=== chriadam|away is now known as chriadam [00:36] mhall119, once there's some documentation on *how*, I'm more than happy to try :) [01:20] aquarius: how to write a scope? === ahoneycutt is now known as ahoneybun === ahoneycutt is now known as ahoneybun [05:59] kalikiana: I wonder why the CI didn't autotrigger itself when it was changed [06:57] akiva-thinkpad: is there a way to make 1 button change 2 properies? [06:58] ahoneybun, yes [06:59] oh? [06:59] ahoneybun, so like, "onButtonPressed: dostuff"? [06:59] thats what you are trying to do? [06:59] akiva-thinkpad: onClicked: dostuff [06:59] should it be onButtonPressed? [07:00] ahoneybun, yep; its been awhile since Ive done qml, but I think just doing && works === ara is now known as Guest58945 [07:00] I want 1 button to change 2 different textfields defaults [07:00] so onclicked: a.color = "black" && b.color = "red" [07:00] that I have set as "property int actualValue = 0" [07:00] oh! [07:00] let me try [07:01] ahoneybun, && means and, || means or [07:01] Expected token `,' [07:01] I remembered that from C+++ [07:01] C++ [07:02] hmmmmm [07:02] onClicked: score1.actualValue = 8000 && score2.actualValue = 8000 [07:02] ahoneybun, try this [07:02] onClicked: { score1.actualValue = 8000; score2.actualValue = 9000 } [07:03] I think && is more about comparisons actually [07:03] so [07:03] onClicked: { score1.actualValue = 8000 && score2.actualValue = 8000 } [07:03] Expected token `,' [07:03] no, I removed && [07:03] yea if 1 is true then the 2 is [07:03] oh [07:04] works! [07:04] i think its actually onClicked: score1.actualValue < 8000 && score2.actualValue < 8000 ? something : something [07:04] akiva-thinkpad: I'm setting modes for common games [07:05] yea ahayzen gave me something similar [07:05] ahoneybun, yah sorry my initial thing is about making comparisons. [07:05] && is and, || is or [07:05] akiva-thinkpad: that was what I was thinking in the beginning as well lol so no big [07:06] ahoneybun, best way to learn is to teach :P (I learned why I was wrong by trying to teach) [07:06] lol [07:07] thanks akiva-thinkpad [07:07] np [07:07] almost 320 users on uBeginner app akiva-thinkpad [07:08] now your just bragging :P [07:08] * akiva-thinkpad cries as he tries to finish autopilot [07:08] akiva-thinkpad: no just happy [07:08] {: [07:08] https://appstore.bhdouglass.com/app/gazeteer.aaronhoneycutt [07:21] ahoneybun, oh this is your scorecard app; yah ! thats neat [07:21] I thought you were speaking about your very first app :P [07:25] akiva-thinkpad: the users is for uBeginner [07:25] 320 [07:25] that is my new app [07:25] :) [07:26] akiva-thinkpad: http://imgur.com/DhKC7ci [07:26] current work === chihchun_afk is now known as chihchun [07:27] Ha! ever play dominion? [07:27] no [07:27] I really don't play any of those games lol [07:27] Magic is on there though :P [07:28] my friend does and he has a app on his iPod Touch and I thought we should have one on Ubuntu as well [07:28] Same guy made dominion [07:28] Ah cool :P [07:28] he plays yugioh [07:28] does dominion have a score system? [07:28] ahoneybun, sort of [07:28] ahoneybun, we'd have to play a game for you to understand. Want to? [07:29] takes like 10 minutes to do one [07:29] online I'm guessing? [07:29] akiva-thinkpad: any exp with including C++ ? [07:29] http://www.cplusplus.com/forum/beginner/1304/ [07:29] ahoneybun, a little [07:30] I want to include a coin flip as you see [07:31] ahoneybun, and so you just need a random number generator? [07:31] yea pretty much [07:31] I'd imagine that you wouldnt need to use c++ for that. [07:31] oh? [07:33] some c++ : srand((unsigned)time(NULL)); [07:33] ahoneybun, lookup qml math.random [07:34] http://doc.qt.io/qt-5/qtquick-usecase-integratingjs.html? [07:35] ahoneybun, try it [07:35] good morning [07:35] hey dholbach [07:35] might need to import math [07:35] hi ahoneybun [07:36] hai folks [07:36] good old 3:30 am! [07:42] akiva-thinkpad: it is not making much sense for me but I think I'm going to stop on