=== ajmitch_ is now known as ajmitch [00:03] renato_: your mp does fix calendar :-) [00:05] balloons, good [00:06] balloons, but the clock still broken? [00:07] renato_: yes I updated the MP with my notes. Essentially clock is unchanged [00:14] renato_: any thoughts on what's happening with clock/ [00:16] balloons, no idea [00:16] I will discuss with szombi tomorrow to see if he can filter the alarms [00:16] on alarm api [00:19] renato_: ok, so we'll hold another day then to see if we can fix clock also? [00:19] thanks for digging into this === tsimpson_ is now known as tsimpson [06:31] good morning === zequence_ is now known as zequence [08:37] zsombi, so to give you some background: we're making the Weather app convergent, and it works well on N4 and N10: we've got a default phone layout and a tablet layout [08:38] However, on the N7 we've got an issue whereby we've got a footer that should expand the whole width of the screen in landscape, but for some reason is stuck at a fixed size in between [08:38] let me show you a pic... [08:40] zsombi, here it is: http://imgur.com/4a8bXJk,h3mrXJM#1 the "The Weather Channel" footer has width = mainView.Width, and strangely enough, it doesn't have that width on the N7 [08:40] dpm: so it should go all along to the edge of the screen, right? [08:41] zsombi, exactly. Strangely enough, that works on N4 and N10, but on N7 it fails as in the picture. Here's the bit of code that does that: http://bazaar.launchpad.net/~dpm/ubuntu-weather-app/n7-tweaks/view/head:/components/LocationTab.qml#L168 [08:42] we have been trying to debug this for a couple of days to get a convergent layout in for MWC, but we were a bit tapping in the dark, [08:42] and I thought you might spot something obvious or perhaps help us think of a workaround === chriadam_ is now known as chriadam|away [09:11] morning popey, have you seen balloons e-mail? Can we get weather and all the other apps uploaded to the store? [09:12] yeah. [09:12] dpm:guys, ConditionalLayout can take only ONE component. Check the documentation: the ConditionalLayout.layout default [property is a Component. While this does not give error on parsing, it may not work properly [09:13] dmp: quote from ConditionalLayout doc: "There must be only a single component defined in the ConditionalLayout which will be instantiated upon activation." [09:13] dpm: in the code I see loads of components declared under the ConditionalLayout, so no wonder the items are not laid out properly [09:14] dpm: has balloons run the AP tests on these do you know? [09:15] dpm: so the code should embedd all those Rectangles and ItemLayout components in ONE single element [09:20] popey, all I know is the e-mail we got from him [09:21] zsombi, ok, thanks a lot for looking. We've got no weather developers around, so I'll try to give it a go [09:22] dpm: ok, good luck :) [09:22] and patience :) not having a device to test makes it even more "fun" [09:23] dpm: you can alter the MainWidth to have 90 GU for instance, then the ConditionalLayout will activate straight away [09:23] popey, the r197 package from Jenkins did pass the tests. My guess is that he built a new .click manually based on that [09:24] zsombi, yeah, to activate the layouts I can do manual resizing on the desktop, what's been challenging has been to trigger the not-full-width footer, as on the desktop it works well [09:25] dpm: that's weird! [09:26] zsombi, wait, all the elements are already inside one single Rectangle, it's just that the indentation is off === mpt_ is now known as mpt [09:26] dpm: ah, ok... the indenting made me think it's not :) [09:26] yeah, just noticed that too, let me quickly fix that [09:31] dpm: in QtC Ctrl+A, Ctrl+I, then Strl+S :) [09:33] argh, too late, already fixed in manually in qt creator :) [09:33] but nice to know [09:33] zsombi, ok, the indentation should now be correct, so I guess it's back to the drawing board: http://bazaar.launchpad.net/~dpm/ubuntu-weather-app/n7-tweaks/view/head:/components/LocationTab.qml#L168 [09:35] dpm: question: doesn't the Page