=== _salem is now known as salem_ [01:45] how are indicators started? [01:45] eg /usr/lib/indicator-messages/indicator-messages-service [01:46] what runs this, and how? === salem_ is now known as _salem === jamesh_ is now known as jamesh [07:27] mzanetti: so what's the attitude towards the "more stable" branch? [07:27] is it really more stable? [07:27] i see a few fails [07:27] but not sure if in the area it was trying to stabilize [07:33] pstolowski: pong? [08:00] tsdgeos: it makes things slightly better... give me more time with that [08:00] oka [08:03] tsdgeos: and if you manage to reproduce this somehow... https://jenkins.qa.ubuntu.com/job/generic-mediumtests-saucy/2789/artifact/unity8.shell.tests.test_notifications.EphemeralNotificationsTests.test_summary_and_body%20%28Desktop%20Nexus%204%29.ogv [08:05] lol [08:05] the thing got stuck? [08:10] hi [08:10] so we see unity8 looping on CPU regularly in automation [08:10] like here: [08:10] https://jenkins.qa.ubuntu.com/job/saucy-touch-mako-smoke-camera-app-autopilot/85/artifact/clientlogs/top_before.log/*view*/ [08:10] thats after boot, before we run the camera autopilot [08:17] asac: you sure this is not still during the startup phase? [08:18] mzanetti: pretty sure [08:18] we wait 10 minutes :) [08:18] or something [08:18] ah ok. [08:18] hmm... I'll see if I can reproduce [08:18] mzanetti: seems to happen on every 4th boot in average [08:19] from the looks [08:19] even less [08:19] but i see it reliably at least once for every image [08:19] hmpf... I hate those.. but ok. will check [08:19] mzanetti: yeah i dont expect a short term fix [08:19] just if something is known to have changed thats cool [08:19] and we could nail it before it gets old [08:20] mzanetti: we could give you better logs during the top if that helps [08:24] asac: well, unity8.log wouldn't hurt for sure, but I'm not sure if we will actually see anything there. but we can try [08:26] let me see [08:26] where is that log? [08:26] ~/.cache/upstart/ [08:27] maybe the whole ~/.cache/upstart/ [08:27] would be good then [08:27] to get as artifacts like: http://reports.qa.ubuntu.com/smokeng/saucy/image/3833/camera-app-autopilot/ [08:27] guess we want to wipe all .cache on every reboot though [08:27] asac: yeah... if it's not too big in size it would actually be good to attach it to test reports [08:27] yeah lets do that ... cool [08:27] yeah... wipe it on reboot [08:27] unless this comes back all the time, i will ping you once we have that info and see the problem again [08:27] so dont bother :) [08:27] thanks [08:30] mzanetti: /home/phablet/.cache/upstart ? [08:30] asac: yeah. that should hold a unity8.log (and some rotated ones too if you never wiped it) [08:30] tsdgeos, mzanetti any of you guys done a mouse test with a ListView before? [08:31] nic-doffay: yeah, what's up? [08:31] :D [08:31] tsdgeos has done like a bazillion of those :D [08:31] mzanetti, HAHA [08:31] tsdgeos, I'm trying to test a signal trigger, but it doesn't appear the mouseEvent is being triggered. [08:32] you're clicking? moving? or what? [08:32] tsdgeos, clicking. [08:32] tsdgeos, and moving [08:33] i'd have to see the test [08:33] moving is tricky [08:33] and we have a helper for that [08:33] tsdgeos, pastebinning it... [08:33] since flickable doesn't detect a single move as a move [08:33] clicking should be relatively straightfoward afair [08:38] tsdgeos, here's the example: https://pastebin.canonical.com/96624/ [08:38] findChild finds the ListView with the objectname listView. [08:40] nic-doffay: hmmm [08:41] now that i think [08:41] i don't think you can click on the list [08:41] i mean you can click on the list [08:41] but it won't be passed to the delegate [08:41] or maybe i'm wrong [08:41] but rings a bell [08:41] nic-doffay: can you try to find the listview children and click on it? [08:44] tsdgeos, I'll give it a try. [08:44] tsdgeos, surely you could