[00:08] popey, content-hub destination support top approved, should land tonight :) ... note it requires the latest mediascanner2 so will currently only work with devel-proposed === chihchun is now known as chihchun_afk === _salem is now known as salem_ === chihchun_afk is now known as chihchun === salem_ is now known as _salem === Mirv__ is now known as Mirv [05:45] nik90: ping, let's get the alarm enabled sorted out today!!! [07:18] hey all [07:19] hey [07:19] justCarakas, working on anything? [07:20] At the moment I'm at work :p but in my free time I'm working on Be Mobile 2.0 [07:20] neato [07:20] working in a database dump of 1G [07:20] kinda slow :p [07:20] long live VIM [07:21] ha ha [07:21] regex to the rescue! [07:21] idd [07:21] deleted all the duplicate lines [07:21] but that apperantly removed too much :p [07:21] so now I need to place stuff back :p [07:22] like the insert queries to split the records up, because there are tables with 7mil records [07:22] oh gosh [07:22] its a database with all the bus and train stops for belgium and all the depart times [07:23] o__o [07:23] :D [07:24] and thats just the data for 1 year [07:27] geepers [07:29] gonne do the import on the server somewhere in the night probably to minimise the problems for the others :p [07:30] * Akiva-Thinkpad tries to find out how to scroll to the bottom of the flickable in a calender app. [07:31] it really is more tricky than I thought [07:31] I realy should learn qml, do you know any good tutorials [07:32] justCarakas, honestly, its actually kind of easy to just pick it up. [07:32] is there a language thats close to it ? [07:32] justCarakas, I would just use askubuntu for any questions I had. It actually has turned out to be one of the funnest languages. [07:32] justCarakas, not even remotely, at least not from what Ive used [07:33] okidoki [07:33] it does use javascript syntax for functions though. [07:33] maybe gonne do my next one in qml [07:33] justCarakas, next app being what? [07:33] no idea yet :p [07:33] justCarakas, :P [07:34] I have some ideas but I do way to much :p [07:34] justCarakas, I'm working on a few. Right now I just want to help ubuntu with getting their apps ready for rtm [07:34] justCarakas, I really want to program a "Pair Programming" plugin for the ubuntu sdk [07:34] nice :) wish I had the time [07:34] Akiva-Thinkpad: I know, remember your research on it :) [07:35] justCarakas, and I wish I had the money :P [07:35] Akiva-Thinkpad: :p [07:35] justCarakas, yah; I might have been complaining about the api here [07:35] :D [07:36] good morning [07:37] dholbach, morning [07:37] hi Akiva-Thinkpad [07:38] dholbach, busy with the rtm? [07:38] :) [07:39] always busy :) [08:09] yippee; think this bug is fixed :) [08:13] Question on programming style. I am going to add an animation for scrolling to the bottom of the page in the calender app. It will be in newEvent.qml ; Because it is an animation, where should I place it in the code? [08:13] on the bottom? [08:13] near the flickable? [08:16] * Akiva-Thinkpad just places it in the flickable. Someone can get me to change it in the merge request [08:44] Good morning all; happy World Mosquito Day! :-D [08:45] JamesTait: wat, why should I be happy about that :P [08:45] I hate mosquitos [08:45] this channel is strangely quite these past few days.. [08:45] nik90, because we know about the link between mosquitos and malaria transmission - that's surely worth celebrating. :) [08:46] JamesTait: that yes..I misinterpreted that [08:46] nik90, tbh, my initial reaction was the same. [08:49] nik90, then my subversive side came out and I thought "Why not celebrate the fact that something so small can cause so much irritation to so many people?" Then I clicked through and understood the real reason for it. :) [08:49] hehe [09:29] hey popey, I was able to find the source of the bug of calculator and to find a solution, but it losts a lot of preicision, so if we don't find any other solution I'll implement it next week. Meanwhile I found Bartosk, that now is on it :-) [09:29] yeah, i saw his comment, thanks === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun [09:55] I'm confused by how long it takes to review a merge proposal for the gallery app which only consists of one line of code... https://code.launchpad.net/~nikwen/gallery-app/video-delete-message/+merge/225065 [09:56] Would someone please be so kind as to look into it? Thanks. :) [10:00] mzanetti: ping [10:00] mivoligo: hey! [10:01] mzanetti: this is me, Michał [10:01] yep [10:01] I remember :) [10:01] :) [10:02] mzanetti: so, do you have time now? [10:02] yes [10:03] mzanetti: can we use hangout, or better to speak here? [10:03] mivoligo: I think IRC is fine... [10:03] ok [10:04] mivoligo: I just figured that I had to change the launchpad project to be driven by a team, instead of me [10:04] mivoligo: so now you should be able to join the team [10:04] mivoligo: ok. added you to the team [10:04] https://launchpad.net/~machineheads [10:04] mivoligo: great name ;) [10:06] mzanetti: what's the procedure for me to push my stuff there? [10:06] rpadovani: hey, do you know how to install Ubuntu.Components 1.1 on trusty? [10:06] mzanetti, nope, I upgraded to Utopic to avoid the search of solution to this problem :D [10:07] mivoligo: well. for a start I'd prefer you to do merge requests... Once you say you're good with bzr and all this you can directly push to the branch [10:07] mivoligo: so, you branch the source: "bzr branch lp:machines-vs-machines local-branch-name" [10:07] mivoligo: then you do the changes, i.e. add designs (bzr add ..) and commit (bzr commit) [10:08] mivoligo: then, you push to launchpad using: bzr push lp:~mivoligo/machines-vs-machines/my-new-thing [10:08] replace "my-new-thing" which something descriptive of what you've done [10:09] and then, on the launchpad website you can click on propose for merging, and select the main branch lp:manchines-vs-machines [10:09] mivoligo: that will trigger some email to me and I will merge it [10:10] mivoligo: you can always ping me here if you have troubles or questions [10:10] mzanetti: hold on, I got this: http://paste2.org/mWBxkjXN [10:12] mivoligo: I think you didn't enter "yes" in line 4 [10:12] oh you did in 9 [10:12] mzanetti: I had my key on previous machine but I have no idea how to import it now from Launchpad [10:12] mivoligo: but this did work with the old branch with had, the lp:~mzanetti/+junk, did it? [10:13] ah, new machine [10:13] yep [10:13] well, I guess easiest is you just add the key for this machine to launchpad too, or copy the old key from your old machine to the new one [10:13] its in ~/.ssh/ [10:14] mzanetti: let me try [10:14] mivoligo: so you can either copy ~/.ssh/id_rsa.pub from the new machine to launchpad, or copy all the contents of the old machine's ~/.ssh/ to the new one [10:14] (which might overwrite this key - not sure if you used it somewhere else already - so think a bit what you're doing)+ [10:16] mivoligo: I pushed a code change to make it work on 14.04 again. The button's green is not the correct one for now though [10:16] but it should work again for you [10:16] once 14.10 is released and you can upgrade we need to fix a few things then. no big issue [10:16] * mzanetti realizes its only 1.5 months to go until 14.10 [10:18] mzanetti: I've copied the old .ssh stuff but it turns out I don't remember my password :( [10:19] :D [10:19] mzanetti: I guess I need the new key [10:19] ok :) [10:19] mivoligo: on Launchpad, click on your name. you'll get a summary which also contains "ssh keys". [10:19] you can add it there [10:20] mivoligo: if you have deleted/overwritten your old key, you can create a new one with ssh-keygen [10:20] mzanetti: ok [10:40] mzanetti: Ubuntu.Components 1.0 is not installed as well :( [10:40] whaaat? [10:40] ok... [10:41] lemme downgrade once more [10:41] 14.04 is ancient :/ [10:42] yeah... totally [10:42] mivoligo: pushed [10:44] mzanetti: I'm pulling but nothing there yet [10:45] mivoligo: sorry... I pushed to the old branch... now it should be in the new one too [10:45] :P [10:46] mzanetti: it's working, the sound just scared me ;) [10:46] :D [10:46] whichone? [10:47] the laser? [10:47] the first tower [10:48] hey how do you add towers? [10:48] I tried yesterday and I see the machines come, but I didnt know how to add towers to kill them [10:48] mzanetti: laser is cool, man :) [10:49] heh [10:49] nik90: hmm... you should be able to click on an empty field and the available towers show up from the bottom edge [10:50] mivoligo: :) [10:50] mzanetti: hmm, when I clicked on an empty field, i see the selection box, but not the available towers [10:50] * nik90 tries again [10:52] mivoligo: I think tower 1 needs larger bullets [10:52] mzanetti: http://imgur.com/AIGetUJ [10:52] nik90: not on the street [10:53] * mzanetti wonders why its so stretched for nik90 [10:53] ah yes that works [10:53] nik90: did it start up so stretched or did you resize the window? [10:54] mzanetti: I stretched it a bit [10:54] ah ok [10:54] mzanetti: let me restart and get a screenie [10:55] mzanetti: we should add some info if someone tries to put a tower on not allowed places [10:55] mzanetti: http://i.imgur.com/UN15Eoe.png [10:55] mzanetti: I added towers [10:55] mivoligo: and yes I was wondering why it didnt show the towers then [10:56] mivoligo: true [10:56] mzanetti, mivoligo: May be some info on the towers when you hover or longpress on them [10:57] nik90: if you ever played a tower defence, you should know you can't put stuff on the path ;) [10:57] both not good for a touch interface... but anyways, yes... maybe we'll just make the street tiles unselectable [10:57] mivoligo: hehe [10:57] mivoligo: well. there are some where you can [10:58] mzanetti: didn't play them [10:58] the enemies just walk around the tower then (remember my first implementation with the, admittedly bad, intelligence for the wnemies) [10:58] mzanetti: oh this is cool. I will report bugs and when I find time try fixing some UI stuff [10:58] nik90: yeah... its still very much in progress [10:59] mzanetti: is the manage tower button supposed to do something? [10:59] nik90: but seen we'll get the proper level backgrounds and then we start creating real, challenging levels [10:59] nik90: ah no... not any more... that's supposed to go away [10:59] we moved that into the gameplay (was for unlocking towers) [10:59] which is now done during the game [11:03] mzanetti: about the level backgrounds: do you want to give me path coordinates and I will design the backgrounds around them, or should I go completely with my own ideas? I'd prefer the former ;) [11:04] mivoligo: hmm... I hoped for the latter :D [11:04] mzanetti: ok, I have some ideas for few first levels [11:04] mivoligo: I guess you can just create some various paths... [11:05] mivoligo: I think the longer the path (and the more corners) the easier it will be [11:05] mivoligo: but we can reorder them later according to difficulty [11:05] mivoligo: [11:05] so I guess we just want all sorts of random things, straight ones, curved ones, etc [11:05] mzanetti: ok [11:06] mzanetti, mivoligo: cross paths and may be obstruction on path like slow :D [11:06] yeah, we have support for cross paths [11:06] don't have support for tunnels etc. but code wise that shouldn't be too hard I think [11:08] mzanetti: what about the size of the game on desktops? Can you do make it static or full screen? So users don't break proportions. [11:09] mivoligo: yeah, guess I can do that [11:09] mivoligo: I think I can also force the proportions, still allowing resize [11:09] but not my highest priority atm [11:10] mzanetti: ok [11:19] mzanetti: I have a problem with pushing to lp: http://paste2.org/EWjjC6OY [11:19] mivoligo: isn't your launchpad nick mivoligo too? [11:20] mzanetti: no [11:20] hmm... mivoligo: did you add your new ssh key to LP? [11:21] mzanetti: yes [11:21] mivoligo: can you paste the output of "ls ~/.ssh/" [11:21] id_rsa id_rsa.pub known_hosts [11:22] mivoligo: ah, and are you sure you copied id_rsa.pub to LP (as opposed to id_rsa) ? [11:23] mzanetti: yes [11:23] ok... then I'm a bit lost atm... [11:23] anyone else seen this before? http://paste2.org/EWjjC6OY [11:23] hmm... maybe ssh-agent is still using (or trying to use) the old key [11:24] mivoligo: try a "ssh-agent -k" and then try again [11:24] mzanetti: same [11:25] * mzanetti often has troubles too with ssh-agent reloading keys... [11:25] mivoligo: did you copy the key from old machine or create a new one? [11:25] nik90: yeah, key should be fine now [11:25] mzanetti: sometimes uploading a new key to the public servers take some time..may be that's the issue [11:25] at the risk of sounding like a windows admin: try rebooting to be sure ssh-agent is restarted properly [11:26] nik90: I've created the new one and deleted the old one from lp [11:27] mzanetti: ok, I'm rebooting in 3...2... [11:28] :D [11:31] mzanetti: it's working :D [11:31] nice [11:32] mivoligo: now remember to use a new branch for every feature/bugfix [11:32] and that's it. you're fully set up for contribution :) [11:32] mzanetti: great :) [11:33] mzanetti: no more emailing [11:33] :) [11:34] mivoligo: I was really struggling in the end to find all the stuff you've sent me in again in my mailbox [11:34] mzanetti: sorry [11:34] mivoligo: no problem :) [11:34] mivoligo: now its solved anyways [11:36] mzanetti: you said the PNG will be magically generated from SVG, right? [11:36] mivoligo: yeah, I'll figure something to auto export png's from svgs [11:37] mivoligo: taking only the "page" part of the svg [11:37] so you can keep other stuff around in the svg file too which won't end up in the png [11:37] at least that's the plan, didn't check it out yet. but inkscape has a command line mode, so I'm quite sure it should be possible [11:38] mzanetti: I wonder if you can also switch layers visibility, so I can do one file for a different tower levels (red dots)? [11:39] mivoligo: inkscape --help [11:39] that's all I know for now. didn't look at it closer yet [11:40] heh... inkscape --shell [11:40] awesome [11:41] we need --export-area-page at least [11:41] and I guess with --shell we could also remove layers and whatnot [11:41] but need to find a documentation for that still [11:42] mzanetti: ok, it looks promising anyway === chihchun is now known as chihchun_afk [12:14] i spy a new clock from nik90 [12:14] popey: can you test the click? [12:14] popey: too fast :P [12:14] yeah, will do [12:15] popey: the issue is described at https://bugs.launchpad.net/ubuntu-clock-app/+bug/1359180 [12:15] Ubuntu bug 1359180 in Ubuntu Clock App "Add world city is often triggered accidentally when swiping up" [High,In progress] [12:15] kk === _salem is now known as salem_ [12:17] nik90: looks good [12:17] popey: the g+ post or the click? [12:18] click [12:18] popey: so the accidental triggers are much less now? Cool [12:19] yes [12:21] are we doing things to prevent something like this to happen ? http://www.howtogeek.com/194993/the-windows-store-is-a-cesspool-of-scams-why-doesnt-microsoft-care/?utm_content=buffer2ce45&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer [12:47] balloons, when u get a moment do u mind checking over https://code.launchpad.net/~andrew-hayzen/music-app/ap-helper-refactor-001/+merge/231466 [12:47] balloons, i managed to figure out why MusicTracks wasn't working, i had to use Page10 instead [13:09] balloons, we keep running into regressions in apps because no one ran them on a device [13:09] balloons, obviously jenkins doesn't do it, but does anyone? [13:09] balloons, at any stage [13:11] popey: we will need https://bugs.launchpad.net/ubuntu-geonames/+bug/1359206 resolved before I can show location in the clock [13:11] Ubuntu bug 1359206 in Ubuntu Geonames "Need ability to search cities by their lat and long" [Undecided,New] [13:12] I was working on the calender app, and finished the bug which I was working on. To help with RTM, is the Calender App still the application which needs the most work? Or is there another application I can throw myself behind? [13:14] mzanetti: oh btw is it possible to run QML unit tests on the device? [13:14] nik90: sure [13:20] geepers, does anyone know what formatting options for text are available on launchpad? [13:20] I tried html, and reddit style; none worked [13:21] I want to a some text [13:22] Akiva-Thinkpad: you cant do that [13:22] nik90, ah snakes [13:22] welp; maybe I should use unicode >:) [13:22] * Akiva-Thinkpad checks if there is strikethrough unicode [13:23] * Akiva-Thinkpad won't be defeated by limitations [13:25] nik90, LOL! https://code.launchpad.net/~akiva/ubuntu-calendar-app/new-event-pan-flickable-down/+merge/231531 [13:26] check out that formatting :P [13:26] cool [13:26] I am a formatting god :P [13:26] hehe [13:49] http://askubuntu.com/a/513830/253579 I expect some upvotes, as I have now revealed to all of you how to format on launchpad [13:52] If i could I would but I need 15 reputation to do that [13:53] justCarakas, ha ha [13:53] justCarakas, Ive actually had a lot of fun playing on ask ubuntu [13:53] the incentive system really works :) [14:06] beuno, jdstrand: if you have a bit of time, I'd appreciate a quick review of https://code.launchpad.net/~dholbach/click-reviewers-tools/1355215/+merge/231569 :) [14:06] popey, ^ maybe you can test it a bit too? [14:06] Hello guys. Do you need some help with development? [14:07] dholbach, ack [14:12] I realise I should have left a few more comments [14:18] dholbach: will do [14:18] added a few more comments [14:18] hope that'll help [14:19] I wanted to make it so that if we add more clickreviews/cr_*.py modules later on, it'll automatically run these as well [14:19] so we don't need to update it all the time [14:19] that's the main reason it got so long and doesn't look like https://bazaar.launchpad.net/~dholbach/click-reviewers-tools/1355215/view/head:/bin/click-run-checks [14:19] others would say "he got a bit carried away", I guess :-P [14:24] https://www.youtube.com/watch?v=NcbE_jgA9No if anyone is interested -- Ubuntu engineering live [14:26] kenvandine: hey, yesterday I had a play with the ContentHub api. You can now generate QR codes on the fly wherever you can import pictures :D [14:27] kenvandine: small feedback on the API: I think its a bit odd that its a singleton object with a signal... Wouldn't it be possible to have a ContentHub {} object with a state? that way it could be already initialized when the app starts and I don't need to undo all the initialisation for the not-content-exchange use case [14:28] you shouldn't need to under anything? [14:28] kenvandine: well, when the app starts I push the scanner component to the pagestack [14:28] kenvandine: then I get the signal for the content exchange [14:28] so I need to pop that page again and push the generatecode page [14:29] which causes flickering etc. so what I do now is to have a timer in the app that waits for 1 ms for the contenthub signal [14:29] and if that doesn't come in, proceed with the normal use case [14:29] which makes me feel a bit dirty though [14:29] oh... i see [14:29] humm [14:30] so imo it'd be nicer to have: [14:30] maybe it would be better to connect to that in the MainView [14:30] you want your app to switch state when it gets a request [14:30] ContentHub { property bool exchangeActive } [14:31] so in Component.onCompleted I could do "if (!contentHub.exchangeActive) doThis else doThat [14:31] thanks popey [14:32] mzanetti, if you connected to the signal in the MainView, you could push the appropriate page when you get the signal [14:32] kenvandine: yeah... the issue is not that it doesn't work [14:32] kenvandine: the issue is that its a bit late... so I either initialize lots of stuff I wouldn't need, or I do the hack, waiting for the contenthub signal [14:33] but whenever the app is running, you need to be listening for the signal [14:33] not only when a specific page is loaded [14:33] kenvandine: yeah sure... that's still possible though, no? [14:33] that registers the handler with the hub service [14:34] so you just want to avoid the overhead of connecting to the hub on app startup? [14:35] kenvandine: no... I would like the hub to have the state already when the app starts up, not only coming in later as a signal [14:35] you also need to handle the signal on app startup, in case your app was started by the hub because of the transfer request [14:35] exactly, that's the issue [14:36] i see [14:36] for the case where its started just for the transfer, the signal is not really nice imo [14:36] it should be a state that's already set when the app starts [14:36] in that case the signal will come as soon as the hub sees the handler is registered [14:36] Yeah, I guess internally there could be some woes realizing it... [14:37] this is just some feedback using the api [14:37] we need the handler registered [14:37] thanks... i always appreciate feedback [14:37] kenvandine: how do you even know if the singal is connected? [14:39] kenvandine: it would be even easier with an object I guess... [14:39] an object for ContentHub? [14:39] dholbach: so, I like this [14:39] because with my suggestion the ContentHub object would be created when the app starts up [14:40] dholbach: it is quite clean [14:40] dholbach: one of the things I've been thinking about is how click-run-checks isn't dynamic and that it should be fixed [14:40] jdstrand, just pushed some small modifications to make it a bit clearer about what it's trying to do [14:41] so this branch gets me thinking about all of that [14:41] dholbach: I saw r224 [14:41] jdstrand, ah... so you think we could probably merge the two and "just" add a "--raw" option which does what click-run-checks did? [14:41] mzanetti, how are you using it now? with Connections { target: ContentHub, onExportRequested:... } ? [14:41] kenvandine: yeah, exactly [14:41] kenvandine: like in the example in the docs [14:42] ok [14:42] that's what i do as well :) [14:42] dholbach: I am thinking along those lines. I don't have a full thought yet [14:42] jdstrand, I just pushed r227 [14:42] mzanetti, you should have seen the old API... you had to call a function to register your handler.. it was ugly :) [14:42] dholbach: here are ideas: add --verbose to show info too. add --json to output report as json [14:43] jdstrand, sure, I can do that [14:43] dholbach: I think that are not contentious [14:43] mzanetti, so you'd prefer to do something like ContentHub { onExportRequested: ... } [14:43] kenvandine: heh : [14:43] dholbach: awesome, thanks [14:44] kenvandine: actually: ContentHub { onExportRequstChanged: ... } [14:44] kenvandine: where exportRequest is a bool property [14:44] kenvandine: because that can be already set to true when the app starts up [14:44] e.g. in the ctor [14:44] mzanetti, the problem is if it isn't a singleton, the hub would see it as multiple handlers [14:44] yeah... so... :) [14:45] each one is a dbus connection :) [14:45] well, you can make it one in the backend [14:45] kenvandine: you can still have the "singleton" but invisible for QML [14:45] and the objects created in qml connecting to that [14:45] i see [14:46] dholbach: then there is click-show-files.... we could also support --module=security (or something) to support running separate modules-- that would allow us to replace the click-check-* with calls to click-review [14:46] mzanetti, however... if that was a bool [14:46] you would need a way to get the transfer object from it [14:46] ah yes, that's right [14:46] onExportRequested includes the transfer to operate on [14:46] not sure I'll get around to doing that today [14:46] dholbach: I wonder if it is important to continue shipping /usr/bin/click-check-* [14:46] kenvandine: well, that was just a quick suggestion, could also be a QObject* property [14:46] feel free to take the branch as a basis :) [14:46] dholbach: no, those are discussion points :) [14:47] kenvandine: so I could still do if (transferproperty != null) ... [14:47] ah, yeah - I like the ideas :) [14:47] mzanetti, and there could actually be multiple transfers queued up for the handler [14:48] but right now we kind of suppress that [14:48] until we get the trusted session/multiple instances sorted out [14:48] dholbach: so, an easy route we could take is you add --verbose and --json. someone could add --module. we stop shipping /usr/bin/click-check*, but do ship click-review and click-show-files. we then ship click-run-checks which does click-review --json --verbose along with calling click-show-files [14:49] dholbach: (and stop shipping click-check*) [14:49] yeah, I'll try to add --verbose and --json today, then we can file the rest as bugs [14:49] and land them one at a time [14:49] dholbach: I happen to know that currently the sdk only uses click-run-checks and parses the output [14:50] we could make it a simple shell script which runs 'click-review ' [14:50] for transition reasons [14:50] mzanetti, thx for the feedback/ideas [14:50] good stuff to think about [14:50] kenvandine: np [14:50] starting to think we need coordination with them before changing the output of click-run-checks [14:50] i love talking APIs ;) [14:51] dholbach: right-- we can change click-run-checks to use click-review so long as the output is identical to what we have now, until the SDK team moves over [14:52] yep :) [14:52] dholbach: the sdk team would probably love 'click-review --json' [14:52] cause they don't have to parse anything other than json [14:52] yeah, I guess they should be able to drop a lot of code :) [14:52] getting there :) [14:52] dholbach: but, the --json output will need to be something like: [14:52] { [14:52] "security": { [14:53] json for security test [14:53] }, [14:53] "lint": { [14:53] json for lint test [14:53] }, [14:53] ... [14:53] } [14:53] dholbach: what do you think? [14:53] I can use the same pprint function you're using [14:53] that way, there is one big json file that they can drill don't into [14:53] yep [14:54] that works for me [14:54] rather than outputting several json documents in a row [14:54] dholbach: what do you think? [14:55] sure [14:56] ok, cool [14:56] dholbach: yeah, I didn't mean to create more work for you, but --verbose and --json shouldn't be too much. we'll figure out the rest later [14:57] dholbach: put another way-- I found your work inspiring :) [14:59] thanks a lot jdstrand [15:00] I wasn't quite sure if I hadn't been carried away a bit :) [15:00] I added --verbose [15:00] and will work on it some more tomorrow [15:00] if anyone wants to contribute to the effort, feel free to do it :) [15:00] yeah, I'm sure I will [15:00] thanks! [15:01] Akiva-Thinkpad: sooo. interested in that terminal merge? ☻ [15:02] * Akiva-Thinkpad looks at it again [15:02] is always interested in old term [15:05] popey, ah interesting merge. [15:05] popey, does it just need a review or testing? or is there more development that needs to happen? [15:06] Akiva-Thinkpad: it needs a bit of qml work to invoke the function [15:06] Akiva-Thinkpad: if you take a look at the file manager there's a button at the bottom of the screen. [15:06] Akiva-Thinkpad: i think the terminal should have an option under the menu in the top right.. [15:07] okay, in an hour, I'll branch it and let you know if or when I could have it working. Trouble is; I don't have a device to test this on. [15:07] ah okay. [15:08] Akiva-Thinkpad: it's okay, I'll find someone who has a device [15:08] popey, okay thanks [15:08] anything else though? [15:11] mzanetti: hey can you check if https://code.launchpad.net/~nik90/ubuntu-clock-app/fix-accidental-worldcity-trigger/+merge/231230 seems like a good qml solution? [15:12] mzanetti: I am afraid design won't approve removing the add world city button at this point. [15:12] nik90: uh... didn't test it... but seems to change listview overshooting behavior [15:13] mzanetti: yes, this way when user flicks and lets go, it will stop at the bounds and not be elastic. [15:13] nik90: I guess it'd be better to just trigger the button when the finger is still *on* while the button is revealed [15:14] nik90: the accidental one mostly happens if you quickly flick, in which case the finger is not on the listview any more, no? [15:14] okay i'll be back in an hour; if anyone has an app that needs tackling; let me know [15:14] mzanetti: yes [15:15] nik90: I disllike if the listviews behave differently everywhere... [15:15] nik90: rpadovani did a similar fix for the reminders app recently [15:15] nik90: we had the same problem [15:16] still in review though [15:16] * mzanetti blames mzanetti for that [15:16] mzanetti: well the add world city button shouldn't be shown let alone triggered when the user does the flick [15:16] since that looks bad [15:16] mzanetti: so onyl when the user drags should it be visible [15:16] right [15:17] mzanetti: I am afraid changing the listview behavior is the only way to go [15:17] mzanetti: although I am thinking, may be I can have a bool variable which stores where the listview is being pressed on by the user or not. [15:17] hmm... [15:17] maybe [15:17] mzanetti: if true then stop at bounds, if false then overshoot [15:18] meh... [15:18] don't stop at bounds :D [15:18] then the add world city button becomes becomes though :/ [15:18] or you think that is okay as long as it does not trigger the page itself [15:19] * nik90 is fine with that [15:19] +1 [15:19] interesting, let me test that out on the phone and see what it feels like [15:32] mzanetti: After testing, I dont like either of the methods :/.. I think a simple add button would be better [15:33] mzanetti: I will try to do a hangout with the designer [15:33] nik90: +many [15:33] nik90: again... you need that thing once in your lifetime (well, once in your phone's settings lifetime) [15:33] why not hiding it more and put the useful things here === salem_ is now known as _salem [15:34] mzanetti: some people use the world clocks frequently [15:34] mzanetti: besides if we change that to alarms, then we would have to also move the alarm add functions to the main page [15:34] mzanetti: it is huge undertaking at this stage [15:34] yeah... guess so [15:34] well, just saying [15:37] balloons, hey [15:37] brendand, howdy [15:38] balloons, i gather dpm isn't around today? [15:39] brendand, no, he's out for a few weeks [15:39] balloons, wow :) [15:39] brendand, what do you need? [15:39] balloons, anyway - i was trying to ping you earlier about merge proposal testing for apps [15:40] balloons, the fact that they don't get run on devices is proving a big gap [15:40] balloons, i've asked CI about this a lot and they say - later [15:40] balloons, anything we can do to mitigate it in the meantime? [15:40] balloons, is it possible to arrange a sign-off from someone with a device on each MP? [15:41] brendand, an excellent discussion; there's some history here [15:42] brendand, hangout? [15:42] balloons, yeah quick one [15:42] balloons, fire a link across [15:43] fired === gatox is now known as gatox_lunch [15:51] ping boiko: we have two branches ready that simplify the tests: [15:51] https://code.launchpad.net/~canonical-platform-qa/dialer-app/fake_url-dispatcher/+merge/230403 [15:51] https://code.launchpad.net/~canonical-platform-qa/dialer-app/custom_proxy_objects/+merge/230687 [15:51] the goal is to run them as autopkgtests, so a little more work is on the way. [15:59] elopio: nice! we are just finishing a set of changes to the telephony-components, as soon as we are done with that I'll go ahead and review this [15:59] thanks. [16:01] balloons: so who is in charge of translations while balloons is gone, I have a pending pot file update [16:01] nik90, mm.. someone else on the team can probably have a look [16:02] balloons: ok, popey is my teammate there :D [16:12] ping artmello: here is a quick fix to get the gallery tests back to green: https://code.launchpad.net/~canonical-platform-qa/gallery-app/fix1358968-update_content_picker/+merge/231480 [16:18] Okay I'm back [16:27] elopio: thx a lot === _salem is now known as salem_ [16:36] ybon: hey, would you be free in a few hours to talk about Qt Location? I am planning to integrate it in the clock app, but need some questions answered first [16:36] ybon: I am a bit busy atm with some MPs. [16:41] So any apps that need working on for rtm (that don't require me to own a device >_>) [16:42] Akiva-Thinkpad: hey! [16:42] Akiva-Thinkpad: http://people.canonical.com/~dpm/core-apps-sponsoring/ is a good place to start ☻ === gatox_lunch is now known as gatox [16:43] cool beans popey [16:44] popey, didn't realise i've got back to top of that list again \o/ only 42 days in queue not too bad this time [16:44] * popey shakes fist at ahayzen [16:44] hehe [16:44] ahayzen: I pushed music to the store this morning [16:44] lol [16:45] popey, basic content-hub destination support landed :) [16:45] ya [16:45] \o/ [16:45] is balloons about? [16:49] popey, we are potentially gonna have the issue again soon where we cannot push trunk to the store until ms2 lands in a promoted image otherwise we'll break in devel [16:50] popey, specifically if this lands https://code.launchpad.net/~andrew-hayzen/music-app/fix-no-music-use-model-status/+merge/231547 which is as a result of the comments in https://bugs.launchpad.net/mediascanner2/+bug/1358275 [16:50] Ubuntu bug 1358275 in mediascanner2 "SongsModel.status is not set as MediaStatus.Ready if rowCount is zero" [High,Incomplete] === salem_ is now known as _salem [16:58] ahayzen, howdy [16:58] balloons, o/ when u have a moment would you be able to review https://code.launchpad.net/~andrew-hayzen/music-app/ap-helper-refactor-001/+merge/231466 [16:59] balloons, i managed to get it working but had to use Page10 for some reason? nik90 said they had to do the same in clock [17:00] balloons, but mainly could you check that the conversion of the new helpers is good as it is the basis to when i start converting all the other tests so don't wanna get this bit wrong ;) [17:00] pfff [17:01] ahayzen: autopilot uses the qml filenames instead of the names of the components.. that's why you have to use Page10 [17:01] and when you switch to Ubuntu.Components 1.1 probably it will need 1.0 [17:01] ahayzen: understood [17:01] ahayzen: I changed a bunch of code to not use the component type, only the objectName, otherwise it has to be changed every time we change filenames internally [17:01] t1mp, but hmmm i have MusicTracks inside a tab ... and that comes from a different file? [17:02] ahayzen: I think it doesn't matter whether you have the Page inside something [17:02] t1mp, hmm ok we need to transition to 1.1 anyway (i thought we had in one merge but think that got deferred) so i'll mp that up :) [17:03] t1mp, as we are like half 0.1 half 1.1 atm lol as we use things like SortFilterModel [17:03] I think the AP guys are working on something to make the select_single use the component name (Page) instead of filename (Page10 or Page11) [17:03] or at least they are aware of the problem [17:03] t1mp, ok good :) [17:04] t1mp, while ur here any movement on https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1341814 ? [17:04] Ubuntu bug 1341814 in Ubuntu UI Toolkit "Using search in the header can sometimes have a text field from a different tab" [Critical,Confirmed] [17:04] ahayzen: self.main_view.select_single(objectName='tracksPage') probably works too [17:04] ahayzen: no, sorry. I've been super busy with other stuff [17:05] t1mp, i guess but i need the methods in the class which i give it eg i'm doing self.main_view.select_single(Page10, objectName='tracksPage') [17:05] t1mp, and then i do .get_track(i) on the returned object [17:05] t1mp, no worries about that bug i understand ur busy :) [17:07] popey, i'll let u know if that lands ... so then u know not to push to store [17:08] t1mp, is there a bug for the Page10 vs Page issue you mentioned? Thanks for the explaination on what's happening [17:09] Question: What part is the header again? Is that the Page Top? https://bugs.launchpad.net/ubuntu-calendar-app/+bug/1357315 [17:09] Ubuntu bug 1357315 in Ubuntu Calendar App "Move the Save and Cancel button in the new event page to the header" [Medium,Triaged] [17:09] ahayzen: what do you mean? What I was saying is that if you remove "Page10, " then it will also work [17:09] hes not talking about a popover is he? [17:09] balloons: I think so, let me search [17:10] nik90: sorry, I'm travelling in boat this week, so not really online [17:10] t1mp, i know it will work but will it then automatically pick my helper class? [17:10] can I help now, [17:10] s/,/?