2014/08/01/#ubuntu-app-devel.txt

=== _salem is now known as salem_
[02:12] <diego7319> hi guys
[03:07] <gerlowskija_home> balloons: I see "revised-fix-for-bug1334883" got merged.  Pretty pumped!...now for the bug that was holding it up...
=== chihchun_afk is now known as chihchun
=== salem_ is now known as _salem
[06:33] <karni> nik90: pong ;P
[06:33] <karni> nik90: about the bug you mentioned
[06:36] <dholbach> good morning
[07:13] <mihir> popey_: ping !!
[07:33] <nik90> karni: hi, can you update that bug with more details
[07:41] <karni> nik90: done :)
[07:56] <dholbach> nik90, popey_: I pushed some changes to lp:~dholbach/ubuntu-clock-app/reboot-packaging
[07:56] <dholbach> nik90, popey_: it still won't build, as the tests are not run in the chrooted build
[07:56] <nik90> dholbach: Hi, I will take a look now
[07:57] <dholbach> nik90, popey_: it looks to me like something in the cmake logic needs to be changed to run the autopilot tests properly...if we want to run them during the build
[07:58] <nik90> dholbach: I am guessing the custom target for autopilot in the cmake file needs to be revisited.
[07:58] <dholbach> yeah, I had a look at gallery-app - not sure if that's the best example though - they seem to use some magic from "include(${CMAKE_SOURCE_DIR}/cmake/autopilot.cmake)"
[07:59] <nik90> dholbach: but do the c++ plugins get built properly according to the arch?
[07:59] <nik90> dholbach: that's one primary diff w.r.t to the old clock app and where I was not sure how to do it
[07:59] <dholbach> nik90, if I get past this point, should I be fine?
[07:59] <dholbach> Linking CXX shared module Timezone/libtimezone.so
[07:59] <dholbach> ^
[08:00] <nik90> dholbach: I would assume so so yes :) there are 2 other libraries (AlarmSettings.so and DateTime.so)
[08:00] <dholbach> yep
[08:00] <nik90> yup that's good then
[08:01] <dholbach> nik90, http://paste.ubuntu.com/7921784/ ← this is where I fail now
[08:01] <nik90> dholbach: can you propose a MR, the code diff will be more visible then
[08:01] <nik90> dholbach: I will investigate how to fix the autopilot issue
[08:01] <dholbach> nik90, https://code.launchpad.net/~dholbach/ubuntu-clock-app/reboot-packaging/+merge/229173
[08:04] <dholbach> nik90, to replicate what I'm trying to do, you could run this once:  "sudo apt-get install ubuntu-dev-tools; pbuilder-dist utopic create; bzr branch lp:~dholbach/ubuntu-clock-app/reboot-packaging; cd reboot-packaging" - and then every time you want to test the build in semi-real-life build conditions:  "bzr bd -- -S -us -uc; cd ..; pbuilder-dist utopic build ubuntu-clock-app_1.2.dsc"
[08:05] <nik90> dholbach: thnx
[08:05] <dholbach> nik90, this way, a utopic chroot is created and it will attempt to build the created source package (that's the 'bzr bd' step), in this chroot - it's all automated and the packages will be cached, so building it the second time will be quicker
[08:05] <dholbach> but yeah... if we could figure out what the de-facto way is of running autopilot tests in build scenarios, that'd be great
[08:08] <dholbach> nik90, if you attempt the build locally, are the autopilot tests run just fine?
[08:09] <nik90> dholbach: I didn't try yet, I was looking through reminder app's cmake files to see if they did anything different
[08:09] <dholbach> oh ok
[08:10] <nik90> also the chroot is being created atm
[08:10] <dholbach> rock and roll
[08:10] <dholbach> nik90, ok... what I just tried locally (without a chroot) was: cmake .; make; make test
[08:10] <dholbach> and it fails the same way
[08:11] <dholbach> so I guess if we can make the above workflow work, then it might also work in the chroot
[08:11] <nik90> so that's how you use it..I was wondering how to use the custom targets in the cmake file
[08:13] <dholbach> no idea
[08:14] <dholbach> nik90, it looks like it tries to run something like this "qmltestrunner -input app/tests/unit/ -import backend/"
[08:14] <dholbach> does that work for you?
