[01:04] beuno, how long until my app shows up under more suggestions? :) === mh0 is now known as away0 [01:52] nik90, if you see this, I believe the last 2 test errors should be fixed; https://code.launchpad.net/~nskaggs/ubuntu-clock-app/fix-1220998/+merge/184218 [02:49] How to create my own graphical desktop for my Ubuntu desktop? | http://askubuntu.com/q/342004 [06:32] morning [06:39] good morning [06:54] тут по советски общаются? [07:42] dholbach: hi [07:42] dholbach: I wanted to upload a click package yesterday night, but failed badly :D [08:00] popey, dpm: I have a temporary fix for https://bugs.launchpad.net/bugs/1191291, https://bugs.launchpad.net/bugs/1201830 and https://bugs.launchpad.net/bugs/1201858. 3 bugs (high priority) can be fixed with one solution. [08:00] Launchpad bug 1191291 in Ubuntu Clock App "Clock app does not honor timezone" [High,Confirmed] [08:00] Launchpad bug 1201830 in Ubuntu Clock App "Clocks need to sync better and use the same backend" [High,Triaged] [08:00] Launchpad bug 1201858 in Ubuntu Clock App "Clock reports time in UTC by default." [High,Triaged] [08:00] oh wow! [08:01] popey, dpm: I can make the clock app default to the system timezone thereby honouring the timezone of the system. If the user changes the system timezone then it will also be reflected in the clock app. It comes at a cost however (temporary) [08:01] nik90|Office, what's the solution you were thinking of? [08:01] ok, I wasn't too quick in asking :) [08:01] popey, dpm: However I *remove* the ability of the user to set a world clock as their current location [08:02] dpm: A user can add a new world city to see the time at that location, but they wouldnt be able to click on a world city and set that as their current location as per design spec [08:02] dpm: Would that be okay? [08:03] dpm: that feature need the clock app to read the /etc/timezone file which cannot be done without a qml plugin. I have already written a qml plugin but I dont think we have the time to integrate the plugin into the sdk or into a click package. [08:04] nik90|Office, ok, gotcha. Sounds good to me as a temporary measure. Can you show me where the WIP plugin lives if you've got it in a VCS, and the current status? Perhaps we can get someone to help [08:05] what I recall from previous conversations was that the plugin was quite small [08:05] dpm: The WIP plugin in hosted at https://launchpad.net/file-qml-plugin. It was written by me and Chocanto [08:06] dpm: it is a standard file read plugin which provides the capability to read any file. [08:06] nik90|Office, thanks, looking. And what it is supposed to do in the context of the clock app? [08:06] dpm: on talking to the security team, as long as I do not write into any file, the clock app will have permission to read a file. [08:07] dpm: using that plugin, I can read the /etc/timezone to detect timezone changes which I need desparetely [08:08] dpm: the problem with the clock app now is that when the user sets a world clock as his current location, when the system timezone changes, the clock app wouldnt know about it. Thereby it does not honour the timezone [08:08] dpm: causing those 3 bugs I stated above [08:09] nik90|Office, you've been talking to them, I think, do you know how the settings app guys read/write the file? [08:09] Laney, perhaps you know? (good morning!) ^^ [08:09] hello [08:09] we use timedated [08:09] dpm: the setting app is written using C++ which has native file read capability [08:09] over dbus [08:10] Laney: good morning. Can we use timedated over dbus in QML and Javascript? [08:10] I doubt that unfortunately [08:10] dpm: there you go ^^^ [08:11] Not sure if it would work with confinement either [08:11] even if you did use some C++ [08:11] Laney: the music app uses the nemo folder plugin which access folders and files using a c++ plugin. [08:12] Laney: I am guessing exceptions can be added to the default core apps confinement rules [08:12] Laney, forgetting confinement for a sec, what do you think we'd need to access the timezone from QML? Is it only DBUS QML bindings we're missing? [08:12] heh, I don't know if that's an "only" [08:13] but if you had that then it's just a property on the interface: http://www.freedesktop.org/wiki/Software/systemd/timedated/ [08:14] Laney: damn that's exactly what I need :D [08:14] well... the code is there in system-settings :-) [08:14] Laney: true but useless for me since clock app is all qml and javascript. I am trying to get a c++ plugin into it. [08:15] mzanetti, what happened? [08:15] mzanetti, did you file a bug? [08:15] Laney, heh, "only" in the context that we wouldn't need anything else. I've no idea whether we've got dbus qml bindings already or how difficult it'd be to make them :) [08:15] dholbach: no... not sure if its just me [08:15] nik90|Office: it's /currently/ only that but I'm guessing that isn't a fundamental restriction [08:15] dholbach: so I uploaded it here: https://myapps.developer.ubuntu.com [08:16] mzanetti, try http://developer.ubuntu.com/publish/ [08:16] dholbach: at the end it says, "Thanks you for uploading... Unfortunately we cannot process your request right now" [08:16] mzanetti, it seems you uploaded to the old store, try the link dholbach gave you, it points you to the software store beta [08:16] as a bonus you can get notified when the timezone changes [08:16] dpm: can you ask around and then let me know if I should go ahead with my temporary solution. The immediate benefits would be that the clock app shows the correct system time. The only downside is that a user cannot click a world city to make it as his default location. [08:16] like if someone sets it from u-s-s while the clock is open [08:17] Laney, do you think we could use your code and make it into a plugin? [08:18] like its own package? [08:18] Laney: yes [08:18] yeah, for the clock app [08:18] dholbach: the next thing is: I'm not really sure what to put in the .desktop file for