[08:46] morning all [09:38] Saviq, don't blame it on proposed migration, that's not an issue there, if bits moved separatly and shouldn't have then your depends/conflicts/... are buggy, if you have components that need to go together the packaging should enforce that [09:39] pstolowski, ^ [09:43] seb128, yeah, agreed... [09:44] seb128, sorry, didn't mean to blame, was cause instead [09:44] let me clarify [09:44] Saviq, no worry, but thanks === Malsasa_ is now known as Malsasa === Malsasa_ is now known as Malsasa [10:26] can we retrigger https://code.launchpad.net/~aacid/unity8/moreAsyncDash/+merge/241524? [10:28] Cimi, trying [10:29] Cimi, nope, CI still down [10:39] Saviq, is tsdgeos off today? [10:39] pstolowski, yes [10:39] pstolowski, back tomorow [10:39] ok [11:08] Cimi: biw, if both sides of == are real ints, there's absolutely no difference between == and === [11:09] mzanetti, wasn't === always skipping the conversion? [11:09] Cimi: yeah, but if you compare "property int x" to 5, there isn't any conversion ongoing anyways [11:10] Cimi: it's different if you compare "property int x" to "5", e.g. when the "5" comes as a string from a textfield [11:10] mzanetti, yes but === is better no? [11:10] not really :D [11:10] I would use == only when needed [11:10] otherwise === [11:10] it's the same, and when it's not the same you gotta know which one you want, there isn't a better or a worse one :D [11:11] Cimi: anyways... I changed it in that branch and added a test as you requested [11:11] mzanetti, hoping ci will run... [11:11] yep [11:19] Cimi, no it won't, CI is down still [11:56] Cimi: one last small thing in here, then I'll approve: https://code.launchpad.net/~cimi/unity8/fix-1368778-2/+merge/242942 [11:56] greyback_, what's your zsh theme? couldn't find it in oh-my-zsh [11:57] mzanetti, I'll add to the silo then [11:57] Saviq: cimi's branch? [11:57] mzanetti, ok [11:57] mzanetti, yeah [11:57] ack [11:57] Cimi, please let me know when it's ready [11:58] Saviq: am using slightly tweaked https://gist.github.com/agnoster/3712874 [11:58] Saviq, well, change will be just in the test so... [11:58] Saviq, if you want to do it now or wait 10 mins [11:58] :) [11:58] Cimi, still, I need to wait for the code to be in LP before building the silo [12:00] greyback_, hmm thought it would be that, the |> bit looks like a bad character :? [12:00] Saviq: did you do the powerline font hack? [12:00] greyback_, did not [12:01] that's it, it patches your font to add the extra characters needed [12:07] now where do I change my fonts.. [12:09] greyback_, that worked, thanks [12:11] Saviq: np, hope you like it [12:11] greyback_, still getting used to zsh, but a few things (like wd) are just awesome [12:11] yeah [12:11] tab completion a bit buggy for me still [12:12] Saviq, mzanetti done [12:15] approved === _salem is now known as salem_ [12:18] greyback_, yeah, could definitely use more of that [12:22] Saviq: hey, bug 1394645 is fixed in trunk in revision 1436. Can you add it to list of things to be backported? [12:22] bug 1394645 in The Webapps-core project "OSK doesn't appear after OA login" [Critical,Confirmed] https://launchpad.net/bugs/1394645 === dandrader is now known as dandrader|afk [12:26] greyback_, right, good catch === dandrader|afk is now known as dandrader [13:09] mzanetti, what's the plan to enable desktop mode by default on desktops? [13:10] Saviq: we didn't agree on anything tbh [13:10] well, conclusion was that it would be great to set the dconf key on login [13:11] but that's not working in dconf yet [13:11] and there seems to be confusion about whether it'll land or not soon [13:11] mzanetti, can you file a bug ainst unity8 (Ubuntu) then so we can track progress? [13:12] yeah... can do [13:15] Saviq: https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1400312 [13:15] Launchpad bug 1400312 in unity8 (Ubuntu) "On desktop, windowed mode should be by default" [Undecided,New] [13:15] mzanetti, tx [13:31] Saviq, mzanetti, easy step 1 would be to set the key in "ubuntu-settings" [13:31] until we get per-session defaults [13:31] that would let desktop installs have it set easily [13:34] seb128, sure, please comment on the bug [13:35] Saviq, I did (just before commenting here) [13:36] ok. [13:37] Saviq, mzanetti, I can land that change if you want [13:37] works for