[08:14] <dpm> dholbach, I'm not sure if the autopilot tests are run at all during the build. I think they are just packaged as a separate package
[08:14] <dholbach> dpm, right
[08:14] <nik90> dholbach: I tried "make autopilot" and that started to run some tests although they did fail due to missing dependency on trusty
[08:14] <dholbach> dpm, right now it looks like they don't run at all :)
[08:14] <dholbach> aha
[08:15] <dholbach> nik90, http://paste.ubuntu.com/7921868/
[08:15] <nik90> dholbach: yup
[08:16] <nik90> dholbach: let me try merging https://code.launchpad.net/~carla-sella/ubuntu-clock-app/add-alarm-test which adds the first working ap test for the new clock app
[08:18] <dholbach> nik90, "ubuntu-clock-app" → "ubuntu_clock_app"
[08:18] <dholbach> ok.. look like that's what Carla did as well
[08:18] <nik90> dholbach: so isn't that enough for our debian packaging?
[08:19] <nik90> instead of make test we run make autopilot
[08:19] <dholbach> sure, I can do that
[08:19] <dholbach> hang on
=== popey_ is now known as popey
=== renato is now known as Guest39756
[08:33] <dholbach> nik90, does add_custom_target pass through a return code?
[08:33] <dholbach> nik90, it looks like "make autopilot" will return 0 no matter what
[08:33] <dholbach> so stopping the build on failed tests is a bit hard :-D
[08:33] <nik90> dholbach: sry not sure since it was there with the cmake template as part of qtC
[08:34] <nik90> dholbach: but wouldn't jenkins do that? Stop merge if tests fail
[08:34] <dholbach> I don't know how this works in jenkins - I was just talking about the package build right now
[08:34] <dholbach> but yeah, I guess that a pre-merge test-run would stop the merge if a test failed - if that's the question
[08:35] <nik90> in the old clock app, does the build fail if a test fails?
[08:36] <nik90> hey also I have a utopic vm. So I tried building in that. I did bzr bd -- -S -us -uc; cd .. to get the .dsc file. How do I proceed from that?
[08:36] <nik90> build filename.dsc doesnt seem to work
[08:36] <nik90> it cannot find the build command
[08:36] <dholbach> nik90, pbuilder-dist utopic build <bla.dsc>
[08:38] <dholbach> nik90, I updated http://code.launchpad.net/~dholbach/ubuntu-clock-app/reboot-packaging - what I get now is http://paste.ubuntu.com/7922007/
[08:38] <JamesTait> Good morning all; happy Friday and happy World Wide Web Day! :-D
[08:40] <nik90> dholbach: in the app/tests/autopilot/ubuntu-clock-app/main.py file, could you comment the self.assertThat() statement. That should make the test pass for now
[08:40] <nik90> dholbach: we have full ap tests in another branch which anyway removes this test
[08:40] <dholbach> ok
[08:46] <dholbach> nik90, am I correct in assuming there's nothing to be installed into usr/bin?
[08:46] <nik90> dholbach: yes, we don't have an executable
[08:47] <dholbach> ok
[08:48] <nik90> dholbach: although that looks bad since the old clock had a executable
[08:48] <dholbach> was it just a quick shell script?
[08:49] <nik90> dholbach: yeah, it eseentially called "qmlscene ubuntu-clock-app.qml"
[08:49] <dholbach> ok, I guess we can re-add that
[08:53] <nik90> dholbach: btw, I think this should help you https://code.launchpad.net/~jpakkane/ubuntu-clock-app/utopiccmake/+merge/225694
[08:54] <dholbach> nik90, brilliant - I was just trying to do this on my own (without knowing much about cmake... ;-))
[08:55] <nik90> dholbach: it was proposed a long time back by Jussi, but we couldn't merge it completely. So we took the cmake fixes from the MP but left behind the debian packaging since he said it wasn't fully working yet
[09:07] <dholbach> nik90, with Jussi's changes, I get http://paste.ubuntu.com/7922204/
[09:08] <dholbach> lp:~dholbach/ubuntu-clock-app/reboot-packaging updated with the latest
[09:10] <nik90> dholbach: that is weird...the alarmsettings.so builds fine otherwise
[09:10] <dholbach> I applied Jussi's changes
[09:12] <nik90> dholbach: I think that's because you also did 1	=== modified file 'CMakeLists.txt'
[09:12] <nik90> dholbach: which removes links to libraries we use
[09:12] <dholbach> nik90, that was Jussi's changes
[09:12] <dholbach> hum
[09:13] <nik90> dholbach: yeah as I said jussi's changes were done some time back after which we added these libraries. That's why jussi's changes remove them
[09:13] <dholbach> ah ok - let me see if I can repair this
[09:13] <nik90> dholbach: can you do bzr revert CMakeList.txt alone. the rest of the files seem okay
[09:13] <dholbach> no no
[09:13] <dholbach> there are other changes which look like we need them
[09:13] <dholbach> files were installed into the wrong place
[09:13] <dholbach> ie debian/tmp/share
[09:14] <dholbach> when it should have been debian/tmp/usr/share
[09:14] <nik90> hmm
[09:15] <nik90> dholbach: okay in that case let me ask sergieusens to do one final check of our cmake file. It seems like if we fix the cmake files, then the debian packaging will also fall in line properly
[09:15] <nik90> I feel like we are searching in the dark due to the cmake stugg
[09:16] <nik90> stuff*
[09:17] <dholbach> nik90, right... I'm just not sure... which changes were you expecting to get from Jussi's branch
[09:18] <dholbach> it was mostly just the CMakeLists.txt stuff, right?