just click the ListView though? [08:45] not sure [08:45] at least i do remember you can't do that in pure QWidgets [08:45] you have to click the listview inner widget [08:45] which is the one that contains all the crap [08:46] if you click the outer widget the fake-event gets delivered wrongly [08:46] of course that doesn't apply to QtQuick [08:46] but i fear something like that may be appening [08:46] i.e. when you click with your mouse [08:46] you don't click on the listview [08:46] you click on the window and then there's some magic that decides where to really deliver the mouse click [08:47] and i would understand if that wasn't the listview itself [08:47] so the listview is never really clicked by anyone [08:47] unless you in the test :D [08:47] this is all speculation of course, i may be wrong [08:48] let's call it "informed speculation" :D [08:50] tsdgeos, sounds like a reasonable train of thought :P [09:04] tsdgeos, trying it on the currentItem instead (which isn't null) still not getting the signal which is triggered onClicked. [09:05] nic-doffay: can you put at console.log in the qml code of the mousearea of the delegate to make sure the click is happening or not? [09:06] and not that let's say the signal is not being delivered or something [09:15] tsdgeos, it's not being clicked at all. [09:15] Just confirmed that at least. [09:16] nic-doffay: can you confirm that if you manually click it works? [09:16] i.e. add a sleep and go there while the test is running and click [09:17] tsdgeos, yeah confirmed that already. [09:18] tsdgeos, not in the test though. [09:18] I've test it externally in another qmlscene [09:18] nic-doffay: ok, if you push it to a branch i'll add it to my queue of things to look at [09:19] can't think of anything else without a closer look [09:29] tsdgeos, here's the branch: lp:~nicolas-doffay/ubuntu-ui-toolkit/list-item-option-selector [09:30] Let me know when you're able to have a look at it and I'll give you more info. [09:43] dednick: Hi! How is the messaging menu port to unitymenumodel coming along? There are some bugs in it... [09:43] larsu: cant go until we have UnityMenuAction in qmenumodel [09:44] larsu: otherwise it's pretty much ready for business [09:44] dednick: nice! I'll look at your patch right away then :) [09:49] dednick: thanks! Approved [09:50] larsu: cool. i'll get the messaging up for MP today [09:51] dednick: nice, thanks. Hopefully that'll solve bfillers bugs :) [09:56] pstolowski, ping [09:59] nic-doffay: pong [09:59] pstolowski, did you get that branch merged yesterday? [10:00] nic-doffay: nope, it needs a review [10:00] btw, https://code.launchpad.net/~stolowski/unity8/all-filter-option/+merge/182659 if anyone has a moment ;) [10:01] pstolowski, on it. [10:01] pstolowski, mind reviewing my filters MP? [10:02] nic-doffay: hmm, I'm really not much into qml [10:02] pstolowski, perhaps a functional review then? [10:02] nic-doffay: sure, that I can do, np [10:02] pstolowski, wicked ta [10:02] nic-doffay: where is your branch? [10:03] nic-doffay: or it's still the same (filter-selector)? [10:03] pstolowski, one and the same [10:03] nic-doffay: ok, thanks [10:13] * greyback moving to office to find better wifi === jibel_ is now known as jibel [10:29] nic-doffay: ok, what do i do with your branch? [10:30] tsdgeos, from root [10:30] qmake [10:30] make [10:30] then make test [10:32] That will show you which test is failing. [10:35] nic-doffay: hmmm [10:35] the only fail i can find is [10:35] FAIL! : components::TextFieldAPI::test_0_contentHeight() contentHeight by default [10:35] that's not the one i'm guessing :D [10:35] tsdgeos, try make test again. [10:35] Sometime random tests fail. [10:36] nic-doffay: same [10:36] it runs them all [10:36] that's the only one that fails [10:36] can you just tell me the name of your test? [10:37] dednick: is it on purpose that the network indicator doesn't have