=== bjf is now known as bjf[afk] === bjf[afk] is now known as bjf === bjf is now known as bjf[afk] [02:46] There is no way you can assign a shortcut key to an application in its desktop file correct? [02:48] I don't believe so, no, although I'm not sure where System→Preferences→Keyboard Shortcuts gets its listings from. [02:49] keynames come from /usr/share/gnome-control-center/keybindings/*.xml [02:49] the shortcuts usually get stored in gconf [02:51] I know about the gconf shortcuts, but was wondering whether apps could be given a shortcut in their .desktop file, similar to what is possible in Windows. [02:52] no [02:52] Ok, thanks. [03:37] kenvandine: if anyone wants to try out the new adium theme in maverick for lucid my PPA has it backported [03:37] bcurtiswx3, i saw that, thx! [03:38] kenvandine: cool, yw [04:01] I don't suppose anyone would be willing to look at an xterm merge for sponsorship? http://sarvatt.com/downloads/merges/xterm/ [04:04] Sarvatt: I'll take a look. [04:05] robert_ancell: hey [04:06] TheMuso: thank you a ton for the help [04:07] Sarvatt: np [04:09] after that xterm and nvidia-graphics-drivers* rebuilds are all thats left for X stuff [04:09] err xdm [04:17] Sarvatt: uploaded. [04:17] thanks! [04:17] np [04:54] anyone here familiar with ureadahead's code? [04:55] i'd like to implement btrfs support for ureadahead, since i'm sick of waiting and nothing's been done === bjf[afk] is now known as bjf [06:16] desrt, hey [07:31] Good morning === almaisan-away is now known as al-maisan === al-maisan is now known as almaisan-away [08:19] good morning [08:26] hey didrocks [08:27] Guten Morgen pitti [08:28] hum 712M for the netbook CD :/ [08:28] well maverick desktop amd64 is approximately the same size [08:28] (not the i386 one, weird) [08:35] ugh, what was added since a1? [08:38] pitti: don't know for desktop. For netbook it's all the removal of maximus/netbook-launcher in favor of mutter/unity [08:38] pitti: but I think this huge delta should be due to something else [08:39] hum, I can remove metacity already [08:50] morning === Amaranth_ is now known as Amaranth [09:05] didrocks, urgh [09:05] didrocks, you completely broke armel netbook [09:06] ogra: with yesterday's update unity should build now on armel [09:06] ogra: I didn't check the status, but I'll refresh the ubuntu-netbook metapackage before the week-end [09:07] I'll add also the -efl, as I guess it was on the CD as a dep of netbook-launcher [09:08] didrocks, nice work with unity [09:08] seb128: thanks :) [09:09] all the transition thing is working well for you? [09:09] yes [09:09] great! [09:10] didrocks, I get no wm decorations though, wanted or gtk csd bug? [09:10] not sure how to close dialogs out of the using the menus which is suboptimal [09:11] seb128: it's wanted, I should check again the gconf key (I think it's the same than maximus used) [09:11] weird [09:11] but yeah, until the titlebar isn't in the panel, not useful, I should change temporary the default [09:11] thanks for the notice, I changed it a while ago there. Anyway, trying on the stock install for lucid [09:12] np, good work again, it upgraded in a flawless way [09:20] bonojur seb128 [09:20] hey pitti [09:20] how are you? [09:21] I'm good, thanks! still catching up with peer reviews [09:22] didrocks, would you mind if i re-added all the netbook bits for armel you just removed for all arches ? [09:22] pitti, oh, I need to do that as well [09:22] meh, the desktop amd64 CD exploded as well [09:22] we dont have any desktop anymore with your last change to the seed [09:23] desktops are overrated [09:23] yeah, arm users should just phone with their HW :) [09:23] what else is it good for [09:23] ogra_: regenerating the metapackage should do it (at least, adding unity to armel) with yesterday's evening build. Then, you have to add explicitely as a recommend netbook-launcher-efl, right [09:24] ogra_: before, it was a dep of netbook-launcher which is removed as just a transitional package now [09:24] will there be an -efl like 2D port of unity? [09:24] didrocks, i'm currently not caring for unity but for having the old 2D desktop working (maximus, go-home-applet, window-picker etc) [09:24] pitti, there will be a panel thats a gtk container for indicators [09:25] but we'll keep the efl launcher in maintenance mode as long as we can for arm [09:25] *nod* [09:25] ogra_: sure, I can do if it you want, add them for armel [09:25] but we could certainly run n-l-efl with upanel [09:26] pitti, not sure how well that works, i havent tried upanel yet [09:26] gnome-panel is definately to heavyweight though which brought up the minimal panel idea at UDS [09:26] afaik ted is working on it already [09:27] isn't that by and large the panel that we see in unity already? [09:27] i. e. which is just an indicator container? [09:27] pitti, afaik that still uses a lot of gnome libs [09:27] ah, I see [09:27] we want to cut down memory footprint and startup speed [09:28] * ogra_ has to move rooms at LT now ... see you [09:28] right, just gtk, dbus, and libappindicator would be great [09:28] yepp [09:30] pitti, ogra_: the issue with unity is not the depends, it's the graphical 3d requirement [09:31] seb128: but that would only go for the launcher, I suppose, not for upanel itself? [09:31] ? [09:31] I mean, upanel certainly doesn't need 3D, does it? [09:31] no, I meant unity does [09:31] both the panel and the launcher [09:32] that's why they need a stock gtk version [09:32] urgh, seems like I was disconnected… [09:32] didrocks, no you are not [09:32] didrocks, you didn't quit the channel at least [09:32] did you received? 10:26:40 didrocks | pitti: go-home-applet and window-picker should be ported to it so [09:32] 10:27:34 didrocks | and upanel is rendered under mutter, not sure how this behave under a 2D driver [09:33] no [09:33] (and then, I recevied your last sentend "they need a stock gtk version" [09:33] so, my server was still connected, but not my laptop [09:33] and as it didn't say "heh, I quit", I don't have the backlog [09:34] didrocks, you didn't miss anything, ogra just said he had to move rooms [09:34] seb128: oh ok, he didn't answered if I can add them for armel only now? [09:34] in any case, we have to work on the fallback system again [09:35] didrocks, when did you drop from the channel? [09:36] seb128: it's ok, I'm reading the log from my bip server now [09:36] ok [09:36] I was going to copy you the log [09:37] seb128: that's fine, thanks :) [09:38] yeah, I confirm the upanel can't work without mutter, that's why it can't be used for -efl [09:39] and window-pickler/go-home-applet are applets, so, don't work (they will need to be rewriten too for the stock gtk bar) [10:28] * Added unity to netbook [armel] [10:28] didrocks: ^ ?? [10:29] pitti: yeah, some armel card support 3D driver (as we had netbook-launcher working on them) [10:29] pitti: so, having a fallback plan, as with netbook-launcher is in range, but won't happen before alpha3 [10:30] yesterday, unity wasn't on armel as the version FTBFS [10:31] didrocks: ah, good to know; it just didn't look deliberate after the discussion from above [10:31] * pitti hugs didrocks [10:31] it was deliberate :) [10:31] * didrocks hugs pitti [10:32] but I would prefer a better fallback solution than the one I added to netbook-launcher, we discussed a little about it with asac [10:32] we'll figure out [11:08] where was the Mozilla maintenance change announced? I'm looking for a URL to link to [11:09] ah, found it: https://lists.ubuntu.com/archives/ubuntu-devel-announce/2010-June/000719.html [11:09] seb128: support != maintenance :-) [11:10] mdz, I copied the blueprint title ;-) [11:10] but right [12:00] seb128: hm, something confuses the hell out of me [12:00] Package: gvfs [12:00] Depends: [...] libgvfscommon0 (>= 1.1.7) [12:00] but libgvfscommon0 is 1.6.1-0ubuntu1build1 [12:00] (on lucid, anyway) [12:00] just got a report about a broken upgrade due to that [12:00] but how did that _ever_ work and doesn't cause complete uninstallability?!? [12:01] argh, ignore me [12:01] 1.7.1 != 1.1.7 [12:01] heh ;) [12:01] i was just thinking that [12:01] so why was it removed for that guy.. [12:01] * pitti asks [12:01] *brown paperbag*, sorry for the onise === MacSlow is now known as MacSlow|lunch [12:07] hi [12:07] is there anything wrong in this uscan line: [12:07] http://download.gnome.org/sources/tomboy/([0-9.]