it for this morning [07:44] thanks akiva-thinkpad [07:50] later [07:50] later [08:37] zsombi: the results are positive, I tested your MP 15-20 times and no crashes :) [08:37] zsombi: oh and good morning ;-) [08:37] nik90: awesome! [08:37] nik90: yeah, I'm already here for >3h ;) [08:37] nik90: good morning to U too [08:39] :) [08:39] zsombi: yeah I have no intention of coming online here at 06:00 AM :P [08:50] nik90: :D [08:50] mhall119, in go, in Ubuntu SDK, without cmake misery [08:53] kalikiana: ogra_: haha... seems qmlscene is also setting the application name before anybody else would do that!!!! [08:53] kalikiana: ogra_: I have now the same problem with the state files [09:00] balloons, can you check my comments on https://code.launchpad.net/~dholbach/help-app/1434415/+merge/253844? [09:00] balloons, I don't know how to best solve them, but I think it'd be good if we could land something soon, so everything builds again [09:19] zsombi: is that a new issue? [09:19] kalikiana: I don't think so, I just encountered this while doing unit test for StateSaver [09:20] kalikiana: I think the qmlscene always set the app name and org name [09:20] kalikiana: before loading the QML file [09:23] kalikiana: what is weird on the state saver is that while the logs clearly state that the state file exists, the QFile::exists(statefile) returns false for the same file [09:24] kalikiana: ah, btw, an alarm MR is waiting for your review ;) https://code.launchpad.net/~zsombi/ubuntu-ui-toolkit/alarmmodel-get-crash/+merge/253981 [09:25] hello! how do I get WiFi signal strength and battery information in QML? [09:26] should I use B2Qt WifiManager? [09:27] zsombi: I'm on it [09:27] kalikiana: thx!!! [09:31] zsombi, ha ! awesome ... [09:34] ogra_: kalikiana: I think it would be the time either to distropatch the qmlscene or to really have our own launcher [09:35] i think long term the latter makes sense [09:35] short term rather the former :) [09:35] ogra_: this has been discussed many many times, and the decision was always to go with qmlscene [09:35] I've stopped counting how often this question comes up… [09:36] yeah, as kalikiana said [09:36] haha [09:36] even upstream doesn't like we use this binary to run the apps... === dbarth_ is now known as dbarth-afk [09:43] do we have a bu open about the fact that anchorToKeyboard isnt respected if you rotate to landscape ? [09:44] (or is there any programming trick to work around this) [09:44] *a bug [09:49] zsombi: did somebody disable ci again? staging has flake8 errors :-( [09:50] ./tests/autopilot/ubuntuuitoolkit/tests/components/test_urihandler.py:22:5 F401 mock imported but unused [09:50] kalikiana: Mirv was also wondering this morning about it [09:51] we really need ci for ci at one point… [09:51] or maybe cuffs for whoever does the disabling [09:51] kalikiana: btw, Popovers vs OrientationHelper... [09:52] kalikiana: have you tried to move OrientationHelper from PopupBase's base type to the foreground only? [09:58] zsombi: I can't see any crashes with staging in ubuntu-clock-app [09:59] kalikiana: just plain staging without zsombi's MP? [09:59] nik90: yes [09:59] which you suggested should be crashing left and right [09:59] kalikiana: that's weird.... especially that the only alarm fix did land beginning of March, and with that Clock was still crashing [10:00] kalikiana: pmcgowan reported crashes every time [10:00] kalikiana: yeah me too..on my vivid phone and desktop, i can make the clock app crash with my mind [10:00] the force must be too weak in me… [10:01] kalikiana: I just got one [10:01] :P [10:01] kalikiana: I didn't know what's up but I ran CI manually for your u1db-qt branch [10:01] kalikiana: so, you need to have alarms, the open the settings, move here'n'there, you will get a crash [10:01] kalikiana: easy way to get it to crash would be to create a new alarm and then do what zsombi says ^^ [10:02] kalikiana: believe me, it crashes [10:02] kalikiana: the weird thing is that the last line in get() still has