the mobile data checkbox any more? [10:38] mzanetti: not sure. you'll need to ask tedg. [10:40] tsdgeos, yeah sure. [10:41] tsdgeos, /tests/unit_x11/tst_components/tst_itemselector.qml [10:41] mzanetti: it's not in the spec, so i guess so. [10:41] dednick: will there be a separate page for mobile network stuff where this is in? [10:42] mzanetti: i think it will be in 'cellular settings' [10:42] dednick: ah ok. cool. thanks [10:45] nic-doffay: tsdgeos_work@xps:~/phablet/list-item-option-selector$ find -name tst_itemselector.qml [10:45] tsdgeos_work@xps:~/phablet/list-item-option-selector$ [10:45] in bzr+ssh://bazaar.launchpad.net/~nicolas-doffay/ubuntu-ui-toolkit/list-item-option-selector/ [10:45] rev 658 [10:47] mzanetti, what "stuff"? [10:47] tsdgeos, ls in the dir to confirm it's not there... [10:48] seb128: the enable/disable mobile data. [10:48] nic-doffay: tsdgeos_work@xps:~/phablet/list-item-option-selector$ ls tests/unit_x11/tst_components/tst_i* [10:48] tests/unit_x11/tst_components/tst_inversemousearea.qml [10:48] tsdgeos, it's not in the branch, you're right. Weird, let me try readd. [10:48] dednick: did you mean cellular settings in the settings app or will there be a indicator tab called like this? [10:49] mzanetti: settings app [10:49] there is a link to it at the bottom of the indicator page, but it's not connected up yet [10:49] mzanetti, dednick: it should already be in there [10:49] tsdgeos, ok added this time, must have been an issue with bzr explorer. [10:49] dednick: we also need some indicator for cellular stuff though... for example "SIM locked. Tap here to enter PIN" or something like that [10:49] if you pull they should be there. [10:50] tsdgeos, confirmed: http://bazaar.launchpad.net/~nicolas-doffay/ubuntu-ui-toolkit/list-item-option-selector/files/head:/tests/unit_x11/tst_components/ [10:50] rev 659 [10:51] mzanetti: there is no design for sep cellular page at the moment. [10:51] dednick: last time I talked to design about this they confirmed that we will have such a thing. [10:53] dednick: https://bugs.launchpad.net/touch-preview-images/+bug/1185056 [10:53] Launchpad bug 1185056 in Ubuntu UX "phone signal indicator" [Undecided,New] [10:55] mzanetti: there is a cellular icon on the network indicator. but the multi-icon support hasnt landed for unty8. tsdgeos < [10:55] mzanetti: if you have time for it [10:55] it'd be awesome if you can have a look [10:55] I see... so it's only a separate icon for the network indicator tab [10:55] tsdgeos: is there already some branch or anything? [10:56] mzanetti: dednick's [10:56] mzanetti: i think the pages are intended to hold all "network" related things. [10:56] mzanetti: https://code.launchpad.net/~nick-dedekind/unity8/indicator.multi-icon/+merge/181862 [10:56] s/pages/network page [10:57] including SIM lock/No SIM, as theres also a label for that next to the network icon. [10:59] mzanetti: ok, just reading the spec now. "If the SIM is locked, the “Network” menu should include an “Unlock SIM…” item immediately before “Cellular settings…”" [10:59] mzanetti: although i guess that's not finished as we dont have any system dialogs. [11:04] dednick: ok... yeah. I don't need it urgently... There's still a lot to do in the background for the SIM pin stuff. I was just checking if everything fits together [11:05] mzanetti: it should just go when the background stuff is finished. dont think there's any work for unity8 === hikiko is now known as hikiko-lunch [11:06] dednick: well, the greeter and unity need to talk to the indicator service to check that value. [11:46] nic-doffay: is this your test? http://paste.ubuntu.com/6043795/ [11:48] tsdgeos, that's correct but test_delegate_clicked_signal isn't being run for some reason. [11:48] which is the offender. [11:49] tsdgeos, judging by your pastebin at least. [11:50] hmmm [11:50] there is ./tests/unit/tst_components/tst_listitems_itemselector.qml