+)/tomboy-([0-9.]+)\.tar\.gz [12:07] bzr bd... doesn't seem to find the 1.2.1 tarball, which is there [12:10] oh, I get another error before that -> dh: unable to load addon quilt: Can't locate Debian/Debhelper/Sequence/quilt.pm [12:10] which package does this file come from? [12:12] rodrigo_: [12:12] pkg -S /usr/share/perl5/Debian/Debhelper/Sequence/quilt.pm [12:12] quilt: /usr/share/perl5/Debian/Debhelper/Sequence/quilt.pm [12:13] dpkg -S /usr/share/perl5/Debian/Debhelper/Sequence/quilt.pm [12:13] ;) [12:13] hmm, I have it installed, it seems [12:13] ah, no, I don't :) [12:13] lol [12:14] but it still can't find the tarball :( [12:16] didrocks: Hi! Did you get a chance to look at my experimental clutter egl packaging? [12:17] alf__: that's next on my list :) [12:17] didrocks: great, thanks :) [12:18] you're welcome [12:18] am i here? ... damn connection - this week is clearly bad [12:18] asac: you are [12:18] ah ;) [12:18] there i am [12:18] asac - i can't see you ;) [12:18] pitti, sorry I was at lunch, seems you sorted it by now ;-) [12:18] chrisccoulson: i think thats a good thing ;) [12:19] i try to not look in a mirror right now ;) [12:19] lol [12:19] rodrigo_, weird [12:19] hey chrisccoulson, asac, glatzor [12:20] seb128, I guess it's getting to an out-of-date mirror? [12:20] hey seb128 - how are you? [12:20] hi seb128 [12:20] rodrigo_, you get no clue in the log? [12:20] chrisccoulson, I'm fine thanks, what about you? [12:21] yeah, good thanks. i'm a bit sleepy today though [12:21] alf__: so you said that the es11 and es22 lib is supposed to be binary compatible? [12:21] worked late again? [12:21] if igts not we should ship it with different soname [12:21] otherwise i dont see why we need to have different test packages for them [12:21] i worked fairly late last night, fixing the few remaining issues with the hardy updates [12:21] seb128, no, just that it can't get the tarball from that url [12:22] rodrigo_, is your debian dir somewhere online? [12:22] chrisccoulson, oh, ok, did you fix the font rendering one as well? [12:22] seb128, no, it's just lp:ubuntu-desktop/tomboyubuntu branched, and added a patch to debian/patches [12:22] seb128 - yeah, that's fixed too [12:23] seb128, I can push it if you want [12:23] rodrigo_, would be useful yes [12:23] ok [12:23] chrisccoulson, well done! [12:23] thanks :) [12:24] asac: Yes they are supposed to be compatible, so yes we could just have one eglx test package [12:24] alf__: right. imo we should have that. [12:24] asac: I am 99.9% sure of this, that is why I have two different test packages now [12:25] alf__: you can compare symbols files [12:25] seb128, https://code.launchpad.net/~rodrigo-moya/ubuntu/maverick/tomboy/proxy-settings [12:25] for the es11 and es22 ... are those identical? [12:25] asac: right, will do [12:25] alf__: if we could get rid of it that would be great. also i think the -tests package shouldnt start with lib... [12:25] we usually only use lib... prefix for libs [12:25] mutter will be usable with no hardware acceleration in MM at least :) [12:26] asac: ok [12:26] alf__: so we could just ship "cutter-eglx-tests" [12:26] or i guess [12:27] clutter-1.0-eglx-tests [12:27] and clutter-1.0-common-tests [12:27] and clutter-1.0-glx-tests [12:27] asac: or clutter-glx-1.0-tests, clutter-eglx-1.0-tests? [12:27] for me the 1.0 refers to the clutter-1.0 source package [12:28] but not sure [12:28] get more opinions ;) ... from didrocks and seb128 and others [12:28] asac: mesa is waiting to go through NEW in debian, I patched it up to work with libdrm 2.4.21 we'll have at the same time so it shouldn't be long. in the process of redoing the mesa packaging so things work when we switch to 7.9 in a few weeks that is very different for the egl/gles stuff [12:28] Sarvatt: cool. did you just keep intel1 enabled for Archi: any ? [12:29] the package raof upped had a problem there ... imo intel1 should just be any [12:29] yeah until someone fixes plymouth :( [12:29] thats fine. i dont see why just intel1 should be