/srv/irclogs.ubuntu.com/2013/06/04/#ubuntu-app-devel.txt

zsombigood morning05:36
=== dpm-afk is now known as dpm
dpmmorning zsombi05:51
dpmand good morning everyone05:51
zsombidpm: :)05:52
=== greyback|away is now known as greyback
MirvKaleo_: I've now briefly tested the Qt 5.1 beta on desktop and it seems to work - so if using saucy, you can start testing/developing with it from ppa:canonical-qt5-edgers/qt5-beta-proper07:03
Mirv(and anyone else specifically interested in Qt 5.1)07:03
Mirvit's not all of the modules, but "all up until qtwebkit", which means I was able to run a simple webkit browser with pure 5.1 beta modules07:04
=== chriadam is now known as chriadam|away
=== hunger_ is now known as hunger
oSoMoNgusch: I was starting to review your MR, but I see that tmoenicke already approved it, so nevermind my comment09:39
guschoSoMoN: I can include the fix in my next MR09:40
oSoMoNgusch: that’s alright, it’s just a harmless typo in a debug message, I was just being picky, as I usually do when I review code ;)09:41
guschoSoMoN: but a valid complain - shall I just fix it in the next MR, or do a separate?09:43
oSoMoNgusch: such a small fix definitely doesn’t warrant a separate MR, that would be a lot of overhead and would be counter-productive09:44
oSoMoNso feel free to fix it in any other MR (better if it’s slightly related with this code, but not mandatory)09:44
guschoSoMoN: ok - so I'll just fix it in the next MR - thx09:44
nerochiarodpm: yesterday you said that in the plugin i'm trying to i18n-ize i would have to set explicitly the i18n.domain, but I'm not sure how to do that, it seems that that property doesn't exist10:11
nerochiarotimp: ^ any ideas ?10:21
timpnerochiaro: i18n.domain property should exist if you import Ubuntu.Components.10:33
nerochiarotimp: i am importing it but no, it doesn't exist10:58
nerochiarotimp: anything obviously wrong: http://pastebin.ubuntu.com/5732238/10:59
nerochiarotimp: ?10:59
timplet me check11:00
Kaleo_Mirv: thanks but I'm not using saucy11:01
timpnerochiaro: no, looks good to me11:03
timpnerochiaro: some things to try, add some property string test: i18n.tr("test") to see if i18n works11:03
timpand try Component.onCompleted: i18n.domain = "your-domain"11:03
* timp standup now11:03
davidcalleHello app-dev! I'm trying to figure out how to read a file line by line in qml. Any hints?11:08
nerochiarotimp: tr seems to work fine, it's just i18n.domain that's not there11:11
timpnerochiaro: weird.11:12
* timp in standup now, I'll have another look after that11:12
MirvKaleo_: ok. all new development is moving to there ~now, which is why targetting also that for qt 5.1.11:12
Kaleo_Mirv: oh is it?11:13
Kaleo_Mirv: and saucy is stable for use?11:13
nerochiarotimp: let me push the branch somewhere so you can check it out there directly11:14
MirvKaleo_: with the daily release quality processes it's supposed to be always usable. certainly enough for a developer.11:15
nerochiarotimp: lp:~amanzi-team/webbrowser-app/webbrowser-app-component11:15
MirvKaleo_: daily-build-next is now moving to build everything on top of saucy, today / this week11:16
Kaleo_https://www.dropbox.com/sh/uzkr674j7zuglbn/jAEZkfnAAo11:16
Kaleo_Mirv: supposed to be is not good enough11:18
Kaleo_Mirv: it's potentially weeks of time wasted for all developers if it isn't11:19
nerochiarotimp: also what's the replacement for the deprecated QuitAction ?11:22
nerochiarozsombi: ^11:22
timpnerochiaro: nothing, closing apps from the app lens11:22
nerochiarotimp: how do i know i'm being closed ?11:22
timpKaleo_: ^ do you know?11:23
MirvKaleo_: well all of ubuntu engineering is using it, I'm just using the word since I can't make guarantees. it's the rolling release talked about, and daily quality has been good during the raring as well.11:23
