[00:06] Woo! Unity's less hateful on dual head displays now! [00:08] Daekdroom: there has been a clear definition of what notification area is for and it didn't stop anyone from abusing it. [00:09] kklimonda, where? [00:09] (and yes, that question is serious) [00:10] gnome hig [00:10] http://library.gnome.org/devel/hig-book/stable/desktop-notification-area.html.en [00:11] there is also a similar document for indicator applets, and for windows notification area (misnamed by some as "tray") [00:12] but developers don't really read this stuff [00:14] That document makes me wonder about transmission and indicator-weather [00:15] transmission devs have added the "tray icon" because "every other bt client had it" [00:15] and users asked for it - it's disabled by default. [00:16] unfortunately removing this small icon meets with a lot of resistance from users. [00:16] we got really flamed for removing it from gwibber [00:17] there are actually requests to enable it by default. [00:17] people wanted a way to "refresh" [00:17] btw, is it just me or is the unity launcher making messaging area pretty redundant? [00:17] it can :) [00:17] messaging area? [00:18] messaging menu [00:18] ah, right [00:18] I'm not sure. Does it integrate in some way with empathy? I don't use it. [00:18] yeah [00:18] Let me see [00:18] Daekdroom, it is also known as the messaging indicator [00:19] I can see integration with gwibber right away [00:19] but not the redundant kind [00:19] in unity we have menus for things like refresh and stuff for gwibber [00:20] Compiz crash *shrughs* [00:21] :) [00:21] in unity we have menus for things like refresh and stuff for gwibber [00:21] Yeah. I noticed. [00:21] but in the messaging menu it displays counts of unseen posts [00:22] xchat does that in the launcher, doesn't it? [00:22] yes [00:22] so does empathy [00:22] kenvandine: doesn't unity launcher displays the same counter? or is it just not implemented yet? [00:22] not for gwibber [00:22] i can only display one count [00:23] but i am not a big fan of putting a single count which is just a total of the streams [00:23] even though that is what i did for evolution :) [00:24] * Daekdroom removes transmission from indicator-applet [00:24] i don't mind transmission being there if it can let you close the window and keep running [00:24] It's the sort of thing that you probably want to run on background, indeed. [00:25] if you close it, you should be able to see it is still doing something [00:25] so i am not opposed to that one [00:25] The only difference between it and the empathy and pidgin is that it doesn't need to warn you about stuff [00:25] kenvandine: in Unity it doesn't matter anyway, you can just minimize it - at least I don't mind another icon in the launcher [00:25] and when it does, it uses NotifyOSD [00:25] Minimizing? I don't do that. I spread things around workspaces and switch between them. [00:25] kklimonda, right, i kind of like having the count there though [00:26] kenvandine: any idea if the launcher api for "decorating" icons will be extended? it could be a nice feature to display up/down speed on transmission icon [00:26] i turn off autohide in the launcher [00:26] so i love seeing a nice count [00:26] ah :) [00:26] I just don't run that many apps [00:26] and i did the progress bar for empathy file transfer [00:26] i love that :) [00:27] yeah, progress bar is a nice feature, but it doesn't fit how transmission downloads files currently [00:27] i.e. there is no way to queue downloads [00:27] so it would have to display a total progress.. not really helpful imo [00:27] It could show down/up speeds. [00:27] Daekdroom: I did think about it a moment ago but I'm afraid it will add too much clutter to the icon. [00:28] Possibly, specially if it's gonna be resizeable. [00:29] empathy does that now [00:29] it does a total [00:30] so if you are transferring 2 and you add a third it shifts down the progress to the total percentage [00:30] But that is such a different situation [00:30] yeah [00:30] ubuntuone does the same thing [00:30] Because when doing the transference through IM, you expect it to finish ASAP. [00:30] it is more telling you that there are files transferring [00:31] not so much that a specific file might be done [00:31] When downloading loads and loads of stuff, some stuff might not matter in the short time. [00:31] ubuntuone could be loads [00:31] i did a fresh install on a box today and it had to download 4G of data from u1 [00:31] it was nice to have that one little progress bar showing me where it was :) [00:42] I did a fresh install on my laptop and the U1 progress bar was a huge and delightful surprise. [00:42] As were the notification