=== chihchun_afk is now known as chihchun [06:24] good morning [07:13] I'm trying to put a function for expanding the window to a larger size and going back to its original size in my notetoolbar component in qt and wanted to know is there a way to do that in QML or should I find a javascript solution [08:50] zsombi: another alarm related bug 1362341, It was there since the old clock app times, but just surfaces when I was implementing the single alarm feature. [08:50] bug 1362341 in Ubuntu UI Toolkit "OneTime alarms are not automatically dismissed or delete after they are triggered" [Undecided,New] https://launchpad.net/bugs/1362341 [08:52] nik90: so, if you dismiss the alarm and open the app, do you see the one time alarm just triggered enabled? [08:53] zsombi: yup [08:53] zsombi: but it doesn't show up in the indicator-datetime [08:53] nik90: then EDS or the notification did not remove the alarm's visual & audible attachments [08:54] zsombi: if ^^ that's the case, then it would show up in indicator-datetime, no? [08:54] nik90: Alarm does not save the enabled state, it builds up based on these two reminders existence [08:55] nik90: not necessarily, if the date is old, then won't right? [08:55] zsombi: hmm, could be. I will check with charles [08:56] nik90: otoh, SDK alarm services do not run when teh clock app si not running [08:56] zsombi: which should remove the visual & audible attchments, EDS or Indicator-dattime? [08:56] nik90: indicator-datetime should initiate the cleanup using EDS [08:57] zsombi: ack. I will then see if charles can fix this on his end [08:57] nik90: if the alarm must disappear *which shouldn't be the case IMO, then indicator must remove the alarm, otherwise shoudl remove audio&visual reminder from it [08:58] zsombi: say a one-time alarm is triggered and then disabled, then when the alarm enables it again in the clock app, I presume clock app will have to check the date and set the correct one [08:58] when the user enableds it again* [08:58] nik90: sure [08:59] nik90: and adjust the date, not the time [08:59] zsombi: I'm setting up sdk from clean on my machine after updating to 14.10.. I setup kits and an emulator using the wizard, but i cannot run a project I checked out from lp [08:59] oops zsombi sorry [08:59] zbenjamin: ^ [09:00] popey: :) [09:00] zsombi: well, if the alarm time is 09:00 and the current time is 11:00, then the alarm should be set not for today but tomorrow which clock app doesn't allow. So in those cases, the time must be adjusted [09:00] popey: what happens? [09:00] zbenjamin: lp:~flscogna/ubuntu-terminal-app/ubuntu-terminal-app branch that, open the cmakelist with sdk and if I try and run I get "[CMakeFile/com.ubuntu.terminal.desktop] Error 127" [09:00] zsombi: according to the designer, all one-times alarm can only be set for today (from the clock app) [09:00] zsombi: any one-time alarm for other days must go via calendar-app [09:01] zbenjamin: https://imgur.com/lAFjn8s [09:01] nik90: ok, but still, if the alarm cannot be set for the current day, that should happen next day, right? [09:01] popey: yeah core apps are not really SDK friendly [09:02] popey: its on my todo list to fix and create MRs for the coreapps but i'm busy with other things atm [09:02] zsombi: yes, but then there would be no indication of that in the UI since for single time alarms we do not show the day since it is set for today. [09:02] popey: but i can have a look at the terminal app if you want to hack on it using the SDK [09:02] zsombi: but anyways this I can check with designer and fix accordingly [09:03] zsombi: one last question, In the date-time picker can I set the minimum to be the current Date() for one-time alarms so that the user doesn't choose an earlier time? [09:03] zbenjamin: well, I'd like to review a bunch of merges, but I'm kinda blocked because I can't even run them ☹ [09:03] (not just that app) [09:03] popey: let me check [09:03] nik90: yes, you can [09:03] nik90: however, not teh seconds :( [09:04] zsombi: that's ok, we don't show seconds anyway [09:04] nik90: so, fine then ;) [09:04] zsombi: I remember trying the minimum yesterday nite, but ran into looping issues with datetime picker date. [09:04] zsombi: will try again now [09:05] nik90: ahh, that can be... I'll have to "rework" that DatePicker a bit, it is too slow... [09:05] popey: go to your projects settings, and put -DCLICK_MODE=on into the cmake arguments [09:05] zsombi: oh, I run into that issue again, I will report a bug and ping you the bug number [09:06] if I run into that isseu* [09:06] popey: after that it just works :) [09:06] nik90: fine [09:06] zbenjamin: nope, it doesn't [09:06] popey: the problem is , all core apps assume debian packaging by default, thats why you have to manually set -DCLICK_MODE=off in every build setting [09:06] popey: huh for me it worked fine [09:07] ah, now it's building for desktop [09:07] thanks [09:08] popey: np, i guess you are the guy to ask for reviews when i hack on the core apps projectfiles? [09:08] ya [09:08] popey: ok :) [09:16] zbenjamin: ok, next problem - I get the same 127 error when trying to run on device. I have an armhf kit.. [09:16] (and have set the CLICK_MODE=on [09:17] popey: what does the compile output say? [09:18] :-1: error: [CMakeFiles/com.ubuntu.terminal.desktop] Error 127 [09:19] popey: nothing to scroll up? [09:19] zbenjamin: https://imgur.com/DFY4VSO [09:19] oops [09:19] http://paste.ubuntu.com/8167314/ [09:21] question: does anybody knows which slot should i implement in qml for listening this event -> UbuntuWindow::handleSurfaceFocusChange(focused=true) ? [09:21] i've tried onFocusChanged() into MainWindow but it doesn't seem to be working [09:21] zbenjamin: thats the only output I have [09:23] popey: let me try again [09:23] popey: what you can do is to delete the build directory and try again [09:26] popey: wow the latest devel-proposed provides support for favouriting scopes and provides more granular control over what you want to show in the scope :) it is awesome [09:27] * nik90 cant bear to run devel or rtm, when devel-proposed has all the new goodies [09:28] nik90: hehe ;) [09:28] nik90, yes, which means the TWC scope will be able to have C/F settings and we will stop having to cross locale, country and a fair dice roll to guess the unit \o/ [09:28] nik90: btw i saw your video for the UbuntuSDK , great tutorial [09:28] davidcalle: yes indeed :D [09:29] zbenjamin: ah thnx :) [09:29] popey: i started from scratch but i cannot reproduce it [09:29] zbenjamin: now I get qmake: could not find a Qt installation of '' [09:29] paste error... [09:29] ":-1: error: No targets specified and no makefile found. Stop." [09:29] * popey closes and re-opens [09:30] i clearly ned more coffee, i just searched for zb in the dash rather than sdk [09:30] zbenjamin: nope, just doesn't work ☹ [09:30] popey, you also can't spell need [09:31] hah [09:31] * brendand gives popey coffee [09:32] ☹ [09:34] popey: you are opening a cmake project right? [09:34] popey: ok nevermind, there