fill the entire MainView client area? [09:36] dpm: I don't see reason to use the mainView property to stick to the app's width [09:36] dpm: if some component in the hierarchy overrides that property, then you'd get in trouble [09:37] dpm: which may also be the case [09:37] Good morning all; happy Friday, and happy International Mother Language Day! :-D [09:37] zsombi, so you're suggesting to use parent.width instead? (I'm not sure if the Page fills the whole area) [09:38] dpm: if the Page does fill it, then yes, use that. [09:38] dpm: but I'd use anchors instead of setting the width [09:38] dpm: I see the developer doesn';t really use those... which is bad [09:39] parent.width works on desktop [09:39] let me try with anchors. Right now we anchor the footer to the bottom, but nowhere else [09:40] zsombi, that works as well (on desktop) [09:41] zsombi, so you're suggesting to use anchoring (left, right) to the parent instead of specifying the width? [09:41] for the footer [09:41] dpm: right [09:41] ok, cool, let's give it a spin. popey, would you mind trying a new .click on the device? [09:42] dpm: and in general in the code, whenever you have parent.width, use anchors [09:42] ok [09:42] dpm: same for the heights. [09:42] dpm: sure [09:42] dpm: this is a really bad practice... [09:43] dpm: also, Label's anchor.verticalCenter will not give the proper result [09:43] dpm: it should be the Label's verticalAlignment property be used [09:44] ok, fixing that too [09:45] zsombi, and the value assigned should also be parent.verticalAlignment? [09:46] dpm: nope, you shoudl anchor it's top and bottom to the corresponding parent anchorLines, then use Label.verticalAlignment: Text.AlignVCenter [09:47] dpm: so, anchors.top: parent.top; anchors.bottom: parent.bottom; verticalAlignment: Text.AlignVCenter [09:47] gotcha [09:48] * zsombi goes out 4 lunch === zsombi is now known as zsombi|lunch [09:53] popey, sorry, my p.c.c account seems to be failing, so no wget. Can you get the .click from http://ubuntuone.com/59GSo0qyNze0m9TGwV70Kj ? [09:53] sure [09:56] hi nik90, reading this: https://bugs.launchpad.net/ubuntu/+source/qtorganizer5-eds/+bug/1282129 - do you think we need a new clock release to limit the date events are returned for? [09:56] Launchpad bug 1282129 in qtorganizer5-eds (Ubuntu) "Clock and calendar tests fail with static void QOrganizerEDSEngine::itemsAsyncListed(ECalComponent*, time_t, time_t, FetchRequestData*) " [High,Triaged] [10:31] dpm popey hey [10:31] hey m-b-o [10:32] popey dpm reading back the logs .... [10:32] m-b-o, good news, would you mind reviewing https://code.launchpad.net/~dpm/ubuntu-weather-app/n7-tweaks/+merge/207616 :-) [10:33] dpm yes :) [10:34] m-b-o, ah, just a minute, I'll remove those print statements. The diff looks big, but it's simply because it fixes the indentation [10:34] dpm: ubuntu-weather-app.desktop is added. ok? [10:35] m-b-o, yeah, I'll remove it to make sure Jenkins doesn't trip on it, but I'm planning to add it post-MWC to make it possible again to do Ctrl+F12 [10:41] dpm: ready? [10:41] m-b-o, ok, it's now ready for review. Please have a look at the description of the MP, which includes a link to a simplified diff to make it clearer to see the real changes between all those indentation fixes. If possible, please test on a N4 as well, to make sure it still works there, and we'll get cwayne to test on a N10 in the afternoon [10:42] okay [10:42] I need to run now, will be back in ~1h [10:42] * dpm is really excited about getting convergence in! === lool- is now known as lool [11:21] dpm: hey good morning [11:22] dpm: Concerning your question about a new clock release, the answer is it depends. If the Alarms API provides filtering support by data, perhaps I need adjust that in the clock as well. If not, then there is nothing much I can from the clock app in which case we do not need a new release. [11:23] s/data/date [11:34] balloons, I have a fix for you to test, are you around? [11:38] nik90, hi [11:38] nik90, do you have the device to test a patch for