[09:52] zsombi: tmoenicke: do you know if there's a way to know when the OSK has completely finished the animation that it does when it slides in ? I noticed that the Qt.inputMethod.visible property changes to true immediately, but the keyboard takes a bit longer to slide in (and if MainView.anchorToKeyboard = true the height of the MainView gets animated) [09:52] zsombi: tmoenicke: i need to do something when i am sure the keyboard has completely appeared [09:57] nerochiaro: at the time visible is true, the osk has already allocated the window and space [09:59] nerochiaro: it doesnt report anything else [10:01] tmoenicke: i just checked and i think you're mistaken. if i print the height when Qt.inputMethod.onVisibleChanged reports that visible = true I get one height, if I wait a few milliseconds and print the height again i get another height. I'm printing it on a listview that's occupying the entire mainView, but let me try to do the same with the MainView itself, just to be sure [10:03] tmoenicke: hmm,interesting. the MainView does indeed behave like you said. so there's a delay somewhere else. thanks [10:06] tmoenicke: zsombi : it's the Page object that takes some time to resize to match the keyboard === chriadam is now known as chriadam|away === t1mp_ is now known as t1mp [10:08] t1mp_: zsombi: do you guys know how can I know when a Page has fully adjusted to the height of the OSK when MainView.anchorToKeyboard: true ? It seems to take some time to animate to the right size and I need to know when it has finished [10:10] nerochiaro: I don't know [10:11] t1mp: do you know who worked on that and might know ? [10:12] nerochiaro: from bzr log I'd say renato [10:38] oSoMoN: i just pushed an update to the MR and added some comments [10:38] oSoMoN: well, adding them now [10:40] oSoMoN: ok, done [10:47] nerochiaro, looking [10:50] nic-doffay: It seems that the tests pass in your https://code.launchpad.net/~nicolas-doffay/ubuntu-ui-toolkit/selector-code-refactor/+merge/191953 branch. Is this ready for merge to trunk [11:00] nik90, yeah just waiting on timp to review it again sometime. [11:01] nic-doffay: okay [11:16] nic-doffay: I added a few comments https://code.launchpad.net/~nicolas-doffay/ubuntu-ui-toolkit/selector-code-refactor/+merge/191953 === _salem is now known as salem_ [11:29] Good morning all; happy Guy Fawkes Day! :-D [11:38] JamesTait: I'm trying to approve an app into the store and I'm getting "Application status could not be updated. Please try again later." [11:38] Also, good morning. [11:39] popey, good morning! Let me see what I can find out for you. [11:40] popey, this is in production, yes? [11:40] nerochiaro, more comments on your MR (and fixes needed) [11:41] JamesTait: i only use production [11:42] popey, http://blogs.msdn.com/cfs-filesystemfile.ashx/__key/communityserver-blogs-components-weblogfiles/00-00-01-32-02-metablogapi/8054.image_5F00_thumb_5F00_35C6E986.png ? [11:43] \o/ [11:43] oSoMoN: the solution you propose to replace the hack will probably not work. it's not the keyboard which has a delay in getting the right size,it's the MainView's contents [11:44] nerochiaro, right [11:44] oSoMoN: i think there's actually a bug in the MainView, i need to talk to renato. That animation should not be there as it's happening when they keyboard is already up anyway, if what tmoenicke says about Qt.inputMethod.visible being emitted only when the keyboard is already up [11:45] nerochiaro, I don’t like this hack though, there’s gotta be a cleaner way [11:45] nerochiaro, ok, so let’s file a bug and fix it as a first step [11:45] oSoMoN: let me talk to renato first [11:46] oSoMoN: that hack is just an interim fix so that at least we get the thing working. before something gets fixed in the SDK and when it gets into the images ages can pass [11:47] nerochiaro, right, but if it’s not a critical bug then we can probably afford to fix it in the SDK first [11:48] oSoMoN: it's High, seems already pretty important to me, but i'm ok either way [11:49] nerochiaro, let’s file the bug against the SDK and talk to Renato first, and then we can decide which route to take [11:50] nerochiaro, and regarding the thing not working in landscape, it may also be a bug in the MainView itself, that’s something to check [11:51] oSoMoN: i was about to say that I'm gonna investigate that one, and make a report with a test case if it's indeed a bug === MacSlow is now known as MacSlow|lunch [12:29] renato_: ping [12:31] nerochiaro, hi [12:31] renato_: hi, i wanted to ask you something about MainView and anchorToKeyboard [12:32] renato_: I see that you have added a NumberAnimation to animate the movement of the MainView content [12:32] renato_: but it seems that it's finishing a bit after the keyboard is fully shown [12:33] renato_: is that animation necessary ? === greyback is now known as greyback|lunch [12:42] nerochiaro, this is the best way that I and Kaleo found to make a smooth transition [12:43] nerochiaro, right now there is no way to follow the keyborad animation, and I do not know if they intend to change that [12:43] oSoMoN: ^ [12:43] but for sure the best solution is follow the keyboard [12:45] renato_: that's not enough. when they keyboard has finished its animation the content is still changing [12:46] renato_: the MainView content doens't have the final height when Qt.inputMethod.visible is true [12:58] renato_: oSoMoN: lunch, let's talk about this in a bit === nerochiaro is now known as nerochiaro_lunch === MacSlow|lunch is now known as MacSlow [13:27] popey, mhall119: Hi, I was just thinking about exposing bite size bugs for the entire core apps project. Currently there is no easy way to showcase these bugs to new contributors. Sure you can tag bugs as "bitesize" etc..but not everyone will take the effort to go through launchpad bug filters to find them. [13:27] popey, mhall119: That said, why cant we use http://harvest.ubuntu.com/opportunities/ ? [13:27] May be add a new filter touch-coreapps which would list all bitesize bugs or i18n bugs in the core apps projects? [13:31] Sounds like a great idea [13:32] could add a bitezise-touch tag just as easily. [13:32] and link to it [13:32] add a bit.ly link all over the place to make it easier to find === greyback|lunch is now known as greyback [13:51] nik90: +1 on using both bitesize tag and harvest, though I don't know how much maintenance harvest has gotten lately [13:56] popey: I tried bit.ly and it is good but it returns links like http://bit.ly/1cLYGG3 which is not intutive..whereas with harvest I could just type go to harvest.ubuntu.com which is much easier to read and remember. [13:56] mhall119: agreed. But it seems like a really good website [13:56] sorry, i just meant any link shortener, doesn't have to be bit.ly [13:56] tinyurl lets you name urls for example [13:57] ooh that I like [14:39] gusch: tmoenicke: when running the qml test case for the onTextChanged bug, what command are you exactly running on the phone ? [14:41] nerochiaro_lunch: qmlscene, and i picked the desktop file from the browser === nerochiaro_lunch is now known as nerochiaro [14:41] tmoenicke: from the browser ? on the device ? [14:44] nerochiaro: I picked the desktop file fom the notes-app - yes, on the device [14:45] gusch: what do you mean that you picked the desktop file from notes-app ? did you launch "qmlscene /path/to/example.qml --desktop-file-hint=/path/to/notes-app.desktop" ? [14:46] nerochiaro: yes [14:47] gusch: when i do that now i get this error "QUbuntu: Could not create application instance" and then the app segfaults [14:47] nerochiaro: I'm currently flashing my phone - afterwards I can check for the exact command [14:47] gusch: that would help, thanks [14:48] nerochiaro: yes on the device [14:49] tmoenicke: the browser doesn't appear for me on the device, i just get "QUbuntu: Could not create application instance" when i launch qmlscene [14:49] tmoenicke: are you launching from adb shell ? [14:52] nerochiaro: nope [14:52] nerochiaro: i was using ssh [14:52] tmoenicke: ok, let me try that [14:53] tmoenicke: how do you enable ssh ? [14:55] nerochiaro: adb shell start ssh [14:58] tmoenicke: hmm, same problem, QUbuntu: Could not create application instance when i run qmlscene from an ssh shell [14:58] tmoenicke: are you on a normal image where you just set /userdata/.writable_image ? [14:58] nerochiaro: qmlscene RotationTest.qml --desktop_file_hint=.local/share/applications/com.ubuntu.notes_notes_1.4.214.desktop [14:59] gusch: you just flashed the image, set up writable_image, uploaded the file and ran that over ssh ? [14:59] gusch: or there's some other step i am misisng ? [15:00] can i ask desktop app related questions here? [15:00] nerochiaro: just reflashed, and now I can see it as well ... [15:01] gusch: i asked sergiusens on the other touch channel but response yet === gatox is now known as gatox_lunch [16:09] oSoMoN: can you please