and ./tests/unit_x11/tst_components/tst_itemselector.qml [11:50] is that ok? [11:51] lunch! === _salem is now known as salem_ === hikiko-lunch is now known as hikiko [12:02] greyback: do you know whom I need to talk to for info about click packages? === alan_g is now known as alan_g|lunch [12:05] mzanetti: jdstrand good person to talk to [12:05] greyback: thanks [12:06] unity8 with qt5.1 still flickering after reboot for me, trying another dist-upgrade and reboot [12:15] tsdgeos, yeah for now. [12:15] tsdgeos, itemselector is the one you're after though. [12:17] * greyback rebooting [12:24] Mirv: hey, having dist-upgraded, rebooted and tried qt5.1 with and without xmir, I still see the flickering on rendering. It's weird, we have the same graphics chip, why do I have problems? [12:42] hm, no boiko today? [12:43] larsu, it's ~9am for him and it's friday, maybe he's not an early morning guy [12:44] seb128: that makes sense :) I'll try to catch him later today [12:51] greyback: now on X11 (I broke my XMir today) I see it! and I'm not sure how it'd be with the new XMir multimonitor and composite bypass that came yesterday/today - how I ran it was with XMir and the old style mirror-mode (that wasn't slow, but no longer available) - maybe it does some syncing that doesn't happen here [12:52] Mirv: hmm, ok. I'm currently using XMir. I tried both Compiz and Metacity, same problem. I can try rebooting into non-xmir? [12:53] greyback: re-read (if I'm not reading wrongly:) - so on "old" XMir I didn't have it, now on X11 I do have the flicker [12:54] Mirv: okay. So we're finding it with both Xmir and without it. Maybe mesa or driver change? [13:01] nic-doffay: something's not cool [13:01] nic-doffay: [13:01] file:///home/tsdgeos_work/phablet/list-item-option-selector/tests/unit_x11/tst_components/tst_itemselector.qml:32:10: ListItem.ItemSelector is not a type [13:01] ListItem.ItemSelector { [13:01] ^ === alan_g|lunch is now known as alan_g [13:03] ok [13:03] now [13:03] got it to run [13:05] nic-doffay: dude, i just added a wait(10000); to your test [13:05] and it's a empty white window [13:05] where am i supposed to click? [13:13] greyback: maybe, it's hard to say but it was sure there was no flicker earlier this week with earlier XMir with my "dummy" mirrored display setup [13:13] hehe, earlier this week, I mean I guess this was yesterday we talked?-) [13:14] Mirv: correct :) [13:14] the days are tiny bit long [13:14] I wish I had that complaint [13:14] mzanetti: have you had time to have a look dednicks' branch or should i? [13:22] tsdgeos, yeah I just saw that after adding a wait too. [13:27] tsdgeos: no, I didn't [13:27] mzanetti: ok [13:40] tsdgeos: mzanetti nic-doffay dednick dandrader paulliu ....just a heads up forgot to mention its a holiday in US on monday (so no mterry racarr kdub or kgunn) [13:40] greyback: ^ [13:40] oki [13:41] Labor Day [13:41] kgunn, mterry enjoy! [13:41] it's always cool how the "work day" is a holiday :D [13:41] kgunn: noted, thank you [13:42] kgunn, ok, enjoy your long weekend! [13:44] funny that's it's no on May 1st... [13:46] dandrader: if you read the wikipedia it explains it is because 1st may is associated with commies :D [13:47] tsdgeos, yeah, I was reading that page :) [13:47] larsu: hi [13:48] boiko: good morning! [13:48] larsu: morning! :) [13:49] boiko: I'm having trouble reproducing bug #1217676 and I was wondering if it could be a problem in the phone app [13:49] bug 1217676 in indicator-messages (Ubuntu) "performing action on menu incorrectly clears and reloads menu" [High,Incomplete] https://launchpad.net/bugs/1217676 [13:49] boiko: can you reproduce it? Or is there a way to insert fake messages into telephony-service? [13:50] larsu: ah, that bug is not on indicators, it is on my side, can you set the project to affect telephony-service and reassign it to me? [13:50] paulliu: seen the CI branch of