me? === zsombi|lunch is now known as zsombi [12:25] dpm: Can you check if the MP https://code.launchpad.net/~charlesk/indicator-datetime/alarms/+merge/204420 has landed on the phone or not. It concerns the alarms notification fix for the indicator-datetime package. [12:26] hi renato_, balloons won't be around until in ~3h time [12:26] nik90, just a sec, just came back from an errand and I need to check how we're looking for Weather [12:28] dpm: np [12:41] fginther, ok, we got weather working with tablet layout across all devices. When you're around, could you please trigger a click build for weather and we'll upload that to the store? [12:42] nik90: hey dude [12:42] fginther, could you tell us for what time of the day the click builds are scheduled? This might save you some pinging (i.e. if I know when they're getting build and the changes are in that time window, I'd just wait for the build to happen) [12:43] zsombi: hi [12:43] nik90: we're trying to sort out the alarm API issues, and seems you are not cleaning the alarms made during the clock app test, right? [12:44] zsombi: yes, I am not sure how exactly to clear the alarms database after a test [12:44] I was planning on working on it with elopio and balloons later [12:44] nik90: you remove all alarms that have special message [12:45] zsombi: special message? [12:45] nik90: when you create an alarm, you give a message text, right? [12:46] zsombi: yeah I give it the alarm a label [12:46] zsombi: as a temp solution, it wouldn't be hard adding a "swipe delete alarm" action after a test is done to clean up after itself [12:46] nik90: awesome, so you give the "test_XXXX" label to it, right? [12:46] zsombi: yes [12:46] nik90: IT MUST BE :) [12:47] nik90: otherwise all test alarms will land in the Alarm DB... [12:47] zsombi: the proper way of doing this as suggested by renato and elopio was not the swipe delete alarm action [12:47] nik90: which is not a good idea to pollute the Alarm DB with test stuff :) [12:47] zsombi: since if the test fails midway, the test alarm still remains in the db [12:48] zsombi: also the alarms AP tests were only enabled 2 days ago :) .. so I can patch it up to delete the alarm after the test [12:48] nik90: ehh.... so there's no emergency cleanup? :( baaaaad test suite [12:49] zsombi: well that's the thing, other stuff in the clock app uses u1db which creates a local db in the application db. This allows for a proper clean up [12:49] zsombi: in the case of alarms, the EDS doesn't make it easier to do this [12:50] zsombi: I just need more time to do the cleanup. It was ignored just postponed for after MWC [12:50] nik90: once the Alarm API will have the ability to create separate collection for each app, that would get solved, as test alarms will land in a separate collection, thus separate DB [12:50] nik90: ok, np, just wanted you to know that the cleanup is an absolute must :) [12:51] zsombi: lol I mean it *wasn't* ignore just postponed :) [12:51] nik90: ok :) [12:51] zsombi: completely agree on your cleanup [12:51] zsombi: I will have a bug report to track this [12:51] nik90: kewl [12:51] thx!!! [12:52] kewl ? [12:53] cool [12:53] lol [12:53] nik90: as daker says :D [12:57] zsombi: just so you know, I was able to fix the convergence "animation" issue by using the default layout option ... thnx for that :) [12:58] nik90: ;) I knew you can make it [12:59] nik90: it is always like that: choose a default layout, then implement the other ones :) [12:59] nik90: for code clarity, you can also have the ConditionalLayouts in different files [13:00] zsombi: yeah I overlooked the documentation there [13:00] zsombi: wow really? I will look into that in the coming weeks..I like separating stuff into smaller files for clarity [13:01] nik90: the tutorial also uses some layouts like that... for clarity, the default layout for instance is in a separate file [13:01] nik90: well, these are all QML elements after all :) [13:01] nik90: so you can declare them in separate files as you wish :) [13:02] zsombi: I guess I am stating an obvious fact