4 do funny things for me on MacOSX too [12:11] like only loading some of the images on a page [12:12] I didn't quite get "black holes", but rather the image with the alt tag displayed though [12:13] asac: ^^^ as a data point -> most likely an upstream issue [12:14] \sh: only happened to me after several days of browsing, and was fixed by a browser restart though [12:23] \sh: thats the zoom level [12:23] we have to use in-source jpeg most likely [12:23] (i know that its fixed that way, but i hoped to get the proper jpeg patch cherry-picked from firefox sources) [12:24] <\sh> asac: so the bug is known... [12:25] yes [12:25] its a blocker [12:25] we will revert to in-source jpeg if no solution is found [12:26] <\sh> asac: thx :) I was wondering if something was wrong on my side :) [12:26] \sh: are those jpegs? [12:27] <\sh> asac: jepp [12:27] ok then its known [12:27] <\sh> btw..is there any tool for metacity to configure its own compositing thingy? === \sh is now known as \sh_away [12:45] azeem: ping? [12:45] azeem: got a #debian-type question for you. [12:50] hi all, could anyone tell me if kubuntu has a set of targeted milestone bugs set up like ubuntu does ( https://launchpad.net/ubuntu/+milestone/ubuntu-8.04 )? [12:50] my guess would be no, but i wanted to check. [12:50] j1mc: it doens't - that's the complete list for all flavours (assuming they're all using it) [12:51] Hobbsee: thanks. [12:51] j1mc: see specs assigned to Riddell for the planned kubuntu stuff, though [12:52] Hobbsee: thanks again. a question was asked about this in xubuntu-devel. this is helpful information. [12:53] j1mc: xubuntu is free to add to that list, and discuss things with slangasek and other members of the release team, too [12:54] i will pass along that information. :) [12:55] but please assign the bugs to some of your guys [12:55] as you're responsible for them [12:56] Hobbsee: understood. thanks. [12:56] at least, that's hte way this all worked in gutsy. i'm unsure if slangasek has changed protocol, as i haven't seen many non-canonical people involved in any areas of RM this cycle. [13:07] Hobbsee: pong [13:08] azeem: /query? [13:08] Hobbsee: sure, but I'm probably not around long :-/ [13:09] azeem: no problem, hopefully it won't take long === \sh_away is now known as \sh [14:50] slangasek: i tried today's mythbuntu alternate disk and the udeb wasn't installed, despite it being Priority: standard [14:53] slangasek: do you think you could take another look? [14:58] slangasek: or.. wait. when is the menu entry in d-i supposed to show up? right at the beginning or is it added later, eg after the base system was debootstrapped? === tjaalton is now known as tjaalton_ [16:26] why do my shift and control keys not wokr anymore/ [16:28] * lamont applies the redmond solution [16:29] seb128: why doesn't my shift key love me anymore// ditto for ctl... [16:29] hey lamont, dunno I'm not an xorg guy [16:29] heh [16:29] it's like gnome... ;-0 [16:30] hrm... emoticons screw up too... amazing how much one uses the shift key... [16:30] * lamont hits it with a rock [16:31] hey seb128 [16:33] lamont: I read something about a fix for the "keys stuck" problem in X [16:33] lamont: there was a patch for that [16:33] Chipzz: ah, cool [16:33] lamont: not sure if it got uploaded though [16:33] lamont: maybe that could possibly have the side-effect you're seeing? [16:34] lamont: the fix would be in the xorg-core package, so try downgrading that (you would also need to restart your X-server obviously) [16:34] lamont: lemme see if I can find the url in the backlog ;) [16:35] This "run this action now" dialog... may at least be putting power in my hands... but it's way too technical. [16:36] Package authentication failure of some sort. [16:36] lamont: https://launchpad.net/bugs/194214 [16:36] Launchpad bug 194214 in xorg-server "Keys get "stuck" down" [Unknown,Confirmed] [16:37] lamont: hrrrm no mention of the patch being applied and uploaded yet; so that's probably not it [16:38] seb128: sometime, I should find out about the rising dominance of gnome-keyring-manager (or whatever the hell it is) and seahorse, and figure out what to add or remove from my sledgehammer-of-a-gnome-session [16:39] maybe it's finally usable out of the box?? [16:39] * lamont wanders off for a while [16:40] lamont: the gnome-keyring new changes are cool ;-) [18:46] laga: ok, trying to take a look; it's possible that I have to set the priority to 'Standard' in the archive overrides rather than relying on what the package thinks, for that change I should probably check with cjwatson and pitti first [18:53] laga: right, it looks like the CD build takes the package priorities from the archive overrides, rather than looking inside the individual packages. In that case, making it Priority: standard for the installer would require making it visible as Priority: standard for other install methods too, so that'll need more discussion; sorry for putting you to the effort [19:33] Riddell: ping [19:33] Riddell: does kde use shared-mime-data and associate programs via desktop files for mimetypes they claim