updates from U1 every x minutes actually. [01:32] DBO you around? === _LibertyZero is now known as LibertyZero [07:53] good morning [08:00] greetings everybody [08:05] good morning === API is now known as Guest28214 [10:14] hello [10:14] how much will Unity 2D differ from Unity? [10:18] DBO, Hi! bug 690096 seems to have come back [10:18] Launchpad bug 690096 in Ayatana Design "Launcher - Implement launcher auto-scroll" [Critical,Fix released] https://launchpad.net/bugs/690096 [11:15] !register [11:15] Information about registering your nickname: https://help.ubuntu.com/community/InternetRelayChat/Registration - Type « /nick » to select your nickname. Registration help available by typing /join #freenode [11:19] /opt/extra/src/canonical/unity/sync-geometries/utouch/unity-mt-grab-handles/src/unity-mt-grab-handles.cpp: In constructor ‘UnityMTGrabHandlesScreen::UnityMTGrabHandlesScreen(CompScreen*)’: [11:19] /opt/extra/src/canonical/unity/sync-geometries/utouch/unity-mt-grab-handles/src/unity-mt-grab-handles.cpp:723:42: error: no matching function for call to ‘GLTexture::readImageToTexture(CompString&, CompString&, CompSize&)’ [11:19] /usr/include/compiz/opengl/texture.h:205:14: note: candidate is: static GLTexture::List GLTexture::readImageToTexture(CompString&, CompSize&) [11:19] I'm getting this, but I'm already with the latest nux from trunk, any idea? [11:19] ah, I need a newer compiz [11:20] smspillaz: ^^ === MacSlow is now known as MacSlow|lunch [11:25] rodrigo_: yeah, update compiz [11:42] rodrigo_: could you give me some advice on ecal [11:42] klattimer, yes, sure [11:42] I'm having trouble with the e_cal_generate_instances [11:42] by looking at the code in eds it appears it generates objects, then generates instances for each object and calls my callback with the following args [11:42] how much will Unity 2D differ from Unity? [11:43] ECalComponent *comp, time_t instance_start, time_t instance_end, gpointer data [11:43] right? so, when I get that in my callback, I check its vtype [11:43] klattimer, right [11:44] it's vtype returns neither event or todo and asserts that E_IS_CAL_COMPONENT failed [11:44] so looks like the callback isn't actually getting a comp [11:44] hmm [11:45] klattimer, do you have only one callback? [11:45] yeah [11:45] just one [11:45] the ECalComponent is unref'ed as soon as the callbacks are done [11:45] ah [11:45] right [11:45] so maybe you are using when it's been unrefed? [11:45] ok, so I need to ref them [11:45] that's good advice and probably what I'm missing [11:45] makes sense [11:46] not in the callback itself, if you just want to look at it there [11:46] but if you keep it around, yes, g_object_ref it [11:46] ok [11:47] was that the case? [11:47] rebuilding now [11:47] ah ok [11:47] will let you know in about 3-5 mins [11:47] :) [11:47] :) [11:47] (it's only a little HP mini 110 doing the build, bless it:) [11:49] rodrigo_: that worked [11:49] cool [11:49] except, the dtstart is always the same [11:49] how can i get the start date time of the instance? [11:50] you can use e_cal_component_get_dt* [11:50] don't remember, but maybe it calls the callback with the dtstart/end you used when calling e_cal_generate_instances [11:50] let me check the code [11:50] rodrigo_: thanks [11:51] rodrigo_: I use e_cal_component_get_dtstart (ecalcomp, &datetime); [11:51] but it always returns the same datetime [11:51] the same datetime for all components? [11:52] rodrigo_: yep [11:54] klattimer, do you have the code isolated enough so that I can compile and run it here? [11:54] or well, pastebin anyway [11:54] rodrigo_: I can push up to launchpad in a minute [11:54] and point you at the code there [11:55] just need a sec to test some stuff [12:18] rodrigo_: tell me what you make of... each item has the same datetime from dtstart here; http://bazaar.launchpad.net/~karl-qdh/indicator-datetime/recurring-and-gsettings-fixes/view/head:/src/datetime-service.c#L592 [12:29] how much will Unity 2D differ from Unity? [12:31] klattimer, I see a few things wrong, but nothing that would explain why you get the same datetime for all ECalComponent's [12:31] rodrigo_: what do you see wrong? [12:32] klattimer, using a global variable (tmpobjects) for temporary stuff [12:32] klattimer, in line 731, you're g_object_unref'ing a GList [12:32] rodrigo_: I know ugly huh, but I was just trying to fit things into the way things were before we hit the recurring problem in the first instance [12:34] hello, i am trying to update a gtk.MenuItem in an indicator (created on the fly and prepended to the indicator menu). However there appears to be no dbus signal sent about the label change so it does not get updated. Sometimes i can reproduce