Mirva general plan on the switching could be good, but at least the device images and daily builds of packages are moving to saucy now11:26
Mirvbut for any generic development raring (or even precise) is fine, it's just if one wants to do something related to device image development directly or future upcoming things11:27
Kaleo_nerochiaro: for now try Component.onDestruction: { in your main QML file11:41
nerochiaroKaleo_: we've been in this conversation already, there's no guarantee of the order components are destroyed in QML and i need to know that everything is still there when i want to save my notes before quitting. it's not reliable if I use Component.onDestruction as things I need might be already gone11:50
timpKaleo_: is the Icon supposed to replace all url-type iconSource properties in buttons, listitems, actions?11:57
=== MacSlow is now known as MacSlow|lunch
guschnerochiaro oSoMoN tmoenicke a short MR https://code.launchpad.net/~schwann/gallery-app/gallery-remove-quit/+merge/167155 and a long MR https://code.launchpad.net/~schwann/gallery-app/gallery-previewmanager-refactor/+merge/167234 take your pick ;)11:59
oSoMoNgusch: I’m getting the short one :)12:04
guschoSoMoN: first come, first serve12:05
oSoMoNgusch: so there’12:06
nerochiarozsombi: one question about InverseMouseArea: i noticed that you have to be *very* precise in how long the touch is if you want the IMA to register it as an onClicked event. If it's too short or too long it doesn't register. I tried it with a self contained example, so I'm sure it's not my app12:06
oSoMoNgusch: so there’s no action left in the HUD?12:06
guschoSoMoN: not at this place12:06
Kaleo_nerochiaro: ok, what do you want to do upon exiting?12:07
nerochiaroKaleo_: save whatever not is expanded12:07
nerochiaroKaleo_: save whatever note is expanded12:07
Kaleo_nerochiaro: ok12:08
Kaleo_timp: not exactly replace, but yeah12:08
nerochiarozsombi: this is the example i'm using http://pastebin.ubuntu.com/5732389/12:19
zsombinerochiaro: on hangout...12:22
nerochiarozsombi: sure, no hurry, just need to have your opinion and report a bug if you think it's a bug. it's certainly afftecting notes-app, lots of people told me about it12:30
Kaleo_nerochiaro: do you have a bug report for the notes app?12:32
Mirvbzoltan: ok the qt creator 2.7.1 for precise/armhf is finally there in the qt5-beta-proper PPA12:32
nerochiaroKaleo_: yes, but it's marked as fix released because i thought i had fixed it in another way. probably better if i reopen it12:32
Kaleo_nerochiaro: what's the importantce?12:33
Kaleo_-t12:33
nerochiaroKaleo_: critical12:33
bzoltanMirv: I know... I am just about to test it12:33
nerochiaroKaleo_: https://bugs.launchpad.net/notes-app/+bug/118181312:34
ubot2Ubuntu bug 1181813 in notes-app "Notes don't collapse correctly" [Critical,Fix committed]12:34
Kaleo_nerochiaro: so there is a hurry12:34
Kaleo_zsombi: ^12:34
nerochiaroKaleo_: yes, but not so much that zsombi can't finish his hangout i guess12:35
Kaleo_nerochiaro: sure12:35
zsombiKaleo_: sorry, still in hangout12:35
=== MacSlow|lunch is now known as MacSlow
nerochiaroKaleo_: this is the notepad but i just reported that is caused by the lack of an equivalent for the HUD quit action https://bugs.launchpad.net/notes-app/+bug/118736912:53
ubot2Ubuntu bug 1187369 in notes-app "Expanded note isn't save when quitting from the dash" [Undecided,New]12:53
oSoMoNgusch, nerochiaro, renato_: got a trivial MR up for review, can someone take a look? https://code.launchpad.net/~osomon/webbrowser-app/do-not-automatically-hide-chrome-while-testing-toolbar/+merge/16726212:58
guschoSoMoN: I take it13:00
oSoMoNgusch: thanks13:00
