[00:41] hggdh, cyphermox yeah think we could scrounge up a Vietnamese reader to test? [00:41] (and/or educate; I was going by assuming the bug reporter knew what he was talking about, but would be helpful to independently verify) === mfisch` is now known as mfisch === mfisch is now known as Guest29044 [01:33] what sets _NET_WM_DESKTOP_FILE on an applications window? [01:53] robert_ancell: You here and not sick today? [01:53] RAOF, yep [01:53] Hurray! [01:54] Hope you're well recovered. [01:54] yeah, it was one of those 24hr things so wasn't too bad in the end [01:55] I think I'm in a position to hand off something you can use for developing the lightdm end of system-compositor integration. [01:55] cool [01:56] With the *tiny* problem that it's not trivially PPAable. [01:56] Because it's based on xserver 1.12, which will require a big song and dance to PPA. [01:56] But! It's very easy and safe to install locally alongside the existing X server. [01:58] RAOF, aren't we getting 1.12 for quantal anyway? [01:58] We are. [01:58] Just not yet. [01:58] approx how long? [01:58] (Or possibly 1.13, depending) [01:59] I guess I could actually do it now; we just had a preference for leaving the current stack so quantal testing is also closer to precise testing for a while. [02:00] And by “now” I mean “post A1”, obviously. [02:01] But if you wanted to start developing now it'd be reasonably easy for you to do so. [02:01] ok, send me the details! [02:03] First you need to clone a couple of git repositories: wayland, libxkbcommon, and $YOUR_GRAPHICS_DRIVER. [02:03] What is your graphics driver, by the way? [02:05] ... [02:05] You know, someone's scripted this. I'll just hunt that down! [02:08] robert_ancell: http://www.chaosreigns.com/wayland/buildscript/ (you're after the "master" buildscript) will get everything interesting bulit and installed; you'll then need to just pull a couple of my branches. [02:08] On inspection, I think that perhaps my “It'll totally be easy for you to start hacking” might be biased by my current setup which already has all the things you'd need to start hacking ☺ [02:10] If you're *not* feeling like building on the order of 6 things from git and installing them locally, it's not going to be trivial for you to start hacking :) [02:11] You can see what the lightdm integration would need to look like here, though: https://github.com/RAOF/weston/blob/system-compositor/clients/simple-display-manager.c [02:14] robert_ancell: Interesting bits are global_handler, where it binds to the wl_display_manager interface, and add_system_client where it spawns an X server. [02:14] aight [02:15] Or, in fact, all of it, because there's not a whole lot to it ;) [04:05] /aw/c [04:16] tremolux, ping [04:57] Hey all [04:58] I've been toying around with a laptop given to me by a friend that found it's early demise at the hands of a wall [04:58] Anger management issues [04:58] :P [04:58] In any case, the only thing busted is the screen, so I'm running it on an external display, installed Ubuntu from USB just fine [04:59] Even though there's no POST or visible BIOS on the external display, booting from USB worked with the display and the first run after installing Ubuntu also worked with the display [05:00] Now here's the catch and the reason I ended up here - after installing Ubuntu - I went on to updating the whole shabang through the update manager [05:01] 80% in - it failed for some reason - I was not really concerned (even though I should've been) so I just took my losses, reported, closed Update Manager and went on to reboot [05:01] And from there on out - no more external display [05:02] FN-Fkey combo doesn't bring it up - spamming possible BIOS buttons turns up blank - booting from USB gives me jack - tried an optical disc, both Windows and Ubuntu opticals [05:02] Nothing [05:02] Nada [05:02] Niente [05:03] Not sure what exact type the laptop is, it's a Compaq CQ62-series, that's about all the info I could gather from the product [05:03] The friend who gave it to me is about as computer savvy as a potato [05:04] So he doesn't know/care [05:04] Google is about as concise and helpful as AskJeeves on the subject [05:05] So I'm kinda at a loss... does