is no pro file ;) [09:35] popey: it fails for all your 3 targets? [09:37] zbenjamin: more detail found when right click and expand! http://paste.ubuntu.com/8167419/ [09:37] (sorry, power went off for a moment) [09:37] I am able to build and run for desktop, not device [09:37] popey: ok i know the problem now [09:37] popey: intltool is not installed in click chroots [09:38] ahh [09:38] popey: go to Tools->options->ubuntu->click and use the maintain button on every target, apt-get install intltool there [09:38] that fails - i get a terminal with "unable to find shell" in it [09:39] popey: arghs, check if the terminal emulator that is set in your qtcreator config is correct [09:39] zbenjamin: https://imgur.com/RK4aFVT [09:40] zbenjamin: it was blank [09:40] popey: check in tools->environment->General Tab->Terminal [09:40] yeah [09:40] popey: /usr/bin/x-terminal-emulator -e is selected for me [09:40] that fails here [09:40] xterm works [09:42] thanks for the help zbenjamin, sorry to interrupt you [09:42] popey: np, anytime [09:42] fginther: hey, in https://code.launchpad.net/~nik90/ubuntu-clock-app/onetime-alarms/+merge/232519 which link am I supposed to see to check the test results? [09:42] fginther: all the links in the jenkins comment don't show the test results [09:43] zbenjamin: hey any cool new features we can look forward to in the upcoming update? [09:43] nik90: yeah scopes support :) [09:44] zbenjamin: cool :) [09:44] nik90: when you are on trusty then you need a device for it, because scopes can not be executed on trusty anymore [09:44] nik90: on utopic desktop + device will be supported [09:45] zbenjamin: assuming device also includes emulator in which case trusty should still be fine [09:45] nik90: of course [09:47] zbenjamin: [10:45:01] /!\ transfer failed /!\ [09:47] * popey manually transfers [09:48] popey: transfer failed?? you explore totally new problems ;) [09:48] \o/ [09:48] popey: reattaching the device might help [09:48] check me out! [09:48] zbenjamin: that's normal for popey [09:48] he's right ☻ [09:55] https://imgur.com/nRzCzUN \o/ (just need a VT100 keyboard skin for this..) /cc Elleo ㋛ [09:59] popey: ======== Testing required =============> https://code.launchpad.net/~nik90/ubuntu-clock-app/onetime-alarms/+merge/232519 (Click provided in dropbox) [10:00] * nik90 grabs something to eat (brb in 5 mins) [10:00] nik90: will do! [10:00] popey: I wrote a manual testsuite in the MP [10:02] popey: heh, nice [10:26] popey: the issue with the OSK not appearing is not a clock app issue since I did not make any changes that could affect that. Must be a random bug you might have hit with OSK. [10:27] yeah [10:27] expect so [10:27] zbenjamin: http://paste.ubuntu.com/8167734/ can't install any click on device from sdk [10:27] popey: as for setting an alarm for midnight, I was told by the designer that clock app single alarms should only target today. [10:27] ok [10:28] how do you set a non-repeating alarm for tomorrow morning then? [10:28] popey: how old is your SDK? [10:28] popey: so not sure if setting a one-time alarm for midnite 00:00 can be set [10:28] popey: you can't [10:28] ☹ [10:28] popey: you need to use the calendar app for that [10:28] thats a really common use case, surely? [10:28] popey: are you on trusty and use the sdk team ppa? or on utopic? [10:29] zbenjamin: latest utopic [10:29] zbenjamin: which package should I give you the version number of? [10:29] popey: yeah so one-time in the clock app target only today while calendar app should target one-time alarms on any day as an event [10:30] popey: how do you start your apps? with the green run button [10:30] ? [10:30] zbenjamin: Build -> Ubuntu -> Install application on device [10:32] popey: ok, it works for me , can you go to the device page and open a SSH connection to the device? [10:32] popey: its on the advanced tab [10:32] Unable to find a shell [10:32] wtf, is your setting empty again? [10:33] /usr/bin/xterm -e [10:36] popey: ok seems this is a old script and uses x-terminal-emulator directly [10:36] popey: i wonder why that does not exist for you [10:36] $ which x-terminal-emulator [10:36] /usr/bin/x-terminal-emulator [10:37] via various symlinks it eventually ends up at /usr/bin/terminator [10:37] sudo update-alternatives --config x-terminal-emulator [10:37] for me it says /usr/bin/gnome-terminal.wrapper [10:37] i use terminator as my terminal [10:38] and if I change it to the same as yours it breaks differently [10:38] alan@deep-thought:~$ x-terminal-emulator -e [10:38] Use of uninitialized value $args[1] in exec at /usr/bin/x-terminal-emulator line 81. [10:38] Failed to parse arguments: Argument to "--command/-e" is not a valid command: Text was empty (or contained only whitespace) [10:40] popey: seems terminator is not compatible with the -e switch [10:40] Is there a place I should go to if I wanted some examples of good QML applications to learn from them? [10:40] zbenjamin: no, i changed it to match yours [10:41] * 1 /usr/bin/gnome-terminal.wrapper 40 manual mode [10:41] popey: without a argument it fails for me too [10:42] x-terminal-emulator -e sleep 10 [10:42] i get a window appear doing that [10:42] exactly what you want [10:43] just in the case of our script it runs a ssh shell instead of sleep [10:43] so i changed the setting in the sdk from xterm to x-terminal-emulator -e [10:43] and still can't ssh to the device [10:45] did you also change what x-terminal-emulator is using? [10:45] yes, to match yours [10:45] still same error? [10:45] terminal window pops up then disappears [10:45] when i press the ssh button in devices [10:46] i have a pin code set on my phone btw [10:46] adb forward --list [10:46] there should be a line that looks like: 0182ffb256cd4354 tcp:10000 tcp:22 [10:46] http://paste.ubuntu.com/8167885/ [10:47] ssh phablet@localhost:10000 [10:47] ssh: Could not resolve hostname localhost:10000: Name or service not known [10:47] ⍨ [10:47] perhaps you mean ssh -p 10000 phablet@localhost [10:47] because that works [10:48] popey: ssh -i ~/.config/ubuntu-sdk/ubuntudevice_id_rsa -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null -p10000 phablet@localhost [10:48] Warning: Permanently added '[localhost]:10000' (RSA) to the list of known hosts. [10:48] Received disconnect from 127.0.0.1: 2: Too many authentication failures for phablet [10:49] popey: please clean up all ubuntudevice_ keys in your .ssh directory [10:49] popey: then reboot :/, restarting ssh agent alone does not seem to be enough [10:49] clean up as in delete them all? [10:49] ugh [10:49] popey: all ubuntudevice keys can go away in .ssh , we store them now in ~/.config/ubuntu-sdk and we have just one [10:49] k [10:49] * popey reboots [10:50] thanka! [10:50] -typos [10:51] popey: works? [10:51] * zbenjamin can't take the pressure ;) [10:55] popey: bug 1362548 [10:55] bug 1362548 in Ubuntu Clock App "Alarms take several seconds to save (~5seconds) which is quite long compared to existing solutions" [High,Confirmed] https://launchpad.net/bugs/1362548 [10:56] ) [10:58] zbenjamin: i can ssh to device now! [10:58] popey: awesome! [10:58] popey: so try again if the SDK now works :) [11:00] hmm, install on device seems to not be giving me a log of it actually doing that [11:00] popey: its in the general messages tab [11:00] popey: is there any way to test MR with different locale ? [11:00] zbenjamin: nope [11:01] zbenjamin: http://paste.ubuntu.com/8167995/ [11:01] popey: http://pastebin.ubuntu.com/8167998/ thats what i get [11:01] mihir: good question.. other than setting the device to a different locale.. [11:01] popey: yup [11:01] zbenjamin: i have a meeting, I'll have to come back to this after, thanks. [11:02] mihir: don't think so. [11:02] popey: sure [11:02] popey: okay , no issues. [11:04] nik90: ping [11:05] nik90: i am using this code, and it triggers the text property function only once , after that it is not triggring , any idea , http://paste.ubuntu.com/8168022/ ? [11:05] mihir: probably http://qt-project.org/doc/qt-5/qlocale.html#setDefault [11:07] twstddev: hmm let me check === _salem is now known as salem_ [11:19] mihir: or try setting a locale locally http://archive09.linux.com/feature/53781 [11:27] mihir: That's right. If you want it to call that function automatically, call it whenever "rule" changes. So onRuleChanged: thishappens.text = eventUtils.getRecurrenceString(rule) [11:27] mihir: otherwise that function will only be called once === MacSlow is now known as MacSlow|lunch [11:47] nik90: got it [12:10] DanChapman: Sorry it's taken so long but this was the first time I was able to install dekko again http://paste.ubuntu.com/8168459/ [12:12] davmor2, hey that's cool. It's failing at the same point it was for ogra_ and ybon. I presume your username doesn't have your domain name in it either? :) [12:14] DanChapman: No just davmor2 [12:14] davmor2, if so http://people.ubuntu.com/~dpniel/dekko/click/com.ubuntu.developer.dpniel.dekko_0.2.5_armhf.click should get it back up and running, I'm just fixing an SMTP issue before i upload it to the store [12:15] DanChapman: I'll give it a go after Lunch [12:16] zbenjamin: actually when running an app on device, I notice that the output shown in the "Application Output" is sparse and missing all the output lines. I see a few of them but they are intermittent. Is this a known issue? [12:18] nik90: yeah, the new release should fix that [12:18] davmor2, The issue was it was trying to decide on an account icon based on the domain in the username. And trying to split on the non-existent @ causes an assert fail. Anyway that click fixes it :-) [12:19] nik90: its not so easy to get the output out of confinement to qtcreator, so i had to write a execution wrapper that dups the stderr and stdout to names pipes i can read from the launcher application [12:22] can someone tell me if we are moving away for qmlproject for core apps? [12:31] DanChapman: you missed me? [12:34] zbenjamin: ah ok === MacSlow|lunch is now known as MacSlow [12:45] Wellark, missed you? I just filed bug 1362592. Sorry i forgot last week. While your here i've been having issues getting libconnectivity-qt1-dev to install in my click chroot loads of missing deps.... ahh i see you just changed the bug status [12:45] bug 1362592 in Network Menu "bandwidth limitation detection not functioning in the connectivity-api" [Undecided,New] https://launchpad.net/bugs/1362592 [12:45] * DanChapman goes to read [12:47] nik90: you've couple of mins ? [12:51] DanChapman: missing deps of libconnectivity-qt1-dev package declaration or the declared packages being missing from your chroot archives? === chihchun is now known as chihchun_afk [12:56] DanChapman: thanks for the bug report [12:56] Wellark, well it's saying unmet depends libconnectivity-qt1-dev : Depends: qtbase5-dev. I'm just building a new chroot so will try again to get the exact issue [12:57] Mirv: is that depend correct? [12:57] DanChapman: try to apt-get install qtbase5-dev [12:57] to see if there are some deps missing for that in the archive [12:58] Wellark, sure i'll give it a try [13:00] Wellark: nothing wrong specifically with qtbase5-dev [13:02] Wellark ok so i now have unmet depends for libgl1-mesa-dev, libgles2-mesa-dev,libglu1-mesa-dev and qt5-qmake [13:02] mihir: yeah [13:05] Mirv: can you help DanChapman --^ ? [13:05] he is trying to build inside chroot [13:06] so seems archives not configured properly [13:06] or we have some bigger issues somewhere [13:13] nik90: i have pushed the changes :) [13:13] let me know we can try to merge our branch [13:14] mihir: anything I need to look out for? Or does everything work as expected? [13:14] nik90: it workes file, i am still testing multiple scenarios and will make fix if it breaks. [13:14] ok [13:14] nik90: i did , create/edit event but still testing more [13:14] nik90: i have added slider :D [13:14] mihir: cool :D [13:18] Wellark: DanChapman: it's hard to say what's wrong if apt install doesn't work. check /etc/apt/sources.list and ensure it has "main universe" at least on the line starting "deb http://archive.ubuntu.com/ubuntu utopic", after that part? [13:18] and apt update to update the repos. to find out more about the error, you can try installing individual packages like apt install libgl1-mesa-dev and see what that complains [13:37] popey: oh man, terminator is freaking awesome...(got to try it while you were discussing it with zbenjamin) :P [13:37] nik90: now you got me curious too [13:37] hah [13:38] zbenjamin: hehe [13:40] nik90: popey: isn't that similar to what tmux does? [13:40] yeah, kinda [13:41] mhall119, ping [13:41] * nik90 looks up tmux [13:43] nik90: popey: i saw some pretty awesome terminal emulators that gave you much more functionality than just terminal. Autocompletion of commands and much more, but i always forget the names ;). All still in early development though [13:44] nik90: tmux is basically the other way around, it runs in the shell, so you can use it over ssh for example. Its basically a more awesome screen replacement [13:44] zbenjamin: final term? [13:44] zbenjamin: ah [13:44] nik90: yeah that was one [13:45] nik90: there is also a project that turns any editor into a shell [13:45] zbenjamin: yeah I tried that one but it still needs some work...its not that fast as a normal terminal or terminator [13:45] https://i.imgur.com/TVSqlQu.png thats how mine looks most of the time [13:45] nik90: which makes it possible for you to move the curser around [13:46] popey: why u have the normal terminal open as well? [13:46] i close that frequently [13:46] i use a normal non-byobu one for lengthy things that I need to scroll back in [13:46] because scrolling in byobu/tmux/screen sucks [13:46] ah [13:51] popey: why does scrolling in tmux suck for you? Ctrl+B, Page Up [13:51] zbenjamin: imagine you want to capture the whole screen to paste in a pastebin? [13:52] popey: hmpf ok ;) [13:52] popey: we should have a ubuntu pastebin tool : dmesg > ubuntu-pastebin [13:53] pastebinit [13:53] yeah, we do ☻ [13:53] I'm usually trying to capture the output