me [13:37] k [14:05] mzanetti, any eta on when this will land? [14:05] 79+14.10.20141007-0ubuntu1 [14:05] [07:25:13] gcollura, same here... 1.1.1279+14.10.201 [14:05] sorry; https://code.launchpad.net/~mzanetti/unity8/desktop-stage/+merge/242140 === salem_ is now known as _salem === dandrader_ is now known as dandrader [14:19] paulliu, would you have some spare time to review https://code.launchpad.net/~dandrader/unity8/properInSceneDialogs/+merge/243998 ? [14:23] mterry, goot morning. Updated the greeterRefactoring branch and replied to all your questions [14:23] s/goot/good [14:23] dandrader, oh awesome, will look [14:23] mterry, made it Showable again :) [14:24] dandrader, I didn't mean to be down on some of your refactors, I just wasn't clear on why some changes were made (and have a vested interest in less churn in that component ahead of my own refactor ;)) [14:25] dandrader: sure. WIll do it after stand-up. [14:25] paulliu, thanks! [14:33] dednick, hey. these indicator profiles. what values are supported and where are they defined? [14:35] dandrader: hm. not sure they're specifically defined anywhere. tedg? [14:35] dandrader: can take a look at the values in the indicator files. /usr/share/unity/indicators [14:35] * dandrader checks... [14:36] dandrader: /usr/share/unity/indicators/com.canonical.indicator.network is probably the most exhaustively defined. [14:37] We've been a bit ad hoc there. [14:37] dednick, so the profiles are the values between [ and ]? like phone, phone_greeter, ubiquity, etc? [14:38] dandrader: yup [14:38] We really only have [phone|desktop][_greeter]? === _salem is now known as salem_ [14:38] Ah, and ubiquity. [14:38] tedg, I don't know. you tell me :) [14:39] I would but you left! ;-) [14:40] dandrader, So I guess the question for me is, what is the list useful for? [14:40] Are you trying to make an enum? [14:40] Or do we need to ensure a list is in all hte indicators? [14:41] balloons: in a silo already [14:41] balloons: depending on how testing goes, between today and wednesday [14:42] tedg, I'm coming up with a device configuration thingy for unity8 which I'm thinking could also contain that indicator profile name instead of having unity8 fetching it from a UNITY_INDICATOR_PROFILE env var [14:43] tedg, to consolidate those device-specific (or form-factor specific, "usage scenario"-spefic) things in a single place [14:44] dandrader, Cool, make sure to base your branch on mterry's greeter branch as I know he changes some of that logic for the greeter modes as well. [14:44] tedg, do you know who sets this UNITY_INDICATOR_PROFILE env var? [14:44] dednick? ^ [14:44] dandrader, So I think we can be flexible to what naming makes sense for you. I'd rather be consistent with the names than use the ones we have. [14:44] dandrader, I imagine that's ubuntu-touch-session today [14:44] dandrader, that branch is https://code.launchpad.net/~mterry/unity8/greeter-profiles/+merge/237155 [14:46] dandrader: this in a config file somewhere, or command line? [14:47] dandrader: the indicator-client takes a -profile commandline option [14:47] mzanetti, ack. I'll follow the silo ty [14:48] dandrader: but i'm presuming the env var intention was for upstart config to set it. [14:48] dandrader, For instance, if you get away from the phone/desktop naming (which I know that Saviq is passionate about) I'd like that reflected in the indicator profile names. [14:49] dandrader, looking at your branch again; it's super annoying that qml doesn't have virtual methods that can be overridden [14:50] mterry, it's a javascript trait I guess [14:59] dednick, I think this UNITY_INDICATOR_PROFILE env var is not being set or used currently [15:00] dednick, couldn't find it mentioned anywhere in /etc or /usr and it's also not in the phone's env [15:06] dandrader: they're set in the upstart conf [15:07] greyback_, where exactly? which file in the device filesystem? [15:07] greyback_, grep could not find anything, as I said .... [15:08] greyback_, and env command also doens't show it. unity8 defaults to "phone" is that env var is not present. I guess this is what has been happening [15:09] s/if that/if that [15:09] dandrader: it's not used on phone, but on desktop [15:09] ahhhh [15:09] UNITY_INDICATOR_PROFILE=desktop is set for unity8 on desktop, am looking to see what sets it [15:10] my mistake, isn't set in upstart conf [15:11] greyback_, couldn't find any mention of it in my desktop either... [15:12] dandrader: /usr/bin/lightdm-unity8-session [15:13] greyback_, don't have that file. what package adds it? [15:13] dandrader: unity8-desktop-session-mir [15:14] greyback_, got it. so that's likely the only use of it anywhere I guess [15:14] yeah === salem_ is now known as _salem [15:25] dednick, Is qml/Panel/Indicators/client/IndicatorsClient.qml dead code? [15:26] dednick, ah, it's loaded by src/Panel/Indicators/client/indicatorsclient.cpp. nevermind [15:26] dandrader: ok. [15:26] dandrader: and the UNITY_INDICATOR_PROFILE env var is for overriding the default "phone" profile. [15:27] dandrader: or rather. if the env var isn't set, it uses "phone" [15:27] dednick, yeah, saw that in main.cpp === jhodapp_ is now known as jhodapp === dandrader is now known as dandrader|lunch === _salem is now known as salem_ [16:03] dandrader|lunch: what's dialogLoader in ShellDialog.qml? [16:04] dandrader|lunch: I mean, line 291 [16:07] dandrader|lunch: Do you need to set dialogLoader as a property? === salem_ is now known as _salem [16:16] bah [16:17] Saviq, bregma, after upgrading my vivid unity8 test machine, I can't run any application on unity8 anymore, is that a known issue? [16:17] willcooke, ^ works for you? [16:17] seb128, _any_ application? [16:17] yeah [16:17] seb128, my installation is b0rked from playing with mlankhorsts ppa - so I'm a bad example [16:18] bregma, I tried webbrowser, notes, system settings [16:18] start with gedit/eog because I wanted to test gtk stuff [16:18] then went back to try default ones [16:19] the unity8 log has "Unable to get cmanager connection" errors [16:19] and "Unable to get pids for unity8-dash to send signal 18" [16:19] that's new [16:19] is it working for you? [16:20] I haven't updated since last week, it will take a while [16:20] k [16:20] let me know if you try [16:20] http://paste.ubuntu.com/9428906/ [16:20] is my unity8.log [16:20] ChrisTownsend, have you had trouble with the latest Unity 8 desktop? [16:21] "Unable to get pids for unity8-dash to send signal 18" - is nothing to worry about really [16:21] k, so maybe something else [16:21] we need to quieten that [16:22] "Unable to get cmanager connection" - that's something I've not seen before [16:22] seb128: ^ [16:23] qtmir.applications: ApplicationManager::onProcessFailed - appId= "webbrowser-app" duringStartup= true [16:23] webbrowser crashed? [16:23] I don't think so [16:23] bregma: No, it seems to work...and this includes the latest Xmir stuff. [16:24] I click on any icon, see a rectangle start sliding from the right for less than a second and then it closes [16:24] ChrisTownsend, what unity8 version do you have? [16:24] seb128: 8.01+15.04.20141205-0ubuntu1 [16:25] seb128: Seems the cgroups are not right and ubuntu-app-launch is rejecting the app. [16:25] seb128: can you launch "QT_QPA_PLATFORM=ubuntumirclient messaging-app --desktop_file_hint=/usr/share/applications/messaging-app.desktop" [16:25] asuming it's installed [16:25] seb128: What version of cgmanager are you running? I have 0.33-2. [16:28] ChrisTownsend, greyback_, Saviq, bregma, ignore that, my fault for switching to systemd as init, seems there is something not working correctly with unity8 (or cgmanager) [16:28] ah, systemd [16:28] seb128: ack:) [16:29] seb128: ah yeah, tha'll do it. [16:29] do we have a bug open for that? [16:30] if not I'm going to open one, we plan to switch this cycle and that should work [16:30] seb128, perhaps unrelated. I just upgraded my u8 machine and now the option to use u8 is gone from the greeter [16:30] willcooke, yeah, unrelated [16:31] willcooke, is unity8-desktop-session-mir still installed for you? [16:33] seb128, seems not. Tried apt-get install unity8 and bad things: [16:33] The following packages have unmet dependencies. [16:33] unity8 : Depends: qtdeclarative5-ubuntu-web-plugin but it is not installable [16:33] Depends: ubuntu-system-settings but it is not going to be installed [16:33] Depends: unity8-common (= 8.01+15.04.20141205-0ubuntu1) but it is not going to be installed [16:33] Recommends: unity-scope-click but it is not going to be installed [16:33] E: Unable to correct problems, you have held broken packages. [16:33] probably because I'm using