Exec and Icon [08:18] if you bind all of the properties and make it nice then sure why not [08:18] generic dbus bindings would be more ideal though [08:19] not that I expect anyone to work on that [08:19] nik90|Office, I think this might be the way to go long-term, rather than writing a custom plugin. I like the fact that it'd get notifications when the system timezone changes [08:19] dpm: +1 ^^ [08:20] well, long-long-term then dbus bindings, yes, but from the comments I understand this is nothing trivial [08:20] and no-one is working on it [08:20] mzanetti, shouldn't the sdk do that for you? [08:20] and i doubt anybody has time to work on it at this stage [08:20] dpm, ^ do you know about mzanetti's question? [08:21] yes [08:21] brb, need to walk the dog [08:22] mzanetti, for Exec, you can use "qmlscene $@ yourmainfile.qml". In any case, when you create the click package with Qt Creator, it will replace it and make sure it's got the right content [08:22] mzanetti, for Icon, relative path to the icon. I.e. if the icon lives at the top of the source tree, just the name of the file will do [08:23] Laney, where does your timezone reading code live? [08:24] lp:ubuntu-system-settings plugins/time-date/ [08:25] dpm: no qmlscene here [08:25] hello. I'm the developer of RamSamSam Reader. I wanted to ask, if someone could check the design of the application. What things should I change to make it better comply with the ubuntu design guidelines. [08:25] thanks Laney! [08:26] np [08:26] dpm: ah, so the working dir when executing is the root of the install dir? [08:26] dpm: so having all paths relative to the working dir should do... [08:27] mzanetti, yes, click packages are installed in /opt/click.ubuntu.com/com.$YOURDEVID.$YOURAPP/current. They are essentially self-contained there [08:27] dpm, nik90|Office: didn't we have that discussion like a month ago? [08:27] seb128, I vaguely remember, but I wasn't really involved in it [08:28] hi DanielBeck_, I don't think there's any designer here yet, but I can try to ping one when they come online [08:29] mehow might be a good person to ask for the visual design when he comes up [08:29] dpm: thanks. [08:29] dpm, nik90|Office, Laney: wasn't https://lists.launchpad.net/ubuntu-phone/msg03701.html tthe same topic? [08:30] seb128: we did but arrived at no solution viable yet :( [08:31] seb128, it seems all the replies address confinement only [08:31] dpm, what else needs addressing? [08:32] dpm, you want an sdk api to change system time, is that really something we expect many apps to be able to do? [08:32] seb128: I cannot used timedated that the system setting app uses [08:32] seb128, how to actually read the system timezone from the clock app's code [08:32] seb128: no I just need to *read* the system timezone ID [08:32] seb128: and also be notificed when the system timezone changes [08:33] seb128, it'd be nice to have an API, but I don't think we've got the time. I'm just wondering if the clock app could just reuse the code from the settings app for now [08:33] dpm, it can if it's not confined, which goes back to that discussion... [08:33] you are stealing "our" code for search, so it's just fair that we steal "your" code for timezone :) [08:34] dpm, nik90|Office: can't you use qtlocation to determine the current location? [08:34] seb128: we are talking about the default clock app for ubuntu touch. I am still not able to appreciate why it is being confined [08:34] nik90|Office, I'm not sure it is/needs to be, that's my point [08:35] seb128: Even if I did use qtlocation (which btw hasnt landed yet), I would only know the location and not the timezone ID of the system [08:35] seb128, yeah, we'd need to double-check with the security team indeed, but from jdstrand's reply, it seems to me it should be ok to do this for the clock app [08:35] nik90|Office, why do you mean "not landed yet"? https://launchpad.net/ubuntu/+source/qtlocation-opensource-src/5.0~git20130805-0ubuntu3 [08:35] nik90|Office, btw, good news: Qt Location is about to land (if it hasn'T already) [08:35] that's in main and using the ubuntu location service [08:35] what else needs landing? [08:35] nik90|Office, there you go :) ^^ [08:36] seb128, I don't think anything else needs landing, I think it is just so new not many folks know it's already there [08:36] dpm, we discussed confinement a month ago, while they prefer to confine things that can be (which makes sense) I think they said it was ok to run default app unconfined in v1 [08:36] seb128, ack [08:37] dpm, e.g there is no issue to get clocks unconfined from what I know [08:37] seb128: I was looking at https://code.launchpad.net/~thomas-voss/platform-api/add-location-service-api/+merge/169778 and https://code.launchpad.net/~ricmm/platform-api/add-location-service-api-take-2/+merge/177468 [08:37] but nobody wrote "I need access to dbus' service" in https://lists.launchpad.net/ubuntu-phone/msg03701.html [08:37] the only thing mentioned in that email is tzdata [08:37] and Jamie replied pointing you can read that file already [08:38] Not sure they knew about timedated before now [08:38] you should maybe follow up saying you need to use the dbus timedated [08:38] seb128: I knew about dbus timedated since this morning from Laney [08:38] Laney, they did, I discussed it with the same people on IRC of an hour a month ago :p [08:38] dpm: hmm... if I launch my app installed from a click package it just stays white. it works when I run the installed binary via the command line. any ideas how to debug? i.e. are there any logs? [08:38] heh [08:38] seb128, I'll check when the security guys when they come up online later on [08:38] nik90|Office, SHRUG, we discussed it a month ago [08:38] seb128: from what I knew, I thought that the timezone info cna only be accessed through tzdata [08:39] seb128: you sure I was part of that discussion? [08:39] I wasn't [08:40] mzanetti, how do you run it from the command line? [08:40] ok