[09:18] <nik90> dholbach: no actually it was the debian/rules file
[09:18] <dholbach> aha
[09:18] <nik90> dholbach: I already implemented what was necessary from his Cmakelist.txt file few weeks back
[09:18] <dholbach> ok ok
[09:19] <dholbach> in that case, I'll back out the change again
[09:26] <kalikiana> nik90: thx
[09:31] <dholbach> nik90, all changes pushed - if you could have a chat with sergiusens later on, that'd be great
[09:31] <dholbach> as I said: debian/tmp/share/ should be debian/tmp/usr/share/
[09:31] <dholbach> nik90, if you need anything else from me, let me know
[09:33] <nik90> dholbach: ok, I will check with sergiusens about this
[09:33] <nik90> dholbach: thxn
[09:36] <dholbach> great
[10:09] <popey> dpm: https://code.launchpad.net/~dpm/reminders-app/debian-copyright/+merge/225646 looks like it needs a little love
[10:10] <victorp> anyone knows how to get a qml app running full screen (no indicators)?
[10:11] <popey> ooh, good question. i want to know this.
[10:11] <dpm> popey, indeed. I've not had the time to look at this and the other branch I've got on the list. I'll have some hacking time this weekend and I'll fix them
[10:11] <popey> k
[10:12] <dpm> popey, shall we get on a hangout and do the branch review?
[10:12] <popey> ya
[10:12] <victorp> Kaleo, ?
[10:13] <Kaleo> victorp, sir
[10:13] <Kaleo> victorp, (I should not be up so early!)
[10:13] <victorp> popey, I see people do it on c++
[10:13] <victorp> Kaleo, ouch, forgot you moved
[10:13] <popey> yeah, camera and gallery do
[10:13] <Kaleo> victorp, http://qt-project.org/doc/qt-5/qml-qtquick-window-window.html#visibility-prop
[10:13] <victorp> Kaleo,  I was wondering if I can go fullscreen with qml, with no indicators
[10:15] <victorp> Kaleo, ? set the visibility prop and that makes it fullscreen?
[10:15] <Kaleo> victorp, with the right value of course
[10:16] <Kaleo> Window.FullScreen
[10:16] <Kaleo> victorp, if that does not work it sounds like something in qtubuntu is broken
[10:17] <victorp> i shall try :)
[10:17] <Kaleo> oSoMoN, I recall you might have tried to make an app fullscreen before no?
[10:17] <Kaleo> oSoMoN, was there an issue like, it would only go fullscreen if you make it fullscreen before showing the window?
[10:18] <Kaleo> https://bugs.launchpad.net/qtubuntu/+bug/1328839
[10:18] <ubot5> Launchpad bug 1328839 in Unity 8 "Toggling fullscreen causes wrong window behaviour" [High,New]
[10:19] <Kaleo> victorp, you are going to hit that bug I believe
[10:19] <Kaleo> loicm, do you know if this bug requires just a fix in qtubuntu or also in other places by any chance? https://bugs.launchpad.net/qtubuntu/+bug/1328839
[10:19] <ubot5> Launchpad bug 1328839 in qtubuntu "Toggling fullscreen causes wrong window behaviour" [High,Confirmed]
[10:20] <Kaleo> loicm, or maybe it's just an issue in Unity?
[10:22] <Kaleo> it looks like the code in qtubuntu is about correct but I'm not sure
[10:28] <karni> Kaleo: hey brolo, I'm trying to make anchorToKeyboard: true work from MainView, but no luck. Any other requirements in my QML code I need to have?
[10:28] <karni> uhh, 7:30 AM for you d'oh
[10:29] <karni> anyone else might know why anchorToKeyboard: true would not work? nik90, maybe you? :)
[10:30] <karni> I pop the keyboard, and nothing changes in the UI. I'd expect bottom of my chat window to surface above the top edge of the keyboard.
