=== chriadam|away is now known as chriadam === chihchun_afk is now known as chihchun [07:16] good morning [07:26] rpadovani: could you please check MR , Jenkins has failed integration https://code.launchpad.net/~rpadovani/ubuntu-calculator-app/1287340/+merge/209411 === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun [09:20] Good morning all; happy Monday, and happy Day of Awesomeness! :-D [10:04] where can be ubuntu app platform notes be located [10:05] where can be ubuntu app platform notes be located === chriadam is now known as chriadam|away [10:06] where can be ubuntu app platform notes be located [10:07] where can be ubuntu app platform notes be located [10:10] sridhar_: http://developer.ubuntu.com/ [10:22] dpm popey: Moin moin [10:22] 3 mp are waiting for review with bug fixes. Should be enough for a fine new working release :) [10:22] m-b-o: hurrah! [10:23] popey: I've removed WorkerScript entirely, now no more crashes [10:23] hey m-b-o o/ [10:23] 0/ [10:24] awesome news for a monday morning! thanks m-b-o! [10:24] popey tgim! ;) [10:25] kalikiana: can you quickly also review https://code.launchpad.net/~kalikiana/u1db-qt/hackingDocs/+merge/205426...I think it is almost good to go..I have commented on it [10:28] m-b-o, nice work. To have some background while looking at the MPs, why did you decide to remove the WorkerScript? [10:28] nik90: by review you mean incorporate the comments? you are the reviewer here :-) [10:29] will take a look in a jiffy [10:29] kalikiana: yeah that's what I intended to say :) [10:30] nik90: I mean… what you said is actually quite rethorically accurate. just the ambiguity of the jargon made me wanna double-check [10:31] how to check version of ubuntu-sdk [10:31] i am using ubuntu 13.10 [10:31] can ubuntu 13.10 be used to develop html5 apps [10:31] ??? [10:31] sridhar_: apt-cache policy qtdeclarative5-ubuntu-ui-toolkit-plugin [10:31] sridhar_: run above ^^ in terminal [10:32] it shows Installed: 0.1.46+14.04.20131216bzr936saucy0 [10:32] that means 14.04 version right??? [10:33] can this version be used to develop html5 apps??? [10:34] sridhar_: yes, but it looks like you are not running the latest SDK version. [10:34] sridhar_: I am not sure about the html5 apps since I am not really familiar with that.. I am more a QML guy [10:44] dpm: two bugs related to WorkerScript [10:44] https://bugs.launchpad.net/ubuntu-weather-app/+bug/1287446 [10:44] Launchpad bug 1287446 in Ubuntu Weather App "Sometimes, the temperature and conditions are not shown" [Undecided,Fix released] [10:44] https://bugs.launchpad.net/ubuntu-weather-app/+bug/1289536 [10:44] Launchpad bug 1289536 in Ubuntu Weather App "Crash opening weather app and switching backends" [Critical,In progress] [10:45] dpm: in the first one, data was disappearing (see last comments) [10:45] second one crashes the app, when the app does HTTP requests in the WorkerScript at startup [10:47] but I can't really say if a http request was needed for crashing, just a speculation === chihchun is now known as chihchun_afk [10:51] kalikiana: btw the u1db-qt project page is pointing to http://developer.ubuntu.com/api/ubuntu-12.10/qml/u1db/overview.html for its documentation which does not exist anymore. This needs to be updated to http://developer.ubuntu.com/api/qml/sdk-14.04/U1db/ === maclin__ is now known as maclin [10:52] nik90: good catch, I'll fix that [10:53] hmm except I can't [10:53] kalikiana: me neither :P [10:53] one would think anyone in the team can edit the descriptions [10:54] pretty sure I used to [10:54] kalikiana: oh btw, I was thinking of triaging bugs for u1db-qt [10:54] kalikiana: can I create new milestones to target bugs which are important as of now and push other bugs to later milestones? [10:55] nik90: yes, please do [10:56] btw the wonderiousFields got merged if you haven't seen, should solve a ton of buglets [10:56] kalikiana: yup I saw that...excited! [10:57] kalikiana: but we need to make sure that these make it to the phone once the qt5.2 dust settles [10:57] I'd love to take a stab at filter functions, though not sure right now if I'll have time for it [10:58] nik90: yeah. problem is it's not a fixed timeframe, all images don't get promoted so it can take days [10:58] so not sure how to reliably find out when stuff hits the phone [10:58] kalikiana: I know..I will do that testing for you since I have a couple of sample apps ready which I can test on desktop and phone to figure that out [10:59] kalikiana: hmm I cannot create milestones since I don't seem to have permissions for that [10:59] nik90: that would be most welcome. [11:00] nik90: you should see "Create Milestone" here https://launchpad.net/u1db-qt/trunk [11:00] kalikiana: i know, but I don't see the "Create milestone" button there [11:01] kalikiana: I created the milestones for the clock app. So I compared with the procedure with it [11:01] nik90: try again now [11:02] kalikiana: yup works now.thnx [11:02] cool. I set a "Release manager" so apparently it is what I hoped it might be :-D [11:03] lol [11:03] kalikiana: apparently someone reported the broken link before at https://bugs.launchpad.net/u1db-qt/+bug/1218549 [11:03] Launchpad bug 1218549 in U1DB Qt/ QML "Link in summary not working" [Undecided,Confirmed] [11:04] nik90: I'll see if I can find out why neither of us can update the descriptions [11:04] kalikiana: okay. [11:16] nik90: I pushed an update, the file now mentions apt-get build-dep [11:16] kalikiana: taking a look now [11:18] kalikiana: approved [11:18] kalikiana: btw I moved all u1db sync bugs to milestone 2.0 since we are other bugs which take priority. [11:26] nik90: hmm not sure I'd agree that sorting/transformation is more urgent [11:26] though I guess I have no fair measure on that [11:27] I'm with nik90 on that. If you can't sort, then you can't display a sorted ListView, which is really annoying. [11:27] kalikiana: you mean sync is more important sorting? or sorting is not important for now? [11:27] transformations are similar: if you can't index on lower() or splitwords() then you can't simulate a phonebook or similar without caring about case sensitivity. [11:28] kalikiana: I believe the music-app is blocked from using u1db due to lack of sorting since it is quite important for them. [11:29] nik90: I assumed sync was more important up until right now. but I don't get much feedback from most app devs on what blocks them outside of specific bugs [11:29] kalikiana: in fact in the clock app, the timer listview is not in any particular order although I would like it to be ordered alphabetically. Ofcourse this is not a priority for the clock app but something still nice to have before sync. [11:30] doing sync also almost requires you to port the test suite. [11:30] syncing is *hard*. there are a lot of corner cases, which the test suite tests. :) [11:30] kalikiana: the thing is Sync is a huge feature in itself meaning it has quite a number of bugs which needs to be fixed before we can ask devs to use it. In the meantime for 14.04, I thought it would be wise to focus on smaller issues like sorting and so on. [11:31] and what aquarius said ^^ [11:31] aquarius: btw good morning ;) [11:31] hey :) [11:35] nik90: I think I agree with that. there's also still some unclear concerns with regard to operator requirements and on-going sync in a world where the shell happily freezes and kills apps [11:35] ;) [11:36] oh btw there's a uds session where that will come up, about storage apis [11:36] I'm hopeful we get some people there who know about the server side [11:37] kalikiana: oh nice..I did not see it. I will try to attend them just to get an idea of what happens [11:38] as I understand it'll be more focussed towards coming features than appdev [11:38] ok [11:39] kalikiana: btw