zsombinerochiaro: Im'm back 4 U13:07
zsombinerochiaro: what do you mean very precise, like the timing you press?13:08
zsombinerochiaro: so on your example, if you press outside the IMA area then you don't always get the rectangle color changed?13:09
zsombinerochiaro: also, does it happen on device or also on PC?13:13
nerochiarozsombi: i can't really "tap" on the pc so it only matters on the device. what i mean by precise is that you have to put your finger and remove it within a certain number of ms or it won't be counted as a tap. but it's a different length of time as the normal taps that happen everywhere else. taps everywhere else (e.g. on buttons) are much more forgiving13:15
nerochiarotimp: any news on that translation issue with i18n.domain ?13:16
zsombinerochiaro: I also found out that longer taps on buttons may not always trigger the clicked signal on the MouseAreas either... hmm... looks like we have some upstream or hybris bug here. If the onClicked is not received it is either because there was some move event in between so the clicked cannot be emitted, or we don't get touch/mouse release event. The worst could be if we get mixed touch press/mouse release events...13:18
nerochiarozsombi: no, i think it's just simpler than that. try it yourself, you will see it's just a case of very sensitive timing. *very* differnet than the buttons13:20
zsombinerochiaro: how to launch an app so that you can see the logs realtime?13:21
nerochiarozsombi: what do you mean ?13:21
zsombinerochiaro: I used to launch from QtC and I can get the logs only after I close the app :/13:22
nerochiarozsombi: not sure, never ran anything from qtc. what i just do is to ps aux | grep notes while notes is running then copy the cmd line replacing then notes-app qml file with my test file13:23
nerochiarozsombi: then kill notes app and run that cmd line13:23
zsombiyak13:24
timpnerochiaro: how do I "run" the webbrowser-app-component?13:24
nerochiarotimp: you don't. you install the two packages and run webbrowser app13:25
zsombinerochiaro: never the less, you seem to be in a need to file a bug. However, you could still work with onPressed to get this thing working13:25
zsombinerochiaro: onPressed is more reliable13:25
nerochiarozsombi: no, because onpressed will trigger even if the user is flicking i think13:25
nerochiarozsombi: and i don't want that13:25
zsombinerochiaro: right :(... do you get onReleased btw?, try to compare the points from onPressed and onReleased, if those are different, then there was a move13:26
* zsombi reenables developer mode...13:27
nerochiarozsombi: that what i was doing before, but then it required a lot more precision because a tiny move would throw it off, and i would have to introduce an offset, etc. Honestly it would be much simpler if we could figure out why IMA behaves differently on taps that Button13:28
timpnerochiaro: I can set Component.onCompleted: i18n.domain = "webbrowser-app" without problems13:29
guschoSoMoN: approved13:29
zsombinerochiaro: for instance, on UITK Gallery, on PC, the Popover page, if I press a bit longer on the Open button, I no longer get the clicked signalled13:29
timpi18n is a context property, not a property of some component, so you cannot say `i18n.domain: "test"` in your Item.13:29
oSoMoNgusch: thanks13:30
zsombinerochiaro: in IMA we have to deal with the touch events also, as we don't always get the converted mouse events...13:30
timpnerochiaro: MainView does the same, it sets i18n.domain in its onCompleted13:30
nerochiarotimp: in my example i was triying to set it as a property, that's probably why it wasn't working for me. I'll try on the onCompleted13:30
zsombinerochiaro: most likely we have to define a threshold which would not be considered as move event...13:31