[10:31] <Kaleo> karni, not that I know
[10:31] <karni> :(
[10:31] <Kaleo> karni, I need breakfast and all now :)
[10:31] <karni> Kaleo: yeah :)
[10:32] <Kaleo> karni, ask timp
[10:32] <karni> k
[10:32] <Kaleo> t1mp rather
[10:32] <karni> t1mp: hey buddy, would you happen to know why anchorToKeyboard: true might not be working as expected? any other requirements on my QML code for it to work?
[10:32] <karni> t1mp: I removed the fixed height property thinkin it could be it, but no luck there
[10:33] <t1mp> karni: are you using a PageStack?
[10:33] <karni> t1mp: yes I am
[10:33] <karni> as a child of MainView
[10:33] <victorp> Kaleo: result!!!
[10:33] <karni> don't tell me I can't do that :D
[10:33] <karni> victorp: \o/
[10:34] <victorp> as I am not toggleing I think is ok
[10:34] <t1mp> karni: I think you are running into this bug https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1349366
[10:34] <ubot5> Launchpad bug 1349366 in Ubuntu UI Toolkit "[PageStack] Pages does not get correct anchored in the PageStack area" [High,Confirmed]
[10:34] <victorp> popey, wrap the app on a window component and do  Component.onCompleted: {
[10:34] <victorp>         this.visibility = Window.FullScreen
[10:34] <victorp>     }
[10:34] <victorp>  works for me
[10:34]  * karni looks
[10:35] <t1mp> karni: I think there was a workaround where you can put the PageStack inside an Item, but I still need to start working on that bug
[10:35] <karni> t1mp: I thought I'd try that, yes
[10:35] <karni> t1mp: subscribed, thank you
[10:36] <t1mp> karni: something like MainView { Item { anchors.fill: parent { PageStack { ... } } }
[10:36] <loicm> Kaleo: it used to work at the time I wrote it, but shell and new fullscreen decisions might have changed all that
[10:37] <Kaleo> loicm, ok, greyback is on the case to check all the other parts
[10:37] <Kaleo> victorp, :D
[10:37] <t1mp> karni: if it works (or doesn't), please leave a comment on the bug. I try to work on a real fix next week
[10:37] <Kaleo> victorp, why in the onCompleted?
[10:38] <karni> ok
[10:38] <Kaleo> victorp, why not Window {visibility: Window.FullScreen}
[10:39] <victorp> Kaleo, why not indeed
[10:44] <karni> t1mp: so, sadly, it doesn't work. do you think it's the same problem with the keyboard? even if the page didn't overlap the example red rectangle anchored at the bottom, there's no quarantee when you pop a keyboard all of the UI is shifted to accomodate space for the input method. should I file a separate bug?
=== vila is now known as vila-lunch
[10:46] <karni> t1mp: Can I detect if a keyboard is popped from QML? Is there a callback?
[10:46] <popey> victorp: nice!
[10:48] <t1mp> karni: there is a Qt.inputMethod.keyboardRectangle that gives (or should give) the keyboard rectangle
[10:48] <karni> t1mp: thank you
[10:48] <t1mp> but I'm not very familiar with it either, I need to figure out how it works before I can fix the bug
[10:49] <t1mp> karni: I think renato had the same problem with anchoring to the keyboard and after investigating he reported this bug, so I thought it is the same
[10:49] <karni> t1mp: ACK, thanks a bundle
[10:49] <t1mp> karni: can you add your findings with the keyboard to https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1349366 and I will try to figure out if they are the same or separate bugs
[10:49] <ubot5> Launchpad bug 1349366 in Ubuntu UI Toolkit "[PageStack] Pages does not get correct anchored in the PageStack area" [Critical,Confirmed]
[10:50] <karni> t1mp: ok
[10:50] <t1mp> thanks!
[11:16] <karni> t1mp: just FYI, I did a clean package build, and it started working. (anchorToKeyboard: true, that is). not sure what was wrong before, but I have no further feedback on the problem (which seemed to be 'Invalid' on my end)
[11:20] <t1mp> karni: oh that is weird. Can you still get a diff that shows what you changed?
=== MacSlow is now known as MacSlow|lunch
[11:21] <karni> t1mp: not really :/ but I think it was one of those dumb situations when the stuff you push to the phone wasn't really the thing you wanted to push (read: I may have not been updating it properly).
[11:21] <karni> t1mp: althouh I can say I removed "other" stuff that I tried, like anchors.bottom: keyboard.top, etc
[11:22] <karni> which may have played a role here (and breaking rather than helping)
[11:22] <t1mp> karni: okay
=== vila-lunch is now known as vila
=== chihchun is now known as chihchun_afk
=== CodePulsar__ is now known as CodePulsar
[11:56] <seb128> Kaleo, thanks for the pointer to the palette, that doesn't seem to cover colors for e.g action buttons though?