a valid Alarm pointer, but once reaches QML, the ret value is null... [10:03] kalikiana: the best case, sometimes crashes straight away, without returning teh value [10:03] kalikiana: nik90: so it crashes somewhere in the middle of the meta object calls [10:04] bzoltan_: I think you should update the UITK gallery in the store :) [10:04] zsombi: good idea [10:05] * kalikiana I wonder if these are related, I get hundreds of them: [10:05] qml: [CLOCK] Error saving alarm, code: 2 [10:05] file:///usr/share/ubuntu-clock-app/common/AnalogClockMarker.qml:40: TypeError: Cannot read property of null [10:06] bzoltan_: for some reason the new list items page is empty... and I guess it's because the API [10:06] kalikiana: that might also be the reason get() returned null [10:07] kalikiana: the error code 2 is EarlyDate, meaning the date for your alarm is not right [10:08] kalikiana: zsombi: hmm you shouldn't see error code 2 either since clock app automatically should adjust it to the next day. So for instance if the current time is 25th March 11:00 AM and if you set an alarm for 10:00 AM, clock app should automatically set it to 26th March 10:00 AM [10:09] well atleast I have a function that checks the validity of the date and does that for the user [10:10] nik90: well, the error is shown at the time teh save is called, so the date adjustment is wrong then. and this is only for one-time alarms [10:10] nik90: never the less all alarms are saved, so the error might be some false alarm somewhere [10:10] or there might be some collision of which alarm object the save is called on [10:11] hmm I guess [10:13] nik90: zsombi with the branch the errors are all gone [10:16] kalikiana: what about crashes? [10:24] nik90: none. I'm guessing the values get garbage collected, and since it's copied in the branch it works fine [10:27] bzoltan_, zbenjamin Holy macaroni, am I dumb! I got the damn thing embedded; my mistake was to manually call the createConfigurationWidget() [10:27] holy hell [10:27] okay this thing SHOULD be done pretty quickly then [10:45] bzoltan_: next candidate for landing https://code.launchpad.net/~zsombi/ubuntu-ui-toolkit/alarmmodel-get-crash/+merge/253981 [11:08] ooh [11:08] zsombi: review, please https://code.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/sphinxSysPath/+merge/252891 [11:08] zsombi: you got ot nailed? Cool [11:09] akiva-thinkpad: nice job, dude :) [11:09] zsombi: note that it's not easy to repro locally because you'd have to uninstall both autopilot and uitoolkit, I used the CI logs to verify it [11:09] bzoltan_, ;_; tears literally in my eyes, I was so happy. Do you want a demo, and you can let me know what else you want to see before sending it over to you for review? [11:10] via a screencast? [11:10] akiva-thinkpad: I love demos :) [11:10] okay stand by [11:10] zsombi: also: feast your eyes on the PASSING CI, I don't remember when I last saw that [11:11] kalikiana: long ago... but all those seem to be mako related [11:14] zsombi: I got used to seeing "unstable" always… [11:15] zsombi: Do you want me to push it to the ongoing landing once it got merged to the staging? [11:18] bzoltan_: let's wait for CI to finish with it [11:18] zsombi: J has approved it [11:20] bzoltan_: I see it's merged, so do the monkey :) [11:20] zsombi: I do that :) [11:21] zsombi: I top approved it ... [11:21] bzoltan_: well, kalikiana did that 51 mins ago :D [11:21] SDK folks, question - how can I turn off this behavior: have window split in two, I scroll one pane to some part of code with the scrollbar, then I focus the other pane -- first pane scrolls back to where the cursor was initially. [11:21] zsombi: https://code.