guschzsombi: can you/I top approve this? Is it still valid? https://code.launchpad.net/~schwann/gallery-app/gallery-remove-obsolete-swipe-switch/+merge/16087813:31
nerochiarozsombi: i'm not really sure but the point is that touched should work consistently with the same speed across all components in the SDK13:32
zsombigusch: done13:32
guschzsombi: thx13:33
nerochiaroKaleo_: ^ agreed ?13:34
zsombinerochiaro: man, I have to say that I can get the red square color changed no matter how long I press outside of the box...13:34
zsombinerochiaro: or tap13:34
zsombinerochiaro: on tablet at least13:34
nerochiarozsombi: i'm trying on the phone and it's really really picky13:35
zsombinerochiaro: I have 151 on it13:35
zsombinerochiaro: I have my phone at bzoltan13:35
nerochiarozsombi: maybe we should get someone else to confirm13:35
timpKaleo_: I finished filling out my parts in https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0ApMAhwCVytVidF9zdDVRNEh0UkJHQ19odnRrSi14YUE#gid=013:36
timpKaleo_: I was asked to "pass it" to you when I'm done13:36
zsombitimp: can you help us with a small test on the phone? http://pastebin.ubuntu.com/5732389/13:36
zsombitimp: if you tap longer outside of the square, do you get the square color changed?13:37
timpzsombi: meeting now13:40
Kaleo_nerochiaro: I don't know13:41
Kaleo_timp: thx13:41
nerochiaroKaleo_: you don't agree that all taps in the SDK should react in a consistent way ?13:45
zsombiis LP locked?14:07
nerochiarooSoMoN: what was the way to test and application in another language ? set some env var when launching ?14:11
nerochiarodpm: ^14:11
oSoMoNnerochiaro: you mean on a device?14:11
nerochiarooSoMoN: no on desktop is fine14:12
oSoMoNnerochiaro: you first need to ensure you have the corresponding locale generated, using locale-gen14:12
oSoMoNnerochiaro: and then run the app with e.g. LANG=it_IT.UTF-814:12
dpmnerochiaro, you can use LC_ALL=it_IT.UTF-8 qmlscene app14:13
dpmor LANGUAGE=it qmlscene appname14:13
nerochiarodpm: oSoMoN: ok seems to work, and with i18nized qml plugin too. yay14:14
dpm\o/14:14
oSoMoNnerochiaro, renato_, gusch: yet another trivial MR to fix flaky autopilot tests in the browser, your review would be much appreciated: https://code.launchpad.net/~osomon/webbrowser-app/better-assertion-for-page-loaded/+merge/16728914:15
guschoSoMoN: ok - I do it14:15
oSoMoNthanks gusch, appreciated14:15
oSoMoNnerochiaro: you could have a Timer that saves the currently open note every n seconds, and reset it every time the note is modified14:25
nerochiarooSoMoN: yeah, something like that, but rather with 800ms or so, 1s tops14:33
oSoMoNnerochiaro: pick whatever value you want for n :)14:33
nerochiarooSoMoN: :)14:34
nerochiarooSoMoN: this is also another feature where the test will take longer to write than the code14:34
oSoMoNnerochiaro: you might wanna run some quick profiling to verify that it’s not too IO-heavy, and adjust n accordingly14:35
oSoMoNnerochiaro: welcome to quality-driven development ;)14:35
nerochiarooSoMoN: where the time you save now will hopefully not come to bite you in the arse tomorrow ;)14:36
guschnerochiaro oSoMoN renato_ tmoenicke still waiting ;) https://code.launchpad.net/~schwann/gallery-app/gallery-previewmanager-refactor/+merge/16723414:36
oSoMoNgusch: yeah, I’ll take it, but not now, need to write some code of my own14:37
guschnerochiaro tmoenicke renato_ any other taker?14:37
zsombiKaleo_: greyback: timp: anyone eager to start reviewing Layouts? https://code.launchpad.net/~zsombi/ubuntu-ui-toolkit/layouts/+merge/16729614:42
guschoSoMoN: seems like I'll have to wait then - thx14:42
greybackzsombi: I'll try to give it a functional review this evening, if I have some time to spare. If not today, then tomorrow14:43
