[06:49] good morning [07:29] morning all === zsombi1 is now known as zsombi [08:35] Good morning all, happy Friday and happy Blood Donor Day! :-D [08:39] JamesTait, happy friday to you too [08:39] * dholbach hasn't donated blood in ages [08:40] * JamesTait neither. [08:40] I really should do something about that. [08:40] yeah [08:40] * JamesTait looks up the next donor session locally. [08:40] the last time was as a student :-P [08:42] Ah, next Sunday, less than a mile away. [08:42] That'll do. [10:08] I've been on a study donating every 8 weeks [10:08] the usual interval is 16 [10:08] but last time (maybe the 5th 8-week donation) my iron was below the level so I got denied :( [10:12] om26er: would be nice if you could review this https://code.launchpad.net/~schwann/gallery-app/gallery-autopilot-toolbar-buttons-fix/+merge/169379 [10:17] dmp: just answered on your question: http://askubuntu.com/questions/306056/how-to-detect-whether-an-app-is-running-on-a-mobile-device-or-on-the-desktop/308096 [10:32] dpm, hrm, what are the differing behaviours? [10:35] I am extremely sceptical of apps which want to know whether they're on a phone or not and act differently :) [10:36] aquarius, after a few days I've come to realize that they're mostly related to size, so when the support for different layouts lands on the SDK I think this should be sorted. Still, I'd like to open the app's window with a bigger size if it's started from the desktop, so I guess I'll still need to detect the screen size somehow [10:36] right -- that's adapting based on screen size, not on device type :) [10:37] dpm: you can set a fixed window size for desktop, on phone it automatically opens full-screen [10:37] aquarius: +1 [10:37] aha, I didn't know that! [10:37] I *think* that if you set a window size, the phone just ignores it anyway, so a window size is only desktopish [10:37] ha! t1mp is both a quicker typist and more knowledgeable than I :) [10:38] that's a very useful bit of wisdom, I'm glad I asked another question and I learnt that! [10:38] note that deciding on your initial window size on a desktop is hard by itself, unless you're prepared to either be clever, screw high-dpi users, or screw netbook users :) [10:38] screw them all [10:39] if our grid-unit size is different on high-dpi then using gu's to define the window dimensions should work [10:39] note the 'if' [10:39] t1mp, ya. the "if" is very very "if", there [10:40] since I don't think we're adapting gu for desktop devices at all atm? [10:41] aquarius: actually, we may be doing that http://developer.ubuntu.com/api/ubuntu-12.10/qml/mobile/resolution-independence.html [10:41] it is explicitly mentioned there :) [10:41] I never tested it though [10:41] I don't have a high-dpi screen and phone is priority now for the sdk [10:42] hm. I do believe that that list is guidance about what *should* be happening rather than what actually *is* happening ;) [10:42] not that that's a complaint; as you say, phone is priority, and I don't have highdpi either :) [10:42] aquarius, btw, I've been doing some more work on the QR code scanner, and I've also come to realize that the JavaScript scanning library does indeed have a performance penalty when processing hi-res images (e.g. 20s to scan a code vs 1s in the C++ plugin). So I either scale all pictures taken from the phone to a lower resolution before sending them to the JS scanning library, or use the C++ plugin -> https://lists.launchpad.net/qreator-discuss/msg00 [10:42] 088.html [10:42] yeah, so in the future, high-dpi should automatically work fine ;) [10:42] t1mp, yes :) [10:42] dpm, interesting. [10:43] dpm, me, personally, I'd downscale. Avoid C++ at all costs ;) [10:43] om26er: jnkins approved already ;) [10:43] aquarius, dpm for image processing I'd use c++ instead of JS [10:44] for performance [10:44] aquarius, on the other hand, I can consider the C++ plugin as a black box that does the magic. I'm really not interested in how the scanning is done if it works, most of the work I put in the code is in the UI/UX anyway, and that's QML [10:44] t1mp, that's because you think that performing faster is more important than easy distribution. Which is a fair thing to think. I do not think it ;) [10:44] a c++ plug-in for qml of course [10:45] t1mp, yeah, that's how I did it, with a plugin. I'm testing