=== chriadam|away is now known as chriadam [04:59] popey, or balloons, or mhall119 ping? I just want to confirm that the documentation about installing the sdk by using the sudo add-apt-repository ppa:ubuntu-sdk-team/ppa -- is no longer needed. [04:59] ratherhttps://developer.ubuntu.com/en/start/ubuntu-sdk/installing-the-sdk/ [04:59] https://developer.ubuntu.com/en/start/ubuntu-sdk/installing-the-sdk/ [05:00] I have been installing the sdk from the repos on 15.xx, and its been working fine. Where as I have seen people with issues when they try to install via using that ppa. [07:00] good morning [08:48] hey, is it possible to filter a model with SortFilterModel on multiple model property (like pattern matching property1 or property2 or property3…) [08:48] ? [08:52] it seems there is no test for this, so I guess not really possible? http://bazaar.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/staging/view/head:/tests/unit/tst_components/tst_sortfiltermodel.qml [08:53] as u1db doesn't support this sort of filtering, is there any best practice to achieve that, or doing the filtering manually will be performant enough? [08:53] kalikiana: any idea? ^ === chriadam is now known as chriadam|away [09:07] didrocks, good question~ [09:09] didrocks: looks like it will only accept a single string as it's property http://bazaar.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/staging/view/head:/src/Ubuntu/Components/plugin/filterbehavior.h#L28 [09:09] which would mean you would have to create a custom implementation of a QSortFilterProxyModel which can filter on multiple roles [09:10] DanChapman: yeah, I think doing that in pure javascript can be a little bit slow though, so having to resort/filter it in C++, let's see if any of the SDK guys have thought/experienced it [09:18] didrocks: I didn't mean do it in javascript :-D rather it would require a c++ plugin to achieve what your after [09:19] balloons: yet another reason why we need on-device testing :) https://bugs.launchpad.net/ubuntu-calculator-app/+bug/1480167 [09:19] Ubuntu bug 1480167 in Ubuntu Calculator App "Calculator no longer starts" [Critical,New] === JMulholland_ is now known as JMulholland [09:23] Hey popey, just a quick yes or now question. You know that fellow who made a click package bundling a java virtual machine; Did he have to statically link the Java Virtual Machine? I'm just kind of confused whether this is the way to do it. [09:23] no* [09:32] kivi: no idea, I didn't build it [09:33] popey, daw... :/ I'm just struggling to figure out that if I need to say, include a python interpreter in my program (because app armor won't let it through), whether I need to bundle one statically into my application. [09:34] well there is no guarantee that python will be on the image, so yeah, you should I imagine [09:34] see checkbox, that's a python app [09:34] popey, ty! [09:35] !cookies [09:35] Cookies are delicious delicacies. [09:47] hum, it also seems that I can't filter on subproperties (like the one brought by u1db model) via SortFilterModel [09:47] like filter.property: 'contents.title', the pattern never matches (contrary to a simple docId) [09:48] I tried as well filter.property: 'contents["title"]' or alike [09:48] kalikiana: when you get some time, this one as well ^ [09:54] I guess that's https://code.launchpad.net/~kalikiana/u1db-qt/indexRoles/+merge/211771 [10:09] popey: you might want to make the orientation fixed in don't crash: http://themcphails.uk/dc.png [10:11] mcphail: uh [10:11] what device is that ? [10:11] krillin [10:12] (I can break _anything_ ) [10:12] popey: automatic orientation breaks on the restart screen [10:13] ugh [10:13] ok thanks [10:18] DanChapman, this flickering trashcan when scrolling is making me crazy, could we fix that at some point ? [10:19] Mission accomplished [10:19] if user = ogra then flash_trash [10:19] haha [10:19] flickering trash icon?? [10:20] in the message view ... there is a trashcan icon at the top right [10:20] got a bug number? [10:20] if you scroll it goes active/inactive all the time [10:20] very distracting :) [10:21] yes - it would be good to be able to press delete when the screen is still scrolling [10:21] ah... gotcha. Yes that prevents segfaults when destroying an oxide webview while the view is still scrolling. [10:22] well, you could make it a no-op without switching the icon state :) [10:22] (and cache the request until the scrolling is over) [10:23] * ogra_ finds it awesome that this is the only bug bothering him btw :) [10:24] That's a nice idea! i'll create an issue to get that done :-) [10:25] yay [11:24] popey: https://bugs.launchpad.net/ubuntu-calendar-app/+bug/1480154 [11:24] Ubuntu bug 1480154 in Ubuntu Calendar App "Calendar not running on desktop" [Undecided,New] [11:25] ta [11:26] user your converged phone then :P [11:26] ogra_: snappy hasn't fix that yet so all your fault [11:26] odd, works here [11:26] what version of package you got? [11:26] popey: it's in the bug [11:26] * popey scrolls down [11:27] dunno what the hell is generating that error [11:27] eds probably? [11:28] popey: also this is a fresh install so I don't know if there are any deps missing that you might have and I don't [11:28] ah, you probably need the sdk ppa [11:28] well fresh install from the ppa I mean not the desktop [11:29] but yeah, probably missing a dependency maybe [11:29] popey: I'll grab the sdk ppa see if that improves things, I think reminder/notes has the same issue, but music and dekko are both fine === JamesTai1 is now known as JamesTait [11:39] popey: still the same :( [11:47] popey: and it is reminders that has the same issue [11:51] sverzegnassi: pushed latest docviewer to store [11:57] popey: fix for bug 1479483 will wait a bit more for being released, but it's ok! thank you! :) [11:57] bug 1479483 in Ubuntu Document Viewer App "Docviewer available only in side stage mode on N10" [Undecided,In progress] https://launchpad.net/bugs/1479483 [11:57] thats okay, nobody uses n10 :) === _salem is now known as salem_ === alex_abreu is now known as alex-abreu [14:01] dholbach: yay for subclassing \o/ [14:01] dholbach: that all looks much cleaner now [14:01] :-) [14:02] dholbach: do you want me to leave a review in LP? [14:02] whichever way you prefer [14:02] not sure why I'm responding to you in here instead of #ubuntu-community-team :/ [14:02] I'm also happy to repropose the branch against trunk [14:03] well, it's got an approval review now, I can top-approve too if you want [14:04] mhall119, what do you think about the pages being recreated with every importer run [14:06] and do you know if dpm pinged the webops folks again about the deployment ticket? [14:07] dholbach: I got anthonyf to bump the priority on the RT yesterday [14:07] https://code.launchpad.net/~developer-ubuntu-com-dev/developer-ubuntu-com/snappy-docs-import/+merge/265004 has the complete work from davidcalle and myself [14:07] mhall119, brilliant! [14:08] I think deleting and re-creating could be dangerous, I'd prefer updating them in-place, but that would require more complexity too [14:09] ok... I think I just remember a case from the LTP from aeons ago where we had a primary key hit MAX_INT because of this practise [14:09] I'll look into fixing that next then [14:10] apart from that I'm happy where we're going - having multiple snappy branches and snapcraft in the developer site should be no problem any more now [14:10] yeah [14:12] if you have any other complaints^Wsuggestions, let me know === dholbach_ is now known as dholbach === chihchun is now known as chihchun_afk [14:34] mhall119, do you know off the top of your head if there's a cleverer way to find articles by URL other than doing something like [a.get_absolute_url() for a in cms.models.Page.objects.all()]? :-) [14:34] mhall119, if not that's fine and I'll dig deeper [14:36] dholbach: there should be [14:36] in traditional django, urlconf resolvers can do it [14:36] I don't know about django cms though, but there must be something [14:37] ok, thanks - I'll take a look [14:44] JamesTait: you know I keep moaning about https://myapps.developer.ubuntu.com/dev/click-apps/reviewer/ ? [14:45] JamesTait: well I've been told there's an app in that list which needs review. I have NO way of knowing which one [14:45] there are 19 in the list. [14:46] popey, you have not other information than "an app needs reviewing"? [14:47] *no [14:48] Not even "My app needs review" from someone identifiable? Or "FOO:app version 1.7 needs review"? [14:49] i have now, yes, because the developer told me [14:49] but it's been sat there with nobody reviewing it [14:49] because we can't tell one that needs review from one that doesn't [14:50] it shouldn't get to the point that a developer has to go round poking people to get a review, the list should show ones needing a review. [14:50] popey, I sent out an email with the details [14:50] on how, why, etc [14:50] popey, allow me to direct you to someone who can bump the priority of that. ^^ 😉 [14:50] so reviewers could clean it up [14:50] also, click reviews should be an extreme exception that usually gets rejected [14:51] I don't think that covers this