of a command that already ran though [14:01] nik90: did you say you're running the rtm channel on your mako? [14:02] popey: I was running rtm-proposed. I just switched to devel-proposed few hours back [14:02] ah okay [14:02] popey: I presume I need to test on devel-proposed for daily merges. And before pushing to store, I will test against rtm-proposed [14:03] this is why I asked, I have two devices and could have one on devel-proposed and one on rtm [14:03] if you need any testing, to save you having to keep flashing [14:03] ah ok [14:04] in that case, I will stick to devel-proposed and when we decide to push an update to store, you can test the latest-trunk on rtm and see if it works as expected [14:04] yup [14:04] I'll flash my other device then [14:04] cool [14:16] nik90: i am comuting to home will be back after 2 hours or so ,then we can merge and fix our branches , is that okay for you ? [14:18] DanChapman: I hit another issue, if I add a folder for sent mail, in my case Sent it blanks that field I get errors and the to address is blanked :( but I'm assuming this is down to the fact that the user setup got screwed over. Is there a way I can blow away my settings and try and setup from the start using the fixed click? === gatox is now known as gatox_lunch [14:19] popey: do you know how can ask about the click not cleaning up things ? [14:20] DanChapman: ogra_: also see the comedy mail I sent to the phone mailing list as a test mail you'll love I'm sure :) [14:20] davmor2, do you mean you get an error when sending? like an APPEND Failed and that cause the recipients to be removed? [14:20] DanChapman: yeap [14:21] DanChapman: also when I goto smtp settings the add to folder is selected but the field is blank, even though I set it to be Sent [14:21] mihir: yup [14:22] davmor2: yo, have you ever see bug 1265250 ? [14:22] bug 1265250 in ubuntu-system-settings (Ubuntu) "Storage shows same app multiple times" [Low,Confirmed] https://launchpad.net/bugs/1265250 [14:22] seen* [14:23] Right well the APPEND failed probably means you didn't give the correct folder path. It's not very clever at the moment :-( If you open your Sent folder and take a note of the page title (So for me it's INBOX.Sent or [Gmail]/Sent Mail) which you have to use that title valueas the sent folder name [14:23] davmor2, yeah I noticed that earlier. The value should still be what you saved it's just not getting reloaded to the UI. I'll fix that now [14:23] daker: nope [14:24] nik90: the text in subtitle are too light, will it be like that only or we can change [14:30] daker: i think that's one for cjwatson, but he's afk AIUI [14:30] ok [14:30] daker: i have seen this conversation elsewhere though, related [14:31] which talks about cleaning up when we uninstall apps [14:35] nik90, I see the problem and should be able to fix it soon. Until then if you go to the master job, you can find the link to the test job via the console link. for example: http://91.189.93.70:8080/job/ubuntu-clock-app-ci/408/console [14:37] mhall119, ping? [14:42] aquarius: was that a question? [14:43] mhall119, have you got a few mins for a chat at some point? [14:44] sure [14:44] mhall119, when's good for you? [14:44] anytime is good for IRC [14:45] hangout would prolly be easier [14:45] then give me 15 minutes :) [14:45] :-) [14:45] no rush :) [14:47] balloons: morning! [14:47] balloons: any idea why http://91.189.93.70:8080/job/generic-mediumtests-utopic/1712/testReport/ubuntu_terminal_app.tests.test_terminal/TestMainWindow/test_font_size_changes_with_mouse_/? fails? (via https://code.launchpad.net/~flscogna/ubuntu-terminal-app/ubuntu-terminal-app/+merge/232215 [14:47] watched the video and it seems to work fine [14:47] * balloons looks [14:49] ahh, yea, I think I see the issue [14:54] hi everyone [14:54] popey, I'll push a quick fix to trunk [14:55] dholbach: have you finished updating UGJ related material on the wiki? [14:56] rigved, the wiki should be up to date - the materials for running an app dev school are not 100% there yet [14:56] balloons: thanks, ping me a link to review if you need it [14:59] dholbach: ok. i have got an app dev school event setup. also, one person will be bringing a nexus 7 (2013) and maybe a nexus 4 as well for demoing apps. [14:59] woooooooooo! [14:59] nice [14:59] I'll let you know asap when things are ready [14:59] awesome! === caribou_ is now known as caribou [15:21] nik90: thought I'd re-visit https://code.launchpad.net/~dpm/ubuntu-clock-app/reboot-desktop-i18n/+merge/230582 - seems fine here === gatox_lunch is now known as gatox [15:28] nik90: https://code.launchpad.net/~nik90/ubuntu-clock-app/onetime-alarms/+merge/232519 also looks good to merge? [15:30] aquarius: available now? [15:30] sho am [15:38] popey: I want to spend the rest of day testing it and making sure the code is clean and good to go. But mostly I also require someone to do a code review of it. ->https://code.launchpad.net/~nik90/ubuntu-clock-app/onetime-alarms/+merge/232519 [15:38] popey: as for https://code.launchpad.net/~dpm/ubuntu-clock-app/reboot-desktop-i18n/+merge/230582, I will take a look. I need to merge trunk into it and see if it is still ok. [15:46] ok [15:49] renatu: hey, do you know if it would be possible to do some optimizations to the alarms save process in EDS for bug 1362548 [15:49] bug 1362548 in Ubuntu Clock App "Alarms take several seconds to save (~5seconds) which is quite long compared to existing solutions" [High,Confirmed] https://launchpad.net/bugs/1362548 [15:50] renatu: when the alarm.save() operation is called in the Alarms SDK, it seems it takes ages to save the alarm. Is this due to EDS or the Alarms API? [15:50] nik90, I do not know I need to investigate it [15:52] nik90, if you find some time you can create a small example using the SDK alarm API and a QtOrganizer QML api [15:52] and see if still the same problem [15:52] this will help me to find the problem [15:53] hmm I have never used the QtOrganizer QML API. I will try [15:53] you can take a look on calendar app code [15:53] ok [15:54] you can ping me if you need help [15:54] nik90, but is the alarm.save() a sync function? [15:54] no its async [15:55] but the issue is still visible since the user waits on the alarm page to see the alarm he created [15:55] ok [15:55] nik90, btw my alarm is not working [15:55] is that expected? [15:55] no [15:56] as in it doesn't ring? [15:56] I can see the alarm on the clock app but it never rings [15:56] do you see it in the indicator-datetime? [15:56] no [15:56] one-time alarm or repeating alarms? [15:56] repeating [15:57] repeat every weekday [15:57] that's strange, I have never reproduced that issue [15:57] the alarm never works for me :D [15:57] I have been waking up to it for the past month :/ [15:57] I thought that was a problem of the old app [15:58] but still not working after install the new image [15:58] did you clean out the old EDS alarms database file? [15:58] normally not [15:58] normally I just install the new image [15:59] can you delete that