anyone have any idea whatsoever? Is there a known keystroke combo I'm forgetting? [05:06] And how does the external display suddenly decide to stop functioning? I've swapped around multiple displays - the only one I haven't tried is jacking it up to my TV through HDMI [05:07] Any help whatsoever will be greatly appreciated - it's a decent piece of hardware and getting it to run properly on Ubuntu would make me such a huge happy camper [05:09] Whoops! relogged by accident XD [05:09] Anybody? [05:29] Well, swapped out harddrives [05:29] reinstalling ubuntu [05:51] GekiritZ might want to try #ubuntu, this channel is for ubuntu development. [05:52] Thanks Jason! Swapping asap [06:11] good morning === tkamppeter_ is now known as tkamppeter [07:58] hey [08:08] salut seb128 [08:08] lut didrocks [08:08] didrocks, cava? [08:08] seb128: nickel, et toi? [08:08] didrocks, nickel aussi ;-) [08:51] chrisccoulson pingus [08:52] ff seems to broken [08:52] youtube aint loading [08:52] Build identifier: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:14.0) Gecko/20100101 Firefox/14.0 [08:52] search ( Ctrl+K) is also broken [08:59] nm, turns out https everywhere is screwing this up [09:24] good morning everyone [09:28] hey chrisccoulson, how are you? [09:28] seb128, yeah, good thanks. how are you? [09:28] chrisccoulson, I'm good thanks [09:33] good morning chrisccoulson :) [09:35] hi didrocks. how are you? === rickspencer3 is now known as rickspencer3_ [10:09] seb128, haha, found the issue with bug #999827 [10:09] Launchpad bug 999827 in indicator-appmenu "Nautilus missing Preferences (incorrect handling of appmenu and menu used together)" [High,Confirmed] https://launchpad.net/bugs/999827 [10:09] larsu, oh? [10:09] larsu, is it nautilus doing weird stuff? :p [10:10] no, the ubuntu_menuproxy plugin seems to override stuff. It's really weird. You get the app menu if you run with `UBUNTU_MENUPROXY= src/nautilus` [10:10] but of course, the window menus will be back in the window in that case [10:11] seb128, are there other gtk apps for which this is working? [10:11] larsu, that being the "shell menu"? [10:11] larsu, or mixing both? [10:11] seb128, the shell menu [10:12] larsu, I don't know of any application mixing both menus, but ted said bloatpad was working [10:12] larsu, yes, you can try quadrapassel [10:12] larsu: or tetravex [10:13] seb128, but that doesn't have any window menus [10:13] larsu, no, that's why I asked if you wanted an app using a shell menu or one mixing both :p [10:13] seb128, oh sorry, mixing both please :) [10:13] larsu, I don't think we had one mixing both, out of bloatpad that mentioned as the one is used for testing [10:14] larsu, gtk-3-examples: /usr/share/doc/gtk-3-examples/examples/bloatpad.c.gz [10:15] seb128, okay thanks. I'll fix it after lunch - just need to find a way to turn off dbusmenu (for some weird reason the menuproxy plugin gets triggered even though nautilus is using GMenu) [10:15] * larsu --> lunch [10:17] larsu, enjoy lunch === rickspencer3_ is now known as rickspencer3 [12:06] seb128, so I was only semi-right before: dbusmenu *does* pick up the window menu of nautilus windows (which do not contain the shell menu) [12:06] but not by error, but because nautilus is only half-ported to gmenu [12:06] i.e. it exports an application menu, but its windows still use a gtk menu [12:06] larsu: oh, they still use old menus and gmenus shell menu next to it? [12:07] yes [12:07] ok, I see, I guess full gmenus would work [12:07] dbusmenu picks up the gtkmenu in the window and gives that to indicator-appmenu [12:07] which is what bloatpad is using [12:07] and indicator-appmenu doesn't care about the gmenu, because it already has a dbusmenu [12:07] ok [12:07] seb128, bloatpad is using gmenu for both app and window menus [12:07] which is why it works [12:08] so the mixed case has not been considered from the appmenu side [12:08] apparently [12:08] hum, k [12:08] I wonder if that's worth supporting, i.e how transient that situation is [12:08] possible fixes: (1) fix nautilus to use gmenu on its windows, too (which seems like a lot of work [12:08] (2) fix indicator-appmenu to support this case [12:09] in either case, I'd like to