both options: pure JS or a separate package with a C++ plugin for qml [10:45] "branch this and run it" is so much easier than "branch this, make sure you've got the dependencies, cmake, make, make install --prefix=build, run it as qmlscene ./build/exe/whatever"... [10:45] aquarius: we *should* have an awesome packaging system that takes care of everything :) [10:45] https://launchpad.net/~qreator-hackers/+archive/qreator-experimental [10:45] t1mp, indeed we should. Given that we currently do not... ;-) [10:45] aquarius: that may be another thing that should work fine in the future :) I'm not an expert on packaging [10:46] gusch, approved. I used that approach for a core-app as well. But then I changed that to what you were doing in the gallery-app.. so I guess we really need that issue fixed in the SDK [10:46] aquarius: I don't want to tell people to "branch" this, but "go to this website and click the button" or "apt-get install my-awesome-app" [10:46] (also, I wonder if emscripten'ed and asm.js things are faster in qml's JS engine?) [10:46] om26er: thx - and yes ... [10:47] aquarius: we have a "run on device" in qtcreator that automatically compiles c++ code and installs it. [10:47] t1mp, absolutely. But then it'll be "go to this website and click the button *corresponding to your architecture*", which is dead annoying. (Agreed on "install it from the Dash" being easy, btu that's not up and running yet) [10:48] t1mp, do you have any idea why the issue mentioned in "Caveats" on https://lists.launchpad.net/qreator-discuss/msg00088.html is happening? [10:48] t1mp, the actual code is simply defining a toolbar action, not sure why it's not working: http://bazaar.launchpad.net/~dpm/qreator/touch-scanner/view/head:/qreator-touch.qml#L95 [10:50] (also, anyone who wants to hack on qreator, even if they just want to tweak the UI, or replace a couple of images, needs to be able to compile it if bits are in C++. But... I've had this argument lots before :)) [10:51] dpm: I don't know, I never saw the issue before [10:51] dpm: can you reproduce it in a simpler program? [10:52] t1mp, I can try, but probably not until the weekend [10:52] ok [10:56] dpm: [10:56] // FIXME: this property seems to have a weird effect: when set to [10:56] // true, the tab contents are always visible, also when other tabs [10:56] // are loaded [10:56] yeah, Tab currently has visible: active [10:56] //visible: root.debugMode? true : false [10:56] so if you override that, the automatic visibility of the Tab (and its page) breaks. [10:57] visible: active && root.debugMode may do what you want [10:57] hmm.. that will still have the tab listed in the header, so probably not what you want [10:58] it may be nice to use Tab.visible to determine whether the tab appears in the header [10:58] t1mp, aha, yeah, that was something else I wanted to ask you. Is there any plan to make tabs "hideable", or is it simply something I should not do? [10:59] no plans [10:59] if you can think of a use case, file a bug report to request the feature. I don't mind having that feature [10:59] whether you 'should not do' it is I think a question for design [11:00] t1mp, my use case is that I want to use the extra tab to output debug messages. Or is there an alternative to show debug messages when running on a device? [11:01] I don't know [11:01] jppiiroinen: ^ ? [11:01] bzoltan1: ^ === MacSlow is now known as MacSlow|lunch [11:02] t1mp, ah, I see there is already a bug for it: bug 1161363 [11:02] Launchpad bug 1161363 in Ubuntu UI Toolkit "[tabs] Tabs component visible property stopped working" [High,Confirmed] https://launchpad.net/bugs/1161363 [11:05] commented on that one with my use case [11:05] dpm: no, that's different. that is about the visible of Tabs, not of an individual tab [11:06] t1mp, ah, I read that he mentioned individual tabs as well, but you're right, that would deserve a separate bug [11:20] how to find whether there is some ubuntu app already exists for some purpose? [11:21] dpm: There is no output forwarding from the device to the QtC right now. It takes a little tweaking .. not impossible, but not done yet [11:21] bzoltan1, ok, thanks. I'll continue with my debug tab hack for now. [11:27] Naga, I'd recommend checking https://wiki.ubuntu.com/Touch/Collection [11:28] Naga, https://plus.google.com/communities/111350780270925540549 