zsombigreyback: thanks, it would be easier for you as partly is based on your prototype14:43
greybackzsombi: hopefully yes :)14:44
dpmballoons, if you've got any issues installing the SDK and Qt Creator in saucy, I'm sure Mirv will be able to help you if you've got questions14:46
balloonsty <314:46
* zsombi eod14:47
zsombicheers, C U tomorrow!!14:47
=== davidcalle_ is now known as davidcalle
oSoMoNnerochiaro: hey, could you please confirm bug #1187432 ?15:15
ubot2Launchpad bug 1187432 in webbrowser-app "Ampersand not correctly escaped in titles when displayed in suggestions list" [Medium,New] https://launchpad.net/bugs/118743215:15
nerochiarooSoMoN: in a meeting15:16
oSoMoNnerochiaro: no urgency, please have a look when you have a moment15:17
Kaleo_bzoltan: can you add uriboni to https://launchpad.net/~ubuntu-sdk-team/+members please?15:42
bzoltanKaleo_: done15:43
Kaleo_thx15:44
nerochiaroKaleo_: how's that going to work for the autopilot tests ? do they stay with the app ? so when the component is modified the app might have its tests failing ?15:57
Kaleo_nerochiaro: it depends15:58
Kaleo_nerochiaro: if the tests in question are testing just that component15:58
Kaleo_nerochiaro: they should move to15:58
Kaleo_too*16:00
nerochiaroKaleo_: i'm talking about autopilot tests16:00
nerochiaroKaleo_: how can they test just a component ?16:00
Kaleo_nerochiaro: why couldn't they?16:09
nerochiaroKaleo_: how would you launch "the app" ?16:09
Kaleo_nerochiaro: qmlscene16:09
nerochiaroKaleo_: qmlscene what ?16:10
Kaleo_nerochiaro: the component16:10
Kaleo_nerochiaro: or if you like another Qml that instantiates it16:10
nerochiaroKaleo_: the component alone won't work, you need some kind of launcher that instantiates it and sets it up properly. Which is essentially the app itself16:11
Kaleo_nerochiaro: what set up?16:12
Kaleo_nerochiaro: if the component does not work on its own then the component is probably not good16:12
Kaleo_nerochiaro: but anyway16:12
Kaleo_nerochiaro: you can always add a tiny qml16:12
nerochiaroKaleo_: "set up" is something like this: http://pastebin.ubuntu.com/5733047/16:14
nerochiaroKaleo_: minus the aliases16:14
Kaleo_nerochiaro: and minus the title16:15
Kaleo_nerochiaro: and minus the oncompleted16:15
nerochiaroKaleo_: ok, basically sometihng that instantiates the component giving it a size16:15
nerochiaroKaleo_: by anchoring it16:15
Kaleo_nerochiaro: well, a component should have an implicitSize set inside it16:15
Kaleo_nerochiaro: which means: minus the size16:16
Kaleo_done.16:16
nerochiaroKaleo_: ok, I guess the autopilot tests need to be moved too. bleh, more work. ok16:16
Kaleo_:)16:16
nerochiarotimp: could you take care of doing that test that zsombor asked you for earlier, on the phone, for the inversemousearea ?16:22
timpnerochiaro: ok, you still around? I have 5 minutes16:27
nerochiarotimp: yes16:27
nerochiarotimp: do you have the test code ?16:27
timphttp://pastebin.ubuntu.com/5732389/plain/16:28
timpnerochiaro: how can I launch it directly on the device?16:29
timpI put it in a qml file there.16:29
nerochiarotimp: you need to pretend to be some other app. i think the best way is actually to overwrite any app's main qml file, with the same name, and then launch that app from the shell16:29
nerochiarotimp: alternatively just copy the desktop-file-hint cmd line option from some other app and run it with that from a console16:30
timpnerochiaro: ok I did it via qtcreator, created a project with that qml file.16:30
timpand then run on device16:30
timpI have it now. what was the test?16:30
nerochiarotimp: you need to verify that just by tapping normally outside of the rectangle you can make the rectangle change color everytime16:31
