[07:05] morning [08:39] larsu: ping [08:57] greyback: ping [08:57] greyback: https://code.launchpad.net/~dandrader/unity-api/splash/+merge/233288/comments/575802 [08:57] shall we wait for daniel or throw the splash stuff out of the silo and try to get orientationlock landed alone? [08:59] mzanetti: I say wait. [09:00] ok [09:10] tsdgeos: hey [09:10] larsu: commented on the bug, emblems was the wrong thing that is missing, is attributes icons [09:11] tsdgeos: okay thanks. I'll have a look [10:17] dandrader: good morning [10:17] dandrader: can you update this one please? https://code.launchpad.net/~dandrader/unity-api/splash/+merge/233288/comments/575802 [10:18] mzanetti, what should I exactly do? [10:18] dandrader: bump the changelog in unity-api [10:19] dandrader: and then update the version in unity8's debian/control to match that version [10:19] dandrader: example: https://code.launchpad.net/~mzanetti/unity-api/dashActive/+merge/235273 [10:19] dandrader: https://code.launchpad.net/~mzanetti/unity8/async-dashcommunicator/+merge/235272 [10:20] dandrader: https://code.launchpad.net/~unity-team/qtmir/fix-some-lifecycle-bugs/+merge/234087 [10:20] so I guess you need to change the depends in the related qtmir branch too [10:29] mzanetti, all done [10:30] dandrader: thanks === MacSlow is now known as MacSlow|lunch [11:09] dandrader: heya, about the qWarning/qCWarning [11:10] there are some messages that I think should not be categorized, in case there's a filter in place that surpresses all categorized messages [11:10] greyback_, I commented on the mp [11:11] greyback_, you can suppress only the debug messages for instance. like "qtmir.*.debug=false" [11:11] dandrader: sure, but what if we ship qtmir.*=false - which I think is reasonable to do [11:11] since nearly all of our debug/warning output is informational at best [11:12] greyback_, say I want to silence all qtmir messages, *including warnings*, because it's bitching about something I don't care right now as I'm working on some unity8 thingy. I wanna be able to do it [11:13] greyback_, we are only losing flexibility by using qWarning instead of qCWarning [11:13] dandrader: sadly I don't care about you that much :D I'm more concerned with a user's desktop, they tap an icon and nothing happens because the desktop file is missing - I want to be sure that message is printed somehow [11:13] greyback_, we can do less with it [11:14] greyback_, then don't silence qtmir warnings on the final configuration. or make it a critical message [11:14] and don't silence critical messages :) [11:15] greyback_, what if we have -DQT_NO_DEBUG and I want the user to know that a desktop file is missing? [11:15] dandrader: I don't like it, it relies on the right configuration being set, which I think is too easy to get wrong. I think some messages should not be filterable [11:15] greyback_, should we use std::cout for it then? [11:15] dandrader: but we don't set that. [11:15] we have control over the compile flags for our app [11:16] we don't have control over the warning/debug config [11:18] dandrader: shall we ask a third party? To give a deciding vote? [11:19] greyback_, yeah. I don't think your reasoning for using qWarning instead of qCWarning makes sense [11:20] dandrader: who would you like to nominate? [11:20] mzanetti [11:21] mzanetti: you have been chosen to enact a peace treaty between myself and dandrader! [11:21] when you've 2 mins, let us know [11:22] Holas [11:22] o/ [11:22] :D [11:22] greyback_: shoot :D [11:23] mzanetti: topic is qWarning versus qCWarning - and the Debug equivalents [11:23] I tend to use qWarning for messages that I think are of interest to a final user - i.e. they are messages which cannot be filtered [11:24] dandrader: want to state your case? [11:25] A warning is a warning. The qC version of log messages (debug, warning and critical) give more flexibility as you can filter them by level and category [11:26] whereas with the original versions you can either install a message handler (thus filtering only by level) or disable all of them (by adding -DQT_NO_DEBUG) [11:27] if someone wants to override any log filtering then it should not use the logging system in the first place. [11:27] so Imo we want to see real warnings in the logs without messing around with filters... [11:27] so I guess if we have a default filter that allows some certain category [11:27] then I would probably align it for qCDebug [11:28] but really depends on the default installed filter I guess [11:28] there probably is value to still suppress those warnings if you want to debug something else... [11:29] greyback_ wants to give a message to the *final user*. Is loggging the right medium for it? [11:29] all I want is to be able to print stuff, and be reasonably certain it is actually