is this bug valid? https://bugs.launchpad.net/u1db-qt/+bug/1198110 [11:39] Launchpad bug 1198110 in U1DB Qt/ QML "Templates in QtCreator should be proper projects" [Medium,Confirmed] [11:39] kalikiana: I dont see any u1db project templates in qtcreator. [11:40] nik90: do you have -examples installed? [11:41] kalikiana: what's the full package name? [11:42] libu1db-qt5-examples [11:44] also, and I know I keep reiterating this, I really do think it would be worthwhile coming up with some way to bridge the python and qt u1dbs so that you can run the python test suite against the qt u1db. The test suite is really comprehensive, for a reason :) [11:44] kalikiana: just installed libu1db-qt5-examples, but I don't see it in qtcreator templates wizard [11:45] aquarius: helping hands are always welcome :-) [11:45] I know,I know :) [11:46] aquarius: btw some index and query high priority bugs got fixed. [11:46] nik90: hmm you do like, File > New, right? and there you see Projects [11:47] kalikiana: http://imgur.com/Zotykkh [11:47] libu1db-qt5-examples: [11:47] Installed: 0.1.5+13.10.20130916-0ubuntu1 [11:47] nik90: now click on Apps, please [11:48] kalikiana: http://imgur.com/zsfkmtB [11:48] hrm [11:48] * nik90 loves screencloud..:) [11:49] what is screencloud? [11:49] kalikiana: it takes a screenshot and automatically uploads to imgur and takes puts the link in the clipboard [11:49] oooh! [11:49] kalikiana: all I need to is then is take a screenshot and past the link to IRC [11:49] i like the sound of that [11:50] popey: I kinda figured [11:50] nik90: Where'd you get it? [11:50] popey: It is in the ubuntu software center store for 12.04, 12.10 and 13.04. [11:50] popey: for 13.10 and 14.04, I need to install a PPA I think..let me grab the link for you [11:50] thanks [11:51] I really must fix my screenshotter [11:51] the thing mine did that I liked was that it screenshots both the whole screen *and* the active window, and then with one button push uploads one or the other and gives you the url [11:52] popey: http://software.opensuse.org/download.html?project=home%3Aolav-st&package=screencloud [11:52] so I never have to care about alt-prtsc vs prtsc :) [11:52] thanks [11:52] nik90: my mistake, the files are in libu1db-qt5-dev [11:52] popey: that's how I installed it on 13.10. It continued to work on 14.04 when I upgraded. So hope it works for you as well. [11:52] kalikiana: trying now [11:52] * kalikiana hugs ubuntu-sdk-libs-dev [11:53] kalikiana: ooh I get http://paste.ubuntu.com/7067205/ [11:54] kalikiana: either way, is there a plan to add u1db templates to qtcreator? since it seems like this package is not installed by default on the developer's machine? [11:54] what is a u1db template? [11:55] nik90: thats amazing! [11:55] aquarius: for qtcreator, I am guessing it might add a sample project where it has code to create a u1db database and so on [11:55] nik90: well define "default", they should be pulled in by the sdk meta package [11:55] hrm. [11:56] popey: it has upload capability to u1db, imgur, dropbox...It is incredible yes [11:56] are there "templates" for other things? Not too keen on this idea [11:56] sounds like you never used qtc [11:56] aquarius: templates are there for html5, qml apps etc and so on in qtc [11:56] nope. I'm not a Qt developer :) [11:56] quite handy actually [11:56] nik90, yes, for project types, indeed [11:56] but "uses u1db" is not a project type like "tabbed UI" is :) [11:57] you're right. alas there is no other way in qtc [11:57] there's no "uses foobar component" in general [11:57] what you're talking about with u1db is not a project type -- it's a snippet, an example piece of code. This is not what the IDE is for -- it's what documentation is for, and possibly the snippets app that someone (ibeliever?) was working on. [11:58] agreed [11:58] I suppose we could bring u1db code examples through that way [11:58] *nod* [11:58] well, better something than nothing and blocking on qtc features for months [11:58] I have not seen that snippets feature so far [11:59] that's how code examples should work. No-one should ever say "I want to create a project with U1DB in it!". That's the tail wagging the dog; it's like asking for a meal with something red in it. :) [11:59] https://plus.google.com/u/0/+JonoBaconProfile/posts/5ckpQxWJBgG [11:59] is about the snippets stuff [11:59] code snippet manager. [12:00] kalikiana: I suppose we can postpone this until a better system lands. Until then we anyway have some good QML docs now in developer.ubuntu.com [12:00] aquarius: I like meals with something red in it [12:01] t1mp, radish surprise for you then, my man :) [12:02] aquarius: meh. that's mostly white if you cut it up :) [12:02] aquarius: but I like radishes :) [12:02] interestingly a comment talks about "how is it better than qtc snippets" what is that supposed to refer to? [12:02] t1mp: do you have any idea what it might be? [12:03] aquarius: hmm.. I just discovered a pattern in the foods that I like, thanks to you. :p [12:03] kalikiana: let me read. I got distracted with google's header on that webpage when scrolling down [12:03] (I'm currently working on the header for uitk) [12:28] artmello: hi, have you noticed tests failing on the device after moving gallery to the click package ? [12:35] nerochiaro: hi, there was (not sure if still happening) an error during autopilot testAlbumView. It was related to a timeout while changing tabs [12:36] nerochiaro: but it was not a gallery issue [12:39] artmello: i am flashing the new image now to see if things get better. but it seems to me that all tests i run start with an album open and fail because of that. === _salem is now known as salem_ [12:41] nerochiaro: humm, I hadnt noticed that yet. Will test here [13:44] popey: It seems there are still core apps build failures -> https://launchpad.net/~ubuntu-touch-coreapps-drivers/+archive/daily [13:44] nik90: ok will take a look.. [13:53] popey: sorry I see that victor has prepared a patch for the PPA build fail at https://code.launchpad.net/~vthompson/ubuntu-clock-app/fix-1284637-clock/+merge/210208 [13:54] awesome [13:54] Victor is awesome ☻ [13:55] indeed victor is awesome [13:55] zsombi: ping [13:55] nik90: pong [13:55] zsombi: In the clock app, do I need to explicitly add qtorganizer5-eds as a dependency for the deb installation? [13:56] zsombi: or does the SDK package already have it as a dependency? [13:56] zsombi: I am asking since it seems that the jenkins machine doesn't seem to have EDS installed when it installs clock app to run tests which doesn't fully simulate the phone environment then [13:56] nik90: if your app depends on that, than yes. SDK doesn't have dependency to that [13:57] zsombi: but doesn't Alarms API depends on EDS which makes SDK to depend on it? [13:57] nik90: the pronblem is that Jenkins machine will die with EDS if I'm not mistaken... AP tests may at least... [13:57] nik90: Alarm API depends on qtpim [13:57] noty on EDS [13:57] not [13:57] I am not exactly sure what qtpim is [13:58] is that the protocol which saves alarms into EDS? [13:58] or any other storage systems? [13:58] nik90: Qt PIM (Personal Information Management) [13:58] nik90: PIM is calendar, contacts, versit, etc [13:58] zsombi: ok..so you wouldn't recommend jenkins installing EDS? [13:59] nik90: well, you could try, but we already have problems building SDK when the EDS backend is installed.... it simply hangs on teh qmldump :/ [13:59] zsombi: the alarm sunday bug was detectable on the phone and my machine but not on jenkins. Hence I was trying to improve the jenkins environment [14:00] nik90: and then renato said sthing that because of dbus limitations EDS used in backend in Jenkins may not be tghe best idea... [14:00] balloons: ^^ [14:00] don't exactly