run this by ted or desrt first [12:09] I would vote (2) since it's likely other apps will do that and we can't assure they will be all "fixed" by release time [12:09] yeah, sure [12:09] larsu: thanks for looking to it! [12:10] seb128, I tend to lean towards (2), too, but don't want to make this decision [12:10] sure [12:10] well at least we know what's the issue is [12:10] we can discuss what to do next ;-) [12:10] seb128, yeah. I'll add my findings to the bug [12:10] larsu, thanks [12:11] seb128, do you use tomboy by any chance? [12:11] is there a trick to get global shortcuts working for it? [12:12] larsu: I use it ... you mean? [12:12] larsu: use the hud? :p [12:13] seb128, hm, that's definitely an option I hadn't considered :) There are preferences for setting a global hotkey for new note which don't work [12:13] but the hud works for me, so thanks :) [12:13] it's actually nicer, because I don't need to remember shortcuts [12:14] larsu, yeah, indeed that seems to not work, thanks for pointing it ... and yeah the HUD is cool, you can access notes directly by typing the name ;-) [12:14] seb128, yeah just noticed. Awesome! [12:21] larsu: bug #685309 for the record [12:21] Launchpad bug 685309 in tomboy "Hotkeys don't work with indicator-application" [Low,Confirmed] https://launchpad.net/bugs/685309 [12:21] * larsu subscribed [12:21] seb128, thanks! [12:25] hey, while starting empathy I get this error: [12:25] (empathy:5802): folks-WARNING **: Failed to find primary PersonaStore with type ID 'eds' and ID 'system'. [12:25] Individuals will not be linked properly and creating new links between Personas will not work. [12:25] The configured primary PersonaStore's backend may not be installed. If you are unsure, check with your distribution. [12:25] this basically means I cannot link people [12:25] now I recall this bug [12:26] and it was fixed by including libfolks by default (or the appropriate telepathy + folks library, I don't remember) [12:26] so, am I seeing an upgrade issue where something in the default config makes it work [12:26] but fails for existing upgrades [12:26] or is this a new bug that ought to be reported [13:05] good morning ubuntu-desktop! [13:07] hey kenvandine! [13:11] hey kenvandine [13:11] how are you? [13:12] good, it was good to take a day off :) [13:13] good morning kenvandine :) [13:13] hey bcurtiswx [13:14] kenvandine, what did you do from your day off? [13:15] ran errands and got an estimate to get my wife's van fixed... she backed into with my car in the driveway [13:15] hate that kind of accident :( [13:15] :-( [13:16] kind of funny though... she had parked the van at the end of our driveway where it is narrow [13:16] walked to the van on her way out with our son and had him get his baseball gear out of the van... then got into the car and backed into the van [13:16] :) [13:16] she had just seen it... hehe [13:17] early or late in the day? [13:17] i will never stop teasing her about that [13:17] afternoon... they were running late for a baseball game [13:17] so a bit frantic [13:17] but i can still tease her about it :) [13:17] gonna cost me $700 to fix though... that isn't so funny [13:19] is there any functional part damaged? [13:19] or is it "cosmetic"? [13:19] well I guess in any case you will want it fixed... [13:19] cosmetic [13:19] not going to rush to fix it... over the summer sometime [13:19] my mirror got broken off two winters ago on a brand new 09' chevy Cobalt.. cost me 400 bucks to fix that.. must be nice for cosmetic damage to be "only" 700 [13:20] bcurtiswx, that is with me replacing the headlight assembly myself... it is just body work [13:20] my car though... a VW passat looks like it didn't hit a thing [13:21] gotta love a german car! [13:21] hehe, I was about to say, german quality ;-) [13:21] crushed the front corner of her dodge :) [13:21] gotta love just owning a German car [13:21] there is a scratch on the VW [13:21] and the van looks terrible :-D [13:21] sad to say i am going to have to sell that car :/ [13:22] had to buy a truck to haul dirtbikes around :) [13:22] i love that passat [13:25] kenvandine, that sync request for telepathy-glib it was updated to 0.19.1 FYI i think it's in