printed (-DQT_NO_DEBUG and custom message handlers aside) [11:29] dandrader: I guess "final user" is QA in this case [11:29] if you ask QA to send you a log, what do you want to have in there without asking them fiddle with filters [11:30] * mzanetti still didn't fully understand the QMessageLogger filtering [11:30] looked a bit weird the one time I tried tbh [11:31] I had planned to ship qtmir filtering everything, since nearly all the messages are informational. And just use unfiltered messages for things user would care about [11:31] sounds reasonable [11:32] well, either that, or defining a category that is (and stays) enabled [11:32] if you create a category that is always enabled, then why bother making it a category at all? [11:32] I tried to shutdown all the qtmir babbling and filtered out all of it. but the babbling continued, although on a lower volume, because random parts of its logging use the old qMessage instead of qCMessage. so the only way to shut it off was to -DQT_NO_DEBUG it, but that meant recompiling it all [11:33] dandrader: random parts? Have you pastein, I'd like to see what comes out. [11:33] but them I lost my own logging which I was interested into. So I had to resort to use plain printf () or std::cout ofr it [11:33] for it [11:34] greyback_, just grep for qDebug, qWarning etc in qtmir [11:34] well, we all agree that plain qDebug() should eventually be migrated to qCDebug [11:35] its really just about error cases in which you are interested always (99% of the cases) [11:35] dandrader: well sure. But I was more interested in the realistic output [11:35] greyback_, I even had a branch to change it all to qCStuff, but for some weird reason it made application_manager test crash on destruction :-/ so will have to get back to it latedr [11:35] eww [11:35] that's unexpected [11:35] anyhoo [11:36] * greyback_ rests his case and awaits the right honourable judge mzanetti's verdict [11:37] if I have to judge something I guess I'm with Daniel :D better flexibility in the long run. a little more effort to make it clean now though [11:37] http://tinyurl.com/ldtjvmj [11:37] aside from losing flexibility when using qWarning instead of qCWarning, we also end up with a heterogeneous logging [11:38] oh, there should be a utf8 char with a judge's hammer [11:38] haha [11:38] true [11:39] * greyback_ files an appeal [11:39] popey: is there ^ [11:39] ? [11:39] j/k [11:39] I'm sure popey will figure something :D [11:39] just wish I shouted OBJECTION at some stage [11:39] always wanted to do that [11:40] if someone decides to filter out all warnings it's because he does not to see any warning messages [11:40] :D [11:40] hmm [11:40] if you want to still print out your message even if the user disabled warnings, then your message should not be a warning [11:41] (╯°□°)╯︵ ┻━┻ [11:41] we should print that [11:41] Mirv: how is https://bugs.launchpad.net/ubuntu/+source/qtbase-opensource-src/+bug/1371472 going? [11:41] :D [11:41] Launchpad bug 1371472 in qtbase-opensource-src (Ubuntu) "Fix crash in QNetworkAccessCacheBackend::closeDownstreamChannel" [Undecided,New] [11:43] dandrader: ok pushed [11:43] greyback_: now one for you: http://bazaar.launchpad.net/~macslow/unity-notifications/dont-filter-newline/revision/219 vs http://paste.ubuntu.com/8402690/ [11:44] greyback_: first one fails with new tags (e.g. "div"). latter one fails with broken xml [11:46] mzanetti: regexp to parse html is bad practice. Using the XML parser is better IMO [11:46] +1 [11:46] however, it fails with this for example: "So broken" [11:46] greyback_: while the regexp actually does it right [11:47] mzanetti: is broken input likely? [11:47] hmm... notifications [11:47] if it's broken going in, why should we try fix it? [11:47] crap goes in, crap comes out [11:47] I tend to agree... [11:47] greyback_, thanks. your thoughts on these: https://code.launchpad.net/~gerboland/qtmir/desktopFileReader/+merge/235223/comments/575368 https://code.launchpad.net/~gerboland/qtmir/desktopFileReader/+merge/235223/comments/575362 [11:48] dandrader: second one I've addressed - I just removed the comments [11:48] dandrader: first one is doing what it's supposed to do IMO - but message is categorized now [11:49] * greyback_ doesn't mind noisy tests [11:50] greyback_, ok, will make a separate MP on that subject later. this review was contentious enough already [11:50] dandrader: :D [11:50] always good to have a third party to settle the wa [11:50] +r [11:52] tsdgeos: gonna add a landing for it now. should it go to rtm first? [11:55] well there's zero silos available for either utopic or rtm [11:58] I think utopic is frozen at the moment, so rtm first might be the way to go === _salem is now known as salem_ === salem_ is now known as _salem === MacSlow|lunch is now known as