[11:56] <seb128> do you know if that's planned?
[11:57] <t1mp> seb128: we have more colors for the buttons now than we have slots in the palette, so we need to evaluate if we add more properties to the palette or solve it in a different way
[11:58] <t1mp> gventuri: ^ are there online docs already giving guidance when to use which color for buttons?
[11:58] <seb128> t1mp, we got a mp on settings to set the color of a button to blue
[11:58] <seb128> the issue is that it codes the color in our source
[11:58] <t1mp> seb128: basically it is now UbuntuColors.green for positive actions, UbuntuColors.red for negative actions
[11:58] <t1mp> seb128: I haven't seen cases yet where it should be blue
[11:59] <seb128> which means the day the designer decide to change the color, we need to go through every application to change them then
[11:59] <seb128> could be green, sorry
[11:59] <seb128> still same issue
[11:59] <t1mp> seb128: yes, that's true. Maybe we need to add property color positiveAction: UbuntuColors.green to the palette
[12:00] <seb128> t1mp, right, see https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1350915
[12:00] <ubot5> Launchpad bug 1350915 in Ubuntu UI Toolkit "Colors should have symbolic names" [Undecided,Confirmed]
=== DanChapman is now known as DanChapman|afk
[12:22] <daker> Kaleo: yo, Question: why on the "share/delete" menu present on the camera app the icons are on the right ?
=== chihchun_afk is now known as chihchun
[12:42] <Kaleo> seb128, the buttons used to use the color palette
[12:42] <Kaleo> seb128, t1mp would know more
[12:43] <Kaleo> seb128, (in general the palette is restricted to a limited set of colours as to ensure visual consistency)
[12:43] <Kaleo> daker, not sure, it looks better? :)
[12:43] <seb128> Kaleo, well, in that case it's to make an action button green without coding the "green" in our source
[12:44] <Kaleo> seb128, I see
[12:46] <daker> Kaleo: inconsistent, the browser have them on the left (https://lh3.googleusercontent.com/--f9KYJ8C4gI/U9rT2g9D-tI/AAAAAAAADII/UNlLyhH4A94/w435-h725-no/browser1.png)
[12:46] <Kaleo> daker, agreed
[12:47] <Kaleo> daker, I forwarded the question to design
[12:51] <t1mp> Kaleo, daker icons for the actions overflow will be on the left also in UITK, see https://www.dropbox.com/s/355g7jptmxtzt0b/Screenshot%202014-08-01%2014.50.24.png
[12:51] <t1mp> (those changes did not land yet)
[12:52] <daker> t1mp: nice
[12:52] <t1mp> oh, and I see the browser screenshot doesn't have divider lines inbetween the itesm
[12:52] <t1mp> Kaleo, seb128 buttons now use a unique grey color that is not in the palette or UbuntColors.. and there are more cases like that with new designs from last week
[12:52] <daker> t1mp: it's a custom implementation
[12:53] <t1mp> Kaleo, seb128 so some components are not using the palette for all colors now, we need to fix that by extending the palettes
[12:56] <Kaleo> t1mp, just a note, we have to avoid extending the palettes at most (all) cost
[12:57] <t1mp> Kaleo: I don't know yet what is the "real" solution
[12:57] <Kaleo> t1mp, yeah I don't know the current colors
[12:57] <t1mp> Kaleo: but for example, there is no place now to put the colors for positive and negative action buttons (green and red)
[12:57] <t1mp> Kaleo: all buttons used to be orange, now we have grey, red, green. And text selection became blue
[12:58] <Kaleo> t1mp, ok, well, yes, the palette is the right place
[12:58] <t1mp> the palette is a bit confusing right now with all the new colors. It is not always clear which color needs to be picked for each part of each component
[12:58] <Kaleo> t1mp, if there is no color in the palette corresponding to these values it's annoying
[12:58] <Kaleo> t1mp, visual designers need to rationalize it I think
[12:59] <t1mp> maybe after rtm we have to re-think the palette
[12:59] <Kaleo> t1mp, and really own the palette definition
[12:59] <t1mp> Kaleo: I added ubuntu-ux to the bug because of that
[12:59] <Kaleo> t1mp, ok
[12:59] <Kaleo> t1mp, who is giving all the new colors?
[12:59] <Kaleo> jouni?