experimental too [13:27] bcurtiswx, ok, and what version do we need for empathy? [13:27] just .19.x? [13:27] dep is on >=0.19.0 [13:28] i guess it'll get synced after a1 [13:28] no rush [13:28] kenvandine, yup no rush, i think A1 is today anyways [13:28] or remind me and i'll do it after we open [13:29] I'm gone to Rochester for the weekend, bachelor party :) will be back Tuesday [13:29] have fun! [13:29] and don't get arrested [13:30] or lock your friend on the roof of a building :) [13:30] kenvandine, speaking by experience? ;-) [13:30] i'll try not to, it'll depend on how much alcohol vs. food i have. Great movie though :) [13:30] hehe [13:30] movie reference... the hangover [13:30] very funny movie [13:30] ah, ok [13:31] highly recommend the movie [13:31] I watched the summary video from UDS yesterday, now I completely get all the "French default language" jokes i've seen recently [13:31] nice seb128 :) [13:31] haha :) [13:31] lol === m_conley_away is now known as m_conley [13:35] desrt, hey, for info my glib 2.32.3 SRU got accepted yesterday [13:36] oh. good [13:36] no playing around with cherry-picks, then? [13:36] desrt, I got a +1 from the TB for the GNOME exception which was enough to get slangasek to ack it rather than ask me do play backporting [13:36] cool [13:36] let's hope we don't get burnt :) [13:36] yeah, I really hope there are no regression because I pushed for it and that would not play well for the "GNOME updates are fine" ;-) === jbicha is now known as Guest48486 [13:39] seb128: http://cheezburger.com/6306451712 [13:41] desrt, lol [13:53] anyone in here use a UPS ? === mfisch` is now known as mfisch === mfisch is now known as Guest83117 === Guest83117 is now known as mfisch === mfisch is now known as Guest28186 [14:21] hey seb128, I see you're listed as the maintainer for launchpad-integration [14:23] https://bugs.launchpad.net/ubuntu/+source/launchpad-integration/+bug/991602 is starting to become a problem as the site is getting overrun with bug reports instead of questions, how can I help get this on someone's radar? [14:23] Launchpad bug 991602 in launchpad-integration "Apport's recommendation of Ask Ubuntu is vague" [Undecided,Confirmed] [14:37] jcastro, hey [14:38] jcastro, where,when do you get that message? [14:42] seb128: it appears in apport, I've only ever seen a screenshot, I don't know how to trigger it [14:44] jcastro, that has nothing to do with launchpad-integration I think, launchpad integration is what adds the menu items in the help menu to translate an app, ask for help or report a bug [14:44] jcastro, we dropped launchpad-integration in q btw [14:44] pitti moved it over to lp integration [14:45] jcastro, yeah, my gut feeling is that's because it's not from apport either, that's why I'm trying to figure how you trigger it to find where it's coming from [14:47] ah bummer, I was hoping for a quick string replace. :) [14:48] jcastro, the string in the bug is the exact one, with right case etc? [14:49] jcastro, I tried to grep and google for it without success [14:50] http://meta.askubuntu.com/questions/3017/why-is-ask-ubuntu-being-linked-to-from-apport [14:50] aha! there's a screenshot [14:57] seb128: aha! where was it? [14:57] jcastro, xdiagnose [14:57] in the xorg hook [14:57] jcastro, [14:58] jcastro, /usr/share/apport/package-hooks/source_xorg.py [14:58] # Anything else should be redirected to technical support channels [14:58] else: [14:58] ui.information("http://askubuntu.com is the best place to get free help with technical issues.\n\n" [14:58] ok so I'm off to bug raof then? [14:58] jcastro, bryceh rather [14:58] k [14:58] I'll update the bug, thanks for your help! [14:58] jcastro, I reassigned there, yw! [15:50] hum Default Apps, why does gedit come up as my default calendar app ? [15:52] bcurtiswx, do you have any calendar app installed? [15:52] seb128, nope [15:52] that's why [15:52] that's a known gtk issue, they didn't consider the case where no option would be available [15:52] or rather it's a buggy case [15:53] thunderbird has addons for calendar, are there plans to use that for default in 12.10 ? [15:55] not that I know of [15:56] but some GNOME contributors are working on a calendar app