the movie thing just failed? [13:50] boiko: oh, you already know it. Awesome!! [13:51] larsu: yeah, it was a limitation on the previous implementation, but that doesn't happen anymore [13:51] tsdgeos: let me check. [13:52] boiko: ah, cool. Will assign to you. Happy Friday :) [13:52] * larsu feels a bit bad wishing boiko a happy Friday after assigning a bug to him [13:53] larsu: haha, well, it is a bug I can already close, so you are fine :) [13:53] larsu: happy friday for you too! ;) [13:53] boiko: thanks :) [13:56] tsdgeos: /tmp/buildd/unity8-7.81.3+13.10.20130829.2/Dash/Movie/MoviePreview.qml:18,1: module "QtMultimedia" is not installed [13:56] tsdgeos: Maybe I need to modify the debian/control for B-D. [13:56] ok, do it then :-) [13:57] mzanetti, is there a way to tell libusermetrics not to show any user-private info? (is it all user-private?) [13:58] mterry: I have no clue about libusermetrics. pete-woods1 might be able to answer that I guess [13:58] mzanetti, oh sorry, I don't know why I misremembered your involvement :) [13:58] pete-woods1, same question to you : ) [14:01] mterry: in the greeter session, can I access dconf to read default values? [14:02] mterry: as they should be the same for all users I would assume that should work... even if we read defaults for user lightdm [14:02] mterry: as in stuff like, battery usage, etc? [14:02] mzanetti, yeah [14:03] mzanetti, you're not using dconf directly are you? via gsettings right? [14:03] mterry: gsettings-qt, yes [14:03] pete-woods1, yeah or how many calls or something. There's a privacy setting that says "don't show stats for me in welcome screen" or something [14:03] pete-woods1, I'm looking at how to implement it [14:04] mterry: I could certainly make it read the "don't collect info about me" setting? [14:05] mterry: that way it would never even get into the service, and i.e. stopped at the input API level [14:07] pete-woods1, well no... I bet it should be on output side, since if user switches this button after a month, they don't want to reveal the past month either [14:10] mterry: okay, is this a systemwide setting, or a user setting? [14:11] and would the sort of user who turns that on, also want the collected data removing? [14:11] pete-woods1, a user setting stored in AS so the greeter can know per-user who has it on [14:12] pete-woods1, https://wiki.ubuntu.com/SecurityAndPrivacySettings#Phone [14:12] pete-woods1, the setting only talks about showing stats on welcome screen [14:12] pete-woods1, maybe the stats could be useful elsewhere? [14:12] so we shouldn't just kill them... [14:12] pete-woods1, plus... if they user turns off the setting in a week, they might want the back dtaa [14:13] mterry: okay, well I don't have strong feelings either way there [14:13] mterry: do you want me to get the infographics output API to read that setting, then? and hide all the user-specific data? [14:14] pete-woods1, well, libusermetricsoutput doesn't know if it's in welcome screen or not. Maybe just add a set_show_private_info(bool) method or something to the api? [14:14] mterry: that sounds like a sensible solution [14:15] pete-woods1, what sort of data is collected that isn't private? [14:15] pete-woods1, or is this just a way to say "turn off entirely"? [14:15] mterry: well at the minute, there's only the camera app, for some reason I haven't had the go ahead to advertise the input API [14:16] so in reality it's just the "videos recorded" and "photos" captured that would dissapear [14:16] pete-woods1, well there's no way to mark anything in the input API as "private" or not, eh? [14:16] pete-woods1, and the output library doesn't have sources of data besides the input API? [14:17] i.e. everything collected is presumed private... [14:17] hey guys... I keep getting random behavior regarding frame_extents and ConfigureNotify [14:17] no matter how hard I try to fix the toolkit, when I think