this with the items that are created upon indicator initialization. Is there a known bug for that? [12:34] klattimer, how many components you get in tmpbojects? [12:34] 7 [12:34] which is the right amount [12:35] as I have an appointment which repeats every day and that's all [12:35] but all of those say tuesday 4:30 === MacSlow|lunch is now known as MacSlow [12:38] klattimer, so, the for loop looks good, ecalcomp points to l->data [12:38] and the components are ref'ed before adding them to the list, so not sure why [12:38] but it looks like you're getting the same component all the time [12:39] can you g_debug and print the address of the ecalcomp for each loop pass? [12:40] rodrigo_: I'll check the address here [12:41] klattimer, g_signal_connect (G_OBJECT(source), "changed", G_CALLBACK (update_appointment_menu_items), NULL); [12:41] maybe that is being called again and again? [12:41] and thus setting tmpboejcts to null? [12:44] ok, I'll remove that too then [12:45] rodrigo_: all the objects have different addresses [12:45] but I'll remove the changed sig and retest [12:52] rodrigo_: nope, it still returns the same date for each [13:13] klattimer, I'll have another look later, now going out for lunch === alecu_ is now known as alecu [13:26] I want to develop an application which shows an (dynamic) image like the sound indicator does, is this possible with libappindicator ? [13:35] JoshuaL: you need to create your own indicator menu item type [13:36] JoshuaL: take a look here; http://bazaar.launchpad.net/~karl-qdh/indicator-datetime/recurring-and-gsettings-fixes/view/head:/src/indicator-datetime.c#L1113 [13:37] this is the code I'm working on for the datetime indicator [13:37] we do a lot of screwy things here like that ;) [13:37] klattimer, hehe [13:38] klattimer, thanks for pointing me in the good direction! :) [13:38] np [13:39] im new to this so this will be a good start to learn it :p [13:51] kenvandine, i still don't see anything listed in the applications and files launcher icons, yet i have place-applications and place-files installed :-\ [13:52] bcurtiswx, you mean you have no icons for them in the launcher? [13:52] or when you activate it you don't see any contents? [13:52] no no, just no contents [13:52] ah [13:53] maybe broken zeitgeist or something? [13:53] njpatel, ^^ [13:54] bcurtiswx, can you take a screenshot of the applications place please [13:54] bcurtiswx, also, make sure unity-applications-place is running (pgrep) [13:54] njpatel, places are rocking btw [13:54] great job! [13:54] bcurtiswx: both places should show something even if Zeitgeist is hosed [13:55] bcurtiswx: files place should show downloads and favorite folders [13:55] njpatel, my only complaint is the font for the icons, looks too condensed to me [13:55] bcurtiswx: and apps place should show installed apps [13:55] kenvandine, it's not using your system font? [13:55] * bcurtiswx will go check [13:55] njpatel, dunno [13:55] * njpatel finds bug in places [13:55] but i see the same thing on a fresh install too [13:56] njpatel: there is a bug when the places don't show up anything before you hover them [13:56] didrocks, yeah, that's what I just saw [13:56] njpatel, pgrep doesn't show unity-applications-place [13:56] well, any processes for it [13:57] kenvandine, screenshot please? preferably with another app using the system default font please [13:57] bcurtiswx, okay, so that would be the reason :) Do you have a local or system install of unity? [13:57] system [13:57] bcurtiswx, but you are getting the icons on the launcher? [13:57] njpatel, yup icons are there.. lemme get a screenshot for ya [13:58] bcurtiswx, run /usr/lib/unity-place-applications/unity-applications-daemon in a terminal and see if it's segfaulting? [13:58] njpatel, /usr/lib/unity-place-applications/unity-applications-daemon: symbol lookup error: /usr/lib/unity-place-applications/unity-applications-daemon: undefined symbol: unity_activation_get_type [13:59] bcurtiswx, is your system fully updated? [13:59] doesn't sound like it [14:00] njpatel, http://ubuntuone.com/p/flc/ [14:00] njpatel, unity: Installed: 3.6.0-0ubuntu1 [14:00] Could not locate content blob for object d4df1d34-c1d2-4d12-b129-cf8ae96e08ae [14:00] bcurtiswx, libunity? [14:01] njpatel, libunity3: Installed: 3.4.6-0ubuntu1 [14:01] hmm [14:02] kamstrup, didrocks ^ [14:02] have you guys seen that error? [14:02] local install [14:02] no, never saw that [14:02] clean your local builds they take over the distro one [14:02] yeah, I think seb128 spot it [14:02] bcurtiswx, you don't have a local unity hanging around? [14:02] i have been installing it from apt, not local [14:02] bcurtiswx, ls /usr/lib/ | grep unity [14:02] you