understood what was the problem there [14:00] didn't I mean :) [14:01] ok [14:01] zsombi: I will contact renato for the exact reason, so that I can provide that same answer to anyone who asks me why the tests didnt fail on jenkins but on the phone :) [14:02] renato: ^^ [14:02] nik90: but now that U R here, I have a MR to you to test [14:02] zsombi: ;) [14:02] launchpad is not so keen to help me out :/ [14:03] nik90: https://code.launchpad.net/~zsombi/ubuntu-ui-toolkit/alarm-fetch-fix/+merge/208749 [14:03] nik90, zsombi, I am not sure why the tests did not fail on jenkins, but the problem is, we can not touch on user database (EDS database), this is not safe we do not know what is there already and we can not write data for tests on there [14:03] nik90: jenkins fails on a test that got fixed in a separate MR, so don't worry in that [14:04] there is already some solutions for that [14:04] zsombi, nik90 , you can export all user environment variables to a temporary user home, and start a new eds server and use that [14:05] zsombi, nik90 , and now that we start implement the sync functionality touching on user database can cause a really mess [14:06] zsombi: which package do I install from the MP? I see two deb links in the last jenkins comment. [14:06] not worried about the AP failure [14:07] hehh? [14:07] zsombi: I see http://jenkins.qa.ubuntu.com/job/ubuntu-ui-toolkit-trusty-armhf-ci/803/artifact/work/output/*zip*/output.zip and http://jenkins.qa.ubuntu.com/job/generic-mediumtests-builder-trusty-armhf/3363/artifact/work/output/*zip*/output.zip [14:07] renato: ok [14:07] nik90 :ah, the first one, the second one must be the AP tests... [14:07] zsombi: ah ok.give me 5 mins === oSoMoN_ is now known as oSoMoN [14:08] renato: I believe elopio is working on running clock alarm tests in a different environment without affecting user space. === chihchun_afk is now known as chihchun [14:18] zsombi: On testing on the phone, https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1283859 doesn't seem to be fixed [14:18] Launchpad bug 1283859 in Ubuntu Clock App "Updated recurring alarm values are not reflected in the indicator until phone reboot" [High,Confirmed] [14:19] zsombi: and https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1285958 was fixed with renato's patch in EDS. So I cannot say if this MP did something or not [14:19] Launchpad bug 1285958 in Ubuntu Clock App "Recurring alarm can't be set for sunday" [High,Confirmed] [14:20] nik90: renato's fix was only working for dates <= 7 days... this fixes longer periods [14:21] zsombi: ah okay..but again I cannot test > 7 days...that said, I can still save alarms on sunday properly. so that's good [14:21] nik90: was the bug #1283859 fixed by renato for you? [14:21] Launchpad bug 1283859 in Ubuntu Clock App "Updated recurring alarm values are not reflected in the indicator until phone reboot" [High,Confirmed] https://launchpad.net/bugs/1283859 [14:21] zsombi: nope. that indicator not refreshed bug still exists [14:22] nik90: and you cannot reproduce it on desktop, right? [14:22] zsombi: haven't tested it on desktop.. [14:22] nik90: pls do [14:24] renato: seems switching from dueDate() to startDate() did not solve #1283859 [14:24] zsombi: on the desktop without your patch, I can reproduce on the desktop as well. [14:25] nik90: hehehe :) [14:25] zsombi: Should I test it on the desktop with your patch? [14:25] nik90: I am working on the date conversion... and I start getting alarm notifications on the desktop :D [14:25] nik90: yep [14:26] nik90: make sure when you build UITK you're setting ALARM_BACKEND=memory env var [14:26] nik90: otherwise the qmldump will hang [14:26] zsombi: I suppose I need to build uitk and then make install it? [14:27] nik90: yep, but when testing, remove the ALARM_BACKEND env var [14:29] zsombi: how do I set ALARM_BACKEND=memory env var [14:29] nik90: export ALARM_BACKEND=memory [14:29] I usual run the source export_modules.sh [14:29] nik90: in a terminal [14:30] nik90: in the terminal you build UITK [14:30] yup, I branches your code and ran qmake; make [14:30] and then export ALARM_BACKEND=memory [14:30] followed by sudo make install [14:30] nik90: nopez [14:30] nik90: branch, set the var then qmake; make [14:30] ok [14:31] otherwise you will not reach to the sudo make install ever [14:31] well actually I got stuck in the sudo make install part, but now I am following the order you gave [14:35] nik90: just to be sure: Alarms API has a test that tests whether an alarm is updated test_updateAlarm_SameType() and test_updateAlarm_DifferentType() [14:35] nik90: both pass with EDS [14:36] nik90: on desktop at least [14:37] zsombi: I keep getting stuck at /usr/lib/x86_64-linux-gnu/qt5/bin/qmlplugindump -notrelocatable Ubuntu.Components 0.1 ../../ 2>/dev/null > //usr/lib/x86_64-linux-gnu/qt5/qml/Ubuntu/Components/plugins.qmltypes [14:37] zsombi: I branches code again, ran export ALARM_BACKEND=memory, qmake, make , sudo make install [14:37] how about I run source export_modules.sh and run the clock app from the terminal? [14:38] nik90: weird... you've branched by MP, right? [14:39] zsombi: yes [14:39] m-b-o, sorry for not answering this morning. I'll have a look at the MRs between today and tomorrow [14:39] dpm: no problem! [14:39] anybody knows abt html5 app development [14:40] should i use ubuntu 14.04 for html5 ubuntu app development [14:40] nik90: does it get stuck when running the make install? [14:40] zsombi: yes [14:40] darn [14:44] Hey has somebody here recently worked with the appindicator bindings for python lately and could maybe point me to some documentation? [14:44] mainly I would need an overview of the functions and properties I have available [14:45] zsombi: anyways I think source export_module_dir.sh should also work. On testing clock app through that method, same thing on desktop. alarm did not update [14:46] nik90: ok, so the alarm is up to date in the clock app, right? [14:47] nik90: only the indicator which is showing wrong time [14:47] zsombi: yes..it updates *only* for recurring alarms on the clock app. [14:47] nik90: I don't get it. So if there's a one-time alarm, it doesn't work? [14:47] zsombi: https://bugs.launchpad.net/ubuntu-clock-app/+bug/1283212 [14:48] which is the new release of ubuntu 14.04 [14:48] ??? [14:49] nik90: huhh???! I have tests which pass, touching the same things you have failures on! [14:49] nik90: test_updateAlarmSameType() actually tests this particular bug [14:49] and passes === chihchun is now known as chihchun_afk [14:51] sridhar_: http://cdimage.ubuntu.com/daily-live/current/ [14:52] zsombi: then I guess somewhere EDS is messing that up [14:52] Launchpad bug 1283212 in Ubuntu Clock App "Unable to edit single type alarms and update them" [High,Confirmed] [14:52] zsombi: does your tests save the alarms to memory? [14:53] nik90: if EDS is in use, it doesn't save anywhere [14:53] nik90: it's up to EDS to do that [14:53] zsombi: I guess that's why your test suite doesn't fail then. [14:54] nik90: I don't get it.... it passes with EDS! [14:55] nik90: so it passes if I use memory or EDS backends of QtOrganizer [14:56] nik90: when I use memory backend, I save all the alarms in a binary file [14:56] zsombi: hmm..you think then it is a bug in the implementation of alarms in the clock app? [14:56] nik90: if you see the updated alarm in the model, then it shouldn't [14:59] zsombi: hmm, I can no longer reproduce the bug on the phone or on the desktop in the clock app [14:59] nik90: :D [14:59] zsombi: ofc the indicator is still stubbornly showing the old time for both single and recurring alarms [15:00] nik90: perhaps you had some "dust" in the db :D [15:00] zsombi: I will check with Pat McGowan again [15:00] zsombi: well it was reported by Pat