to open? [19:48] tjaalton: ugh, 195953 should have gone through an FFe... [19:48] tjaalton: are you tracking this package then, to make sure there are no regressions? [19:50] hi [19:50] can someone look at this bug? https://bugs.launchpad.net/ubuntu/+source/usplash/+bug/188764 [20:01] slangasek: so are you going to talk to cjwatson/pitti or should i bug them myself? [20:04] laga: I'll talk to them (on Monday, I expect - I don't think either of them are around this weekend) [20:04] slangasek: thanks, that's much appreciated [20:46] does the livecd environment currently warn the user when he doesn't seem to have enough ram? [20:48] hm, I'll better ask in -installer [20:49] slangasek: yeah.. sorry about it :/ [20:50] slangasek: I'll monitor any issues with it, and actually the upload was broken so I'll fix it asap [20:52] tjaalton: ok :) [20:55] huh, this is strange.. it included wacom_drv.o, when it should have built .so [20:55] but building it locally generates only .so [20:55] heh [20:55] maybe it's the default CFLAGS etc biting me again? [20:56] that would be strange if so [20:56] a strange misfeature of the upstream build rules [20:56] yep [21:02] found the bug.. now figuring out how to fix it [21:03] fails to find the module dir [21:21] hmmh, is 'elif -z "$FOO"' bashism? the wacom configure script seems to fail with '-z: command not found' [21:24] Probably missing 'test' or [ ]. [21:24] yep, test should do it, since all the other tests have it [21:25] quality [21:26] I don't think that's even a bashism, I think it's invalid shell [21:26] right, bash fails too [21:26] seems that no-one is using latest wacom :) [21:27] or at least not reporting these upstream [21:31] well that's encouraging [21:34] hehe [21:34] anyway, a fix uploaded [21:35] the release is a month old, so the actual code should be quite solid :) [21:36] how do we know that, if no one's actually managing to build it successfully? [21:38] slangasek: maybe they were built by hand?-) anyway, upstream is adamant about this being much better for xserver 1.4.. [21:38] better release [21:47] <\sh> keescook / bryce: are you pushing inkscape 0.46 to hardy? editing pdfs sounds awesome :) === \sh is now known as \sh_away === mjg59` is now known as mjg59 [22:00] I'm trying to track down a bug regarding Cyrillic input, but don't have a cyrillic keyboard...what's a good way to emulate that? [22:01] setxkbmap ru? [22:01] ion_: the trouble is I don't really know what buttons to push, heh [22:02] The GNOME keyboard layout setting thing shows the mapping when you select it. [22:05] Fujitsu: ahh ok. I should learn to look in obvious places, thanks. === juliank0 is now known as juliank [22:52] Keybuk | Mithrandir | doko: please give back haddock and hscolour on lpia, thanks [23:01] sistpoty: can't do that: it's MANUALDEPWAIT [23:01] doko: what does that mean? [23:03] sistpoty: It'll automatically be retried when the dependency is available. [23:03] Fujitsu: ah, thanks [23:04] MANUALDEPWAIT isn't manual at all. Isn't it great? [23:05] oh... haddock is already built... cool, but I don't see manualdepwait for hscolour on lpia? [23:05] Just for the record I'd like it noted that I'm self-censoring right now. [23:07] Fujitsu: any clues about hscolour, and where I could see it being manualdepwait? [23:07] sistpoty: Looking. [23:08] thanks [23:09] doko: hppa and lpia failed to build; they're not MANUALDEPWAIT. [23:10] doko: ok, then please give back hscolour on lpia (hppa is no use to give back, we don't have ghc6 there) [23:10] Fujitsu: according to https://edge.launchpad.net/ubuntu/+source/haddock/0.8-2 both are "Dependency wait" [23:11] doko: Ah, I speak of hscolour. [23:11] doko: that's haddock, and it's for gutsy (which I misread in the first place as well *g*) [23:11] (and for hppa...) [23:12] sistpoty: done [23:12] doko: thanks! [23:22] Is there a reason why /usr/share/X11/locale/locale.dir lists ru_RU.UTF-8/XLC_LOCALE ru_RU.UTF-8 rather than en_US.UTF-8/XLC_LOCALE ru_RU.UTF-8 ? [23:26] because of KOI8, according to the diff. [23:28] \sh: yawp, been done: inkscape | 0.46-0ubuntu1 | hardy/main [23:28] slangasek: Which package is that? It may be causing a rather old bug in Wine ~ cyrillic input: https://bugs.edge.launchpad.net/wine/+bug/68594 [23:28] Launchpad bug 68594 in wine "No cyrillic input in apps under wine. " [Undecided,New] [23:32] YokoZar: hrm? I'm talking about diff -u /usr/share/X11/locale/{en_US,ru_RU}.UTF-8/XLC_LOCALE [23:33] I think you're unlikely to make a persuasive case that a wine bug should be fixed by *not* using the Russian KOI8 map [23:34] and I'm not aware that XLC_LOCALE has anything to do with input anyway? [23:35] slangasek: I'm not sure why making the change fixed the problem earlier either. Now, it looks like I can type Cyrillic characters in Wine's notepad (which is a UTF-8 app), but I don't have an ANSI app to test this on. Wine's free fonts just started supporting Cyrillic letters too. [23:37] if the problem was display of Cyrillic characters, then XLC_LOCALE would have some bearing [23:38] slangasek: that makes sense. I'll close the bug