never built a local version from source? [14:03] i did, but i remember doing a make remove or make uninstall after i was done.. one sec [14:03] try to run ldd on the binary which raise the error [14:03] bcurtiswx: ls /usr/local/lib/ rather [14:03] * kenvandine never trusts make uninstall [14:03] Ok there are files in there.. give 'em the ol 'rm' treatment? [14:04] yes [14:04] njpatel, did you see my screenshot? is that what you expect it to look like? [14:05] bcurtiswx: it's because libunity broke abi... sorry that was my fault [14:05] so the place daemons need a rebuild against the new libunity abi [14:05] no prob, i appreciate the help.. will i need to reinstall from apt or just re-run unity? [14:05] kamstrup, it's ubuntu versions they don't need a rebuilt, it's just a local install taking over the official ubuntu lib [14:05] just rerun [14:06] kenvandine, I did, I typed something out and then I forgot to hit enter! [14:06] kenvandine, it's wrong, the fonts are not right, could you file a bug please and assign it to me? [14:06] ok [14:06] kenvandine, I think I know the issue, what DPI do you use? === m_conley_away is now known as m_conley [14:06] 96 [14:06] that did it, muchas gracias :) [14:06] alrighty [14:06] njpatel, it looks identical on my desktop box [14:06] which was a fresh install yesterday [14:07] cool, so there's just an issue somewhere [14:07] I think maybe the painting of the text is wrong [14:07] good [14:07] kenvandine: lp:~karl-qdh/indicator-datetime/recurring-and-gsettings-fixes [14:07] i thought that was intentional :) [14:07] There's a tonne of fixes in there [14:07] we might be not allocating enough width for the texture [14:07] klattimer, awesome [14:07] i'll grab it [14:07] please test, we still seem to have all recurring appointments happening on the same day/time though [14:07] njpatel, yeah, it looks squeeze === Guest28214 is now known as apinheiro === dbarth_ is now known as dbarth-otp [15:14] njpatel: are there plans for shortcuts to the dash [15:15] for example, alt-f1, and then hit the right arrow [15:15] On what condition do gtk labels update/menu redraw fail to be propagated underlying in appindicator? [15:15] jcastro, shortcuts to the dash? Super? [15:15] right [15:16] but if I want to select something IN the dash [15:16] oh, keynav? [15:16] for example, super, type "chrome", now what? [15:16] right [15:16] yes, it's broken but iit's there [15:16] ah ok [15:16] definitely we will have keynav === m_conley is now known as m_conley_away [15:54] any cmake expert around? === _LibertyZero is now known as LibertyZero [16:16] bug 694596 seems not fixed - I'm reopening the unity project task too [16:16] Launchpad bug 694596 in unity (Ubuntu Natty) "unity not working on rotated displays" [High,Triaged] https://launchpad.net/bugs/694596 [16:17] hmm, actually I can't. lamalex - could you open it? [16:21] bdmurray: done [16:22] didrocks: thanks - oh by the way I'm still running that cronjob that opens upstream tasks as it still seems to be finding some to open [16:23] bdmurray: you can stop it, I run mine delay [16:23] bdmurray: and it does that in addition to the rest :) [16:23] didrocks: okay, fair enough [16:23] daily [16:25] didrocks, did you see my email about incomplete/needs-design tag? [16:25] is there a specific channel for Unity, or is this the place? [16:26] lamalex: yeah, I don't really care about what workflow to use, just agree with design :) [16:26] ok :P [16:27] well it matter to you because you'd have to update your script [16:27] all design cares about is that they get an ayatana task [16:27] lamalex: sure, I just wait for the resul :) [16:27] result* [16:27] so you can you update the script to set unity to incomplete and add needs-design tag? [16:27] lamalex: basically, I copy design's task [16:27] lamalex: not for that case, in fact [16:28] so, ok, when there is something incomplete [16:28] my script look if there is an ayatana-design and tells if there are none "oh oh, I want to set that as incomplete but no ayatana-design task" [16:28] so, that's a manual task [16:28] (there is generally one every 3 days) [16:29] if there is already one, it set incomples (if the ayatan-design task is NEW or triaged) [16:29] when design set they task to fix committed/released, the script set the unity tasks to triaged [16:29] for existing ones now [16:29] yah [16:29] I always set the task to incomplete personnaly [16:30] kenvandine, what package do I need so that when I right click on an image file and i go to ubuntuone, that it allows me to share ? [16:30] lamalex: you meant that updating those which have ayatana-design task as NEW/triaged and unity ones as complete? [16:30] confirmed* [16:31] bcurtiswx, that