anyways first things first, seb128 do you happen to know if there are qml API docs for qtlocation. I will start using it if it has some. [08:40] nik90|Office, dpm: thanks for making me have to dig IRC logs to prove myself: [08:40] http://irclogs.ubuntu.com/2013/08/09/%23ubuntu-touch.html#t14:26 [08:40] was the discussion [08:41] nik90|Office, dpm: http://irclogs.ubuntu.com/2013/08/09/%23ubuntu-touch.html#t14:37 was you being pinged directly [08:41] dpm: ./binaryname --desktop_file_hint=`pwd`/appname.desktop [08:41] dpm: I assume running it from the shell still appends the desktop_file_hint stuff [08:42] secondly I will send a reply to that email conversation about the need to use dbus timedated to get the system timezoneid [08:42] seb128: looking at http://irclogs.ubuntu.com/2013/08/09/%23ubuntu-touch.html#t14:26, I wasnt part of it :) [08:43] seb128, still, I didn't discuss about timedated + dbus, I was just asked why we need world clock [08:43] * dpm hugs seb128 (and hides) [08:43] dpm, "jdstrand, hey, security question: are (confined) click apps going to be able to call dbus functions (I guess not)? specifically asking about getting the timezone through datetimed over dbus" [08:43] I see that I mentioned it in the comments on that linked post [08:43] * Laney goes off to fix dgettext instead :P [08:43] dpm, the "specifically asking about getting the timezone through datetimed over dbus" was for you guys, because you pinged me there and I moved to -touch to include the security guys [08:43] shit, you can't fool seb128 [08:43] lol [08:43] lol [08:44] :) [08:44] * seb128 hugs dpm back [08:44] don't worry, I won't remember this the next time [08:44] * Laney sieve [08:44] same here :) [08:44] dpm, that's fine but please don't call me a liar saying that conversation didn't happen :p [08:44] I'll ask again in a month [08:44] oh, I didn't mean that, I just didn't remember I was part of it [08:44] k [08:45] well anyway [08:45] back to the drawing board [08:45] nik90|Office, yes, please follow up on the list saying you need dbus [08:46] and yes, qtlocation should have qml bindings [08:46] but check with tvoss maybe to be sure [08:47] nik90|Office, the backend got merged in https://code.launchpad.net/~thomas-voss/platform-api/add-location-service-api-take-3/+merge/178667 [08:48] nik90|Office, as per the original question, I think it's more important that the clock is in sync with the system time than the ability to change the city for now (given that we can't change the timezone in the clock) [08:49] seb128: sure I will take a look and check with tvoss. Thanks [08:50] dpm: agreed. In which case I am going to start on the temporary solution once I finish the alarms. [08:50] nik90|Office, so I'd say +1 on your temporary solution to fix the bugs. To be sure I understand, though. You said: "If the user changes the system timezone then it will also be reflected in the clock app" - how will you actually do this in QML with your solution? [08:51] dpm: everytime the user opens the clock app (cold start), the clock app will obtain the system time using qtdate. Thereby it will always stay in sync with system time [08:52] mzanetti, let me find out the command line to run the app confined, that might give you more output [08:52] dpm: also now that qtlocation has landed, I can show that as the current location of the user [08:52] dpm, nik90|Office: you might be able to read /etc/timezone without having to use dbus btw [08:52] if you just need the current tz [08:52] dpm: I just managed to run it [08:52] seb128: I agree +1 which is also what I suggested to dpm [08:52] I'm not even sure reading that is something blocked by the confinement [08:52] dpm: seems only qmlscene is supported? [08:53] seb128: however it will not provide notifications when the timezone id changes on the fly. [08:53] seb128, yeah, we discussed that, there is a plugin to read files already, and we were talking about whether just to use that or doing what the settings app does [08:53] nik90|Office, you can probably put a file monitor on that file to get notifications on when it changes [08:53] seb128: what dpm said ^^ [08:53] k [08:55] mzanetti, you might need to ask jdstrand or mdeslaur when they come online. What I did to run (QML) apps confined was: [08:55] aa-exec-click -p com.ubuntu.developer.dgalg.wordchain_wordchain_0.3 -- qmlscene $@ wordchain.qml [08:55] taking the Wordchain app as an example [08:55] I'm not sure about the support for binaries, but it is something we need to find out [08:57] dpm: ok... with this app I'm actually fine with using qmlscene as the main.cpp didn't do anything. the logic is in a plugin [08:57] nik90|Office, on another subject, regarding Qt Location docs, apart from the conversation I had with Zoltan this morning to get an example out there, the only thing I've found is http://qt.developpez.com/doc/5.0-snapshot/location-positioning-qml/ [08:57] mzanetti, nice [08:57] dpm: but my other apps were ported from qt 4.7 and still have lots of needed stuff in the "main.cpp" [08:58] dpm: fancy testing my package? [08:58] mzanetti, ack. Let's have a chat with the security guys later on, this is important. Or perhaps dholbach knows something about running confined binaries in click? [08:58] mzanetti, sure. Where is it? [08:59] dpm: ubuntu-fitbit-app (you won't see much unless you have a fitbit account tho) [08:59] dpm: need to upload it somewhere. gimme a minute [08:59] mzanetti, ok, cool === chriadam is now known as chriadam|away [09:01] dpm: http://notyetthere.org/data/com.ubuntu.developer.mzanetti.ubuntu-fitbit-app_0.1_all.click [09:01] oh... I see the _all is actually wrong... it should be _armhf.click [09:01] dpm: thanks. I will try it later on desktop and phone to see if the flickr example on that website works [09:02] mzanetti, not sure we've got architecture-dependent click packages yet. Is the plugin in there compiled for arm? [09:02] if so, I think that should work [09:03] dpm: yes... it works... but one can install it on desktop and it wont work there [09:03] mzanetti, ah, does your .desktop file have a qmlscene -I path_to_your_plugin argument? [09:04] dpm: yes [09:04] dpm: well, its just "-I ." [09:04] dpm: as the plugin is shipped with the click package in the same directory [09:04] cool. does it have a qmldir file as well? [09:04] dpm: yes [09:04] ok, so I think it should all work [09:04] * dpm downloads the click package [09:05] dpm: yes... except that I just noticed that the icon doesn't show up in the dash :/ [09:05] is the icon installed in .local/ on your phone? [09:06] and have you killed unity8? [09:06] dpm: I've rebooted [09:06] that should have definitely killed it with fire :) [09:07] * mzanetti thinks so too [09:07] dpm: where should the icon be? [09:08] dpm: its in /opt/click.ubuntu.com/com.ubuntu.developer.mzanetti.ubuntu-fitbit-app/0.1/ubuntu-fitbit-app.svg [09:10] dpm, that's a question for Jamie or Colin I'm afraid [09:21] Good morning [09:24] xnox, about the tumblr upload - is net.launchpad what we use? [09:24] I might have to check what mhall119 used for the G+ webapp [09:27] mzanetti, let me have a look where the other icons are, but I could get your app to run :) [09:27] Although I can only see the login page (I've got no fitbit account) [09:28] mzanetti, I've noticed that to get apps to load from the command line I need to have another app running in the foreground, but I don't actually know why [09:29] Here's what I ran to start it: [09:29] PATH=$PATH:/usr/sbin aa-exec-click -p com.ubuntu.developer.mzanetti.ubuntu-fitbit-app_ubuntu-fitbit-app_0.1 -- qmlscene -I . $@ qml/main.qml [09:29] I'm told the PATH is no longer necessary, but I still need to reflash my phone [09:30] dpm: so this is what is executed when clicking on an app in the dash? [09:32] mzanetti, that's my understanding, yes, we got that from the security guys as the instructions on how to do the initial manual reviews until everything is automated [09:34] dpm: any ideas what's wrong with the icon? [09:35] I don't know, sorry [09:35] cjwatson might [09:35] or perhaps some unity8 guys on #ubuntu-touch or #ubuntu-unity [09:37] Good morning all, happy Friday, and happy Fight Procrastination Day! :-D [09:40] dpm: doc viewer is missing from the build and oddly when I install it, it doesn't show up [09:41] popey, oh, I hadn't noticed it. I know there were some dependency problems with poppler in the past. Perhaps they weren't resolved? [09:41] kinda screwed me up when making screenshots for hackday blog post [09:42] bummer [09:42] * dpm looks at QA dashboard [09:43] yeah, it doesn't even appear here: http://reports.qa.ubuntu.com/smokeng/saucy/image/4012/ [09:44] * dpm tries to run it on the desktop [09:45] What is the best way to learn QML if i don't already know C or javascript? [09:45] popey, it seems it doesn't run: [09:45] $ qmlscene ubuntu-docviewer-app.qml [09:45] file:///tmp/ubuntu-docviewer-app/ubuntu-docviewer-app.qml:38 ToolbarActions is not a type [09:45] gah [09:45] it needs updating to the newest Toolbar API [09:45] And Actions API [09:45] * dpm has a go at fixing it [09:46] Uh-oh! dpm is doing to try some Science right here! [09:46] :) [09:52] that fix is easy enough, but I'm on my 13.04 laptop and I'm not sure there is a poppler version for 13.04 [09:58] dpm: can you chuck me the fix and I can test here? [09:58] (and get a screenshot for my post) [09:58] if you have a branch I can pull / build / push / run :D [09:59] popey, sure, on it [10:00] popey, lp:~dpm/ubuntu-docviewer-app/toolbar-fix [10:00] the app will run, but I think to get it to do something useful, it needs to have the poppler plugin installed, which oddly enough we don't list on the package dependencies [10:06] ok [10:09] dpm: it doesn't show up at all [10:18] popey, hm, not sure why. I can get it to run on the desktop, though. The thing is, I've not really been following docviewer closely, and I don't know where the org.docviewer.poppler plugin lives [10:21] hmm [10:23] perhaps qtdeclarative5-poppler-qml-plugin [10:23] I've noticed the docviewer dev is working on a new branch since yesterday, I'll merge it to my branch. In the meantime, still searching where that plugin lives [10:25] ah, it's on the core apps PPA! [10:25] of course ☻ [10:26] but not for raring, a bad time to for my power supply to die on my saucy desktop :/ [10:28] I've installed a click I made, but the icon doesn't show up. What could be wrong? Does it have to be name after $appname or last portion of $appid? [10:29] xnox, mzanetti and I were having the same problem with icons, not sure where they are supposed to go. It worked a couple of days ago for my app. [10:30] dpm: or just icon cache not update or some such. Ok, so it's not just me. E.g. clicks that are installed from the lens, do show icons. [10:31] popey, can you try to pull the last revision from my docviewer branch and see if that helps? [10:32] ok [10:37] dpm: still doesn't show up in the dash .. [10:38] hmm, no desktop file [10:38] dpkg-query: no packages found matching ubuntu-docviewer-app [10:38] uh [10:38] ah, that might be a hint :/ [10:39] popey, how are you installing it/running it? [10:39] * popey checks his mad script [10:40] and does the poppler plugin get installed as well? [10:40] balls [10:40] dpkg: error processing /tmp/tmp.QpdOphy8up/ubuntu-docviewer-app_0.1.1_amd64.deb (--install): package architecture (amd64) does not match system (armhf) [10:40] because it's not a qml clean package? [10:41] * popey builds on the device === MacSlow is now known as MacSlow|lunch [10:43] weird, the package should be QML, the arch-dependent package is the plugin, which is a separate one [10:43] hm [10:43] I am just branching your code, and then doing a debuild -uc -us [10:44] g++ -c -m64 -pipe -g -O2 -fstack-protector --param=ssp-buffer-size=4 -Wformat -Werror=format-security -D_FORTIFY_SOURCE=2 -Wall -W -D_REENTRANT -fPIE -DQT_NO_DEBUG -DQT_QUICK_LIB -DQT_QML_LIB -DQT_WIDGETS_LIB -DQT_NETWORK_LIB -DQT_GUI_LIB -DQT_CORE_LIB -I/usr/share/qt5/mkspecs/linux-g++-64 -I. -I/usr/include/qt5 -I/usr/include/qt5/QtQuick -I/usr/include/qt5/QtQml -I/usr/include/qt5/QtWidgets -I/usr/include/qt5/QtNetwork -I/usr/include/qt5/QtGui - [10:44] not QML [10:45] http://paste.ubuntu.com/6069891/ build log [10:45] (on my desktop) [10:47] dpm: managed to do some screenshots: http://wstaw.org/m/2013/09/06/plasma-desktopWa2323.png http://wstaw.org/m/2013/09/06/plasma-desktopkr2323.png [10:54] dpm: ok, got it installed, and it appears in the menu, and starts but I just get a white screen. So suspect it's missing something.. probably poppler [10:55] mhall119: hi, FYI I've prepared a very simple example of how to write touch apps with python: http://people.ubuntu.com/~mitya57/pyqmldemo.