/ [17:10] ybon: yeah sure [17:10] ybon: I want to know if it is possible to detect if the GPS is enabled or not on the phone [17:11] ybon: if GPS is not enabled, I can show a message that "Location cannot be detected" or even show the last known location [17:11] ybon: is this possible? [17:11] ahayzen: ok [17:11] Akiva-Thinkpad: the bit at the top with the arrow in it. [17:11] thanks [17:11] * popey goes afk for a while [17:12] nik90: I think you can check the "valid" property [17:13] balloons: this seems related, but it is not it https://bugs.launchpad.net/autopilot/+bug/1337004 [17:13] Ubuntu bug 1337004 in Autopilot "Make it easier to select a custom proxy object with a class name different from the QML type" [Undecided,New] [17:13] elopio: ^ do you have a bug for the Page10 vs Page in select_single? [17:13] nik90: http://developer.ubuntu.com/api/qml/sdk-14.10/QtLocation.PositionSource/#valid-prop [17:13] ybon: Sweet [17:14] ybon: yeah I will work on this and see how far I get with it. [17:14] t1mp: like this one? https://bugs.launchpad.net/autopilot-qt/+bug/1341671 [17:14] Ubuntu bug 1341671 in Autopilot Qt Support "Versioned QML classes are not recognized by their public type name" [High,Confirmed] [17:14] ahayzen: oh. I don't know if it will pick the helper class [17:14] elopio: yes, thanks [17:14] balloons: ^ there is the bug :) [17:14] elopio, t1mp thanks [17:14] elopio: I was searching in autopilot, not autopilot-qt [17:15] t1mp, idk either i'm new to the helper classes...attempting to move music-app at the moment...hence my questions ;) [17:15] nik90: next thing I'd like to plug in is https://mozilla-ichnaea.readthedocs.org/en/latest/api/index.html#service-api [17:15] balloons: t1mp: in many of the cases, what we are missing is the helper. If we had a helper for Page on the toolkit, we could make it match Page10 and Page11 with the tools we currently have. [17:15] and make sure it keeps wokring once we have better tools to handle that case. [17:16] elopio: how would we do that? for select_single we don't use a helper [17:16] ybon: ah that's AGPS offered by mozilla [17:16] elopio, but not every page is the same.. we ran into that on clock app.. Yes, they are all 'Page', but we extend them different;y [17:16] ah yes, if you extend them some times the type changes for AP [17:17] perhaps we should select only by objectName? [17:17] balloons: for that case we are already overwriting them on the app, so the update for the new version must be on that project. [17:17] that would be an intense dbus search [17:17] or is that super slow or something? [17:17] ah [17:17] I started doing that in a few cases : [17:19] sometimes it's the way to go. Until they fix bug #1337004, it's the only way to match a class that has a different name. [17:19] bug 1337004 in Autopilot "Make it easier to select a custom proxy object with a class name different from the QML type" [Undecided,New] https://launchpad.net/bugs/1337004 [17:27] Which import QtQuick version should I be using these days? === _salem is now known as salem_ [17:29] ahayzen: I saw renatu use Qt quick 2.2 in his pagewithbottomedge component [17:30] nik90, yeah should i be using 2.2 as all of ours are 2.0 at the moment? ...thought i would bump while doing the Ubuntu.Components [17:30] ahayzen: yeah I was thinkking of doing the bump as well [17:30] ahayzen: just never got to it [17:30] nik90, i'm gonna do it :) see what happens [17:33] ahayzen: hope your comp doesn't explode :D [17:34] nik90, damn qtcreator wouldn't find and replace across the project....guess because it is a cmake? [17:34] nik90, had to do it based on 'files on the file system' [17:34] ahayzen: I never knew qtcreator could even search across the entire project [17:35] nik90, yep Edit->Find and replace->Advanced Find->Current Project [17:35] nik90, quite useful when it works :P [17:35] ahayzen: wow, cool [17:35] nik90, makes version bumps easy hehe [17:37] ahayzen: it works for me [17:37] nik90, magic [17:37] indeed [17:39] nik90, ah i'm on trusty with probably an old qtcreator? guess ur on utopic? [17:39] ahayzen: utopic vm [17:39] utopic is remarkeably stable even as an install [17:39] well other than the suspend issues [17:40] Akiva-Thinkpad: I know, I intend to stay on trusty until next LTS [17:40] Akiva-Thinkpad: VM seems to be working fine, so rather keep it that way [17:40] nik90, but this is cutting edge... err well not sure what is actually different in utopic :P [17:40] I say this but I have a feeling I might upgrade when utopic comes out [17:40] :) [17:40] same [17:41] i need to really test new mediascanner and media-hub versions on desktop as well as device [17:41] * Akiva-Thinkpad runs it as his main. Cause' hes edgy [17:41] geys stay away from me; my middle name is riskay [17:41] err, Rickey [17:41] its richard actually :/ [17:49] nik90: why use a vm when an lxc container is much more lightweight if you need it ;-) [17:49] sergiusens: because I have no clue how to do that ;) [17:50] sergiusens: I need to read up on lxc containers, updating them, etc etc etc [17:50] nik90: here's a good pointer https://www.stgraber.org/2013/12/20/lxc-1-0-your-first-ubuntu-container/ [17:51] * nik90 reads up === salem_ is now known as _salem [18:12] sergiusens: damn it looks really powerful.. I will see if I can set up everything required and then create a video for this. it will be so powerful for app devs who want to stick with 14.04 while still wanting to develop on 14.10 and later [18:13] nik90: that's what I'm doing [18:13] at least; that's how I played with the clock app builds [18:13] sergiusens: oh [18:13] aweseom [18:13] and this is how we run the android in a container ;-) [18:15] sergiusens: so I need to figure out how to set up QtC, necessary ssh keys, share storage etc on LXC [18:15] sergiusens: I will start with some baby steps by creating LXC containers and then slowly working from there === _salem is now known as salem_ [18:33] Yawn~~~ I think i'm done for the day. Thanks all. [18:33] o/ [19:01] popey: when you are back, please review and top approve https://code.launchpad.net/~nik90/ubuntu-clock-app/update-potfile-aug18/+merge/231198 if everything is good. Just a pot file update [19:10] Who would know a bit about working with WebView components? [19:10] I'm looking to increase default zoom level (i.e. change default canvas size I guess) [19:32] bfiller: I added a fix for https://code.launchpad.net/~phablet-team/gallery-app/gallery-app-restrict-pick-content-type/+merge/230776 . I decided to optimize my previous code yesterday and didn't test toroughly enough, sorry for the mistake. I will try to add unit tests on Monday to catch other problems with these modifications [19:33] bfiller: but it should be good to go for now [19:33] nerochiaro: ok thanks, I will test it [19:33] bfiller: thanks [19:35] hi all, so I just created an app with the sdk that has tabs, and I got the old school tabs, did I do something wrong? [19:35] * rickspencer3 wants new tabs [19:37] useDeprecatedToolbar: false [19:41] Toolbar? [19:41] ok [19:41] daker, what object is that property on? [19:41] nm, found it, I think [19:42] new tabs, toolbar actions in the header [19:42] muhc better [19:42] http://developer.ubuntu.com/api/qml/sdk-14.10/Ubuntu.Components.MainView/ [19:42] I wonder if we shouldn't add that to the app