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/sphinxSysPath/+merge/252891 [11:22] I know this is a QtC question, but maybe you'll be able to help. [11:22] zsombi: I reloaded it and it was still "need review" ... but whatever [11:23] karni: It is indeed QtC question... I suggest you the #qt-creator channel [11:23] bzoltan_: ah, that one... I had not checked that, I thought you talked about the alarm stuff [11:23] bzoltan_: thanks! [11:23] bzoltan_: but yeah, that's ok as well [11:29] bzoltan_, https://www.youtube.com/watch?v=Ox7pj5MjGec&feature=youtu.be [11:29] i don't think its private :P [11:32] akiva-thinkpad: awesome! MR it right away :D [11:32] :) [11:32] bzoltan_, anything you want added before that? === _salem is now known as salem_ [11:36] akiva-thinkpad: By looking at the video it is good. Let me see the code :) [11:36] okay, in a bit === chriadam is now known as chriadam|away [12:37] Hi, could someone tell me how I would me able to present a file dialog to a user to be able to open a file for an app (I already know about the ContentHub, but can't see a type more specific to text files) ? [12:39] The normal QtQuick.Dialogs' FileDialog also works on desktop, but not the emulator - am a bit stuck [12:50] dholbach, I see https://code.launchpad.net/~dholbach/help-app/1434415/+merge/253844 is happy :-) [12:51] balloons, yes :) [12:51] balloons, the question is: what to do about the Exception you mentioned [12:51] balloons, the tempfile question I think is solved [12:51] (did you see my comments?) [12:51] dholbach, I don't see any comments [12:52] weird [12:52] it says "unsaved comment" [12:52] ok... [12:52] I did see the pep8 proposal [12:52] I'm not quite sure how to best do this... if we raise an exception, we will have to instantiate our Markdown() object everywhere in try/except stanzas. [12:52] ^ this is what I replied to your exception question [12:52] regarding tempfile: http://paste.ubuntu.com/10677461/ [12:53] balloons, if you think it makes sense to use exceptions, I'll look into it [12:53] dholbach, ahh yes. So I raised the question because it appeared like you were reusing the object in tests. So it would make sense to raise an exception and send it back. The print isn't the biggest deal; it's the sys.exit() to go with it [12:53] right [12:53] I know this isn't the only case with them, so I wouldn't hold this mp for it. Something to think about though [12:54] ok... shall I file a separate bug for it? [12:54] I guess that might be best so we don't forget, heh [12:54] done... bug 1436316 [12:54] bug 1436316 in Ubuntu Help App "Don't sys.exit() in Markdown() instantiation" [Undecided,New] https://launchpad.net/bugs/1436316 [12:54] thanks balloons for the reviews! [12:55] cool, so I'll approve [12:55] I approved both. Jenkins should merge it all :-) [12:55] oh... [12:56] is it going to autoland the branches? [12:56] dholbach, yes.. Do you not want that? [12:56] no no that's fine [12:56] easier that way methinks [12:56] absolutely :) [12:56] nice! [12:56] popey, in light of bug 1436110 as stated in the comments why can't the file manager call url-handler for anything in /sd_card/Music/ or ~/Music and content-hub for anything outside ? ... i see bug 1413821 has been marked as won't fix :/ [12:56] bug 1436110 in Ubuntu Music App "Music refuses to play file opened via file manager" [Undecided,New] https://launchpad.net/bugs/1436110 [12:56] bug 1413821 in Ubuntu File Manager App "Music files are imported into the Music app, rather than simply opened" [Undecided,Won't fix] https://launchpad.net/bugs/1413821 === chihchun is now known as chihchun_afk [12:58] ahayzen: hmm [12:58] ahayzen, https://bugs.launchpad.net/developer-ubuntu-com/+bug/1436317. I didn't forget about you [12:58] Ubuntu bug 1436317 in Ubuntu App Developer site "Add pyflakes / pep8 documentation" [Undecided,New] [12:58] balloons, sweet thanks that'll be useful for pointing new devs to :) [12:59] popey, the problem with content-hub is not that it is slow ... it is the fact that we import that music again ... so say you open a file that is already in ~/Music via content hub...you'll then have two inside ~/Music and a duplicate in your library :/ [13:02] ugh === dbarth-afk is now known as dbarth [13:06] balloons, do you know how long it roughly takes until the landing happens? I know I'm getting overly