{py,qml} [10:55] libpoppler-qt5-1 libpoppler43 did get installed... [10:55] maybe it'll make sense to put it to some place on developer.u.c [10:55] I'll have another look after lunch [11:03] nik90_, ping [11:04] PaoloRotolo: pong [11:05] wow, nice one mzanetti [11:06] nik90|Office, Hi, I tried to use a "Border Image" with an pink .svg rectangle. It works, but when you press the button the button the rectangle doesn't change the color... [11:06] I think I can't do anything... [11:06] popey, ah, yes, I missed the .pro file in there. ok [11:08] PaoloRotolo: I havent used borderImage before [11:09] PaoloRotolo: can you try finding some app code examples which use them to check if you are doing it correct/ [11:12] nik90|Office, it's only an image on the edge of the button [11:13] Unfortunately, it doesn't change color when the button is pressed [11:13] PaoloRotolo: http://www.qt.gitorious.org/qt/ritts_qt/source/f5e3e21e95275a8cf31cddf2063dfa497e92872e:examples/declarative/border-image/content/MyBorderImage.qml [11:13] PaoloRotolo: might help. not sure [11:15] nik90|Office, that BorderImage is used on an Item. I can add an item instead of a button: in this case the item would not change color/size if you press it... [11:16] PaoloRotolo: are you adding a BorderImage to add the shadow effects? [11:17] PaoloRotolo: can you post a screenshot as it is right now and also the one where you removed the anchors and reduced with the width of the button. I want to see both of them to make a decision if this is all worth changing a button into an item [11:18] nik90|Office, no, I'm adding a BorderImage to fix the rounded corners. The BorderImage is a simple pink rectangle designed with Inkscape [11:19] it should be superimposed on the rounded corners [11:20] PaoloRotolo: alrite try it with an item instead of a button [11:20] nik90|Office, ok [11:28] nik90|Office, http://img42.imageshack.us/img42/2389/xjyu.png [11:29] PaoloRotolo: try anchors.leftMargin: -units.gu(1) [11:29] also set height: search_field.height [11:30] PaoloRotolo: also in the pink rectangle you created in inkscape, add a shadow to it on the left [11:37] nik90|Office, http://img7.imageshack.us/img7/7715/4k3n.png [11:40] nik90|Office, so, no animation when you press the button [11:45] PaoloRotolo: why did the image go to the right? I want it to overlap over the search field [11:45] PaoloRotolo: so that it appears like the design spec [11:45] PaoloRotolo: also it is okay that there are no animations. THe search is instantaneous anyway [11:49] nik90|Office, the search field is transparent [11:50] nik90|Office, ops, I thought that it went under the search field. [11:51] nik90|Office, http://img46.imageshack.us/img46/6842/qzbk.png [11:52] PaoloRotolo: can you increase the height of the image to be the same height of the textfield [11:52] PaoloRotolo: also add shadow in your inkscape image [11:52] PaoloRotolo: but it is getting there [11:53] nik90|Office: hi. we have identified the problem with the HUD when running apps from QtC and it will be fixed ASAP [11:54] Wellark: thnx a lot! [12:00] nik90|Office, I don't know why but UbuntuSDK doesn't accept the .svg file with the shadow [12:00] I added a shadow but the icon in Ubuntu SDK is not updated... [12:01] PaoloRotolo: hmm that's really weird [12:01] PaoloRotolo: did you replace the icon properly? [12:02] nik90|Office, I tried to change the color/the position of the icon and it works [12:02] but when I use an Inkscape filter the icon isn't updated [12:02] PaoloRotolo: May be it doesnt detect the filter [12:03] nik90|Office, yep... I'm adding the shadow manually [12:03] PaoloRotolo: how about a dark gray line on the left side? [12:06] nik90|Office, http://img829.imageshack.us/img829/2996/wqwp.png [12:11] PaoloRotolo: perfect! [12:11] PaoloRotolo: now try searchingn a city and see if it integrates well [12:11] with the activity indicator [12:12] nik90|Office, yes: http://img853.imageshack.us/img853/9192/4nxw.png BTW, why London xD? [12:21] PaoloRotolo: lol no idea [12:21] PaoloRotolo: but it looks really nice. Nice work! [12:22] nik90|Office, thanks. Now I've only to set the action "onClicked" [12:22] nik90|Office, I've to lunch now, bye :) [12:25] dpm, popey: http://img829.imageshack.us/img829/2996/wqwp.png [12:25] handy work by Paolo === _salem is now known as salem_ === gusch is now known as gusch|brb === MacSlow|lunch is now known as MacSlow [12:56] seb128, I'd like to test qt location with a small QML example using GPS. Does the settings app already enables gps? === gusch|brb is now known as gusch [12:58] nik90|Office, nice! [13:06] dpm, try asking tvoos [13:06] tvoss [13:08] hm it seems he just left [13:09] dpm, I think he said he has a swap day today [13:10] yeah [13:13] seb128, but do the switches on the settings app enable anything at all? If you don't know off the top of your head, I can check the code, no worries [13:14] dpm, what switch? [13:14] dpm, we disabled most of the UI elements that have no backend [13:14] e.g they are there but you can't toggle [13:16] oh [13:16] I guess that answers the question, then :) [13:16] dpm, some