disabled on non intel archs ;) [12:29] it makes no sense to be any but it needs to be any because plymouth needs it on all arches.. [12:29] because its just wasting space on non intel arches :) [12:29] so it makes sense ;) [12:30] yeah well. but afaik we also ship X drivers for hardware that will probably never exist for arm or sparc deployments ;) ... so its in line [12:31] rodrigo_, I think it looks for the newest version and find only the 1.3 serie [12:31] they aren't always going to be installed actually wasting space though unlike libdrm-intel1 :) [12:31] seb128, oh, why? it should be looking in the 1.2 directory, right? [12:31] rodrigo_, dunno [12:31] rodrigo_, you don't specify a serie in the watch it uses the newest [12:32] rodrigo_, get the lucid tarball? [12:32] swrast in mesa 7.9 can run compiz and mutter though so that'll all be usable on arm [12:32] rodrigo_, if you have a lucid-update deb-src it will just work [12:32] seb128, well, this is for maverick, does maverick have the 1.3 version already? [12:32] no, 1.2.1 [12:33] rodrigo_, no, but it might be a bug in the download code which is fixed in maverick and not lucid [12:33] seb128, I'm on maverick [12:33] oh [12:34] well dunno [12:34] seb128, can I just download the tarball and place it somewhere so that it doesn't try to get it? [12:34] it's a bug [12:34] you can [12:34] if you have a maverick deb-src it will just work [12:34] either make the watch point to the serie you use [12:35] or download the tarball yourself [12:35] or configure an ubuntu deb-src so it uses it [12:36] so, place it in the parent dir? [12:36] yes [12:37] wget http://archive.ubuntu.com/ubuntu/pool/main/t/tomboy/tomboy_1.2.1.orig.tar.gz [12:39] ok, that works, thanks! [12:39] asac: you already have something to work on with asac's comment, don't you? I would just say in addition to this: don't change the packaging too much before ensuring debian will take it (like dh5 -> dh7, remove of debian/control.in and using debian quilt format 3) [12:39] upss alf__ ^^ [12:40] didrocks: well. [12:40] didrocks: current package i cdbs [12:40] so if it moves somewhere it should move to dh7 [12:40] directly [12:40] asac: ok, is there a discussion with the DD responsible for that package? or do we want to diverge forever? [12:41] seb128: who was the guy with the gtk+3.0 package again? did you take a look at his package already? [12:41] didrocks: our goal should be to get the same features packaged in debian so we can sync at some point. we should start that discussion. if they want to go for cdbs we can work with them on that still and drop our packaging if it comes to it [12:42] seb128, well, that patch does not apply to 1.2 :( [12:42] if they want to adapt our packaging, even better ... [12:42] seb128, can I upgrade to 1.3? [12:42] having a patch like this maintained as a downstream diff is painful in anyway; so either they adapt our features or we will be alone imo [12:42] seb128: this here? https://launchpad.net/ubuntu/+source/gtk+3.0 [12:42] asac: agreed, just to ensure that the discussion has already began to have an easier adoption :) [12:42] rodrigo_, no [12:43] seb128, ok [12:43] didrocks: hoped you opt in to that ;) [12:43] slomo, right, the ppa there, ricotz [12:43] seb128: thanks [12:43] slomo, he's on #ubuntu-devel [12:43] seb128, are you waiting for something before updating the 2.31/ tomboy 1.3? [12:43] didrocks: by ensuring you are happy with the direction it might be easier to our stuff upstreamed [12:43] e.g. we are one time contributors, while you maintain a constant relationship ;) [12:43] (hopefully) [12:44] rodrigo_, we don't plan to update to 2.31 this cycle [12:44] anyway, if not, we will ensure htis comunication happens [12:44] seb128, ah, why? [12:44] asac: great :) [12:44] rodrigo_, we don't want gtk3 in the default installation, not reliable schedule, and the transition will take over a cycle [12:45] alf__: can you do asac's changes and then ping me back? I will do a closer review [12:45] cool [12:45] seb128, hmm, ok, we'll have to backport the patch to 1.2 then [12:45] rodrigo_, it conflicts with gtk2 at runtime so we would need