so maybe we will use that if it turns out to be ready [15:57] seb128, I think it would be nice to have it. Evolution was mail and calendar, but now thunderbird that replaced it is just e-mail. Would be neat to have it handle both by default since it's capable of it right now. [15:57] bcurtiswx, what is capable? tb? [15:57] seb128, sry, yes [15:58] bcurtiswx, the calendar extension is not officially supported by mozilla and with the 6 weeks cycle it's hard to maintain [15:58] bcurtiswx, i.e any of the 6 weeks update could break calendar [15:59] bcurtiswx, which would put us in a difficult position, we don't want to have to block tb updates nor we want to maintain the calendering extension [15:59] seb128, yeah. I'm happy to hear about a calendar app, should take care of that missing feeling of calendar [16:00] seb128, has a name been given to this calendar app ? [16:00] bcurtiswx, "Calendar" [16:00] seb128, how original.. lol [16:00] bcurtiswx: http://iloveubuntu.net/meet-gnome-calendar-fresh-new-calendar-application [16:00] the new GNOME way: Web, Chat, Clocks, Contacts, Documents, ... [16:01] seb128, lol but empathy = chat.. unless i missed the memo of that name changing [16:01] jbicha, hey, how are you? Congrats on the MOTU membership btw ;-) [16:01] seb128: aloha [16:01] bcurtiswx, some people would like empathy to be renamed chat, not sure that's going to happen though [16:02] seb128: what do you think of bug 1005057 [16:02] Launchpad bug 1005057 in ubuntu-meta "Please remove gnome-nettool from ubuntu-desktop" [Undecided,New] https://launchpad.net/bugs/1005057 [16:04] jbicha, no strong opinion, I'm not so much of an admin and I'm unsure how useful those tools are [16:04] cyphermox, ^ do you have any opinion? (asking you since that's sort of network related) [16:06] it's useful. it belongs to ubuntu-desktop since it's gnome and desktop; but it doesn't have to be installed by default, clearly [16:06] out for lunch, bbl :) [16:06] personally i use the console version anyway [16:09] cyphermox: well it could be demoted to universe [16:11] jbicha: no issues at all with that [16:19] * didrocks waves good evening [16:30] jbicha, do you think it's not useful on the CD? [16:31] seb128: I think gnome-nettool is not useful on the Desktop CDs, if I want to ping or whois or whatever I use the command line [16:31] I don't think normal people use ping, finger, nmap, etc. [16:32] jbicha, it's meant as an admin tools I think [16:32] jbicha, the same reason we have i.e a rdp client [16:32] seb128: synaptic, users-admin were admin tools too ;) [16:32] jbicha, well, we kicked those out because there we duplication [16:33] we have update-manager and software-center to replace synaptic [16:33] we have the user panel to replace users-admin [16:33] not sure what we have to replace gnome-nettool [16:34] but it's always hard to draw the line, especially when that's something that can be useful to some people and didn't show up as being confusing for others or being an issue to maintain [16:34] we have ping, whois, traceroute [16:34] personally I'd like to keep whois on the CD, but nmap is too large and not core to the desktop experience [16:34] well, do we want to tell sysadmin they can work from a modern environment through ui or do we want to tell them they need to use the command line? [16:35] jbicha, feels like an ubuntu-devel list sort of discussion [16:35] i.e I would like to hear from admin using ubuntu-desktops [16:35] System Settings>Network has some info [16:35] to know if they appreciate having graphical tools [16:35] I've never heard of anyone using a graphical tool for ping :) [16:36] Sure, I'll ask on the mailing list [16:36] thanks [16:36] jbicha, I'm not a sysadmin, but whoever started gnome-nettool probably though it was a good idea ;-) [16:37] gimp's a good idea too, lol [16:37] yeah, as said it's hard to draw the line [16:38] we had similar discussion when we kicked xchat-gnome out of the CD :p [17:37] hilo [17:37] how can I install paid apps in ubuntu Q [17:37] ? [17:50] pavolzetor, enable the p source for them? [17:58] yep [17:58] I bought humble bundle [17:58] but I cannot install it as I use ubuntu q [18:02] pavolzetor, edit the sources file