[13:00] <t1mp> I get them from jouni
[13:00] <Kaleo> t1mp, ok so he has to be systematic about using and defining the palette
[13:00] <Kaleo> t1mp, not just giving colors
[13:00] <Kaleo> t1mp, let's let him know
[13:01] <t1mp> he's not online now. We briefly discussed it last week but since we were in a hurry to get other stuff done for rtm we postponed it
[13:02] <t1mp> if we decide to simply add more colors to the palette it should not be too much work
[13:02] <t1mp> but the visual designers need to re-think which colors we need in the palette
=== _salem is now known as salem_
=== chihchun is now known as chihchun_afk
[13:03] <Kaleo> t1mp, yes that's the thing
[13:03] <Kaleo> t1mp, taking that easy solution of extending the palette is not good in the long term
[13:03] <Kaleo> t1mp, it needs to be done only if strictly necessary
[13:04] <Kaleo> t1mp, good luck!
[13:04] <t1mp> Kaleo: do you see a different way of getting red (negative), green (positive) and blue (text selection) in the palette?
[13:05] <Kaleo> t1mp, there are only 2 possibilities:
[13:05] <Kaleo> t1mp, either one existing palette value has the right color (or is made to have the right color)
[13:05] <Kaleo> t1mp, or extension
[13:05] <t1mp> those all used to be selected.foreground (orange)
[13:05] <Kaleo> t1mp, by rationalizing the color usage the first possibility might work
[13:12] <dpm> dear $DEITY, please someone change that alarms sound!
[13:12] <dpm> I've got it going on on my phone and laptop now
[13:13] <nik90> dpm: ?
[13:14]  * nik90 pretends to not have read it :P
[13:15] <Snipes> Is the list of the icons available for Ubuntu touch online somewhere?
[13:15] <dpm> the thing is, it does not stop
[13:15] <dpm> you're in a hangout, your phone out of reach, and it keeps ringing the alarm tone forever! :)
[13:15] <nik90> dpm: well it rings for a default of 30 mins
[13:16] <nik90> dpm: unless you change that default value in the alarm settings to a minimum of 10 mins
[13:16] <nik90> dpm: and also that's what will make you wake up in the morning :D (atleast in my case)
=== chihchun_afk is now known as chihchun
[13:33] <om26er> t1mp, Hi! I didn't find much clear documentation on head.actions, I am using this code in an app http://paste.ubuntu.com/7924024/
[13:33] <om26er> whenever the app starts the search field and backbutton are visible.
[13:46] <nik90> om26er: check out http://developer.ubuntu.com/api/qml/sdk-14.10/Ubuntu.Components.PageHeadState/
[13:46] <nik90> om26er: it has exactly what you want to implement
[13:47] <nik90> om26er: basically you need to use PageHeadState to switch between the search and normal mode
[13:50] <Snipes> Is there any way to know what icons are available to use in the header?
[13:50] <nik90> Snipes: yeah, open the Ubuntu SDK Gallery where there is a section "icons" which shows all the available icons
[13:51] <Snipes> In the SDK itself?
[13:53] <nik90> Snipes: yes. Qtcreator -> Tools -> Ubuntu -> Showcase Gallery
[13:53] <Snipes> Oh nice. Of course I'm on my android phone and away from the computer. Is this online somewhere also?
[13:54] <nik90> Snipes: not that I know of .. sry
[13:54] <Snipes> That's alright. As long as its somewhere.
[13:54] <elopio> jgdx: here are the changes I mentioned for your branch: https://code.launchpad.net/~canonical-platform-qa/ubuntu-system-settings/1219793-reset-page_object/+merge/229148
=== chihchun is now known as chihchun_afk
[14:03] <popey> balloons: here?
[14:03] <balloons> popey, lol, sure
[14:03] <popey> ☻
[14:03] <popey> balloons: I'm buzzing through http://people.canonical.com/~dpm/core-apps-sponsoring/
[14:03] <popey> https://code.launchpad.net/~andrew-hayzen/music-app/listitem-actions/+merge/220290 for example
[14:03] <popey> seems to be failing spectacularly
[14:06] <balloons> wow, that is STILL in progress?
[14:06] <jgdx> elopio, I saw, great stuff. Will review in a bit
[14:07] <balloons> and yea, I've been chattering with andrew about the rewrite for music app
[14:07]  * balloons builds new calendar click
[14:07] <elopio> jgdx: thanks. Let me know if you want me to correct something, because I'll keep sending some refactors and I'll just copy the style we agree on.
[14:08] <popey> balloons: any ideas?
[14:09] <balloons> popey, none off the top.. likely he's changed much of the UI
[14:09] <balloons> popey, the music app tests are pretty custom and inflexible, so they need to be brought in line with the rest
[14:09] <balloons> I'll pull it
[14:10] <popey> balloons: the merge updates the tests too
[14:10] <balloons> yea, but it suffers from the same problem..