confirm that launching any qml file with qmlscene from adb or ssh shell (with or without --desktop-file-hint) will crash and display the following error: QUbuntu: Could not create application instance ? [16:10] oSoMoN: gusch has the same problem and I asked 5 more people but no one is paying attention [16:11] nerochiaro, give me a min [16:11] oSoMoN: no rush [16:12] nerochiaro: it's on my list (not that I saw it myself ;) [16:14] nerochiaro, nope, it works here [16:14] nerochiaro, with desktop_file_hint [16:14] nerochiaro, are you logged in as the phablet user? [16:19] oSoMoN: yes [16:19] oSoMoN: today's image ? [16:20] oSoMoN: build 10 === gatox_lunch is now known as gatox [16:42] nerochiaro, I got build 11 [16:42] nerochiaro, I don’t think it would change anything [16:44] oSoMoN: ok, thanks for checking [16:45] oSoMoN: just verified that anchorToKeyboard doesn't work in landscape even in a minimal example. submitting bug now [16:45] * popey tickles mhall119 with https://myapps.developer.ubuntu.com/dev/click-apps/243/ [16:47] nerochiaro, excellent, thanks [16:48] oSoMoN: it was already reported it seems (with a slightly different symptoms but i believe the cause it the same) https://bugs.launchpad.net/ubuntu-keyboard/+bug/1236360 [16:48] Launchpad bug 1236360 in Ubuntu UI Toolkit "MainView.anchorToKeyboard doesn’t resize the contents in landscape orientation" [Undecided,New] [16:48] renato_: please have a look at https://bugs.launchpad.net/ubuntu-keyboard/+bug/1236360 when you come back [16:49] oSoMoN: regarding the hack to make sure the content finishes resizing before getting the cursor into view, there doesn't seem to be any way to know when the animation is done, according to renato [16:53] :/ [16:57] oSoMoN: so either we convice them to expose that animation somehow so we can know when it's finish, or expose a signal when it's finished, or keep the hack [16:58] nerochiaro, or we sleep on it and wake up tomorrow with a slightly less hacky hack :) [16:58] oSoMoN: allright, let's do that [16:58] popey: we're up to 243 apps? [16:59] not quite [16:59] some are rejected [16:59] oSoMoN: one option would be to know what the keyb height should be in the end, and wait for the content to reach that height [16:59] oSoMoN: but it's just as fragile as the timer hack, i think [16:59] nerochiaro, sounds much less fragile to me, I like this approach better [16:59] popey: but 243 submitted apps [17:00] nerochiaro, using the KeyboardRectangle, we know when the OSK is fully shown, so we know its height at this point, and therefore we can calculate the height of the view [17:01] oSoMoN: i'll have a look at how it can be done [17:01] oSoMoN: it's still relying on the fact that we know some internals, but yes, probably a better way [17:05] mhall119: -1 for hello world, sure ☻ [17:06] popey: that's valid app! [17:06] ☻ [17:11] nerochiaro, hey, I just submitted this trivial MR: https://code.launchpad.net/~osomon/notes-app/select-many-noorder/+merge/193966. Care to have a quick look? [17:11] oSoMoN: looking [17:12] oSoMoN: was that order giving problems or is it just to increase robustness ? [17:12] nerochiaro, just to increase robustness, because I just found out that the order is not guaranteed by select_many, I’m writing an e-mail about it to the team [17:13] oSoMoN: it looks good to me anyway, and i think that if jenkins is ok with it you can top approve when it's done [17:13] nerochiaro, thx [18:41] how does the phone actually start a click app that's installed? So I can do the same thing from the command line -- if I just run qmlscene myself from an ssh connection, then it's not being managed by upstart and it's not confined [18:43] aquarius: I don't know how much this reflects current reality vs the intention, but try the "start application APP_ID=foo" thing mentioned here: https://wiki.ubuntu.com/SecurityTeam/Specifications/ApplicationConfinement#Launching_applications [18:43] sarnold, aha, that looks useful! [18:45] looks like tedg is involved with this: I shall ping him too === salem_ is now known as _salem [22:51] Elleo: cutespotify crashes for me... Log in: "No error" [23:32] popey: oh, is this with a recent build? [23:32] or the version I gave you ages ago? [23:32] hmm, the one on my phone, might be old [23:32] popey: hang on, I'll upload a newer one for you to have a poke at [23:32] I'll have to play tomorrow.. being dragged away [23:33] but ping the link and I will look in the morning [23:33] okay