and use precise [18:02] where it has extras.ubuntu.com [18:02] change it from quantal to precise [18:04] kenvandine, are you sure? Paying stuff comes from private PPA, I had to go to Launchpad, look at my private PPA subscriptions and add them to my sources. [18:04] oh [18:04] i could be wrong then [18:04] free stuff that is in software center does that [18:04] i guess you need to edit whichever file in /etc/apt/sources.list.d/ is for that app [18:04] change quantal to precise [18:05] I even cannot find extras :/ [18:07] sorry [18:07] empathy crashed [18:07] pavolzetor, i was wrong, it isn't extras [18:07] not for paid apps [18:07] it'll create a new file in /etc/apt/sources.list.d/ [18:07] pavolzetor, report that crash too ;) [18:08] it crashes a lot :) [18:09] https://bugs.launchpad.net/ubuntu/+source/empathy/+bug/1010139 [18:09] pavolzetor: Error: bug 1010139 not found [18:09] argh it is private [18:13] I am going to windows (sigh, LIMBO is slow at ubuntu) :'( [18:40] yay, opal builds [18:41] cyphermox, hey [18:42] cyphermox, can you look at bug #1009928? [18:42] Launchpad bug 1009928 in ghostscript "Ubuntu Desktop Netboot install failed: unmet dependency - poppler-data Breaks cmap-adobe-japan2 (<= 0+20090930-2)" [High,New] https://launchpad.net/bugs/1009928 [18:42] craphouse [18:42] sure :) [18:42] cyphermox, basically ubuntu-desktop recommends cmap-adobe-japan2 and should stop doing that [18:43] cyphermox, which I think it's the only issue left in that bug [18:43] aye. [18:43] cyphermox, thanks ;-) [18:43] np :) [18:58] seb128: I'm familiar with the branch to update, but what's the exact process for doing the change? update, commit, push then notify someone? [18:59] cyphermox, https://wiki.ubuntu.com/SeedManagement [18:59] ah, thanks [19:01] cyphermox, the wiki says "run the update script in the appropriate source package and upload it " [19:01] I'm reading [19:01] didn't do that in ages though so if you have specific questions maybe ask on #ubuntu-devel ;-) [19:02] trying to decide at the same time whether it needs to be replaced with poppler-data or if it's pulled in by something else [19:02] * cyphermox will take that opportunity to play with germinate a little again [19:06] ok; seems it's just a removal [19:10] cyphermox, yeah, sorry about that, I discussed it a bit with cjwatson earlier and we agreed we should probably just drop it [19:11] well, I totally agree [19:11] the seed mentioned it was just there to force a prefered alternative and that's not required with one choice [19:11] cyphermox, good to have a second opinion (or third one) ;-) [19:11] common includes ghostscript-x, which depends on ghostscript, which depend on libgs9, which Depend on poppler-data | cjk-resource [19:12] hehe, chain of depends! [19:12] so AFAIUI poppler-data should be getting pulled in anyway [19:12] well, poppler-data was not seeded anyway for space reason [19:12] and as a worst case it would be by language-selector if I understand some other comments properly [19:12] that's why language-selector pulls it in only for specific locales [19:12] ok [19:13] well, now it might get pulled in anyway no? [19:13] not sure if the space reasons still stand with the new iso limit [19:13] but we have room [19:13] if we get to fill it we will then figure what needs to be dropped [19:13] but it was something useful we pushed back on in the past due to the CD [19:14] so if it's getting in no worry it's a good thing ;-) [19:14] sure [19:20] ok; so now to look at -meta [19:23] scary ;) [19:27] what does accountsservice do on ubuntu server exactly? [19:32] dobey, is ubuntu-server using,installing lightdm? [19:34] seb128: it's not installed, no [19:36] dobey, ok, dunno then, it's using by the control center user panel and lightdm for desktop [19:36] dunno for server then [19:36] i suspect nothing [19:36] unless one installs all that stuff [19:36] right [19:36] is it installed by default? [19:37] seb128: it seems ubuntu-standard pulls it in, yeah [19:38] or well, ubuntu-standard -> language-selector-common -> accountsservice [19:38] ah, yes [19:38] which seems a bit weird [19:39] dobey, that's because they put some perl script to deal with locales in there [19:39] which