work, like online queries for the dash, enable/disable wifi [13:17] dpm, what did you want to know exactly? [13:17] basically I want to write a very simple qml app that gets the position from GPS [13:17] to test that the Qt Location works [13:18] as bzoltan mentioned it's not working on the example he wrote [13:18] so I want to make sure the sensor is enabled at all [13:18] or perhaps I can enable it via Qt Location directly [13:18] we don't have gps in system-settings [13:19] it didn't work when I tried it [13:19] yeah, I just found out, it's moved to the indicators now it seems, at least on today's --pending image [13:19] I was complaining on ubuntu-phone at the time [13:19] might have changed now === oSoMoN_ is now known as oSoMoN [13:37] dpm: file:///usr/share/ubuntu-docviewer-app/ubuntu-docviewer-app.qml:38 ToolbarActions is not a type [13:40] dpm: take a look at http://bazaar.launchpad.net/~nik90/ubuntu-clock-app/add-hud-support/view/head:/clock/ClockPage.qml [13:41] dpm: I added the toolbar actions after talking to Wellark and bfiller about it. [13:41] dpm: for toolbar actions reference btw [13:48] popey, weird, that's the one bug I fixed originally in my branch [14:07] Is there a way to apply a custom stylesheet to a webview? [14:20] is there some api to make notifications? [14:20] MacSlow, ^ [14:21] cwayne_, libnotify... and there are examples too (see lp:unity-notifications in the examples directory) [14:27] hi, has anyone had an issues with the keyboard not showing when a "TextField" is selected by the user? [14:32] bobws: I have it sometimes when I launch an app through qtcreator [14:41] bobws: do you have the issue when you try it through qtcreator? [14:41] bobws: if yes we have a common issue and can file a bug report [14:57] nik90|Office: Yup, launching the app via qtcreator [14:57] glad im not the only one [14:57] bobws: me too :) [14:57] is it true that bindings aren't evaluated for pages not on top of the pagestack? [14:58] bobws: lets create bug report :), shall I? [15:00] bobws: https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1221776 [15:00] Launchpad bug 1221776 in Ubuntu UI Toolkit "OSK does not appear when an application is launched on the phone through qtcreator" [Undecided,New] [15:00] bobws: can you confirm it pls [15:00] popey: ^^ [15:01] hm [15:02] nik90|Office: lemme try [15:02] hi [15:02] nik90|Office: have to walk away from the computer for a sec [15:02] how can i use native QT or GTK code in my cordova app? [15:02] bobws: np [15:07] nik90|Office: confirmed [15:08] popey: thnx [15:09] popey: do you have a script to delete all manually installed apps? [15:09] popey: correction, "any magic scripts" :) [15:13] nik90|Office: hah, no, I would just re-flash usually [15:13] cleanup = wipe [15:13] will consider it though [15:13] anyone with a good packaging skills https://launchpadlibrarian.net/149507572/buildlog.txt.gz ? [15:13] popey: ah okay :). btw when phablet-flash --wipe is what I need rite? [15:14] popey: it will get the newest image and remove any installed apps as well [15:14] not completely [15:15] phablet-flash cdimage-touch --wipe [15:15] or [15:15] phablet-flash ubuntu-system --no-backup [15:16] oh yeah I was saying it from memory..I also type cdimage-touch [15:17] dpm: btw, regarding your qtlocation example testing, shouldnt it also work without gps using wifi [15:17] dpm: using geoIP? [15:18] nik90|Office, I guess in theory, but I don't know which backends it supports. I thought GPS would be the most obvious, but you're right === matzipan is now known as Matzipan === Matzipan is now known as matzipan [15:48] xnox, sneak attack of apps! [15:48] beuno: =))))) [15:49] beuno: that's the whole point of click-webapps, I'm gonna blog about it and crowd-source usable mobile websites ;-) [15:49] beuno: e.g. not many people know about m.xkcd.com for example which also has touch-friendly way to look at img alt text ;-) [15:49] which is like super important for xkcd followers. [15:50] indeed [15:50] beuno: well it's only up to click-app id 40, so it can't be that many yet =) [15:52] nik90, nik90_, nik90|Office ping [15:52] iBelieve: lol, pong [15:52] btw its only nik90_ or nik90|Office [15:53] nik90|Office, I see in the App Showdown list that one of the apps has Click packages marked as Available. Is that field being used for apps once they're in the app store? [15:53] iBelieve: yes :) [15:54] I should rename that column name [15:54] iBelieve: Do you mind doing that. Rename column to "App Store" [15:54] iBelieve: I am leaving work to go home. Will talk later [15:54] nik90|Office, sure, I can do that, and mark some of the apps as available? [15:54] nik90|Office, bye, thanks for the info [15:55] nik90|Office, sure, I can do that, and mark some of the apps as available (wasn't supposed to be a question) [16:03] so whos gonna port this to utouch? [16:04] http://imgur.com/Iah20Ei === lordsame is now known as SergioMeneses [16:25] Hi all -- I've submitted feature request #1220533 (with patch!) to ubuntu-ui-toolkit. I know y'all must be busy, but could someone take a gander at it? It's nearly trivial, so it shouldn't take more than two minutes to accept or reject. Thanks. [16:26] rschroll, ooh, nice [16:26] rschroll, can you do one similar to let us access the color of a progress bar? :D [16:28] cwayne_: Probably. If I'm understanding things correctly, it's rather trivial. But I'd like someone who understands QML better to check my understanding. [16:28] mhall119, any chance you could take a look at the above patch? [16:29] rschroll, having the ability to change color of progress bars would be an enormous help for my app :) [16:32] It'd be a little different, since the progress bar is a StyledItem. But I've been playing with the same sort of idea with sliders, and have something (sort of) working there. [16:35] The hud seems to show keywords instead of description if both are present. That's not good [16:44] why i always