here, but nice conditional layout tutorial [13:02] nik90: thx, glad it's understandable :) === gatox is now known as gatox_brb [13:33] charles, me and zsombi are discussing about alarms on EDS, we have a problem to query for alarms because right now the alarm API query for events without a limit of dates [13:33] this is causing a problem because every alarm is a recurrence event and this cause the function to take longer to process and return "unlimited" events [13:33] charles_: ^^ [13:36] zsombi, can we go with my weekly patch until we decide about that? [13:36] nik90, balloons could you guys test it: https://code.launchpad.net/~renatofilho/ubuntu-ui-toolkit/limit-alarms-fetch/+merge/207629 [13:37] renato_: ok, just make a big FIXME note that this is a temporary patch till we get a better setup [13:37] popey, dpm , do you know any EDS expert that can help us to decide about that? [13:38] renato_: ted? [13:38] (I don't know, no) [13:39] renato_: based on the structure, I do not see the way to set occurrences to the Alarm... [13:40] zsombi, ECalComponentAlarms ha a list of ECalComponentAlarmInstance [13:41] renato_: yep... hmm... all these would also be stored in a DB, so not sure whether we solve anything with it :( let's see when charles_ gets online... [13:41] renato_, I'd hint at ted, kenvandine or seb128. Or perhaps desrt [13:42] renato_ or if not today, then let's try to catch up with him next week, maybe popey could get us on a meeting, like the one we had this week [13:42] renato_, someone on #ubuntu-desktop might be able to help with EDS [13:42] renato_, dpm: kenvandine is probably your best guess [13:44] zsombi, popey yes a meeting would be nice [13:46] ok, I'll schedule it [13:48] done === Pici` is now known as Pici === gatox_brb is now known as gatox [14:24] dpm, I'm building weather now. I'm also trying a change to the job where it should check trunk hourly and build a new package if trunk has changed [14:25] fginther, that sounds great, thanks! [14:30] popey, ping [14:31] ahayzen: hey [14:31] fginther: i bounced 197 weather back too because the key.js was missing the api key [14:31] popey, has the move to cmake broken the script u gave me to test the music-app on device? http://paste.ubuntu.com/6721815/ [14:32] yes [14:32] ⍨ [14:33] ahayzen: http://pastebin.ubuntu.com/6949462/ can be used to build a click locally [14:33] popey, ok, how do i then deploy it? [14:34] popey, dang... [14:34] ahayzen: http://paste.ubuntu.com/6971167/ [14:34] popey, thanks i'll try tht :) [14:35] fginther: so please can 197 be resubmitted first, with key, so I can test and accept that as fallback, then the later version as per dpm request? [14:35] np [14:36] popey, sure, let me figure out what's up first [14:36] do you need the rev197 branch I used popey? [14:36] by default the rev197 branch isn't exactly what you are looking for [14:37] popey, what I would do is rebuild with revno 197 of trunk [14:43] balloons: i don't know what fginther needs, but it needs to be the 197.2 + key.js, thats all [14:45] balloons, I need some more context if there was a special branch used [14:50] popey, tht doesn't seem to work for me it complains tht grilo isn't installed, is this because we aren't bundling it within the click package? [14:50] ahayzen: that would be https://bugs.launchpad.net/music-app/+bug/1282723 [14:50] Launchpad bug 1282723 in Ubuntu Music App "Need cmake rule to build the grilo plugin" [High,Fix released] [14:51] popey, yep which is 'fix released' :/ [14:51] sergiusens: is there an easy fix for the music devs to be able to do that? [14:51] yeah, but fixed on backend, no good for you [14:53] popey: fginther I included the cmake file and icon from newer versions so everything built properly [14:53] balloons, ah, so this was a one-off package that you built? [14:53] popey: fginther I pushed it here, the diff is small. lp:~nskaggs/ubuntu-weather-app/custom-rev-197 . That said, the new version is pushed for your review, including key this time [14:54] fginther: yes, the straight build of 197 popey rejected