is an hack and meant to go away this cycle [19:49] * dobey submits the MIR for ubuntuone-dev-tools [20:39] hi, can someone review the patch on bug 811475 please [20:39] Launchpad bug 811475 in rhythmbox "Descender letters (g j p q y) get their tail cut off in current song" [Low,Fix committed] https://launchpad.net/bugs/811475 [20:55] AlanBell: I'm concerned that this works for the default font size but doesn't for any other [20:55] (or actually, for any bigger) [20:56] I can't test this until later though, all I have is a quantal box here and no space for VMs atm [20:56] I will test that, you are probably right [20:56] but at least it would work in one font size then :) [20:57] that is totally the wrong fix [20:57] it is a bit of a workaround for the bug in elipsise [20:58] it isn't an elegant fix, certainly. I would love a better fix. [20:59] what is the bug in ellipsize? [21:00] the string in question is something like "song title by artist from album" [21:01] the ellipsize method works on the string and then requests space for it, but doesn't take into account that part of the string is [21:01] seems like we can fix gtk+ or pango (whichever one is the right one) [21:01] so the vertical size it requests is enough for the regular text, but that is pushed down a bit by the big bit [21:02] http://developer.gnome.org/gtk/2.24/GtkLabel.html#gtk-label-set-ellipsize is the bit that gets it wrong [21:03] I don't know where the code is to fix that, and I don't know the impact of fixing it on other applications [21:06] i don't think that specific function is what gets it wrong. it just sets a property on the widget that it should ellipsize if needed, and in what way it should do that [21:08] AlanBell: have you tested just commenting out the set_ellipsize() call? [21:08] yup, works a treat [21:09] until you have a long song title that should in fact be ellipsized [21:09] it stops you resizing the window below the size of the title as I recall [21:09] it is there for a reason, but as soon as that method is called it scrunches up the space and the descenders get chopped off [21:10] right [21:10] so it seems to be a problem with pango ellipsizing vs. markup [21:10] so my inelegant workaround was to make the space a touch bigger, until we get the new layout that doesn't have this problem, but that won't land in Precise I think [21:11] the upstream fix is a redesign that looks like this https://plus.google.com/109365858706205035322/posts/RX3yGBCaRwA [21:12] i know, i've been using it since it landed a couple months ago [21:12] jasoncwarner_, ping [21:13] having daily builds does that :) [21:13] :) [21:13] and the new design has other problems [21:17] maybe some other way of boosting the space by a percentage, or split the string into the big part and the not big part would work [21:18] or someone can dive into the scary parts of GTK and fix the root of the problem [21:19] well if i'm right, i'm looking at the code that's breaking it, right now [21:19] oh cool, where is it? [21:20] http://git.gnome.org/browse/gtk+/tree/gtk/gtklabel.c#n3235 [21:20] somewhere in there [21:20] yeah, that looks about right === m_conley is now known as m_conley_away [21:51] AlanBell: dobey: I have a patch [21:52] how does it work? [21:52] it doesn't involve changes in gtk ;) [21:52] pango? [21:52] gtk_misc_set_padding() [21:53] crude, but effective [21:53] ah, right, so adds a bit of padding all round? [21:53] yep [21:53] also just a hack [21:53] better than my hack, certainly [21:53] tried with 10px with good results, I set it to 5 for being conservative [21:53] it needs to be fixed in gtk+ [21:54] dobey: well, not really so much of a hack. it does feel to me like there is a distinct lack of any padding or adjustments between these widgets [21:54] cyphermox: lack of spacing between the widgets is a spearate problem though [21:54] this works afaict for all cases (font sizes) and doesn't look too insane [21:54] cyphermox: and the right way to fix that is to change the spacing [21:55] cyphermox: does it work if you set padding to 0? [21:55] I haven't tried [21:55] the default should be -1, presumably "more" than 0 [21:55] aka: what's needed [21:56] well, the default should work; but