is in the nautilus extension [16:31] (oh, also if ayatana-design set them as Opinon or won't fix, the script sets the unity tasks as Opinion or Won't Fix as well) [16:31] but i think the file has to be in a folder already shared [16:31] kenvandine, OK [16:33] didrocks, I mean instead of setting the unity status to opinion [16:34] i guess this isn't really related to your script [16:34] so nevermind actually [16:34] lamalex: not really I think :) === m_conley_away is now known as m_conley [17:06] lamalex: did your weird multimonitor bug get fixed? mine did, I'm at feature parity with 10.10 now [17:17] ok, got the cmake magic for glib-genmarshal!! [17:17] and people complain about auto*, but I find it easier than cmake [17:19] jcastro, i haven't actually plugged into my second monitor in a while [17:19] ugh, didn't get it fully :( [17:19] my desk is covered in crap so I've been working on the couch [17:19] rodrigo_: apinheiro: feedback from my weekly call: take some time to investigate making places accessible [17:19] dbarth-otp, ok [17:19] rodrigo_: apinheiro: i'll get back to you shortly [17:20] dbarth-otp, ok, I will take a look tomorrow morning [17:22] didrocks, ping... do you have any idea on what could be triggering this: http://susepaste.org/22607851 [17:23] nmarques: hey, oh? you're still packaging unity? As the fact that you told you were fed up made the headlines :) [17:23] didrocks, I don't might be a cultural issue there [17:24] didrocks, I said 'for the time being', which in Portuguese means, I'm leaving it behind for some time [17:24] didrocks, doesn't mean I ditched it for sure [17:24] nmarques: oh ok, I was surprise about it :) [17:24] didrocks, and press != opinion makers [17:24] nmarques: I'm at you in 10 min :) [17:24] just finishing something first [17:24] didrocks, np [17:36] nmarques: you need our patches in compiz [17:37] didrocks, I've also noticed my utouch stack isn't updated ;) [17:38] didrocks, updating it now, meanwhile is it possible to know which patches are _really_ required on compiz? [17:39] njpatel, https://code.launchpad.net/~rodrigo-moya/unity/sync-geometries/+merge/51929 <- could you have a look at the Geometry retrieval? I am getting wrong coordinates for some of the indicator objects/entries [17:40] njpatel, when you have time, I'm leaving now, so no hurry :) [17:40] nmarques: all the git ones 10* [17:41] didrocks, kk, will do [17:45] how much will Unity 2D differ from Unity? [17:47] Since they aren’t based on the same code with a layer of abstraction for Compiz+OpenGL and something-2D… Too much. :-P [18:01] http://ubuntuone.com/p/fmo/ anyone have that issue who runs chromium web browser ? [18:05] BTW that is a maximized chromium browser [18:06] bcurtiswx: same here, I reported a bug about it [18:06] omer confirmed [18:07] the only one who don't have it is smspillaz :/ === dbarth-otp is now known as dbarth [18:11] didrocks, OK good to know it's reported :) thx [18:16] since Unity 2D will be a lot different from Unity, I'd like to use Unity, but what are the minimum specs for Unity? [18:16] no wait, not minimum, the recommended specs [18:16] to have a smooth experience [18:19] bewdieloomba: you mostly just need to have a well-supported video card with 3D acceleration. [18:19] i got ati [18:20] It doesn't need to really be a powerful card, per se. My 6-year-old ATI Radeon 200M is doing pretty well so far. [18:20] which to date has worked out of the box on ubuntu, but unity might change that [18:20] why is unity 2d being developed then? [18:20] bewdieloomba: there are no 3D drivers for ARM. [18:20] ah [18:21] so unity 2d is a temporary solution to run it on ARM till ARM supports 3D? [18:21] bewdieloomba: I don't think Unity 2D even works on non-ARM. Someone feel free to correct me. :) [18:21] YESSS [18:21] then I can run Unity [18:21] bewdieloomba: I don't know if it's temporary, but yes it is a workaround. :) [18:21] omg i cant wait [18:22] My question for the Unity developers is: how does Unity determine the backlight color for launcher panels? [18:22] Unity 2D works pretty well [18:22] on i386 and x64 [18:22] the only thing i fear for 11.04 will be that the appmenu won't support libreoffice and chromium and other important apps like pidgin and eclipse [18:23] there is nothing to support in chromium [18:23] kklimonda: ooh. Will it be available as an automatic fallback? [18:23] nhaines: no idea, for now there is an additional Unity 2D session [18:23] bewdieloomba: LibreOffice appmenu support will be available as an optional package in Unity, but should be default by 10.10 I'd imagine. [18:24] I hope it will be always available - 2D is much better for battery on my