is also a good place to check for existing apps [11:34] dpm, the list seems to be small!! Is that updated one? https://wiki.ubuntu.com/Touch/Collection [11:35] Naga, the list is growing and up to date. With your help we can make it bigger ;) [11:35] Sure.. [11:36] am looking in ubuntu dvlpmnt for last 2 days. hopefully, let me make it bigger.. taller.. stronger... :p [11:38] :) [11:39] I saw it supports app in c++. So, can't it load the existing apps written in c++, straight away? === mzanetti is now known as mzanetti|lunch [12:06] Kaleo, ping [12:06] weekly === mzanetti|lunch is now known as mzanetti [12:30] victorp: can you please again check https://code.launchpad.net/~schwann/qtubuntu-camera/camera-torch-support/+merge/168468 ? [12:31] gusch, sure - I guess I need to flash first the latest image? [12:31] victorp: better to do so === MacSlow|lunch is now known as MacSlow === _salem is now known as salem_ [13:17] Ubuntu for arm processor (Raspberry) | http://askubuntu.com/q/308164 [13:41] boiko, Hi! [13:42] hey om26er [13:42] boiko, I fixed the tests, please review. https://code.launchpad.net/~om26er/phone-app/fix_autopilot_suite/+merge/169426 [13:42] om26er: awesome! thanks a lot! [13:43] boiko, I am skipping one test there but I'll fix that after this branch gets in... Also I will test these on the phones so you should expect a second branch afterwards. [13:44] om26er: nice! ok, I'll give it a try in a couple minutes [14:46] gusch, hey! what is really up with the gallery tests I wonder. is that happening video recording branch ? [14:47] gusch, /usr/lib/arm-linux-gnueabihf/qt5/bin/qmlscene: symbol lookup error: /usr/lib/arm-linux-gnueabihf/qt5/plugins/mediaservice/libaalcamera.so: undefined symbol: _ZN12SharedSignal8instanceEv [14:47] still not working [14:48] om26er: I did not update the autopilot tests for the changes in the code - now works [14:48] victorp_: strange - I'll test again - thx [15:05] Hi [15:08] Is there anyone? === salem_ is now known as _salem [15:24] hi nicolacorti, there is plenty of people. If you've got an actual question to ask and if someone can answer it, we'll be happy to help :) [15:26] Yes i've got [15:26] I'm looking for more information about PyGTK [15:27] but i think that http://www.pygtk.org is down right now [15:27] is it true? [15:35] nicolacorti: www.pygtk.org doesn't work for me [15:38] nicolacorti: for ubuntu app development there is a bunch of information on http://developer.ubuntu.com/, but no PyGTK [15:39] dpm: on http://developer.ubuntu.com/ under (1), quickly is mentioned, but there is no quickly on the page that it links to. [15:40] dpm: and the link to "quickly" at the bottom of the page under "resources" is broken [15:45] t1mp, thanks for the heads up, I'll fix that [15:46] Thank you anyway, i've downloaded python-pygtk2-doc package [15:48] nicolacorti, if you want to develop for Ubuntu we recommend using the Ubuntu SDK at www.developer.ubuntu.com/get-started - but if you still want to use Python + GTK you can also go to http://python-gtk-3-tutorial.readthedocs.org/en/latest/ [15:49] nicolacorti: note that PyGTK is currently not supported for Ubuntu Touch [15:52] Ok thank you [16:48] om26er: hey, just tried yoru branch, but tests are failing, it seems selecting a button of the toolbar is broken [16:48] om26er: but the tab switching is working fine [16:49] om26er: I guess the toolbar thing might be related to what gusch was telling earlier today [16:49] boiko om26er tim changed the toolbar implementation yesterday [16:50] ouch [16:50] boiko, can you pastebin the logs [16:51] om26er: http://paste.ubuntu.com/5765281/ [16:51] gusch, how did you worked around that ? [16:51] boiko, yep that's what gusch was seeing [16:53] om26er boiko my fix is here https://code.launchpad.net/~schwann/gallery-app/gallery-autopilot-toolbar-buttons-fix/+merge/169379 [16:54] I now use the select_many on a sub-tree [16:54] a new feature of autopilot since about 3 weeks [16:54] gusch: that's clever and probably more reliable [16:55] om26er: so, want me to approve that branch and fix the toolbar button problem in a separate MR? [16:55] boiko, yeah that sounds fare [16:55] om26er: ok, I'll approve your changes then [16:55] boiko gusch om26er my changes shouldn't break anything [16:56] what broke? [16:56] t1mp: autopilot