file and create a new alarm now and check? [15:59] ok [15:59] I will try [15:59] the alarms experience for me has being quite stable for me to the point where I am happy at where it is [15:59] great [16:04] aquarius: your demo doesn't like getting a json array with a single item [16:05] mhall119, it does not [16:05] mhall119, I discovered that. [16:05] mhall119, the actual generic scope will need to actually code defensively about that sort of thing :) [16:05] nik90, ping [16:05] mhall119, demos don't get defensive coding. [16:06] Also, a pox on people who write an API which returns a list of objects if there are multiple hits and just an object if there's only one hit. [16:06] it's still a list! it's just a list with one item in it! grrr! [16:06] nik90, just talking through the alarm volume scenario with rsalveti...we decided that we need a volume slider for alarms in system settings and not in the clock app, but you can leave that for now until we land the backend pulse support for volume roles [16:06] aquarius: http://loco.ubuntu.com/services/teams/?lp_name__contains=florida returns a list of a single item, your demo doesn't handle that though [16:07] it appears to treat the single item's keys as a list of results [16:07] jhodapp: pong [16:07] jhodapp: ack. I will have to check with designer though. [16:07] mhall119, hrm. that might be wrong usage of jsonpath by me... [16:07] nik90, we were talking with mpt about this as well [16:08] jhodapp: i think what we might do is show a button in the alarm settings which takes the user directly the system settings app where you can set the alarm volume. But yeah let's first what mpt thinks and go from there. [16:09] mhall119, jsonpath.js seems to always return a list of one item, where that one item is the list you wanted [16:09] nik90, that might be reasonable...I know that iOS users are used to looking for settings like that that apply more to a system context instead of just an app context would always be in system settings [16:10] mhall119, so if you do jsonPath({results: [1,2,3]}, "$.results") you get back [[1,2,3]], not [1,2,3]. So I always just take the first item in the returned list and treat *that* as the result list. But that might be wrong, and this is a symptom of it [16:11] jhodapp: ok [16:11] nik90, talk to mpt about it for clock app, but don't worry anymore about it until we land the backend [16:12] nik90, that should be in the next few days [16:12] popey, https://code.launchpad.net/~nskaggs/ubuntu-terminal-app/ap-cleanup/+merge/232590 [16:12] jhodapp: ok. I will also have to check with the clock app designers about this. Once the backend has landed, I wil try to organize a hangout with mpt and the clock designers [16:12] nik90, sounds good, thanks [16:13] popey, you're too fast [16:13] balloons: thanks, just got the lp mail ☻ [16:13] dholbach: sorry ☻ [16:13] that's all right :) [16:24] popey, if you can, please also approve https://code.launchpad.net/~nskaggs/ubuntu-filemanager-app/add-lxml-depends/+merge/231457 [16:25] balloons: done and top approved [16:25] balloons: just reviewing your other ap one [16:31] balloons: why is jenkins not picking up https://code.launchpad.net/~nskaggs/ubuntu-terminal-app/ap-cleanup/+merge/232590 ? [16:36] am i just impatient? [16:37] popey, yes I believe so [16:38] hah, just finished [16:38] * popey pours red bull into jenkins [16:39] top approved === boiko_ is now known as boiko [17:15] mzanetti, o/ If you want I have ~30 minutes right now for the hangout === jgdx is now known as jgdx_ [17:38] mhall119, OK, have fixed the issue that was stopping the loco team thing from working [17:40] hey balloons any idea what's going on here http://91.189.93.70:8080/job/generic-mediumtests-utopic/1718/? ? via https://code.launchpad.net/~rpadovani/reminders-app/updateStrings/+merge/231125 [17:41] common question today . .. [17:41] haha [17:41] WHAT IS THIS MADNESS! [17:41] aquarius: cool, I've got a lead on a way to re-use an existing binary that I think will work [17:41] mhall119, oh? do tell [17:42] thoughts of cpp warnings [17:42] *lots [17:42] mhall119, note that the real problem with the web demo is that it doesn't hit the URL itself; it fetches it through YQL, to avoid cross-domain problems, and that's what was causing the issue :) [17:42] popey: https://imgur.com/w9j8DIY [17:43] nik90: are you showing me that for the hilariously named town? ☻ [17:43] aquarius: Unity evidently supports multiple "scope runners", the current one looks for and loads a .so, but there's going to be another that supports Go scopes, and we can have a GenericJsonScopeRunner that only looks for .ini files and launched a .so from the system [17:43] mhall119, aha! sneaky! [17:43] popey: :P the location detected is about 4KM away from where I live which is not bad [17:44] popey: it should become more accurate with AGPS [17:44] so we'll need an actual scope binary .so, and also a new scope runner, but that will do what we wanted [17:44] aquarius: so the first order of business is to write the generic scope [17:44] mhall119, brilliant. And one will select which scope runner one's scope requires with a key in the ini file? [17:44] Or is this done with a click hook? [17:44] popey, the output is pretty wild. Let's see if we can repeat it. Currently the account fakery isn't working so well with reminders. Leo is looking at it, as you know [17:45] right [17:45] it appears it's coming to bear on those results [17:47] aquarius: the click hook I think [17:47] I'm still trying to find out how that works, since htere's only one kind of scope in use today I'm not sure that implementation detail has been worked out yet [17:48] mhall119, fair enough. The scope creator can give you a little downloadable zip which you can unpack with the relevant files in [17:48] it could actually give you a whole click package reasonably easily, but getting the icons into it might be annoying. [17:48] step 3 is that it uploads the click package to the store for you ;) [17:51] aquarius: I'm thinking we just make it a new template in QtCreator and let that handle the rest of the steps [17:52] popey: hi, I got this comment from you in the review of an app: "'ubuntu-sdk-14.10-qml-dev3' is not a supported framework" [17:52] yes [17:52] alecu: adb shell ls -l /usr/share/click/frameworks [17:53] popey: but when I upload the click to the store, I didn't get any error [17:53] s'pose. I rather liked the idea that one could create a scope without having to install all the SDK stuff and be at an Ubuntu machine -- you could do it from your phone. But fair enough :) [17:53] alecu: http://paste.ubuntu.com/8170972/ [17:53] those are valid [17:53] mhall119, ^ [17:53] alecu: you would have if you'd waited, it was the store that gave that error, not me [17:53] mhall119, not sure if qtcreator can show the "mockup" of a dash and a preview in a template, though [17:54] mhall119, and "type stuff into these boxes, which is dead easy" stops being dead easy if your "did it work" cycle is "compile a click package, install