to duplicate lot of sources to have gtk2 and gtk3 builds [12:45] right [12:45] rodrigo_, which is an issue for cds and a non trivial transitions [12:46] rodrigo_, we plan it for next cycle, gtk3 will go to universe this cycle [12:46] seb128, ok, good to know, we'll backport the patch then [12:46] rodrigo_, if the new tomboy will not depends on gtk3 we can update though [12:46] seb128, hmm, don't know, I'll check [12:46] rodrigo_, you might want to try to figure what upstream wants to do [12:46] yes [12:46] thanks [12:47] slomo, I didn't check his work no [12:54] asac, didrocks: It seems that clutter es11 an es20 are not compatible after all, at least symbol-wise... unless the differing symbols are internal and exposed by error. I'll check with upstream... [12:55] alf__: what are those symbols? [12:55] cogl_ stuff? [12:56] asac: yes [12:56] asac: eg cogl_fixed_vertex_shader_* [12:58] thats coming from cogl/driver ... with some luck thats just internal and can be hidden [12:59] alf__: so check with upstream [13:01] alf__: you could try to add #pragma GCC visibility push(hidden) [13:01] to the start [13:01] and [13:01] #pragma GCC visibility pop [13:01] at the bottom of all .h files in cogl/cogl/driver/*/*.h [13:01] well start == after all other includes [13:01] asac: sure, I am waiting for upstream to confirm [13:02] right. but if tests etc. stil build fine with that it should be ok [13:02] (and if the symbols are really gone ;)) [13:08] didrocks, Hi! the latest upload of ubuntu-netbook-unity-default-settings in UNE ppa removed Unity session from GDM list, is that expected? [13:08] om26er: hum, no, it's not [13:09] apt-cache policy ubuntu-netbook-unity-default-settings && dpkg -L ubuntu-netbook-unity-default-settings [13:09] I've tested the migration on my netbook and it worked [13:10] hmm, I actually downgraded and then it worked === MacSlow|lunch is now known as MacSlow [13:10] om26er: can you try upgrade to latest version and dpkg -L, please? [13:10] I confirm there that the session is installed [13:11] asac: although the correct way for a library would be to for everything hidden by default and push(default) explicitly for public symbols... [13:11] asac: but such is life :) [13:11] didrocks, sure brb === gnomefreak76 is now known as gnomefreak [13:13] didrocks, upgraded and its still there. [13:14] mean, now unity session works [13:14] om26er: hum, that's weird, unfortunately, doesn't seem it's reproduceable so :) [13:14] seems we are all good so :) [13:14] s#:)#:/ [13:15] didrocks, great, Thanks :) [13:15] om26er: you're welcome [13:16] om26er: and thanks for testing unity and your bug triaging work on UNE btw :) [13:18] alf__: true. [13:25] didrocks: so how do we do mutter with gles? [13:25] e.g. how can i test/try that? [13:25] guess i need to spin that aganst the egl clutter ... but what do i do with that then? [13:25] asac: I would think, take the package in maverick, link against gles in a ppa, and launch it with mutter --replace [13:26] didrocks: you mean i could run that in my -efl sessioN? [13:26] what would happen :) [13:26] asac: with a device that doesn't support glx, right? [13:27] yes [13:27] so, yes, you can just run that in you -efl session [13:27] it will replace metacity by mutter [13:27] didrocks: so we need mutter + unity built against gles/egl i guess? [13:27] you can see it really running as it have some little effects [13:27] is there anything else we need to prepare to test the full stack? [13:28] asac: right, but I think beginning with mutter only is the first step to see how it behaves [13:28] no, mutter and unity is everything you need [13:28] didrocks: sure. however, plan is to get the full stack packaged and then let vendors and upstream developers collaborate on fixing things [13:28] all the remaining things are just library without any rendering [13:28] just want to undersatnd how much work this will be on packaging side [13:28] but seems we are quite close with just unity and mutter missing [13:29] asac: of course, you will need the other librairies, but nothing special for egl [13:29] didrocks, its reproducable on the live cd(lucid) and also on