template? [19:42] bzoltan1, what do you think? ^ [19:48] Does anyone know how do I install Boost Spirit in Ubuntu ? Can't seem to find a separate package for this Boost library === renato is now known as Guest60224 [20:15] rickspencer3, i thought that property would be false by default soon so all apps would use the new toolbar? [20:19] ahayzen, ah [20:19] that makes sense [20:19] rickspencer3, but i'm sure one of the SDK guys would be able to confirm that for you [20:19] s'all good [20:19] sounds like it's under control [20:20] hopefully :) [20:21] rickspencer3, ah there is the mail about it... https://lists.launchpad.net/ubuntu-phone/msg09272.html [20:21] rickspencer3, that suggests you aren't using Ubuntu.Components 1.1 then? [20:22] ahayzen, I am using what the template creates for me :) [20:22] 0.1 [20:22] heh === salem_ is now known as _salem [20:22] rickspencer3, i wonder if that will be updated when utopic is released and then that sdk version is 'stable' [20:23] dunno, I'm on Utopic now [20:23] it could just be that the template is behind [20:23] yeah probably [20:30] rickspencer3, hmmm just tried removing the setting on a version of the music-app where i'm bumping the Ubuntu.Components version and i get the old toolbar [20:31] huh [20:32] ahayzen, I have to leave soon, but I'll follow tomorrow [20:32] make sure all the pieces are lining up :) [20:32] rickspencer3, cool thanks i'll continue playing as well [20:35] t1mp, is what we've said above correct? [20:36] ahayzen, funny that you say continue playing because I am currently mutli-tasking [20:36] practicing for my music lesson [20:36] I wondered if I had left a mic on? [20:36] :) [20:36] :) [20:36] rickspencer3, what are you learning? [20:37] mandolin [20:37] the world's most annoying instrument [20:37] hah [20:37] I have gotten obsessed with old timey music, it's very weird [20:37] never would have imagined [20:37] hah [20:38] looks like an intriguing instrument === fginther` is now known as fginther [20:58] sergiusens: were you able to run QtC in a lxc-container? [20:58] I plan on following https://www.stgraber.org/2014/02/09/lxc-1-0-gui-in-containers/ to do that, but it seems some apps might require some special config tweaks. [20:59] I created my first lxc container for utopic amd64. Haven't done much after that [21:01] nik90: I can search and help; but stgraber will pinpoint you faster than me [21:01] sergiusens: which channel is usually on? [21:02] sergiusens: does lxc have its own irc channel? [21:02] nik90: ubuntu-devel [21:02] sergiusens: ah ok. thnx [21:02] nik90: no idea [21:16] ahayzen, having a look at your mp, thanks for packaging it up [21:16] balloons, awesome thanks :) if it is good then i'll start working on converting the tests [21:18] balloons, obviously i will then need to add more helpers as i go along... basically all i converted in that mp was populate queue and how the starting of it all works [21:18] ahayzen, MainView still needs split up obviously.. so many methods [21:18] balloons, yep as u said before i'm gonna do in many mps so it is easier to review...and slowly remove the methods [21:18] yep yep [21:19] I'll leave my notes inline on the mp [21:19] balloons, the only bit which i didn't like was the properties for main_view, player and pointing_device ... they are half there for backwards compatibility [21:19] balloons, awesome thanks :) [21:31] balloons, thanks for the review...whats the difference in python between single and double quotes? [21:33] balloons, hahah looks like my copy and pasting failed then :P [21:36] sergiusens: can you check if the changes I made to cmake at https://code.launchpad.net/~nik90/ubuntu-clock-app/update-project-name/+merge/231618 are correct? Its only 4-5 lines cmake code change. [21:36] ahayzen, yes, the self.main_view, etc I assume might go away [21:36] balloons, yeah if it get get it all in emulators [21:36] sergiusens: I changed the project name to com.ubuntu.clock to prep for the replacement [21:36] sergiusens: I also added a missing url_dispatcher file. Although not sure if my install path for click and non-click mode are correct [21:37] sergiusens: here is the old clock source directory https://bazaar.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/trunk/files [21:37] sergiusens: and here is the cmake file for the old clock app https://bazaar.launchpad.net/~ubuntu-clock-dev/ubuntu-clock-app/trunk/view/head:/CMakeLists.txt [21:37] nik90: it looks good; does jenkins build this? [21:38] sergiusens: it will build in a few minutes [21:38] sergiusens: is the url_dispatcher being installed to the correct locations? I always mess up that :P [21:38] nik90: I'll recheck after it builds [21:38] sergiusens: ok. I will ping when the build is done [21:39] nik90: for the deb, I don't really know; for the click; I don't see the hook in the diff [21:39] let me open the file [21:40] sergiusens: for click isn't it set(URLS_DIR ${CMAKE_INSTALL_DATADIR}) [21:40] ahayzen, either is fine, but string literals have migrated to using just ' [21:40] sergiusens: and it is installed with install(FILES ${URLS_FILE} DESTINATION ${URLS_DIR}) [21:40] nik90: yeah you are missing the hook here: http://bazaar.launchpad.net/~nik90/ubuntu-clock-app/update-project-name/view/head:/manifest.json#L9 [21:40] balloons, i thought u could use either? https://docs.python.org/3.4/library/stdtypes.html#textseq [21:40] nik90: a "urls" hook, like in https://wiki.ubuntu.com/URLDispatcher [21:41] balloons, or is this just a standard u guys are using to make the ap tests consistent? [21:41] ahayzen, yes you are correct. you are free to use either [21:41] I tossed that in there so elopio would get a laugh if he had a look [21:41] nik90: it's only value is the path to the dispatcher file [21:41] \o/ lol [21:41] the mp would be huge if you did a replace ;-) [21:41] ahayzen, but yes, he's migrated to 'blah' [21:42] select_single('ListItemWithActions, objectName="tracksTabListItem") [21:42] nik90: which is ${CMAKE_INSTALL_DATADIR} [21:42] balloons, i leave for now...and maybe do an mp later on to make them all one type for fun once i have converted all the tests [21:42] nik90: /thefilename.json [21:42] ahayzen, pure troll.. love it [21:42] I support the giant mp after you're done [21:42] https://www.irccloud.com/pastebin/tdogQ2XH [21:42] sergiusens: ^^ [21:42] balloons, ok i'll push a fix to my minor typo of mentioning reminders app then it is ready to land? [21:43] ahayzen, yes, I'll approve [21:43] nik90: btw, you don't need to discrimitate URLS_DIR for click and for deb; you can just use the same one and reference it from the manifest as share/url-dispatcher/urls/file.json [21:43] balloons, pushed! [21:43] sergiusens: ah I just took this from the old clock app to avoid my personal mistakes [21:44] ahayzen, night [21:44] night? [21:45] sergiusens: should I leave it as it is? [21:45] balloons, one question i sense soon the populated property that we do a wait_for(True) will disappear soon we will have to compare .status to a enum of a QML class however u can only access it by calling the QML class directly...how would we access it through autopilot? [21:45] balloons, or maybe i should check how the ms2 tests do it first... [21:45] nik90: it's your pref really; but for desktop file I think I used a similar logic [21:46] nik90: that's why