it's all working, the weirdest behavior happens [14:17] mterry: indeed, there was no distinction between a user's different data sources, I figured anything about a particular user was "private" [14:18] mterry: some point soon, I'd really like to take the time to connect it to upower, etc for system-type data [14:18] which, I *think* is perhaps what's meant by non-private? [14:19] pete-woods1, probably. OK. So there is theoretically room for non-private data, so the API would make some sense [14:20] mterry: agreeed [14:26] tsdgeos, sorted out the rest of the problems, that wait was handy. However I'm having issues using consecutive mouseClicks. The signal is emitted only after clicking when the list is expanded. I can the list expanding, however I can't see the affect of another click. === dandrader is now known as dandrader|afk [14:32] nic-doffay: not sure i understood that sentence [14:32] tsdgeos, nm think I wasn't estimating the distances correctly. [14:41] mzanetti: https://code.launchpad.net/~mzanetti/unity-api/application-api/+merge/182692 didn't land. But before I mark approve, I suspect it's missing a pkgconfig file === alan_g is now known as alan_g|ea === alan_g|ea is now known as alan_g|tea [14:44] mzanetti: well I'm more confused. There are already pkgconf files for unity-shell-{launcher,notifications}, plus a more general one libunity-api.pc which actually includes linker info. [14:44] Wellark: could it be that QIcon::fromTheme isn't doing the fallbacks according to spec? [14:45] mzanetti: so honestly I think all I need is libunity-api.pc - but I don't get the point of the others [14:46] Wellark: like, it doesn't turn a-b-c into a-b if a-b-c isn't available [14:46] greyback: not sure I can follow === dandrader|afk is now known as dandrader [14:47] mzanetti: in the unity-api-dev package right now, there are 3 pkgconfig files. One general libunity-api one, but also 2 more, one for unity-shell-launche, one for unity-shell--notifications [14:48] mzanetti: I'm wondering if we're expected to add one for unity-shell-applications [14:48] mzanetti: but mainly I don't see the point of the 2 unity-shell-* pkgconfig files at all [14:50] greyback: it should be autogenerated [14:50] mzanetti: aha, really? I didn't realise [14:50] greyback: the point of having them split is that they will be packaged independently [14:50] so you could only pull single plugins [14:51] mzanetti: okay. Sometimes I think packages can be too small, but anyway :) === alan_g|tea is now known as alan_g [14:51] greyback: :D [14:59] tsdgeos, if you have a moment do a pull on my branch and attempt to run that test. You'll notice that no matter which of the two selectors you input for the mouseClick the same selector is activated. Strange. [14:59] it's easily visible with a wait. [14:59] It's never the right one though? [15:00] You'll notice that mouseClick(selector, 0, 0, Qt.LeftButton); [15:00] selector is the one with the TEST label. [15:00] You'll see the other is always triggered? [15:02] greyback: also, when you're editing debian/control. it's a bit tricky... check out how the launcher does it === boiko_ is now known as boiko [15:23] sil2100, about the intel box, which kernel is on the host and in the container? [15:25] jibel: let me check to make sure, the container has 3.11.0-4.9, the same for the intel box [15:26] Wellark: hey, just wondering if that desktop file parser library might be arriving any time soon? [15:58] greyback: by Tuesday [15:59] Wellark: oh cool :) === dandrader is now known as dandrader|lunch [16:37] * greyback eod [16:44] So apparently adding a wait() to batch test creation solved a fail although it worked perfectly in qmltestrunner. wtf of the day. === alan_g is now known as alan_g|EOD [17:29] larsu: ping === dandrader|lunch is now known as dandrader [17:57] seb128: do you know if this is the right place to put the default config for the launcher