hardware. [18:24] kklimonda: Oh, maybe I did see that one... I'll have to look closer, thanks. :) [18:24] ok [18:24] hopefully it'll be nailed down by 12.04 LTS [18:24] i gtg now but ill bbl [18:24] thx === m_conley is now known as m_conley_away [18:48] nhaines: the light takes like an average of all the colors in the app's icons [18:48] or something to that effect === ivanka is now known as ivanka-train [18:50] jcastro: that's what I thought although several icons looked suspiciously uniform in backlight color. [18:51] And the WorkSpace switcher seems very purple for all the grey in there. :) [18:51] well the app and files place are black and white on purpose [18:51] yeah, that's a special one too [18:51] I meant the File Manager, Firefox, and Ubuntu One icons. [18:51] jcastro: omg h4x! [18:51] heh [18:51] I turn the backlight off actually until the app is running (the option is in ccsm) [18:52] otherwise it's a rainbow over there [18:52] haha [18:52] Well, U1 dropped and the backlight turned red when the red emblem appeared, so I can believe it's an average. [18:53] my U1 icon hasn't done anything yet [18:53] the U1 control panel is beautiful though. :) [18:53] klattimer, that branch failed to build for me earlier, but i didn't look at fixing it yet [18:55] Looks like that train has left the station. [19:12] hey im back [19:27] is there a PPA for the appmenu? [19:27] id like to try out the latest version [19:35] didrocks: ping [19:42] DBO, you around [19:42] where do i find the pinned .desktop files [19:42] gconf or some file on the dekstop [19:42] seiflotfy, gsettings [19:50] kenvandine, adding the right click option for the launcher to gwibber, is there a reference site to add that functionality to empathy? I want to make it so on a right click if there are new messages that they will show there like the do in the indicator applet [19:51] as well as new messages and that sort [19:51] https://wiki.ubuntu.com/Unity/LauncherAPI === m_conley_away is now known as m_conley [19:51] i added the static ones, last i tried the dynamic quicklists didn't work [19:52] which is what you would need [19:53] i could do static ones for new messages.. [19:53] nope [19:53] the static ones are just in the .desktop file [19:53] so you need a way to launch the app into what you want the menu to do [19:53] let me rephrase, "Create New Message" [19:54] right, empathy doesn't have a way to do that from a command line [19:54] gwibber does 'eh [19:54] yeah :) [19:54] and evolution [19:54] it runs like evolution -c calendar [19:54] for example [19:55] and gnome-screenshot [19:55] has args to take a screenshot of just a window or full desktop [19:55] etc [19:56] there is an API for creating the dynamic quicklists though [19:56] but last time i tried it, it failed silently :) [19:56] maybe you will have more luck === ogra is now known as Guest42737 === Guest42737 is now known as ogra_ [19:59] i doubt if you have no luck that I will :P [20:00] guys, where's the best place to leave a patch for compiz (unity-window-decorator) which tweaks a bit the install ? [20:00] basically implements DESTDIR [20:03] kenvandine, empathy folks think GApplication should do most of the work to make empathy command line accessible [20:03] nmarques, file a bug and attach the patch [20:04] kenvandine: will do [20:06] tedg: ping [20:06] Howdy m_conley [20:07] tedg: hi! I was wondering - is show-hide-server in libindicate supposed to be showing and hiding the *visibility* of the server in the Messaging Menu? [20:07] tedg: because, as it stands, it does not. While it says it is hiding the test server, the visibility persists. [20:08] tedg: or am I misunderstanding the meaning of libindicate's server_show and server_hide? [20:08] m_conley, Yes, it's broken. No one was using it, so I hadn't prioritized fixing it. Do you need it for something? I can push it up in priority. [20:09] tedg: it's not an earth-shattering, stop the presses priority. Were you hoping to get it pushed for alpha 3? [20:09] m_conley, No, A3 is already frozen. [20:10] m_conley, We'd need to be killing kittens at this point to get it in A3 :) [20:10] tedg: ah, k. Wasn't sure how close you were to getting alpha 3 out [20:10] kenvandine: LP #728015 [20:10] Launchpad bug 728015 in compiz-fusion-plugins-main (Ubuntu) "unity-window-decorator cmake install doesn't use DESTDIR" [Undecided,New] https://launchpad.net/bugs/728015 [20:11] ups [20:11] wrong place :/ ffs [20:11] tedg: is there a bug # for server_hide/show? [20:11] tedg: if so, maybe I'll just monitor that. [20:12] m_conley, One sec, let me find it. [20:14] m_conley, Uhg, I can't seem to find it. It must be misfilled :-/ [20:15] I doubt someone fixed it for me :) [20:15] tedg: Ok, I'll take a look around and see if I can find the bug. [21:01] kamstrup: how do I run the python place example? [21:02] mhall119: so there are a few steps at this point in time unfortunately - but here goes [21:02] Download http://bazaar.launchpad.net/~unity-team/dee/trunk/view/head:/bindings/python/Dee.py [21:02] I got that [21:02] And sudo cp Dee.py /usr/lib/pymodules/python2.7/gi/overrides/Dee.py [21:03] Then makre sure you have libdee >= 0.5.10 and libunity >= 3.4.6 [21:03] and the packages gir1.2-unity-3.0 [21:03] gir1.2-dee-0.5 [21:03] okay, I'm running Natty daily as of yesterday [21:04] mhall119: that should be good enough [21:04] mhall119: so now bzr branch lp:~unity-team/unity-place-sample/unity-place-python [21:04] got that [21:04] I followed along in the session [21:05] Then follow the README in there [21:05] or see it here http://bazaar.launchpad.net/~unity-team/unity-place-sample/unity-place-python/view/head:/README [21:05] copied the .svg and .place [21:05] mhall119: cool [21:05] oh, I didn't restart unity [21:05] mhall119: yeah, that's the crucial point [21:05] mhall119: we want to make it so unity discovers new places on runtime, but that's not yet done [21:06] so a restart is required [21:06] mhall119: there is also the slight gotcha that the place daemon needs to be running when you restart [21:06] mhall119: because we hack this daemon into the system really - it can't do proper dbus activation [21:07] which unity expects [21:07] so just keep it running and unity will think all is good :-) [21:07] ah, cool, it's working now [21:07] mhall119: wow, that's awesome to hear! [21:07] are places always grey in teh launcher? [21:07] mhall119: not necessarily [21:08] mhall119: i just tweaked the icon to be in line with the rest of the places [21:08] ok [21:08] mhall119: but it's awesome you have it working! [21:08] mhall119: all this code just came together today - so this is the bleeding edge of the bleeding edge [21:08] gah, unity keeps crashing on me [21:08] but hopefully everything will be in the repos by monday [21:08] Is anyone else having a problem with rhythmbox and unity? Whenever I minimize/close it, unity treats it still as the focused window, despite it not showing up. [21:09] mhall119: oh, bugger... i'd guess it's unrelated to the new place? [21:09] kamstrup: probably, it's been crashing on me regularly since I installed [21:09] running on an EeePC 701 netbook [21:09] mhall119: since you installed the place daemon? [21:09] since I installed Natty [21:10] mhall119: phew - you had me scared :-) [21:10] Bleh. Same thing happens to banshee [21:10] not your fault [21:10] mhall119: which video card and driver? [21:10] well, maybe your fault,but I can't be sure [21:10] jono, ping === marcos_ is now known as Daekdroom [21:10] andreasn, hey [21:10] mhall119: lol - you never know :-) [21:10] kamstrup: Intel 915GM [21:11] mhall119: hmm - i think I recall some bugs mentioning i915 [21:11] I'm also running at 800x480 res [21:11] i have i945 here and it works very well [21:11] jono, hi! Is the correct name "Messaging Menu" or "Messaging Indicator"? [21:11] mhall119: yeah, i know there was some performance issues on i915 [21:11] performance has been good, actually [21:11] andreasn, hmmm not 100% sure, I think it is Messaging Indicator [21:11] jono, my Internet suggests both [21:11] it's just the crashing [21:12] and things being cut off because of my low screen res [21:12] tedg, ^ [21:12] can you answer andreasn [21:13] andreasn, Messaging Menu -- though we're sadly very inconsistent. :( [21:13] kamstrup: this should get me started at least,thanks [21:14] mhall119: most welcome [21:14] tedg, thanks! [21:15] kamstrup: okay, now to make the python sample stop appearing, I have to remove the .place file? [21:15] andreasn, Basically the whole thing is the "Menu Bar" with the various menus on it. But the packages are all named "indicator" as that's what we called it before someone with naming skillz (mpt) got ahold of us :) [21:15] mhall119: yep, that's all [21:15] aha [21:15] mhall119: i'll just put that in the README, it's good info [21:16] kamstrup: woo! [21:16] then setsid unity again? [21:17] mhall119: that should do it [21:17] kamstrup: I am thinking for tomorrow, since it's A3, a nice blog post on how to do this, step by step would be awesome [21:17] and it did, thanks [21:17] and then we can get people started on doing Places [21:18] jcastro: yeah, I want to do that [21:18] jcastro: I also need to throw mad props to the PyGI team [21:18] man, alpha 3 already, I need to get cracking on Qimo [21:18] kamstrup: I'll remind didrocks about making the quickly template do /opt or whatever so we can get people