proposed [16:33] willcooke, those are fun [16:33] willcooke, yeah [16:34] :) [16:34] willcooke, usual way is to append things it lists on the install command, until it gives you a real error [16:34] I'll give it a go [16:34] like "apt-get install unity8 unity8-common qtdeclarative5-ubuntu-web-plugin ubuntu-system-settings" [16:34] and keep going if it gives you other similar errors [16:34] bingo: [16:34] qtdeclarative5-ubuntu-web-plugin [16:35] when it gets bored, apt give you a real reason :p [16:35] has no installation candidate [16:37] that's weird, "apt-cache policy qtdeclarative5-ubuntu-web-plugin" ? [16:39] qtdeclarative5-ubuntu-web-plugin: [16:39] Installed: (none) [16:39] Candidate: (none) [16:39] Version table: [16:39] 0.23+15.04.20141202.1-0ubuntu1 0 [16:39] 100 /var/lib/dpkg/status [16:39] seb128, ^ [16:41] willcooke, do you have a vivid source enable? or did you replace it by proposed? [16:41] willcooke, proposed is not a full archive it's an overlay [16:41] it contains only some packages [16:41] I replaced everything with -proposed [16:41] you need a vivid source [16:41] yeah, that's your issue [16:41] seb128, gotya [16:41] thx [16:41] yw [16:41] greyback_, bregma, just for the record, opened bug #1400394 [16:41] bug 1400394 in systemd (Ubuntu) "Unity8 fails to start applications under systemd init (cgmanager issue?)" [Undecided,New] https://launchpad.net/bugs/1400394 [16:42] seb128: ack === _salem is now known as salem_ [16:50] seb128, kinda fixed - in that I can now at least select a U8 session from the greeter and u-s-c looks like its started, but no desktop [16:50] seb128, I'll worry about this tomorrow, I've got other things to be getting on with# [16:51] willcooke, k, if you want/can share your .cache/upstart/unity8(-dash).log [16:57] greyback_, just saw your comment, systemd being used for init doesn't prevent upstart to work/be used in the session, user session keeps working fine [16:57] seb128: ah fair point [16:57] greyback_, are you sure the issue is ubuntu-app-launch? [16:57] hadn't thought of that [16:58] that seems an orthogonal issue to me [16:58] seb128: well I assumed we were only using 1 init system :) [16:58] we need to make ual use systemd, but it should work with upstart meanwhile [16:58] we are [16:58] upstart is used as a job manager for the session, not as an init :p [16:58] but yeah, I see what you mean ;-) [16:58] you could call it a session init :) [16:58] but yeah, I'm wrong [16:59] feel free to correct me in the bug so [16:59] just did so [16:59] thanks [16:59] thank you === dandrader|lunch is now known as dandrader [17:11] paulliu, replied in the MP === alan_g is now known as alan_g|EOD [18:08] ping Saviq, or somebody else: is there a way to start the edge demo on demand? [18:08] I want to initctl start unity8, and then start the edge demo on that window. [18:10] mterry, all fixed. [18:11] dandrader, ok looking! [18:12] elopio, there's a gsetting [18:13] elopio, phablet-config enables it [18:14] Saviq: thanks. Will try that. [18:14] elopio, actually, not a gsetting, accountsservice property [18:16] yes, looking at it. [18:17] dandrader, looks good code wise! will just test around with it and should be good [18:20] dandrader, we should "tease" on a drag start too -- last time we talked with design they wanted that, I believe because a new user may accidentally do a baby drag and not understand what's going on or what would happen if they finished the drag [18:22] dandrader, and I'm seeing a local failure in TabletShell::test_leftEdgeDrag [18:34] mterry, is it possible to manually reproduce "TabletShell::test_leftEdgeDrag(with demo)" ? [18:34] mterry, I mean, with "make tryTabletShell" [18:34] oh let me try [18:38] dandrader, you can but you have to futz with the AccountsService mock to have it show the demo by default. But I see the problem. [18:38] dandrader, swipeFromLeftEdge starts the drag at x=2 [18:38] mterry, it's not. would have to add a button for "AccountsService.demoEdges = true" below the "Show greeter" one [18:38] dandrader, if you make that larger, it works [18:39] mterry, ahhh, right [18:39] dandrader, looks like there's a dead zone on the left (for the launcher?) [18:39] dandrader, yeah that button would help for sure [18:39] mterry, I added a margin on the left for the Greeter DragHandle so that it doesn't