for various reason, like the icon :-) [14:56] popey, I can craft something, but is it for now now? [14:57] ahayzen, what's the plan for grilo btw; is it going away? [14:57] jim has made excellent progress on the media hub and it may land in the archives next week [14:57] sergiusens, yep we should be moving to the new mediascanner API and media-hub soon [14:57] (not image though) [14:58] 19:02:29 < jhodapp> hey guys, so media-hub is fully working for music-app playback now [14:58] 19:02:54 < jhodapp> I'm working to get it pushed up to launchpad and then put into the build, although it won't get used by default until I land the video portion too [14:58] 19:03:14 < jhodapp> but I'll let you know how you can utilize it from music-app if interested [14:58] ^^^ what he said yesterday [14:58] popey, he put the landing task upon me though ;-) [14:58] popey, thts awesome, i saw the G+ post :) [14:58] hah sergiusens [14:59] sergiusens: the problem is we have bugs and the developers ( ahayzen ) can't debug on device. [15:00] ahayzen, how are you building now? [15:01] sergiusens, i just tried the script popey gave me to build locally http://pastebin.ubuntu.com/6949462/ [15:01] sergiusens, and then something similar to this to deploy http://paste.ubuntu.com/6971167/ [15:02] popey, have you tried click-buddy --dir [trunk] --provision ? [15:02] it does all that [15:05] woah, nice [15:05] yeah, do that ahayzen ☻ [15:05] ahayzen: http://www.theorangenotebook.com/2014/02/click-buddy-and-you-how-to-test-your.html [15:05] \o. [15:05] popey, sergiusens, IIRC i tried tht yesterday to test dpm's branch but i'll try again :) [15:05] still doesn't solve the plugin thing... [15:05] balloons, hei, could you test the fix on SDK for the clock app? [15:06] I can give you an instant gratification hack if you want [15:06] ahayzen, ^^ [15:06] balloons, https://cohttps://code.launchpad.net/~renatofilho/ubuntu-ui-toolkit/limit-alarms-fetch/+merge/207629de.launchpad.net/~renatofilho/ubuntu-ui-toolkit/limit-alarms-fetch/+merge/207629 [15:06] renato_: sure thing [15:06] sergiusens, wht would this hack be? [15:06] balloons, the correct link: https://code.launchpad.net/~renatofilho/ubuntu-ui-toolkit/limit-alarms-fetch/+merge/207629 [15:07] ahayzen, after installing the click on the device do [15:11] sergiusens, would it make things easier if we merged the plugin's branch into the app's branch so that it gets built from a single project in LP? === charles_ is now known as charles [15:13] ahayzen, http://paste.ubuntu.com/6971350/ [15:13] dpm, yes, it would; like reminders? [15:13] sergiusens, yes, exactly [15:13] dpm, that it would; but you said music is soon going to be a pure breed app :-) [15:14] dpm, wouldn't mind it for terminal and filemanager though [15:14] sergiusens, yes, good point. Let's leave it for terminal and filemanager, which will definitely need it [15:14] renato, i-datetime has the same problem as qtorganizer5-eds there; both are using an unbounded query in the ECalClientView [15:15] ahayzen, did that solve the issue? [15:15] it's a little easier to tweak this to meet our needs in the C alarm API and in datetime, since we know we only care about (for example) [now ... now + 1 month] [15:16] sergiusens, it doesn't appear to have fixed it :/ [15:16] renato_: looking good so far [15:16] sergiusens, this is how i'm installing http://pastebin.ubuntu.com/6971374/ [15:16] renato_: but is there an API in qtorganizer s.t. ubuntu-ui-toolkit can do the same kind of thing, where its alarm model can specify a range [15:17] ahayzen, lines 9 through 13 aren't needed [15:18] ahayzen, well, 9 through 19 aren't [15:18] sergiusens, does the --provision install on device for u? [15:18] ahayzen, after installing the click, did you go through my pastebin? [15:18] ahayzen, yes, the click and the tests [15:18] sergiusens, i did... i'll try again [15:19] renato_: so I ran ubuntu_clock_app.tests.test_alarm and no more eds errors. I would say the fix worked. I can check the other apps to make sure it didn't break them. I'll run calendar at least [15:20] balloons, ok nice [15:21] I will fix the comments on the MR which should not cause any regression, and then we can do the last test before ask to release it [15:23] ahayzen, if you want a quicker have 'mount -o remount,rw / && apt-get install qtdeclarative5-qtgrilo0.1 qtdeclarative5-qtpowerd0.1 -y && mount -o remount,ro /' [15:23] sergiusens, nope does it make a difference it is running from /opt/click.ubuntu.com/com.ubuntu.music/1.1.350/ [15:23] ahayzen, do find /opt/click.ubuntu.com/com.ubuntu.music/1.1.350/ [15:23] and pastebin that [15:23] ahayzen, how are you running ? [15:23] sergiusens, clicking on it from the dash [15:24] sergiusens, http://pastebin.ubuntu.com/6971404/ [15:24] ahayzen, perfect; with a previous search to refresh the desktop files? [15:25] sergiusens, ah....thts wht the pkill unity8 was doing? [15:25] ahayzen, so you created /opt/click.ubuntu.com/com.ubuntu.music/1.1.350/lib/arm-linux-gnueabihf but copied nothing into it :-) [15:25] ahayzen, hmmm; add a -r to the cp ;-) [15:26] sergiusens, do i need to modify the preinstalled to be 350? cp /usr/share/click/preinstalled/.click/users/\@all/com.ubuntu.music/plugins/* lib/arm-linux-gnueabihf/ [15:26] sergiusens, or should just the -r work? [15:26] ahayzen, just the -r [15:27] sergiusens, magic! [15:27] sergiusens, it works, thank you so much :D [15:27] ahayzen, so depending on the ETA for music to migrate; what would you prefer? [15:28] hmmm, I know what to do as a transparent hack at least [15:28] sergiusens, i don't mind doing this hack each time (i have a bash script to automate it) but others may [15:29] sergiusens, do we know wht the eta of the mediascanner/media-hub are because we'll move to it as soon as it is available and has feature parity [15:31] ahayzen, no idea about media scanner [15:31] don't even know who works on that [15:31] popey, do you know? [15:31] renato_: so calendar passes fine as well. I approved both the mp's on that basis and linked the original bug to them [15:35] sergiusens: someone under thostr_ I think [15:35] fginther: balloons any ETA on weather 197+fun? [15:36] ooh, i see 197.3 [15:36] * popey hugs fginther and balloons [15:42] balloons: fginther weather 197 published to store [15:42] popey, would you rather have .$(date %YYYYMMdd) instead of revno? [15:43] I'd rather have revno_date [15:43] well, we would need to add min and sec [15:43] fginther, ok, now that we've got the fallback safely in the store, can we get a build of trunk? Then we'll test that and we can determine if it's good enough for MWC [15:43] fginther, that'd be for weather again ^ [15:44] dpm, building [15:44] awesome [15:44] balloons: fginther file manager has the same armhf/all issue that sergiusens fixed for music app with https://code.launchpad.net/~sergiusens/music-app/arch/+merge/207635 [15:44] should do same for file manager? [15:44] (and terminal)? [15:44] yes, I am aware [15:45] I need to do that once konsole plugin finishes building [15:45] ok, magic [15:45] like a well oiled machine [16:13] So when i take a branch of the music-app in QtCreator it doesn't show any of the qml/js files just the CMakeLists and the folders, anyone know wht is going on? [16:37] fginther, can I use this weather click for testing? https://jenkins.qa.ubuntu.com/view/All/job/weather-app-click/120/ - I wasn't sure because the build is marked as red [16:38] dpm, let me fix that and retry [16:41] ok, cool [18:21] balloons, could you run the test plan for this MR: https://code.launchpad.net/~renatofilho/qtorganizer5-eds/limit-number-of-events/+merge/207542 [18:21] and if works do the top approval? [18:21] I want to release this today [18:32] renato_: I can't top approve; but yes I will run through it again [18:33] balloons, just paste the test plan [18:33] I will ask other person to top approval [18:38] renato_: pasted [18:39] balloons, thanks [19:39] dpm, I hope you already saw this, but the last build of weather-app-lcick was green [19:39] fginther, I did, thanks, trying to upload to the store now === _salem` is now known as _salem