should add `pkg-config --cflags --libs gtk+-2.0` to compile a gtk app ? [16:45] please [16:46] * rschroll wanders away from the computer for a bit, but will read responses on return [16:46] an you answer me ? [17:12] hi @ll, has anybody managed to run the sdk with native oxygen style under kde? [17:17] mistty, I used to run it in KDE, I don't remember if it worked or not, but I think the theme wasn't native. [17:18] mistty, I think the reason is that the SDK is Qt 5, and Oxygen only themes Qt4/KDE apps [17:18] thx [17:22] but it's really weird, because i have harmattan and sailfish sdk installed in my home directory. and the harmattan sdk is qt4 but doenst run with oxygen either (even with -style parameter added) and remains plastiq [17:22] is the theme "build-in" so that i have to recompile the sdk? [17:23] Listmodel in QML: How to retain Listmodel values on page switch in page stack | http://askubuntu.com/q/342262 [17:24] rschroll: it would be even better if you could push your branch to launchpad and file a merge request. do the commit with "bzr commit --fixes lp:1220533" and your branch will be linked to your bug report [17:29] rschroll: cwayne_: best to ping bzoltan about changes to the ui-toolkit, [17:29] rschroll: thanks for the patch! [17:38] mzanetti: I'm aware, but that would mean downloading bzr and downloading the source and learning how to use bzr. If you want this for the merge, I'll do it. But to just check whether this is a step in the right direction, a simple patch was a lot less work for me. [17:38] mhall119: Thanks; will do. [17:39] rschroll: no problem. I assumed you have the source already as you have a patch... [17:39] I have the deb, but not the bzr repository. [17:40] rschroll: yeah... I forgot for a moment that we don't compile stuff any more nowadays [17:41] bzoltan: If you get a chance, could you take a look at #1220533 ? If it seems reasonable, I can make it into a proper bzr branch. [17:43] rschroll: I'm not the one to decide, but I'm afraid this is not the right way. if you want to change the style, you should make it themeable, which is a bit more efforts I thing. But just exposing internas of a component for this purpose probably won't get accepted [17:46] mzanetti: Noted. Is this related to StyledItems? They completely confuse me. [17:46] rschroll: yes [17:47] why does gallery-app set a background as an image instead of using the theme? [17:48] rschroll: so basically what happens is that you split the item into 2 parts. one only holds the API and the other the actual implementation [17:48] rschroll: the one with the implementation is the theme [17:48] so with a different look they can still keep the same API [17:49] rschroll: so your project would provide a theme component OptionSelectorDelegateStyle.qml which has a label with a different color in there and gets loaded by the sdk [17:50] but I agree this can be confusing in the beginning [17:50] mzanetti, but what if you want to change it on the fly? [17:51] as the Style is then in your repository you can do whatever you want with it [17:51] mzanetti: I sort of get how it works, but I don't see an easy way to change the appearance of just one widget. Presumably you'd inherit and modify the WhateverStyle.qml thing, and set that as your style. [17:51] But many of the styles aren't exposed by qmldir [17:52] rschroll: yes. that's not complete yet [17:52] rschroll: those will be exposed as the SDK team sees the need for it [17:53] I've been fighting with Sliders here, and I ended up modifying the Slider.__styleItem.thumb property, which can't be an intelligent approach. [17:53] Those underscores probably mean "Don't touch!" [17:53] yes. sort of [17:53] let me find an example [17:54] Here's where I commit my crimes against QML: https://github.com/rschroll/beru/commit/927b2fd377d7cc86b0633a2707b03c3a79a330db [17:55] rschroll: this is how I made the Popover themeable: https://code.launchpad.net/~mzanetti/ubuntu-ui-toolkit/themable-popover/+merge/181078 [17:56] rschroll: and how to use it: https://code.launchpad.net/~mzanetti/unity8/theme-quicklist/+merge/181223 [17:56] rschroll: the Popover already supported the styled item though... [17:56] rschroll: you'd need to see how that is done there and do the same in the OptionSelectorDelegate [17:58] mzanetti: Thanks. Finding examples is difficult -- this should be a great help [18:00] Hi everybody! [18:00] Hi nik90_ :) [18:00] mzanetti: If I'm reading that right, this approach assumes that you're using Themes.Ambiance. What if you're not? [18:01] rschroll: everything is using Ambiance [18:02] Forever? [18:02] no [18:02] rschroll: no. but for example suru just inherits from Ambiance and changes colors [18:03] If I want to use theme colors and change the font, then what do I do? [18:03] cwayne_: and in your custom theme you inherit that just again and change colors. no need to do the full implementation again if you can get that from ambiance [18:03] yeah, that's true [18:03] rschroll: check this https://code.launchpad.net/~mzanetti/unity8/theme-quicklist/+merge/181223 [18:04] rschroll: it just inherits the default style and changes colors and one image. [18:04] rschroll: could do the same with font [18:05] mzanetti: Right. But what if the user is using Suru, or something else that changes the colors. I want to use those colors, but change the font because I'm writing a font selector. [18:06] rschroll: haven't thought that much about it... the above merges are the only thing I've ever done with themeing [18:06] mzanetti: OK. I realize you're not responsible for everything. [18:06] These are more general questions that I'm having. [18:07] Is there a mailing list I should be posting to? [18:07] rschroll: yeah, there is the ubuntu-phone list [18:07] ubuntu-phone@lists.launchpad.net [18:07] Thanks. I'll try to write up my confusion and send it there. [18:08] rschroll: prefix your mail's subject with [SDK][Theming] or something like that