rolling on them even when we start to freeze [21:26] jcastro: the issue here is dbus activation... [21:26] jcastro: normally dbus only looks for .service files under /etc [21:27] jcastro: so that's a headache for the desktop team :-) or maybe platform... [21:27] I thought he had a solution for that, anyway, I'll ask him tomorrow [21:27] we'll worry about that, you just make sure it works. :) [21:27] jcastro: maybe it's not an issue anymore - that would be awesome [21:27] and if we can't sort it maybe we can ppa it or something === m_conley is now known as m_conley_away [22:08] Ok. Where do I go to make unity suck less on multiple heads? === rickspencer3_ is now known as rickspencer3 [22:18] RAOF, I thought that was all fixed? [22:19] Not really. It is slightly less hateful, in that Unity now knows where the edges of my screen are. [22:20] But it still explodes spectacularly if you try to *change* the screen geometry, and there are a bunch of other miscelaneous problems if your primary monitor doesn't happen to be left-most. [22:20] DBO, you around [22:21] seiflotfy, no [22:21] RAOF, baby! [22:21] Aha! There's the man who knows where to point me to make things work! [22:21] DBO, i am having issues with the python bindings for dee [22:21] for some reason my system cant find them [22:21] although they are copied and installed from source [22:21] seif@seif-VirtualBox:~/Projects/unity-place-python$ ./unity-place-python.py [22:21] ERROR:root:Could not find any typelib for Dee [22:21] Traceback (most recent call last): [22:21] File "./unity-place-python.py", line 5, in [22:21] from gi.repository import Dee [22:21] ImportError: cannot import name Dee [22:22] seiflotfy, asking my about python? you might as well be talking about ancient greek.... [22:23] RAOF, okay so your problem lies in the fact that unity makes no attempt to deal with screen geometry changes yet [22:23] it shouldn't be *that* difficult [22:24] ISTR a compiz hook that should be usable to make that happen. [22:25] even better [22:25] hold on [22:25] So, which part of the unity stack do I need to prod? Unity itself? Nux? [22:26] RAOF, unityshell.cpp [22:26] OnSizeChanged [22:26] soooo from there you should be able to start debugging :) [22:27] oh and yes... I know it's not much of a lead, but in fairness, I have not begun debugging that yet [22:28] RAOF, so here is the second part of the issue [22:28] in nux, NuxGraphics/GraphicsDisplayX11.cpp [22:29] case ConfigureNotify: happens [22:29] seiflotfy: do you have gir1.2-dee-0.5 installed? [22:29] except it filters out foreign events (when running embedded mode this is true) [22:30] ROAF, in reality this needs to allow to happen when foreign is false OR the window in question is the m_X11Window [22:30] this will cause nux to resize its internal buffers [22:30] Which will presumably stop the new screen being a fetching shade of black :) [22:30] more than likely it will cause a segfault [22:30] thats my bet [22:31] then comes the fun part [22:31] its owkring now [22:31] kklimonda, thanks for helping seiflotfy [22:32] no problem [22:32] DBO, kklimonda i got a surprise for you guys :) [22:32] seiflotfy, put your pants back on [22:32] cookies? [22:33] DBO, but i dont like my pants [22:33] kklimonda, your mind doesn't go bad places immediately, what a blissful world this must be for you [22:36] DBO: if you only knew. I've just learnt not to write the first thing that comes to my mind ;) [22:36] kklimonda, a hard learned lesson I intend to never learn [22:50] Man a lot of GraphicsDisplayX11 is commented out :) [22:59] Man, this is going to be much more fun when my ccache is populated with nux objects. [22:59] RAOF, at no point in time is fun to be had whilst working on nux [23:00] if you find yourself having fun, remember I can always ask you to fix some race conditions we have with XDND [23:00] It looks like it should probably be using intel_swap_events where possible, too. [23:00] wtf is intel_swap_events? [23:00] They're basically acks that a buffer swap has completed. [23:01] So you don't drive your rendering pipeline at > SwapBuffers fps. [23:01] that would be done in compiz then wouldn't it? [23:01] i mean thats where all of our paint calls come from [23:03] fooooood time [23:03] Probably. But the comments around GraphicsDisplay::SwapBuffer that suggest that someone would like something like swap_events. [23:04] Woo! We have a nux build. [23:11] DBO: Well, that was easier than you thought it would be. [23:13] I mean, apart from unity having no idea about how it needs to change the layout, of course. [23:13] But at least the second monitor actually displays stuff :) [23:39] By the way, cmake is terrible. [23:40] And I hate you!