excited :) [13:08] popey, surely file manager can just have a whitelist of paths that if a audio/video file is tried to be opened from, it then uses url-handler rather than content-hub? [13:15] dholbach, to be fair we haven;t tested that jib yet, hehe [13:15] ahayzen: for the small set of paths, potentially, yes. [13:15] dholbach, http://91.189.93.70:8080/job/help-app-autolanding/ [13:15] balloons, https://code.launchpad.net/~dholbach/help-app/run-pep8-and-pyflakes3/+merge/254059 failed, but that was expected [13:15] popey, /media/*/*/Music and /home/*/Music ? [13:15] bah, it ran them in the wrong order [13:15] i tried to be sneaky [13:15] :) [13:16] it appears to be doing something :) [13:19] ahayzen: yeah, worth a bug in file manager :) [13:19] popey, or do we reopen that one that was closed lol ? [13:19] oh, we have one already, don't we. Yeah, a comment on that. [13:19] popey, bug 1413821 :) [13:19] bug 1413821 in Ubuntu File Manager App "Music files are imported into the Music app, rather than simply opened" [Undecided,Won't fix] https://launchpad.net/bugs/1413821 [13:21] popey, commenting :) [13:21] thanks [13:21] balloons, landed! :) [13:21] can't believe it [13:24] popey, ah bugger i can't reopen the bug lol [13:24] popey, do you have permission to flip it from won't fix -> new? [13:24] ya [13:25] done [13:25] thanks popey :) [13:25] dholbach, excellent :-) [14:07] I upgraded to 15.04 yesterday and could build Ubuntu Touch app with qmake [14:07] with 15.04 click target [14:08] however, when deploying, it says it can't find the runtime for 15.04 on the device (or similar, don't remember the exact error message) [14:08] ubuntu-core-15.04 armhf or the like [14:08] any idea how to enable this? [14:10] xhoch3: obvious question, are you trying this on the devel images on the phone? [14:10] nik90, no, stock image [14:11] when will those be able to to run on ordinary devices? [14:11] xhoch3: ha that's the issue then..the current stock image (RTM) is based on utopic [14:11] xhoch3: the stable phone images will switch over to the 15.04 framework only when it is deemed fit and stable and without regressions [14:11] ok, since I bought the Aquaris E4.5, I'd like to work with the vanilla environment [14:12] xhoch3: as a developer if you want to target users with the Aquaris E4.5 which runs the stable images, you will have to use the 14.10 framework [14:12] xhoch3: and then when the E4.5 switches to 15.04, you can update your app and make use of the new features that 15.04 brings [14:12] ok, since I ultimately need qmake, I also need 15.04, so my question is when the 15.04 image is expected to appear? just an estimate [14:13] unfortunately 14.10 is feature-incomplete for me, so I cannot use this [14:13] xhoch3: too hard to say..but the current focus is to get 15.04 stable and ready. so some say the ETA is between 5-10 weeks [14:14] xhoch3: there are lot of bugs that need to be fixed in vivid before it will make it to the stable images [14:14] nik90: wow https://i.imgur.com/7jdN4y4.png [14:14] daker: you got lxc setup? [14:15] yes [14:15] ok, and in order to work with 15.04, I would have to erase my device? Or can I switch to 15.04 on my phone like I did with my desktop? [14:15] by accepting the latest dev version [14:15] daker: excellent! enjoy! [14:15] xhoch3: no you wouldn't need to erase your phone [14:16] xhoch3: you can switch to vivid easily. However in big caps "THINGS MIGHT BREAK" .. so obviously it cant be used as a daily phone any longer [14:16] s/cant/shouldnt [14:17] ok, no risk no fun [14:17] :D [14:17] how do I do it? [14:17] xhoch3: if you run "ubuntu-device-flash touch --channel=ubuntu-touch/devel" in the terminal after connecting your phone to your laptop, you should be upgraded to 15.04 [14:17] xhoch3: I presume you know how to restore it if something goes wrong? [14:18] too late [14:19] :D [14:19] * nik90 brb in 15 mins [14:19] xhoch3: same command but with the channel name ubuntu-touch/stable or bqaquarius_en I think [14:19] ppl in the #ubuntu-touch channel will know for sure [14:24] 2015/03/25 15:21:59 Rebooting into recovery to flash [14:24] Failed to enter Recovery [14:25] now I have the orange ball spinning [14:25] can I safely shutdown the phone? [14:26] oh it rebooted [14:28] nik90, ok, cool I have 15.04 now [14:28] thanks [14:32] 15.04 is snappier, isn't it? feels smoother [14:50] charles: ping [14:51] nik90: do you remember, have you agreed with charles about any EDS tags to be set for application ID and alarm ID? [14:53] zsombi: https://code.launchpad.net/~charlesk/indicator-datetime/lp-1426519-add-tag-for-menuitem-activation-url/+merge/253997 [14:54] zsombi: I guess if alarms specify com.ubuntu.clock along with the alarm ID in the X-CANONICAL-ACTIVIATION-URL EDS tag that should be it? [14:54] charles: can you confirm ^^ [14:56] nik90: charles: uhm... how should I encode it? $(appid):///$(alarmId)? [14:56] hmm that was the discussion we had with ted, but I cannot remember the outcome we decided upon [15:19] nik90: checking charles' comments on teh bug, I think we should decide what to place in the URL [15:19] zsombi, oops, didn't see your pings from earlier, looking now [15:19] nik90: so url-dispatcher would expect sthing like $(appid):///URL [15:20] zsombi, nik90, we should also get renatu in on this converation, he had some opinions on this yesterday [15:20] nik90: charles: where URL should be somethiong containing the alarm ID [15:21] nik90: charles: reading this https://developer.ubuntu.com/en/apps/platform/guides/url-dispatcher-guide/ the format is well defined: $(appId):///$(URI) [15:21] where $(URI) is something the app should understand [15:21] iirc renaut suggested not use appid as the url scheme, rather something like 'alarm' [15:22] charles: how do we then know which app to open? [15:22] charles: well, the thing is url-dispatcher needs appid as protocol [15:22] nik90: exactly [15:22] ah. :) [15:22] nik90, yes, confirming "X-CANONICAL-ACTIVATION-URL" is the x-prop to use [15:22] zsombi, no you can register your app to handle any url prefix [15:23] renatu: true, but if there are multiple apps registering for "alarm" url prefix then we get into trouble [15:23] renatu: right, but then any app can override the url dispatcher... [15:23] for example the dialer app register itself for dilaer-app:/// and cal:// [15:23] exactly [15:23] zsombi, renatu has a patch at https://code.launchpad.net/~renatofilho/qtorganizer5-eds/fix-1426519/+merge/253985 to generate x-prop entries in ical from QOrganizerItemExtendedDetail in QOrganizer [15:24] charles: yes, I am making the changes on Alarm based on that [15:24] renatu: the idea here is to allow alarms created by app-1 to be opened by *only* app-1 [15:24] zsombi, my opinion is that if the app create the event/alarm and it is a app especific thing then the app should set the event url [15:24] zsombi, cool :) [15:24] renatu: if we use the standard "alarm" then app-1 could get app-2's alarms [15:25] but if it is a generic stuff like a google calendar event then the user should be able to choose which app to use to open it [15:25] I thought we wanted to be able to let 3rd parties make their own clock, calendar, etc apps to supplant the default ones? [15:25] nik90: I see some rational in renatu's idea, because then it would be easier to "replace" Clock app [15:25] +1 [15:26] zsombi: I agree but I remember having this discussion some time back where we wanted to make each app's alarms to be accessible by the concerned app only? [15:27] charles: renatu: nik90: the question is shouldn't we support the ability for multiple apps to create their own alarms? which should be in a separate DB than the clock app? [15:27] nik90: right, as I wrote :) [15:28] nik90, if you want the alarms created by the clock app to accessible only by the clock app you can set the url by yourself during the alarm creating [15:28] creation [15:28] nik90: renatu: charles: so, if separate apps can register their own alarms, then appid is the one we need [15:28] zsombi, I