overlap with the Launcher one [18:45] mterry, so, is it a valid use case to swipe the greeter from he left edge with the demo running? [18:45] mterry, ie, with launcher disabled [18:45] dandrader, yes, that's what the first page of the edge demo asks the user to do (swipe the greeter away) [18:45] dandrader, i mean [18:45] dandrader, it asks from the right edge [18:45] mterry, the first page asks for a right edge swipe [18:46] mterry, exactly [18:46] dandrader, I haven't confirmed with Design if they care about allowing a left edge swipe in that case. But we always have allowed it there [18:46] mterry, so, should I make a "right edge swipe with demo on" instead? [18:47] dandrader, I suppose we should at least capture that... But left edge is also good to test. I think we just used the left edge tests as stand-ins for both edges [18:47] dandrader, but if we only have one, maybe you're right that it's better to test right edge [18:48] mterry, the left *edge* belongs to the launcher [18:48] dandrader, well fine... the test captures swiping right from the left [18:48] That's a matter of the test name [18:48] That's why it starts at x=2, because it's specifically *not* testing the left edge, but "from the left" [18:49] oh wait [18:49] maybe it is testing the left edge [18:49] and is meant to be a launcher drag [18:49] * mterry looks at tests closer [18:49] * dandrader checks what happens in trunk [18:49] dandrader, yeah. that test is specifically for doing launcher drags [18:49] dandrader, it has a comment about "to give time to handle dash() signal from Launcher" [18:50] I don't know why they picked x=2 then instead of x=0. But I guess they were always equivalent [18:51] dandrader, so for some reason in your branch, the launcher is disabled in that situation [18:51] mterry, just tested with trunk [18:51] ok, what does it do? [18:51] mterry, you can do a drag from the left *edge* with demo on and there will be no launcher [18:51] mterry, and there's no such use case I think [18:52] dandrader, oh right because in demo mode we disactivate launcher when we aren't demo-ing it [18:52] dandrader, what do you mean no use case? It caught the fact that your branch has a dead zone on the left, which it shouldn't have [18:54] mterry, being able to dismiss the greeter with a left egde drag when the demo is asking the user to dismiss the greeter with a right edge drag [18:55] dandrader, that doesn't bug me, but that's a question for design. That's how trunk and rtm work right now. I'd rather make that a separate/intentional change, than a by-product of a refactor. I personally think it's better to have the greeter act like it would normally (i.e. allow the rightward drag) [18:56] mterry, ok, I will make it work again [18:56] dandrader, Design is actually dropping that tutorial screen anyway [18:57] dandrader, the next version of the tutorial doens't have it anyway [18:57] I wonder if they are done with that yet [19:00] mterry, fixed. [19:23] Does anybody else occasionally get "/home/mike/Work/code/unity8/greeter-refactor/builddir/tests/mocks/Unity/Indicators/libIndicatorsFakeQml.so: undefined symbol: _ZN14UnityMenuModel12setModelDataERK8QVariant" when running ./run.sh -f? [19:24] I feel like I'm building wrong, but it stays even after a fresh build [19:37] Trying to xcompile unity8, whats the deal with libconnectivity-qt1-dev ? [20:58] josharenson, what about it? /me tries [20:59] saviq, fails when installing === salem_ is now known as _salem [21:35] josharenson, looks like there's a bad dep chain there :/ [21:37] saviq, any remedy to that? [21:37] josharenson, let me try one thing [21:40] josharenson, doesn't look like it, it's a bigger issue [21:40] :-/ [21:40] ok [21:41] josharenson, fwiw, bug #1400502 [21:41] bug 1400502 in connectivity-api (Ubuntu) "Can't install libconnectivity-qt1-dev on multiarch" [Undecided,New] https://launchpad.net/bugs/1400502 [21:42] thanks [23:21] There is no way to connect to dbus with qml? [23:54] josharenson: the primary way we do it is to write a small qml plugin in C++, which is a thin wrapper over the Qt dbus api bits. The plugins directory has many of them, plugins/Unity/DashCommunicator is a nice one to get started with [23:55] greyback, thats exactly what I'm doing, just didn't want to unintentionally overengineer [23:55] thanks [23:55] np