the manifest says: "desktop": "share/applications/ubuntu-clock-app.desktop" [21:46] sergiusens: ah ok [21:47] nik90: wrt the paste; that file is fine; it's just missing referencing in the manifest as a hook for the app [21:47] sergiusens: I will try to mimick the dektop then and also correct the manifest file [21:47] nik90: similar to desktop: .... ; you are missing "urls": "path to url dispatcher file" [21:47] right [21:47] yup [21:48] ahayzen, I'm really confused but your qurstion [21:48] so yea, I guess check the other tests.. [21:48] balloons, i bet u are...i was when it wasn't working...so say you have SongsModel { id: myModel } ... songs model has a status which becomes ready [21:49] balloons, you need todo myModel.status == SongsModel.Ready [21:49] balloons, you can however not do myModel.status == myModel.Ready [21:50] balloons, so how would ap access SongsModel.Ready? as SongsModel comes from directly in mediascanner2 [21:50] balloons, unless we just put a property in there for ap like property bool populated: status == SongsModel.Ready [21:51] ...mediascanner don't have any ap tests /o\ [21:51] balloons, anyway we'll cross that bridge when we come to it as that mp hasn't landed yet :) [21:52] balloons, but i think just adding in a property for ap will be the easiest way [21:53] sergiusens: how does it look now? [21:54] balloons, anyway thanks for the review :) onto the next set of autopilot conversions :D [21:54] nik90: looks good! [21:55] sergiusens: btw since the big deb packaging fix, I am unable to the main qml file ubuntu-clock-app.qml file listed in QtC [21:55] sergiusens: should I add a custom target to it in the app/CMakeList.txt file? [21:56] nik90: yeah, I think that does the trick [21:56] sergiusens: I think add_custom_target(ubuntu-clock-app_QMlFiles ALL SOURCES ${QML_JS_FILES}) didnt do the trick [21:56] nik90: fwiw, I just copied what mzanetti did for reminders; I'm not really a Qtc person [21:56] why screen resolution isn't that good for so many widgets [21:56] and I got used to vim a long time ago [21:56] :-) [21:56] hehe [21:57] sergiusens: but to be clear, add_custom_target() only affects QTC, it should affect the install, right? [21:57] shouldn't* [21:57] nik90: right; unles someone invokes the custom target (which qtc might) [21:58] sergiusens: hmm ok [21:58] sergiusens: I will include my custom target inside if(NOT "${CMAKE_CURRENT_SOURCE_DIR}" STREQUAL "${CMAKE_CURRENT_BINARY_DIR}") to make sure that doesnt happen [21:59] nik90: yeah [21:59] sergiusens: can you quickly approve my MP for the cmake changes. I will ask balloons or popey to evaluate the rest and merge when ready [22:00] nik90: I was going to do a local test build and try it out [22:00] nik90: but if eye balling was all you wanted, I'm fine too ;-) [22:00] sergiusens: ah sure [22:00] sergiusens: no no I cannot afford this to cause any issues now :) [22:00] Anyone know how to make tactile feedback when you click objects? [22:01] ...or haptic feedback [22:01] ahayzen: http://doc-snapshot.qt-project.org/qt-mobility/qml-hapticseffect.html [22:02] nik90, hmm interesting is there a generic Ubuntu one? so i can be consistent with when eg the header buttons are clicked? [22:02] ahayzen: why don't you encapsulate your object inside a AbstractButton{} [22:03] ahayzen: it provides haptic feedback like you want [22:03] ahayzen: and no I don't recall seeing any ubuntu one [22:03] nik90, hmmm but i want it when u click on listitems [22:03] nik90, all i see is https://developer.ubuntu.com/api/qml/sdk-14.10/QtFeedback.HapticsEffect/ [22:03] nik90, which is exactly the same doc :P [22:03] nik90, basically pretty much anywhere in the music app when u do onClicked it should vibrate [22:04] ahayzen: are you sure? [22:04] ahayzen: that would break consistency with other apps [22:04] ahayzen: and too much battery usage [22:04] nik90, ...thts what unity started doing but it keeps changing [22:04] nik90, hmmm ok maybe not *everywhere* but it is inconsistent with itself eg selecting the back button in the header vibrates [22:05] nik90, hmm maybe a question for design then [22:06] ahayzen: I suppose [22:13] sergiusens: suprisingly, I see add_custom_target(ubuntu-clock-app_QMlFiles ALL SOURCES ${QML_JS_FILES}) however I cannot find the variable ${QML_JS_FILES} defined anywhere in that file [22:21] popey: who is the next translation expert after dpm? [22:21] uh [22:21] popey: I would like to check if translations has been properly enabled in launchpad for clock reboot. I have seen any new po files yet [22:21] popey: since dpm is on vacation [22:21] nik90: set architecture to armhf now [22:21] haven't seen any po files* [22:21] possibly ara [22:22] sergiusens: ok [22:22] nik90: or get fginther to do dual builds [22:22] and set it as a list [22:22] sergiusens: list? [22:22] architecture: ["armhf", "i386"] [22:22] or [22:22] sergiusens: if I set it as armhf, will jenkins still be able to build it for amd64 for it to test? [22:22] architecture: "armhf" [22:23] popey: ok [22:23] nik90: yes it will; it will only fail review ;-) [22:23] ok :P [22:23] nik90: all is technically incorrect anyways [22:23] "all" [22:24] done === _salem is now known as salem_ [22:27] nik90: also, can you rename 86 "ubuntu-clock-app": { [22:27] to just "clock" [22:27] the -app is redundant [22:28] and the ubuntu we try to avoid in app names [22:29] sergiusens: well the app is called "Clock" but the package name is "ubuntu-clock-app" since otherwise it is too generic [22:29] nik90: just for the click; not the deb [22:29] nik90: your package is com.ubuntu.clock (so ubuntu is already there) [22:29] ah ok [22:29] the app name should be clock (without ubuntu or app, as it's the app name) [22:30] nik90: and there's one more thing missing [22:30] nik90: "com.ubuntu.clock.devel", [22:30] nik90: that's in the autopilot tests [22:30] I promise I won't run them! [22:30] :-) [22:30] hehe [22:30] oops I forgot them totally [22:31] nik90: asie from that, the clock build runs fine [22:32] nik90: Setting BZR_SOURCE to lp:ubuntu-clock-app/reboot [22:32] that needs updating too [22:32] to the actual trunk [22:32] sergiusens: why? [22:32] sergiusens: I don't think we are moving [22:32] nik90: oh; then no need [22:33] sergiusens: I added 33 +file(GLOB QML_JS_FILES *.qml *.js), hope that is okay [22:33] sergiusens: It then shows the main qml file in qtc for me [22:34] nik90: should be [22:34] hmm although I should indent it [22:45] sergiusens: awesome click-buddy creates click as com.ubuntu.clock_3.1.71_armhf.click. Looks good to me [22:45] nik90: great; clock runs fine for me as well; just need the tests fixed [22:46] sergiusens: I fixed that in the latest push [22:46] sergiusens: I had to just change the variable in __init__.py file in the autopilot folder [22:47] nik90: yeah, the running of the tests I leave to you or balloons :-) [22:47] I'll approve [22:47] I just ran the tests, works as expected using make autopilot [22:48] nik90: running the tests for the click app is a different test path though [22:49] but jenkins runs it after installing the deb [22:49] also atm the clock app tests dont work on the device due to an upstream autopilot error [22:54] balloons: in prep for the clock app replacement, i made the necessary changes to the click, local path names which used com.ubuntu.clock.devel.