it, go to the dash, search"...? [17:54] mhall119, possible integration with unity-scope-tool, though? [17:54] popey: hmm.... the very latest framework in your paste looks like the one that got me rejected [17:55] popey: I got the same list installed on my device [17:56] alecu: hmm [17:56] aquarius: the unity-scope-tool could maybe do the "mockup" stuff [17:56] I'm not sure if it's compatible with alternate scope runners though [17:56] mhall119, yeah, if it can be integrated into a qt creator template [17:57] alecu: lemme download it and test locally [17:57] mhall119, I really do want it to be as easy as "type something in the box, instantly see the results of doing that". That instant feedback loop is what makes it fast to use. [17:57] first things first, I need to see if I can get this metadata from the scope [17:57] alecu: hah, passes here! [17:57] alecu: passed it.. [17:58] beuno: https://myapps.developer.ubuntu.com/dev/click-apps/1045/ - fails automated test (erroneously) [17:59] popey: why do you say that "it was the store that gave that error, not me"? The comment I got in the review had your name :-) http://pasteboard.co/2CPj7UTd.png [17:59] anyway, it says "Ready to publish" now. [17:59] thanks! [18:01] popey: ======== Testing required ========== > https://code.launchpad.net/~nik90/ubuntu-clock-app/implement-location-finding/+merge/231793 (testing instructions in the MP) [18:01] * nik90 goes to prepare dinner [18:01] popey, automated reviews landed just now [18:02] popey, the scripts seems out of date [18:02] will fix soon and keep in sync [18:07] alecu: because the store says "failed" and gives me the error message which I paste in and give to you. [18:07] alecu: if it passed, I'd download the click and confirm locally, but on this occasion I didn't because it failed in the store, no point me locally checking (hah) if it fails in the store [18:09] nik90: ok! [18:18] nik90: i didnt get a request for authorisation for location [18:18] nik90: but it does say "Retrieving location" [18:25] popey: Did you start the location services? [18:25] manually that is [18:25] yes [18:25] ah, hang on [18:25] looks like it dies [18:25] phablet@ubuntu-phablet:~$ start ubuntu-location-service-trust-stored [18:25] ubuntu-location-service-trust-stored start/running, process 3426 [18:25] oh no, its there [18:25] phablet 3426 0.0 0.2 52988 3828 ? Ssl 19:15 0:00 /usr/bin/trust-stored-skeleton --remote-agent DBusRemoteAgent --bus=system --local-agent MirAgent --trusted-mir-socket=/var/run/user/32011/mir_socket_trusted --for-service UbuntuLocationService --store-bus session [18:26] was this a clean wiped phone? [18:26] no, not by a long shot [18:26] I am just wondering if you gave clock app location permission when I asked you test this MP few days back [18:26] quite possibly [18:26] ok [18:27] nik90: hi [18:27] just leave the phone near the window or something and come back to it in a few minutes [18:27] hmm [18:27] you might want to turn the screen on every 30 secs. [18:27] hang on... [18:27] mihir: hi [18:27] http://paste.ubuntu.com/8171254/ [18:28] nik90: can we merge branch ? so that we can work on that together ? [18:28] http://paste.ubuntu.com/8171258/ is the full log [18:28] popey: I have already included location in the manifest file [18:28] mihir: yes [18:28] mihir: I already added afew commits to your branch. did you look at it/ [18:29] ? [18:29] * mihir looks at nik90's commits [18:29] popey: let me check if i got those as well [18:31] popey: http://paste.ubuntu.com/8171286/ [18:33] nik90: if i do , bzr update it should get me a latest version of my branch right? [18:33] mihir: do you have any local changes in your branch? [18:33] nik90: nope [18:33] nik90: it says , Tree is up to date at revision 422 of branch /home/mihir/Documents/apps/revamp-newevent-page [18:34] mihir: then yeah I think so [18:34] bzr update should do it [18:34] it is not doing it , let me try with branch name [18:34] mihir: try bzr pull [18:35] mihir: just a tip, make sure your code is well indented. Also I cleaned your code a bit to follow qml code conventions at http://qt-project.org/doc/qt-4.8/qml-coding-conventions.html [18:35] I generally group id and objectName together [18:35] nik90: yup, i forgot to indent that , thanks for that [18:35] nik90: i'll take care , it was messy, i'll take care now [18:36] mihir: oh btw kunal posted a comment in the MR. Check if is fixed in your code or not [18:36] nik90: bzr pull didn't work [18:37] nik90: I'll reboot and put my phone on the windowsill while I make dinner [18:38] nik90: worked [18:38] popey: :) [18:38] popey: btw do you have a script to keep the screen on, if not the clock app might get suspended [18:38] mihir: cool [18:39] mihir: if all fails, then delete your local copy and the just branch the MP to get your work [18:40] nik90: i think you can poke powerd [18:40] nik90: don't you think we should make font little darker ? [18:41] nik90: specially , the subText of This Happens [18:41] mihir: on the phone it looks alright. Its just on the desktop it looks a bit faint [18:41] mihir: the only way of changing the font color would be to use our own custom component [18:42] mihir: let me share a screenshot of it running on the phone [18:44] mihir: http://imgur.com/HRkOkNo [18:44] nik90: (y) [18:44] nik90: can you test one thing on phone , [18:44] nik90: create and event , put recurrcens after certain date , and then check if ends after that date. [18:45] sure, 2 mins [18:45] i am not sure, i have messed up with my system , or i have some wrong logic [18:46] nik90: powerd-cli display on [18:46] popey: that's it? [18:46] wfm [18:46] popey: wow [18:46] * nik90 looks silly since he kept looking at his phone every 15 secs few meters away :P [18:47] haha [18:47] don't worry, nobody knows, it's not like you said it in a logged channel [18:47] hehe [18:49] mihir: wow it doesn't [18:49] mihir: it is created infinite number of times (all days) === renato is now known as Guest73069 [18:51] Guest73069: renato did you check if clearing the alarm database worked for you? === Guest73069 is now known as renatu [18:51] nik90, not yet [18:51] I will do it later [18:51] np [18:52] renatu: but do keep nagging me if it doesn't work since it is critical if it doesnt work for you. [18:52] * mihir wonders what's an issue [18:53] mihir: so yeah I set the event to end on sep 09th. But it keeps going until dec and further. [18:53] mihir: the "after x occurrences" works as expected [18:53] nik90: yup, it happened with me too. [18:54] nik90: the irony is , if i get limit value, it will be 9th Sept [18:54] nik90: check event details [18:54] mihir: yup. When I edit the series, it shows the correct end date [18:54] nik90: :( [18:56] nik90: i face same with trunk too [18:56] mihir: oh [18:56] i just run, calendar-app from terminal [18:56] it did the same [18:56] mihir: ok in that case we merge and then ask kunal to fix the issue in trunk [18:56] mihir: since it isn't a regression [18:57] nik90: yup , lets get it merged, [18:57] renatu: any idea