[14:10] <balloons> the new tests are in the same style
[14:12] <balloons> anyways, I'll ping ahayzen when he gets on again.. I assume you want to land the branch asap yes?
[14:13] <balloons> popey, anything else you are concerned about?
[14:14] <popey> yeah, its been outstanding for a while.
[14:14] <balloons> I've been working a lot with the calendar guys trying to land some massive branches and clean up bugs.. We filed some more bugs, but I think there's progress
[14:14] <popey> https://code.launchpad.net/~pkunal-parmar/ubuntu-calendar-app/def-calendar/+merge/228837 that one
[14:15] <balloons> yea, that's the last one I was wanting to land. Well, I wanted to land mihir's https://code.launchpad.net/~mihirsoni/ubuntu-calendar-app/1338956/+merge/226362 as well, but we discovered https://bugs.launchpad.net/ubuntu-calendar-app/+bug/1351024
[14:15] <ubot5> Launchpad bug 1351024 in Ubuntu Calendar App "Date & Time picker is not working on device." [High,Triaged]
[14:17] <Snipes> Has the bottom edge for the dialer app been implemented and is this code available to view somewhere?
[14:17] <t1mp> Snipes: bzr branch lp:dialer-app
[14:19] <balloons> popey, so what else? For clock, I've been watching https://code.launchpad.net/~carla-sella/ubuntu-clock-app/add-alarm-test/+merge/229084
[14:22] <popey> those were the main ones
[14:22] <popey> music & calendar
[14:27] <balloons> ok, I'll be pushing calendar once the build finishes
[14:31] <popey> great
[14:31] <popey> thanks balloons !
[14:38] <balloons> woot..
[14:42] <balloons> popey, https://myapps.developer.ubuntu.com/dev/click-apps/156/changerequest/
[14:43] <popey> approved!
=== Thiago_ is now known as Guest65944
[14:47] <Guest65944> Bom dia!
[14:48] <Guest65944> Good morning!
[15:03] <popey> beuno: will we have a "whats new" category in the click scope do you know?
[15:03] <beuno> popey, I can't tell anymore
[15:04] <beuno> those designs have been chaotic and unpredictable to me
[15:04] <beuno> we certainly have the data
[15:04] <popey> ah
[15:04] <beuno> alecu may be able to answer
[15:12] <alecu> popey: the designs include a category of "new apps": point 3 here: https://sites.google.com/a/canonical.com/unity8dash/scopes/store/store-home
[15:14] <popey> thanks
[15:14] <alecu> beuno: if we have that data, then it's a matter of adding a highlight with it. It should be a highlight without a department, like we discussed recently, if it needs to be shown on the store home.
[15:15] <alecu> popey: also, we are working on having department landing pages contain *every app* in the department: point 2 here: https://sites.google.com/a/canonical.com/unity8dash/scopes/store/store-home
[15:16] <alecu> that's the next big task for the apps scope, and it's bug #1337864
[15:16] <ubot5> bug 1337864 in unity-scope-click (Ubuntu) "Apps missing in some store categories" [High,Confirmed] https://launchpad.net/bugs/1337864
[15:18] <elopio> ping pmcgowan: dobey has replied to bug 1332629. The abort doesn't come from u1 credentials. Do you know who from system settings could look at it?
[15:18] <ubot5> bug 1332629 in ubuntu-system-settings (Ubuntu) "lrt test found crash file _usr_bin_system-settings.32011.crash" [Critical,New] https://launchpad.net/bugs/1332629
[15:19] <nik90> balloons: found a small issue with the locale stuff. I see carla did self.useFixture(fixtures.EnvironmentVariable('LC_ALL', newvalue='C')) as you recommended. However when I open the clock app, the time shown is 17:19:00 while normally only 17:19 is shown. Why does it show the seconds while the Qt locale doesn't?
[15:20] <nik90> balloons: the test still passes, but it is different from real world usage
[15:21] <pmcgowan> elopio, not sure, I asked chris to poke at it some more and see if he could reproduce manually based on some theories
[15:21] <elopio> ok, that sounds good.
=== HoloIRCUser1 is now known as Snipes
[15:28] <beuno> alecu, highlights don't auto-generate atm, they are all manual
[15:29] <beuno> (atm, we will in the future)
[15:32] <pmcgowan> elopio, dobey analyzed it there its the same as the bug I pointed to - a qfatal call in bluettoh
[15:32] <pmcgowan> tooth
[15:34] <alecu> beuno: I can live with that coming post rtm (but probably before launch!)