nerochiarotimp: tap like if you were tapping on a Button component16:31
timpI cannot do it every time16:31
timpdoubleclick some times changes it, some times changes it twice16:32
nerochiarotimp: perfect, then it's confirmed16:32
timpsingle tap usually does nothing, some times changes color16:32
nerochiarotimp: zsombi need to fix things16:32
nerochiaro;)16:32
timpnerochiaro: is there a bug report where I need to confirm?16:32
nerochiarotimp: not directly one in the SDK16:32
timpok even doubleclick not always changes it16:32
oSoMoNtimp, Kaleo_: could one of you please confirm bug #1187476 ?16:32
ubot2Launchpad bug 1187476 in Ubuntu UI Toolkit "Removable list items should be swiped up/down to be removed from an horizontal listview" [Undecided,New] https://launchpad.net/bugs/118747616:32
nerochiarotimp: i'll submit one tomorrow after hearing zsombi's comments16:33
nerochiarotimp: thanks16:33
timpnerochiaro: np16:33
timpoSoMoN: on desktop the app seems to work, on phone (deployed via qtcreator), I only get a blank white screen16:35
timpoSoMoN: sorry, no time to investigate further now, I need to leave16:35
oSoMoNtimp: were those last two comments really addressed to me?16:36
timpoSoMoN: yes, I quickly tried to run the code that comes with bug #118747616:40
ubot2Launchpad bug 1187476 in Ubuntu UI Toolkit "Removable list items should be swiped up/down to be removed from an horizontal listview" [Undecided,New] https://launchpad.net/bugs/118747616:40
* timp eod16:41
oSoMoNtimp: ok, well there’s no need to run on a device, really, the code is architected so that it considers only vertical list views, so it breaks for horizontal ones16:41
oSoMoNas simple as that16:41
timpahh.16:41
timpyeah in the horizontal list I had to swipe horizontally for removing it16:42
timpI'm in a hurry so I didn't read that well.16:42
* timp off now16:42
odraencodeHello16:59
=== netcurli_ is now known as netcurli
=== Borborygmi is now known as Casmo
inashdeen hi there, I am learning how to build a custom app using fogger. have a look at this website https://web.samsungchaton.com/  I created a fogger app using it. Then, I want to write a script to remove (or mask) the two panels/bar. one above, with the logo samsung, and one below the black colour bar written terms and condition. how do i do it?19:26
timpdid anyone here try to create an ubuntu touch app with unity (the 3d engine)?19:33
jonotimp, I haven't seen anyone do that yet19:51
jononot a touch app19:51
timpjono: ok. I'm interested in trying some 3D interaction where I use our toolkit for QML for certain UI elements, and for the 3D interaction, and "something else" (perhaps Unity3D, VTK, OpenGL) for rendering the 3D scene20:06
jonotimp, that would be awesome :-)20:07
timpjono: yeah it would be :) unfortunately the 3d libs that I checked don't have qtquick bindings so it will be some work (and learning more of the qt5 scenegraph than I know now) to get it to work together20:14
timpif someone already did it that can save me some time :) that's why I'm asking here20:15
jonotimp, you can't just dump a unity3d canvas on a qt canvas?20:15
jono:-)20:15
timpI have only been trying with VTK (vtk.org) because I already have a bunch of code that uses that, but VTK and Qt5 don't play nice together yet.20:19
timpunity3d is just an idea and I'm not sure that it does what I want, but it seems worth having a look at20:19
odraencodeHey again :D20:20
GuidoPallemanswill this be made into the default time picker? http://blog_uploads.s3.amazonaws.com/wp-content/uploads/2013/05/clock_addalarm.png20:39
odraencodeHELLO21:45
odraencodeWhy nobody says hello? :c21:45

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!