on this ? [18:57] nik90: i did some more cleaning [18:57] mihir: did you push? [18:58] i just did [18:58] ver 427 [19:17] nik90: it seems we got some updates in Qt organizor [19:18] nik90: did you propose merge ? [19:18] popey: are you facing that issue in latest image ? [19:24] mihir__: erm no not yet [19:24] mihir__: one sec [19:24] nik90: no issues [19:31] nik90, ping [19:31] twstddev: pong [19:32] nik90, have you got time to check location search? [19:32] twstddev: the url isn't active though rite? [19:32] oh yes [19:32] it's dead since yesterday [19:32] I checked this morning [19:33] I closed the connection [19:33] twstddev: also I am trying to merge 3-4 other branches, so too busy to make the necessary changes [19:33] I'm going away for couple of days, so I can just propose a branch and probably popey can take a look when he has time [19:33] ok [19:34] ack [19:35] twstddev: Once I get the branch rdy, I can ask popey to test it locally on his comp with your branch as well [19:35] nik90, ok [19:41] mihir_: I merged your branch [19:42] nik90: great. [19:42] nik90: i am trying to move the remmber option to another page [19:47] mihir__: cool [19:52] nik90, beautiful clock [19:54] pmcgowan: thnx :D [19:54] nik90, there should be a daily repeat no? [19:55] nik90, mihir__ , do you need any help? [19:55] pmcgowan: you mean as an option in the repeat page? [19:55] pmcgowan: Saviq was requesting for Weekend and Weekday option :).. [19:56] twstddev: i feel we have bugs listed , you can take any of them [19:56] nik90, yeah that makes sense === CodePulsar is now known as SkyLizard [19:56] bug #1362089 [19:56] bug 1362089 in Ubuntu Clock App "Should have shortcuts for "Weekdays", "Weekends" for Repeat" [Wishlist,New] https://launchpad.net/bugs/1362089 [19:56] +1 [19:56] mihir__, sure [19:56] pmcgowan: 1362089 [19:56] Saviq: too quick [19:56] nik90, that's my middle name [19:56] he's everywhere [19:57] pmcgowan: Can you add the daily option into that bug report. I will have to ask the designer about it [19:57] nik90, I defer to the better suggestion you have [19:59] nik90: no location ☹ [19:59] popey: still says "retrieving location" ? [19:59] yes [20:00] nothing location related in the log [20:00] nothing location related in dmesg either [20:00] popey: you got a spare freshly wiped device? I am using QtPosition which is a standard widget + it worked on my device. [20:00] popey: what I could do is do a clean wipe of my phone and try again. [20:01] nik90: i can try that on my other phone [20:02] popey: pls do. I will keep testing this on my phone as well. This branch requires more testing than a code review. [20:02] ok, will do [20:03] which channel would you prefer I test on? [20:03] popey, one thing you might try is https://bugs.launchpad.net/ubuntu/+source/location-service/+bug/1362769/comments/2 [20:03] Ubuntu bug 1362769 in unity8 (Ubuntu) "Location trust store causes dash to hang" [Undecided,Incomplete] [20:03] will try Saviq thanks [20:03] jdstrand: ping (bug 1362779) [20:03] bug 1362779 in Ubuntu Clock App "Please support concept of one time alarm in clock reboot" [Undecided,New] https://launchpad.net/bugs/1362779 [20:03] popey, the location prompt seems to be a bit moody [20:04] nik90: hey [20:04] jdstrand: hi, so I have a branch already which adds one-time alarms :D [20:04] jdstrand: however with one caveat [20:04] nice! :) [20:05] jdstrand: you can only set oone-time alarms for today [20:05] nik90: by 'today' do you mean within the next 24 hours? [20:05] jdstrand: i was told by design that if you want to create one-time alarms for other days of the week then it is an event and should be done via the calendar app [20:05] or literally, today [20:05] jdstrand: yes [20:05] jdstrand: as in literally today [20:05] s/literally/actually/ [20:06] I meant 2014-08-28 [20:06] * jdstrand is kidding [20:06] jdstrand: so the time for me is 10:05 PM here. I cannot create an alarm for 08:00 AM tomorrow since it is the next day. [20:06] hmm [20:06] popey, https://code.launchpad.net/~rpadovani/ubuntu-calculator-app/1357983/+merge/232630 [20:06] nik90: so, if it is 10pm, and I want to set an alarm for 7am tomorrow, just this one time, I cannot? [20:06] jdstrand: you cannot [20:06] popey, I talked with gang65, no way to have in Javascript precision > 30 digits, more than 5 calcs in history and speed [20:07] nik90: that doesn't seem weird? [20:07] jdstrand: it does. I didnt personally agree with the design decision. [20:07] rpadovani: understood. [20:07] nik90: I shouldn't have to wait until after midnight to set my alarm clock [20:07] jdstrand: I will create a bug report now, and then you, popey and the rest can confirm it. [20:07] rpadovani: thanks for working on that. [20:07] jdstrand: +1 [20:07] and I shouldn't have to go to my calendar to set my alarm for tomorrow morning [20:07] add ubuntu-ux [20:08] nik90: sounds good [20:08] give me 1 min [20:08] if I set an alarm for any time up to now-1 it should allow me to [20:08] from now to 23:59 is today, 00:00 to now-1 is tomorrow [20:09] the calendar and the clock have similar functionality but are thought of differently. if I open the clock and see I can add an alarm, I think of it as an alarm clock [20:09] the calendar has my appointments [20:09] yeah, two apps for alarms is ludicrous [20:09] not when I wake up [20:09] (I realise this was not your decision nik90 ) ☻ [20:09] * jdstrand does too [20:10] popey, also, https://code.launchpad.net/~gang65/ubuntu-calculator-app/ubuntu-calculator-app-slowdownfix/+merge/232631 === SkyLizard is now known as DrSkyLizard [20:10] popey, tonight calculator app will have some performance improvements ;-) [20:10] so clock should be able to create one-time alarms for any day of the week (obviously not like calendar on any day of the month but just the current week) [20:11] yeah, that makes sense to me [20:11] I personally found android alarm fairly easy to use and understand quickly [20:12] same here [20:14] rpadovani: woop woop! [20:15] nik90: I like the new clock btw. it is pretty :) [20:15] rpadovani: :D [20:15] jdstrand: thnx. more to come :) [20:18] bug 1362803 [20:19] popey, jdstrand ^^ [20:19] bug 1362803 in Ubuntu UX "Clock should support creating one-time alarms on any day of the current week" [Undecided,New] https://launchpad.net/bugs/1362803 [20:19] erm ubuntu bot [20:19] there you go [20:21] confirmed [20:30] commented [20:31] rpadovani, popey any luck on the bug #1357983 [20:31] bug 1357983 in Ubuntu Calculator App "Continuos calculation doesn't work with irrational results" [Critical,In progress] https://launchpad.net/bugs/1357983 [20:31] renatu: yeah, there's a merge in flight [20:31] popey, thanks [20:32] renatu, ops, I didn't linked it, sorry [20:49] nik90: is there anything you could point out on new event design ? [20:50] mihir: ? [20:54] twstddev: you about? [20:54] popey, yup [20:54] twstddev: I am testing https://code.launchpad.net/~twstd-dev/ubuntu-geonames/search-by-coordinates/+merge/232626 on my server [20:55] right [20:55] http://146.185.188.87/?query=Paris [20:55] existing functionality isn't returning results. [20:55] interesting [20:55] wondering what I broke [20:55] have you reindexed data? [20:55] yes [20:57] popey, sorry for a silly question, but what command did you run? [20:58] i followed the readme.. [20:58] sudo indexer --all --rotate [20:58] my server already had this setup [20:58] no [20:58] sudo indexer geonames [20:58] you need to refresh index [20:58] ok [20:58] there should be to indexed lists [20:58] ok, running [20:59] oh, sphinx not running either [20:59] %) [20:59] i had to fudge the conf file because my vps is utopic [20:59] and the pid file moved [20:59] make sure it has permission to write in /var/run [21:00] http://paste.ubuntu.com/8172436/ [21:00] hmm [21:01] run again, happens sometimes [21:02] same [21:02] have you given permission to write in /var/run? [21:02] well i had this running previously [21:02] drwxr-xr-x 2 sphinxsearch sphinxsearch 60 Aug 28 21:52 sphinxsearch [21:03] thats /var/run/sphinxsearch [21:03] hmm [21:03] everying in /var/lib/sphinxsearch is owned by root [21:03] that can't be right [21:05] ok, got it [21:05] still no results from the web [21:06] so you changed permissions and reran indexer? [21:06] yes [21:06] no [21:06] ☻ [21:06] changed permissions and restarted sphinx [21:07] so stop, re run indexer, chown, restart? [21:07] when you run indexer it will restart sphinx [21:07] it doesnt here [21:07] just give permissions before running indexer [21:07] if you put --rotate it should [21:07] ok, reindexed, restarted [21:08] well, it was stopped when i indexed [21:08] so guess it leaves it in that state [21:08] it restarts for me [21:08] still no results for me. [21:08] WARNING: index 'geonamescoordinates': preload: failed to open /var/lib/sphinxsearch/data/geonamescoordinates.sph: No such file or directory; NOT SERVING [21:08] that looks bad [21:09] indexer have not completed [21:09] let's try again :) [21:09] heh [21:09] make sure that permissions are right [21:09] then run sudo indexer --all --rotate [21:09] and that should be it [21:10] http://paste.ubuntu.com/8172520/ [21:10] popey, what about https://code.launchpad.net/~rpadovani/reminders-app/updateStrings/+merge/231125/comments/566980 [21:10] ? [21:10] en_us or en_gb ? :D [21:11] * rpadovani takes popcorn [21:11] he's right [21:11] our default is en_US [21:11] we then translate it [21:11] popey: why not fight for en_gb? [21:12] popey, it should index "geonamescoordinates" after [21:12] twstddev: do i need to re-create, as geonamescoordinates has never been an index before? [21:13] nik90: because we have a company and project default of en_US, have had for years [21:13] I can't win that one [21:13] :P [21:13] even if everyone else is wrong [21:13] popey, I don't think so. I did not have to [21:13] it should pick it up automatically [21:13] the problem is I am going to have to ask IS to deploy this, and I need to be sure I can make it work first. [21:14] which I am failing at ☻ [21:14] popey, absolutely [21:14] so you ran indexer, what was the output? [21:15] twstddev: http://paste.ubuntu.com/8172520/ [21:15] that's all? is still hanging? [21:16] *is it [21:16] no [21:16] it finishes [21:17] thats the end of the output [21:17] oh hang on! [21:17] [Thu Aug 28 22:10:23 2014] Out of memory: Kill process 7556 (indexer) score 436 or sacrifice child [21:17] [Thu Aug 28 22:10:23 2014] Killed process 7556 (indexer) total-vm:356084kB, anon-rss:224960kB, file-rss:0kB [21:17] hah! [21:17] my poor vps! [21:18] i run it in vm with 384MB of RAM [21:19] * popey adds a bit of swap [21:20] bah [21:21] 256MB swap not enough, gave it 1GB [21:21] which is mad [21:21] indexer eats it all up [21:21] i have no swap enabled o.O [21:22] it's not event mounted [21:22] its got up to 500MB RAM and 700MB swap now [21:22] (this is utopic) [21:22] which needs updating, so could be a bug [21:22] hm. [21:22] i run precise [21:23] ok, its moved on to geonamescoordinates now [21:23] fuf [21:23] fuf indeed! [21:23] %) [21:24] boom, done [21:24] http://paste.ubuntu.com/8172604/ [21:24] http://146.185.188.87/?query=Farnborough still broken ☹ [21:25] if i try searching by coordinates it feels like it doesn't event process it [21:25] would be silly of me to ask again [21:25] does apache need a restart? [21:26] if geoname-modpython.py is where it should be ;) [21:26] heh, reasonable question [21:26] i directly replaced the one i previously had [21:26] let me put back the old one [21:27] right, old one back [21:27] http://146.185.188.87/?query=Farnborough works [21:27] that's not even funny [21:28] Could you put PythonDebug on into your virtualhost file [21:28] sure, where does that go? [21:29] where did you specify path to the script in apache? [21:29] in ubuntu it should be in sites-enabled [21:29] /etc/apache2/sites-enabled/ [21:29] hmm [21:29] probably default [21:30] unless you created your own [21:30] oh, its in my default apache conf [21:30] http://paste.ubuntu.com/8172633/ [21:30] like that [21:30] add PythonDebug on [21:30] into that element [21:31] then restart apache [21:31] and check if it throws anything [21:31] k [21:32] python script, not apache [21:32] uh [21:32] put your new one in place [21:33] where does the log go? [21:36] popey, what does....'with the music-app failures on krillin still staying a mystery'... in the landing mail mean? [21:36] popey, what is 'krillin' ? [21:36] excellent question [21:37] its a device === salem_ is now known as _salem [22:08] balloons, ping [22:17] hey all [22:20] rpadovani, popey, you can have high precision with JavaScript for the calculator. Just use a BigNumber library, rather than native JS numbers. See http://stuartlangridge.github.io/webutils-calculator/ for an example calculator I built which does exactly that (and should work on Ubuntu phones too); code is at https://github.com/stuartlangridge/webutils-calculator and it uses https://github.com/MikeMcl/bignumbe [22:20] r.js as its bignumber library [22:22] aquarius, we use a bignumber library yet, the problem is a bit more complicated.... [22:22] rpadovani, oh! My apologies, then -- I was a bit surprised :) What's the issue? [22:23] aquarius, sorry, playing dota right now, I'll explain in ~30 minutes :-) [22:24] :-) [22:25] * aquarius reads the bug [22:25] Right, first, 5/3 is not irrational -- irrational means "can't be expressed as a fraction" ;-) [22:26] but I see the issue; if I do "1/3" on my calculator I get 0.33333333, but if I then do "x3" I get 0.999999..., not 1. [22:26] hrm. [22:27] I wonder if anyone's written a better bignum lib? I just used the first sensible-looking one I found :) [22:27] aquarius, tbh, calculator needs someone who is good with code design. I mean, I could write code, but I have no idea on how develop a calculator - I'm not the original authour [22:32] balloons, ah no worries fixed it found field.write(text) instead of keyboard.type(text) [22:52] * aquarius asks a question about actually-infinite-precision libraries on G+ :) === jgdx_ is now known as jgdx