=== gatox is now known as gatox_lunch
=== renato is now known as Guest43934
=== gatox_lunch is now known as gatox
[16:40] <achiang> how can i set an environment variable for a click app?
[16:41] <achiang> to debug, that is
[16:41] <achiang> can i just set it on the Exec= line in my desktop file?
[16:45] <achiang> nope, that doesn't work
=== DanChapman|afk is now known as DanChapman
[17:01] <balloons> nik90, interesting observation. We could choose a different locale. Off the top I would assume the different is the default locale vsalue
[17:02] <balloons> achiang, you need to pass it via upstart app launch
[17:03] <achiang> balloons: do you have an example?
[17:04] <dpm> popey, balloons, did you have that chat about all the branches that need QA attention?
[17:05] <balloons> dpm, we did, minus a mic which didn't work on my edn
[17:05] <balloons> dpm, if you have any you'd like to highlight, go for it
[17:05] <balloons> calendar, which got a new version into the store, yay, and music were the bit two
[17:08] <balloons> achiang, I'm thinking
[17:10] <dpm> balloons, sounds good. I was just thinking of the two ones that we've got in the sponsoring page:
[17:10] <dpm> https://code.launchpad.net/~nskaggs/ubuntu-clock-app/switch-emulators-to-proxy-object/+merge/227103
[17:10] <dpm> https://code.launchpad.net/~nskaggs/reminders-app/fix-1347905/+merge/228917
[17:10] <dpm> it'd be good for a QA person to approve them
[17:11] <dpm> as I don't feel I have the AP knowledge to review them
[17:16] <balloons> dpm, ahh gotcha
[17:16] <balloons> yes, reminders needs to land, I'll have someone +1 it
[17:18] <dpm> thanks balloons
[17:18] <balloons>  on that note plars is still reviewing the mp for francis to put it on the dash
[17:50] <naxuroqa> Hello! I am writing a gtk based application in vala/c and wanted to ask if it's possible to use the appindicator library via dbus / without linking to it at compile time
[17:52] <naxuroqa> Is this the right channel to ask questions like this?
[18:04] <dpm> hi naxuroqa, yes, this is the right channel, but everyone is really focused on the phone and the Ubuntu SDK, so you might not find many vala experts right now
[18:06] <beuno> alecu, likely won't be before launch  :/
[18:30] <doflaherty_> I'm trying to run on the device from QtC - but instead of my app, i'm just getting "snow"
[18:36] <doflaherty_> it seems to have stopped working since QtC got updated
[19:23] <Letozaf_> balloons, hello
[19:37]  * popey pokes balloons with ahayzen 
[20:37] <ahayzen> popey, whats up?
[20:43] <popey> ahayzen: i think balloons wanted to speak to you about failing music tests
[20:43] <popey> https://code.launchpad.net/~andrew-hayzen/music-app/listitem-actions/+merge/220290 those ones
[20:43] <ahayzen> popey, ah ... most (if not all) are nothing to do with ap and are todo with us/mediascanner2
[20:44] <ahayzen> popey, and are the same as the ones failing in CI
[20:44] <popey> yeah
[20:44] <ahayzen> popey, jamesh has suggested some ideas...i'm gonna flash the later image and see what is going on...and then either fix/create mini app to show any ms2 problems to jamesh
[20:46] <popey> ah okay.
[21:09] <balloons> mmhmmm
=== salem_ is now known as _salem
[22:08] <romanzonov> Hello everybody.
[22:08] <romanzonov> I have a question about SDK.
[22:08] <romanzonov> Why you cutted possibility to test apps width "Ctrl+F12"?
[22:08] <romanzonov> And how I can test app now?
[22:10] <nik90> romanzonov: hi
[22:10] <nik90> romanzonov: what kind of app is yours? Pure QML app? or one with c++ plugins?
[22:11] <romanzonov> nik90, hi
[22:11] <romanzonov> Pure QML
[22:11] <romanzonov> After last upgrade...
[22:12] <romanzonov> I tried to start app with "Build->Ubuntu->something like run on device", but that was cutted too
[22:17] <nik90> romanzonov: hmm
[22:17] <nik90> romanzonov: I will test this tomorrow to see what happens
[22:17] <nik90> romanzonov: have you tried adding a kit and then choosing the device or emulator
[22:20] <nik90> romanzonov: I remember they were trying to make it make the execution similar to cmake projects where you just press ctrl+r to run it on device, or emulator or desktop
[22:21] <nik90> romanzonov: the update is meant to make it easier
[23:13] <ahayzen> Hey anyone around with an image <161 that could quickly test an app for me?
=== _salem is now known as salem_
=== salem_ is now known as _salem