there's a bug in gtk+ :) [21:58] dobey: I don't agree. pango is supposed to ellipsize, not mess around with heights unless it's to make the text span two lines rather than one [21:58] cyphermox: pango isn't necessarily messing around with the heights [21:58] cyphermox: which is why i said gtk+. gtk+ is messing around with the height [21:58] s/pango/gtk [21:59] the ellipsizing thing I mean [22:00] in fact, i'm pretty sure http://git.gnome.org/browse/gtk+/tree/gtk/gtklabel.c#n3315 is where the bug is [22:02] go nuts ;) [22:02] I avoid touching gtk, it's truly usually not the broken thing, most of the time [22:03] except for when it is :) [22:08] anyway, gotta run [23:08] hmm... [23:08] how do I find out of libsigc++ has changed between precise and quantal? [23:10] thumper: hasn't yet, needs a merge [23:12] micahg: so... is the current quantal libsigc++ built with the new toolchain? [23:13] nope [23:13] bugger [23:13] I'm guessing that is my unity problem then [23:14] thumper: poke broder (he touched it last) [23:14] huh? [23:14] libsigc++-2.0 needs a merge :) [23:15] hmm. i think i'm unlikely to get to it before next wednesday, given stuff currently on my plate [23:16] thumper: rmadison is your friend [23:18] hmm... [23:18] I have didrocks and seb128 jumping on me right now [23:18] as unity on quantal fails [23:19] bit it fails inside libsigc++ calls [23:19] so my current theory is that we need to make sure libsigc++ is built with the new gcc [23:19] broder: you mind giving up the merge to the desktop team? [23:19] because the same code is fine on precise [23:19] micahg: oh yeah. no problems with that [23:19] sorry for not being clear [23:19] people should feel free to take [23:19] thumper: ^^ there you go :) [23:20] micahg: you are assuming that I know what to do with that :) [23:20] thumper: ask didrocks or seb128 to do the merge :) [23:20] * thumper wonders if I could convince someone closer... [23:20] * thumper looks at RAOF [23:21] You're looking for me to merge libsigc++? [23:21] RAOF: maybe, is it easy? [23:22] Dunno. Let's find out! [23:22] heh [23:26] ok, so... here is a question... [23:26] is my laptop going to be 'orribly broken if I upgrade to quantal alpha? [23:26] in particular I'm trying to debug this unity quantal issue [23:26] however I don't want to be non-productive normally [23:26] like email and stuff :) [23:27] * thumper is waiting for thomi's machine to install the quantal image [23:27] I've been running quantal since the middle of UDS; it's not been broken. [23:28] thumper: Several of us have been running quantal for a few weeks now, and I don't think anybody has experienced any major breakage, but of course it depends on what applications you use on a daily basis. [23:28] RAOF: Wow you were brave risking a broken state when you were away. [23:29] I had a spare laptop, and the local mirror was tempting. [23:30] Yeah well there is that. [23:30] bbl dental appointment. [23:37] RAOF: and unity is working for you now? [23:37] thumper: Yes; it's been working all cycle, as far as I'm aware. [23:37] RAOF: I think it is rebuilding everything that causes the problems [23:37] With that exception, obviously :) [23:37] RAOF: the issue didrocks is having is building a new nux and unity [23:37] RAOF: and the rebuilding then brings in the new libc, new template stuff [23:38] RAOF: and seems to be conflicting with the older sigc++ [23:40] Oh, boo. The bzr importer has failed on Debian's libsigc++ . [23:45] RAOF: is there an easy way to rebuild the existing libsigc++ with the new toolchain to check? [23:45] we finally have quantal installing on a fresh machine [23:45] for this testing [23:45] Yeah, that'd be trivial. Does the build failure occur in a sbuild environment? [23:45] thomi's machine has its guts lying all over the table [23:45] it is a runtime failure, not a build failure for unity right now [23:46] let me get to the stage where we have unity and nux building on quantal from trunk, and I'll get back to you :) [23:46] sudo apt-get build-deps libsigc++-2.0 ; sudo apt-get -b libsigc++-2.0 should build you those packages. [23:47] RAOF: awesome... I'll be on it shortly [23:47] obviously I'd like to reproduce the failures first