problem. [14:52] there are 19 things in this list https://myapps.developer.ubuntu.com/dev/click-apps/reviewer/ [14:52] one of them needs a review, 18 do not (I believe) [14:53] it is erronesly classifying things that _once_ failed as being failed, and shows them as needing our attention [14:53] https://myapps.developer.ubuntu.com/dev/click-apps/2998/review/0.1/ for example [14:53] or does it? I can't tell. [14:53] as I explained in the email [14:54] just reject the ones that don't apply anymore [14:56] I don't know if I'm talking gibberish, but that's not the problem at hand. [15:00] beuno, maybe we can talk through an example case here? [15:00] if you look at the list of reviews, there's 0.4 of wifitransfer [15:01] while 0.5 is in the store already [15:04] mhall119, it's cms.utils.page_resolver.get_page_queryset_from_path() <3 [15:04] dholbach, yes, 0.4 needs rejecting [15:05] beuno, it looks like 0.4 had a negative auto-review [15:06] should the store have auto-rejected it? [15:07] dholbach, it was requested as a manual review [15:07] hence, on the list [15:07] the store will auto-reject soon, if a newer version is approved [15:08] ok, so for some more days reviewers will have to clean up the queue every now and then? [15:18] dholbach, there's the current queue to clean up, and then it should be pretty light touch, until it's back to what it used to be-ish [15:19] thanks a lot - it looks like auto-rejections will help [15:19] * beuno nods [15:19] and again, reviews need to be an exception [15:19] thanks. [15:19] if it's common, there's something else to fix [15:20] I'm seeing a lot of people just asking for manual review on rejection instead of addressing the issues [17:03] just saw this on Android and kinda wondering if one day we can have that in the apps scope https://play.google.com/apps/testing/com.ideashower.readitlater.pro/join [17:04] it's the same app, if already installed it'll be updated, and you get a development release [18:08] kalikiana: that link is useless if you don't have a Google account :) [18:22] JanC: sorry about that. have a look http://geraldfield.org/google-play-store-pocket-beta-opt-in.png [18:24] so I sort off guessed right what it was about (I think Valve has something similar in their Steam store?) [18:24] that would be nice indeed [18:24] *sort of* === _salem is now known as salem_ [18:25] kalikiana: I don't think "regular" users should see beta versions though [18:25] by default, I mean [18:27] JanC: yeah. the way Google does it is actually by way of joining a G+ community beforehand [18:28] so you wouldn't see it in the store out of the box [18:28] JanC: although plenty apps simply put their beta in the actual play store because they clearly disagree with having to require that extra step [18:29] well, ugh about having to join G+ etc. [18:29] in fairness if you use the playstore you already have an account anyway [18:30] but personally I see nothing wrong with just making it a button in the store [18:30] could be semi-hidden in a "hold icon"-activated screen [18:30] possibly requiring a "secret" key if developers prefer a closed beta [18:31] yeah if you have something like croudfunding or internal testing that'd be required [18:31] a voucher code maybe [18:31] and then it only shows up in the store if you have that [18:32] or if you don't want 10k people to comment on your app based on a buggy beta release :) === salem_ is now known as _salem === _salem is now known as salem_ [19:42] if I am writing a tool to be packaged in a click or a snap, should I stick to the XDG guidelines for finding data (eg. check $XDG_DATA_HOME etc) or is there some better way? [19:42] I seem to be using Python for the tool [20:11] hello I need some help for a html5 application build for armhf [20:12] it is giving me this error :-1: error: Unknown module(s) in QT: webkit [20:12] the same app for desktop is fine === salem_ is now known as _salem === _salem is now known as salem_ [20:49] bregma: yes, xdg is your best option. all paths are defined in terms of those variabes and will work across devices [20:49] cool, thanks [20:49] it's especially important to respect them as under confinement you won't be able to read/write arbitrary folders [20:50] eg. ~/.config/myappname only works if you get the correct foldername [20:51] bregma: webkit is deprecated. you should be using oxide [20:51] not sure off head where the docs for that are [20:51] somewhere on developer.ubuntu.com === salem_ is now known as _salem === _salem is now known as salem_ === _salem is now known as salem_ === salem_ is now known as _salem