McGowan [15:00] zsombi: which I was able to reproduce as well [15:00] :D [15:00] nik90: you have my deb package in the phone? [15:00] still? [15:00] zsombi: yes [15:01] nik90: then perhaps its because of that [15:01] nik90: the indicator might be a different issue... [15:01] zsombi: oh come on...don't tell me I need to flash my again to test the bug again :) [15:01] flash my phone* [15:01] nik90: nopez, :D === dholbach_ is now known as dholbach === Ursinha is now known as Ursinha-afk === bfiller is now known as bfiller_afk === Ursinha-afk is now known as Ursinha === kyleN is now known as kyleN-afk [18:42] Hey mzanetti, how are you? :-) Do you have any news on cameraImprovement for reminders? [18:47] mhall119, hey, do you know where should bugs about developer.u.c reported? [18:48] seb128: many of the pages have in barely-legible grey text at the bottom "report a bug on this site" [18:48] seb128: the link on the frontpage heads here: https://bugs.launchpad.net/ubuntudeveloperportal/+filebug [18:48] sarnold, lol, it's light, I didn't even notice it [18:48] sarnold, thanks [18:48] seb128: ^^ what sarnold said [18:48] seb128: I do'nt think that's an accident :) hehe [18:49] I see it's already reported [18:50] https://bugs.launchpad.net/ubuntudeveloperportal/+bug/1191200 [18:50] Launchpad bug 1191200 in Ubuntu App Developer site "Broken links (404) on Resources page" [Undecided,Confirmed] [19:03] Hey popey :-) [19:05] popey, about bug #1281745 I did a brach with a lot of console.log. Could you try to launch the branch on your Nexus 7 and click on white area? I hope there will be some output in console, so we can understand what component is broken [19:05] https://code.launchpad.net/~rpadovani/ubuntu-calculator-app/testFor1281745 [19:05] Launchpad bug 1281745 in Ubuntu Calculator App "white rectangle on calculator sidestage on N7 landscape" [High,Triaged] https://launchpad.net/bugs/1281745 === jackson is now known as Guest57400 [19:47] rpadovani: sure [19:48] what should I use for a date picker ? an option selector ? [19:57] justCarakas: for what ? [19:57] QML ? [19:58] dakar not html5 app [19:58] can I use the default html5 elements for that ? [19:59] justCarakas: well technicaly you should use an input type="date" but that's not yet plugged with the native date picket like in other OSs [19:59] bug 1170136 [19:59] Launchpad bug 1170136 in webbrowser-app (Ubuntu) "[browser] Add support for HTML5 date input" [Medium,Confirmed] https://launchpad.net/bugs/1170136 [20:00] and bug 1170138 [20:00] Launchpad bug 1170138 in webbrowser-app (Ubuntu) "[browser] Add support for HTML5 time input" [Medium,Confirmed] https://launchpad.net/bugs/1170138 [20:00] dakar ubot2 but if I need it for the app showdown , than what should I do [20:03] i am not really sure... [20:04] i'll try to ask the browser-app devs [20:04] me nether, I've been asking it for a couple of days this is the first time I get an answer and I've been waiting to continue with my app [20:04] thx [20:05] if my connection drops its because we have some internet probs [20:25] rpadovani: no output [20:26] !@# [20:27] popey, ok, I have to think to other debugs :-) [20:27] popey, thanks for the time! [20:27] np [20:27] rpadovani: file:///usr/share/unity8/Panel/MenuContent.qml:89:21: QML Loader: Binding loop detected for property "indexActive" [20:27] in my unity8.log.. related? [20:27] popey, no idea, I'll investigate it, thanks [20:28] k [20:32] dakar: have you been able to reach them ? === kyleN-afk is now known as kyleN === jhodapp_ is now known as jhodapp|brb === jhodapp|brb is now known as jhodapp [22:14] daker: is there an easy way to reference system icons in an HTML5 app? [22:16] mhall119: Anthony was/is working on that [22:18] mhall119: report a bug to track the progress [22:36] mhall119: and since I have no idea of the roadmap... [22:41] mhall119, can't you just use