to get the appropriate peoples attention [18:08] Will do. Thanks for the help and examples! [18:09] np [18:19] PaoloRotolo: hi :) [18:20] nik90_, I pushed all in my branch :D [18:20] http://bazaar.launchpad.net/~paolorotolo/ubuntu-clock-app/search-button/revision/187 [18:22] PaoloRotolo: I remembered that there could be some more improvements related to the search box. I will try to code that and send it to you. You can then add that to your branch and merge to trunk [18:22] PaoloRotolo: is that okay? [18:24] PaoloRotolo: it has nothing to do with your work. Your icon stuff looks great. [18:24] nik90_, ok :) [18:25] PaoloRotolo: Is that okay If I send it to you by tomorrow? Would you be available on irc? [18:26] PaoloRotolo: Can you convert the search icon (the one that looks like a magnifier) in the text field into a svg file? [18:42] nik90_, of course :) [18:44] anyone from gallery-app hangin out here? [20:20] nik90_, I take it you saw the tests land and fix everything right? hurray! [20:24] balloons: I saw your MP :) [20:24] balloons: the tests passed on jenkins before. It failed on the phone which I need to check the qa dashboard to check [20:24] but thanks a lot for it [20:25] yup the qa dashboard shows 100% pass for clock app [20:25] hurray! [20:28] beuno: it's interesting how my apps are reviewed out of order =) [20:47] What's the best way to learn QML if i don't know any C or Javascript [20:48] Noskcaj: QML alone wouldn't help. You would need to know Javascript or C++ in combination with QML [20:50] Noskcaj: My advice would be to get started on the tutorials in developer.ubuntu.com to know how to write a basic app and run it. Then start a small project just for yourself and start coding it. If you are stuck ask in this channel, the web or look at sample application code [20:59] nik90_, ok [21:41] nik90_, ping [21:42] iBelieve: pong [21:42] nik90_, got a couple questions related to my app and to the showdown rules. [21:43] iBelieve: will try my best to answer :) [21:43] nik90_, first, could you take a screenshot of the problem with the tags popover you described last night? I can't really picture what's happening [21:43] iBelieve: sure [21:43] nik90_, thanks [21:44] iBelieve: I will have to reflash my phone to a fresh start. For some reason I am having problems using your app alone. For instance I see 2 shortcuts to your app on the dash. Second issue is there is no icon for you app. Third issue is that the tasks are stored only temporarily. On restarting the app, all tasks are deleted [21:45] iBelieve: but continue with your questions. I'll tackle them at the end [21:46] nik90_, I wonder about the tasks being temporary. Maybe the U1db path isn't working under confinement? [21:46] iBelieve: where does u1db store its tasks? [21:47] iBelieve: but I suspect that could be the issue as well. I will try confirming after the reflash [21:47] nik90_, in my app I have path: "ubuntu-tasks.db". On the desktop they're stored in ~/.local/share/Qt\ Project/QtQmlViewer/ubuntu-tasks.db, but I assumed it would use the correct path on the phone [21:48] iBelieve: ah okay. we will have to take this bug up with others to get more information. Since the click packages are hot out of the oven, I am unaware of its issues. [21:50] nik90_, so my next question is since URL opening doesn't work on the phone, how will that affect the judging? Will the judges treat this a problem with the SDK and test Trello integration using the desktop, or do I need to find a temporary solution for the showdown? [21:52] iBelieve: interesting question. I speak for myself and not the other judges. I will consider the working of that feature *alone* on the desktop as well. Since it is a SDK issue which you have no control over, I would do that for all the apps. [21:53] iBelieve: Since this is an important issue, I will raise this with the judges by email this weekend. [21:53] nik90_, okay, thanks. Hopefully it will be working by then, but I was just wondering [21:55] iBelieve: I understand your concern. But the judges will also look from a 3rd party developer's point of view as well. [21:56] It is amazing at what is being done with the current state of the sdk beta. [21:56] nik90_, what do you mean by "the judges will also look from a 3rd party developer's point of view as well"? [21:58] iBelieve: I meant that from a 3rd party developer's point of view, we can see that some issues like the one you pointed is out of your control and hence we will take those into consideration while reviewing your app (in the positive sense). [21:58] nik90_, that's what I thought you meant, but just wanted to be sure. [21:59] nik90_, once the judges decide on that, it might be nice to post that somewhere so other developers aren't worried when they find out URL opening is broken on the phone [22:00] nik90_, so I see that the submission form has been posted. Should I fill that in now, or does that need to wait until I'm done submitting updates to the app? [22:02] iBelieve: I will see if I can get dpm to post something about this or otherwise I will do a blog post myself. [22:02] nik90_, okay, thanks. That's all I have for questions. Thanks a lot! [22:02] iBelieve: can you paste the link? [22:03] nik90_, https://docs.google.com/forms/d/1a1Owwl6EVQ8165Wy_TsSaDGGWmYJaT6INm2P4DgZWAA/viewform [22:03] nik90_, In step 4 of http://developer.ubuntu.com/showdown/ [22:03] nik90_, Fill in the participation form › [22:05] iBelieve: as per the instructions there, I think you can go ahead and fill in the form while updating your click packages until the deadline. [22:05] iBelieve: the form has basic questions, which wouldn't really change in the next few days. [22:06] may be the programming language could change due to a new c++ plugin perhaps [22:07] nik90_, I'll just wait a few days just to be safe [22:07] yeah sure === salem_ is now known as _salem [22:33] can some one help me figure out this error http://pastebin.com/qsxteUcS