tests [16:56] t1mp, though it would be really cool if we be able to assign objectNames to action items ;) [16:57] om26er: you can assign an objectName to anything right? [16:57] t1mp, yep, works for pretty much everything [16:57] ... or to get the so called helper functions to interact with UITK components in autopilot [16:58] me wants this - relly wants this ^ ;) [16:58] :) [16:58] I think jppiiroinen was working on that [16:58] was?!? [16:59] jppiiroinen: ^ ? [16:59] hmm its 8pm in finland, I guess it is eow for him [17:01] bzoltan: ^ you know stuff :) [17:01] I am here [17:01] what is the question [17:01] 18:57:33 < boiko> ... or to get the so called helper functions to interact with UITK components in autopilot [17:02] is someone working on that? [17:02] dunno [17:04] maybe Kaleo knows about the helper functions to interact with UITK components in autopilot ? [17:04] is it the "abstraction layer" mentioned in https://blueprints.launchpad.net/ubuntu/+spec/sdk-qa ? [17:11] t1mp: I think it is [17:12] om26er: I'll fix the toolbar button thing === _salem is now known as salem_ [17:20] om26er: do you have some spare time? === seb128_ is now known as seb128 [17:26] gusch, what is it about ? [17:26] boiko, cool ;) [17:27] om26er: adding a test, that calls the "edit -> auto enhance" in the full view [17:27] om26er: no need to rush - but if you are busy for a while, I'll do that [17:28] gusch, about that, its one of those tests which I remember a few months back when I was writing autopilot tests for the gallery, I didn't knew what change happened on auto enhance in the autopilot vis window [17:29] gusch, I'll do that [17:30] om26er: https://code.launchpad.net/~boiko/phone-app/fix_getting_tool_buttons/+merge/169507 [17:30] gusch: I know that a few weeks ago I gave them to jppiiroinen to implement [17:30] gusch: no more [17:30] gusch: (just after the sprint in fact) [17:30] om26er: well - the ActiviIndicator will be visible, but only for a short time (not reliable to test) [17:30] sorry [17:31] om26er: maybe test if afterwads the undo is enabled [17:32] Kaleo: let's ask jppiiroinen next week [17:32] om26er: maybe test if afterwads the undo is enabled [17:32] ouch system crashed [17:32] *unity got new bugs* [17:32] there's a reason I use KDE :p [17:32] gusch, yes sure, I am heading out now but I'll have a branch ready in a few hours [17:33] om26er: I'll check it on monday - no need to rush [17:33] it's just a nice to have [17:34] gusch, there is one more, If there is only one photo in the gallery and you open the photo and you delete it, you kind of get stuck there, the back button won't work [17:35] too much "you" :D [17:36] om26er: https://bugs.launchpad.net/gallery-app/+bug/1131802 [17:36] Ubuntu bug 1131802 in gallery-app "Deleting the last photo leaves a black gallery" [High,New] [17:37] om26er: and that's the reason I want the test https://bugs.launchpad.net/gallery-app/+bug/1180345 [17:37] Ubuntu bug 1180345 in gallery-app "Auto enhance crashes gallery app and deletes photo" [Undecided,New] [17:37] om26er: but I wasn' [17:38] om26er: but I wasn't sure if that's really teh case (but now that I see the bug...) [17:38] om26er: seems like I'll have to fix a bug ;) [17:38] oh look I in the bug report as well [17:38] *I am [17:38] gusch, yep, you got a new one [17:39] gusch, is that something to be tested with autopilot so it doesn't regress in future ? [17:40] om26er: correct [17:42] om26er: if you want to push me, then add a test for https://bugs.launchpad.net/gallery-app/+bug/1131802 as well ;) [17:42] Ubuntu bug 1131802 in gallery-app "Deleting the last photo leaves a black gallery" [High,Confirmed] [17:45] gusch, sure... === boiko_ is now known as boiko [18:11] boiko: you eagerly wanted to review? https://code.launchpad.net/~schwann/gallery-app/gallery-mediafactory-unittests/+merge/169514 [18:12] gusch: yep, I can take that one :) [20:07] Hello :D [20:08] Hey! [20:49] http://www.jonobacon.org/2013/06/14/the-ubuntu-app-developer-cookbook-announced/ More goodies for App Developers being added to developer.ubuntu.com [20:55] Do you could use python in ubuntu phone? | http://askubuntu.com/q/308332 === salem_ is now known as _salem [22:51] Is it just me or this is a mobile only channel? [23:23] where to find Qt5 downloaded pack in my 13.04 | http://askubuntu.com/q/308373