file:///usr/share/icons/ubuntu/whatever.svg, or whatever the path is? [22:41] or did "easy" mean "without hardcoding a path that may well change"? :) [22:42] justcarakas: there are about a zillion input type="date" polyfills; just have a google around and choose your favourite. [22:42] ah, darn, he's gone. If he asks again, someone tell him :) [22:51] aquarius: ya native implemtation is good [22:52] But* [22:52] daker, who has a native implementation? I thought only Opera 12 did and that was it [22:52] we don't, do we? Oxide won't even, afaik [22:52] there must be a decent touch-friendly polyfill; just googling, there are about one meelion date polyfills ;) [22:52] Chrome/FF [22:54] well stone me [22:54] I stand corrected [22:54] also, blimey. Didn't know that. Last time I checked Chrome didn't do it :) [22:56] They do [22:58] yeah, I've just tested :) [22:59] ours doesn't work in the ubuntu browser, though. :( [22:59] might do with oxide, though. [23:00] The sys icons should be defined with classes OE data- [23:00] Yes [23:00] 19:59:51 daker bug 1170136 [23:00] Launchpad bug 1170136 in webbrowser-app (Ubuntu) "[browser] Add support for HTML5 date input" [Medium,Confirmed] https://launchpad.net/bugs/1170136 [23:01] 20:00:01 daker and bug 1170138 [23:01] Launchpad bug 1170138 in webbrowser-app (Ubuntu) "[browser] Add support for HTML5 time input" [Medium,Confirmed] https://launchpad.net/bugs/1170138 [23:02] ooh, nice, http://gabrielecirulli.github.io/2048/ works perfectly on the phone browser [23:03] I was going to test it [23:03] I saw it on HN :) [23:06] it works. fun, too! [23:06] I mean, it's a great big Threes ripoff, but it's still fun :P [23:10] scroll handling needs a bit of a kicking, but it's playable === salem_ is now known as _salem === jhodapp is now known as jhodapp|afk [23:37] aquarius: yeah, "easy" means without a long absolute path [23:38] aquarius: daker: I've already started a port of it to a native Ubuntu HTML5 app [23:38] mhall119, fair enough. Not sure how that'd happen, though: injecting a magic URL scheme such as ubuntu-icons:// makes it hard to test your app in an ordinary browser. [23:38] plus, custom url schemes in HTML are evil ;) [23:39] aquarius: no idea, but it's done in QML using the theme [23:40] yeah, but that's because custom URL schemes in QML aren't quite as evil ;) [23:40] idea: [23:40] that wouldn't be too hard to do [23:40] and wouldn't be too evil, either. [23:41] and then UbuntuUI.IconTheme would just return file:///whatever/whatever/etc/settings.svg but the path would be kept up to date. [23:43] That could be a solution too [23:45] hm, not quite [23:45] it can't return the URL; that doesn't work, I don't think [23:45] no, it doesn't, I've just tested it. [23:46] So it'd need to have a magic hook to load the file and then return it as a data url [23:46] either that or each UbuntuUI widget which has imagery knows how to parse the magic theme: URLs (that is, the browser doesn't handle those URLs; UbuntuUI does, and rewrites them to be file URLs) [23:47] man, I try to lose the game early by pressing the same sequence (up, right, down, right) over and over, and I get my highest score ever :( [23:47] which means everything which takes a URL now changes to call doMagicUbuntuParsing(url) instead [23:47] mhall119, yeah, a bunch of people on reddit said that, too ;) [23:48] and an onload parser would work, too. [23:48] better, a dom mutation events handler, but I bet we don't do that yet [23:49] mumble mumble s/events/observers/ [23:51] Why parser? You can style data- directly [23:57] daker, yeah, but you can't style it to a computed attr value because nobody supports computed attr() yet, do they? [23:58] i am not really sure [23:58] but you can do : [23:58] div:after { [23:58] content: attr(data-blabla); [23:58] } [23:59] ya, but that will only write the provided string [23:59] only for text [23:59] yes :)