new install( where I first got the problem). installing the previous version of *-default-settings and then upgrading works [13:29] asac: you can depend on the lucid UNE ppa [13:29] sure. but those are hopefully avaiable [13:29] didrocks: well. i need to copy this stuff to the native ppa i guess ... or is lucid UNE with arm ? [13:29] but i will check ;) [13:30] didrocks: do you have url for that lucid UNE ppa at hand? [13:30] om26er: ubuntu-netbook-unity-default-settings is the package having all the dep, this one have to be installed [13:30] asac: https://launchpad.net/~canonical-dx-team/+archive/une/ [13:30] thx [13:31] hmm. what is plasma-* stuff doing there ;) [13:31] om26er: the instruction to install unity is now: 1 - add the UNE ppa, 2- install ubuntu-netbook-unity-default-settings [13:31] is that ppa state to be considered stable? [13:31] asac: yeah, dx team plays with the ppa too [13:31] asac: right, for unity stuff at least [13:31] e.g. could it be that it doesnt work at all? [13:31] I upgrade it once a week [13:32] no, it's working, I'm trying it [13:32] didrocks: do you know out of your head what stuff i need from there that doesnt do rendering? [13:32] asac: sure, one sec [13:32] coool thats helpful ;) [13:33] asac: bamf, dee, libunity-misc, ubuntu-netbook-unity-default-settings, unity-asset-pool [13:33] asac: then, you will need for rendering: clutk, mutter and unity [13:36] didrocks: right. let me copy all those for now and then we improve the rendering bits [13:37] asac: ok, so you will stay on that version for now, right? [13:37] (it's quite fresh, from yesterday's evening :)) [13:37] if there are updates coming soon we should wait for i can do that [13:37] but in the end i hope it doesnt make much of a difference; we can rebase the packaging changes required [13:38] didrocks: so the gtk menuproxy thing isnt needed? [13:38] asac: no, this new version is only if you want your menu in the panel [13:39] as indicator-datetime which is installed by default in UNE, I don't think you want to bother with that for now [13:39] didrocks, on today's(20100611) UNE daily build there is only gnome session which might make it the same bug? [13:40] ok lots of things spinning now: https://edge.launchpad.net/~asac/+archive/armel1/+packages [13:40] stay tuned [13:41] om26er: I don't understand, did you try what I told you on a fresh install? 1 - add the ppa ; 2 - install ubuntu-netbook-unity-default-settings? that's how unity should be installed [13:41] asac: and be afraid? ;) [13:41] yeah!!! [13:42] didrocks, yes exactly [13:42] om26er: and then, you don't get the unity session? [13:42] didrocks, no [13:43] hum, let me try [13:44] oh,, installing unity dont install default-settings? [13:45] om26er: that's what I'm telling you from 15 minutes [13:45] :) [13:45] for* [13:45] oh [13:46] in maverick, ubuntu-netbook is the metapackage installing unity and the default-settings (as before, it was installing netbook-launcher and the default-settings) [13:46] of course, we don't have a metapackage in the ppa [13:46] so, it's the default-settings which contains the session which also have those deps [13:47] * didrocks setup a wiki page [14:02] hm, with yesterday's -proposed updates the keyboard indicator now has a different font and looks hideously blurry [14:02] https://wiki.ubuntu.com/DesktopTeam/Unity [14:11] pitti, right [14:11] pitti, it used to use cairo directy and is now pango rather... [14:13] hm, why was that done in an SRU? appending an "1" for previously identical layouts doesn't seem to need a change in the font rendering, just in the displayed strings? [14:13] no [14:13] but he used cairo directly because he didn't know how to do otherwise before [14:14] ah, so it's an independent change [14:14] but cairo has limitations [14:14] it doesn't respect some of the pango rendering and font settings [14:14] and there was crashers in the code [14:14] didrocks: mutter seems to not like us using clutter 1.3.x ;) http://launchpadlibrarian.net/50131541/buildlog_ubuntu-lucid-armel.mutter_2.29.0-1ubuntu1dx3%2Br3435_FAILEDTOBUILD.txt.gz [14:14] didrocks: guess i have to hack that out [14:14]