think we'd want them to all go through the same alarm API; otherwise, they wouldn't get notifications [15:28] renatu: I do not want to provide complete URL specifying freedome to the alarm user... [15:28] zsombi, I don't think datetime is something we'd want to supplant, because even after we have custom indicators, datetime still is responsible for hw wakeups on alarm [15:29] renatu: you mix Alarm API with QtOrganizer one [15:30] charles: yes. What I planned is to provide Alarm API to access the alarmId, and to have a lookup for it, and use APP_ID to encode the URL [15:30] charles: so in that sense each app can have its own URL [15:32] whats happen if the app get removed? [15:33] charles: nik90: renatu: it would be too ugly if the develoiper would need to set the protocol for each alarm saved... [15:33] who will remove the alarms created by this app [15:33] renatu: same question with a generic "alarm" protocol [15:33] renatu: what if I remove the Clock app, and I don't install anything else? [15:34] clock app can not be removed [15:34] renatu: so then? [15:34] renatu: if you cannot remove, how do you override? [15:34] zsombi: well if we go with a generic "alarm" protocol, deleting a 3rd party app wouldn't affect the user since he would then use the clock app to handle those alarms created by the other app [15:35] I am saying if I intall a new clock app would be nice if this new app can handle the allarms created by clock app [15:35] nik90: but wo decided which app to open in case you install a 3rd party Clock app? [15:35] decides [15:36] we do not have support for that yet :D. But the url-handler will provider a dialog where you can choose which app to use and set the default one [15:36] hmm true..we don't have a mechanism like the content-hub which shows which apps can handle a specific url [15:36] exactly... [15:36] if more than one app register the same url prefix [15:36] I do not know whats happen today [15:36] renatu: will I get that dialog each time it is invoked? that'd suck, sorry to say [15:37] zsombi, you can set the default one [15:37] like android does [15:37] and it will never appear again [15:37] until you reset it [15:37] zsombi: considering that the use case where multiple apps register for the same url prefix is something that's going to become common..(multiple reddit apps etc) .. the dialog allowing the user to choose the app to open and set a default is inevitable..UT definitely needs it [15:38] how do you revert in bazaar, other than deleting and rebranching it? [15:39] today I do not know whats happen if more then one app register the same url prefix, and I think there is nothing blockin the user to install a app with the same app id from a already installed app [15:39] nik90: I understand that, and it sounds right, the problem is that it feel the app developer to provide the protocol and the format every time it saves the alarm si too much [15:40] zsombi, we can define that alarm:// is the default value the devel can chage it to the app id if he wants [15:40] renatu: well, I'd need to expose some AlarmConfig API then... [15:40] renatu: or I could give it as a property to each alarm... [15:41] zsombi, for alarms I do not see much benefit on it, but for calendar events I think this is very useful [15:41] renatu: well, the idea was to let several apps to be able to create alarms individually, and I do no tmean overriding teh Clock app [15:42] zsombi, will the alarms created on others app visible on clock app? [15:43] if not they are individual alarms and should use the app-id on it [15:43] renatu: no, clock app should only see Clock alarms, they should never collide [15:43] zsombi: renatu : All this boils down to one question, "Will apps creating alarms using the same db as clock app or not? " [15:43] s/using/use [15:44] nik90: well, this may hopefully get solved when charles will start the new Alarm backend :) [15:44] yeah... :) [15:44] I'd like to that sooner rather than later, but it's not happening this sprint [15:45]