items? https://launchpad.net/gsettings-ubuntu-touch-schemas [18:00] mzanetti, I would make https://launchpad.net/ubuntu/+source/ubuntu-settings build a ubuntu-touch-settings [18:02] seb128: hmm... so there is already a default for the launcher items. but we need to have an override for the phone. [18:03] mzanetti, right, I'm about to kick a discuss about the 'default/value per form factor', I was planning to send that email after dinner, I can Cc you on it if you want [18:04] seb128: ah, interesting. [18:05] seb128: fyi: ssweeny has prepared this: http://bazaar.launchpad.net/~ssweeny/+junk/ubuntu-defaults-touch [18:05] seb128: that would create overrides for the other defaults. but we're unsure if this is the right way [18:05] seb128: can you put us both on CC? [18:06] mzanetti, sure can do [18:06] seb128: background is, the delivery team needs yet another set of overrides, for OEM integration. so such overrides seem sensible to me. but I'm not sure where to put them [18:06] mzanetti, that package looks like it's done using the ubuntu builder tools [18:06] that's for derivatives usually [18:07] it might be too much for us [18:07] we aim at convergence [18:07] so we should have that much difference [18:07] need to go for dinner [18:07] but I'm going to move that to the email discussion [18:07] thanks. enjoy the dinner [18:14] dednick: what's up? [18:15] larsu: hey. just sent you a email. indicator-messages [18:15] state not coming out a UnityMenuModel because the parameter type of the root action is set to a{sv} [18:16] dednick: just read the mail. I think you're right [18:16] larsu: also, reply not workig for me anymore. did something change in how we activate? [18:16] dednick: not that I know of [18:17] muhahah I can blame ted for that one :D [18:18] larsu: i'm getting telephony-service-sms.msg-actions.XXXX.quickReply in the xCanonicalMessageActions field, but it seems to be a non-existing action. [18:18] dednick: do you have a branch I can try out on my desktop? Lots of stuff has changed in messages while I was gone. I don't have a good overview yet. [18:18] larsu: how do you get a message on your desktop?! [18:18] that would so help debugging. i'm resorted to using phone [18:18] dednick: by running unity8 in a window :) [18:19] larsu: yeah, but how do you get a txt message? [18:19] dednick: without the phone app. I use a python script [18:19] ah [18:19] give me a sec and i'll get a branch [18:20] looks like there's an "indicator." missing in front of that action name [18:20] I wonder why though, unitymenumodel can handle the action-namespace stuff [18:20] I'll figure it out though, don't worry about it [18:21] but maybe on Monday ;) [18:21] larsu: lp:~nick-dedekind/unity8/unitymenumodel.messages [18:22] larsu: yeah, no worries. I'm off now anyway [18:22] dednick: have a nice weekend! [18:22] you as well [18:22] thanks :) [18:46] * mzanetti is surprised that image url's like this actually work: "image://gicon//usr/share/calendar-app/calendar64.png" [18:49] mzanetti: ya, please don't use them :) [18:52] larsu: I thought the same, but the more I think about it, the less convinced I am. [18:53] I mean, this doesn't work by accident. the one implementing this image provider must have thought about this and catches it correctly [18:53] was it you? [18:54] noooooooooooo [18:54] let's forget about image://gicon [18:54] it died [18:54] it's rotting [18:54] you can smell it [18:55] it's awful [18:55] remove it [18:55] mzanetti: no, but I'm the one that's on a quest to get rid of it :) [18:55] * larsu pats mhr3 on the back [18:55] wait. so we're getting rid of the whole gicon provider? [18:55] yes [18:56] oh... what should I use then? [18:56] file:///.... [18:56] what kind of icon do you want to show? [18:56] file icon? file:// [18:56] larsu: dialer-app [18:56] themed icon? image://theme/ [18:56] for example [18:56] the app icon? [18:56] yes [18:56]