MacSlow === boiko_ is now known as boiko === _salem is now known as salem_ [12:53] OK giving Unity a shot (Ubuntu user since 4.10, XFCE user since sundown of Gnome 2.32...) - Are silly gadgets like CPU utilization and thermal monitors coming to Unity at any point? === Malsasa_ is now known as Malsasa [13:11] tedg, hey there... [13:14] MacSlow, Howdy! [13:15] tedg, I was wondering if you meant gnome-settings-daemon in you comment here... https://code.launchpad.net/~macslow/indicator-sound/synchronous-notification/+merge/229061/comments/575109 [13:16] MacSlow, I think I meant unity-settings-deamon? I mean th g-s-d fork that we did. [13:17] seb128, Could probably help with the name ^ [13:17] It's u-s-d, not sure what the U is :-) [13:17] tedg, name is correct [13:17] tedg, https://code.launchpad.net/unity-settings-daemon [13:17] MacSlow, seb128 is your hero! ^ [13:17] :-) [13:17] tedg, just trying to figure out where to look for the notification-relevant code... and compare to see if there are potential issues or not for your described scenario [13:18] seb128, thx... now that you said it, it seems obvious :) [13:19] tedg, anyway I will look into it and provide my findings as feedback on the MP === dandrader is now known as dandrader|afk [13:25] seb128: hey, here's something for you: https://code.launchpad.net/~mzanetti/ubuntu-system-settings/update-reset-launcher/+merge/234309 [13:26] mzanetti, they, thanks, adding to my review list [13:26] seb128: one note: the required change in unity-schemas did not make it to rtm yet [13:26] seb128: its released to proposed though [13:26] k [13:29] mzanetti, we don't publish the launcher config anymore for the greeter/lock screen benefits? [13:29] seb128: not atm, as we don't have split greeter anyways currently [13:29] seb128: that will come back, but the main config will stay dconf [13:29] mzanetti, "currently" [13:29] seb128: accountservice will be a read-only copy of the current state inside the user account [13:30] right, which means we should still write there no? [13:30] if you reset dconf, the user's launcher plugin will notice and sync over to AS [13:30] that's the plan at least [13:31] k [13:35] tedg, see https://code.launchpad.net/~macslow/indicator-sound/synchronous-notification/+merge/229061 and hopefully approve that MP if you're happy with it. Thanks in advance! [13:37] MacSlow, Okay, so notify-osd can handle the fact that they're two synchronous ones? [13:37] there are [13:38] tedg, yes... to notify-osd it just two sync. notifications coming in... and first come first served [13:38] MacSlow, K, cool. [13:42] tedg, btw... volume-notification from those branches at work here... https://www.youtube.com/watch?v=rwEYLXNaL0Q === dandrader|afk is now known as dandrader [14:04] kodiak11, there's indicator-multiload [14:06] dandrader: ah - cool - I see I need to add repos - I poked around but was a bit lost [14:09] mterry, could you take a look at (review) https://code.launchpad.net/~macslow/unity-notifications/synchronous-notification/+merge/234476 when you have a free slot? [14:10] MacSlow, what is the value hint about? [14:11] mterry, it's a way to pass a 0-100 number to a notification in order to make it display a indicator-bar (think: volume, brightness etc) [14:11] MacSlow, I'll look at it today, sure [14:12] MacSlow, it doesn't need a x-canonical prefix? [14:12] mterry, great thanks... it's one of three related branches... but I don't want to dump all three on one person... they can be dealt with on an individual basis [14:13] mterry, that has legacy-reasons [14:14] mterry, if the hint-name would be changed it would break existing apps using synchronous notifications with attached values. [14:15] MacSlow, ah ok, this is an old hint [14:15] mterry, correct [14:18] mterry, fyi... this is what it looks like when all three related branches are at work... https://www.youtube.com/watch?v=mi-Z2Rn7Fxo [14:19] MacSlow, nice [14:20] dandrader: I found this and thought it was useful - thanks for the lead! http://askubuntu.com/questions/30334/what-application-indicators-are-available [14:21] kodiak11, yw [14:24] saviq and thostr still on holiday? [14:30] greyback_, a tiny one: https://code.launchpad.net/~dandrader/qtmir/optionalTestsBuild/+merge/235467 [14:31] mhall119: saviq is anyway === jamesh_ is now known as jamesh [14:43] Cimi: any chance of that card-visual-tweaks branch landing this week? [14:44] greyback_, thanks for the quick review! [14:44] dandrader: does the job :) [14:46] mzanetti: when you get time, I left comment here: https://code.launchpad.net/~mzanetti/unity8/async-dashcommunicator/+merge/235272 [14:46] greyback_: I have to keep the dep to AppManager [14:46] hmm... maybe not [14:46] mzanetti: why> [14:47] but seems fragile [14:47] greyback_: because if I connect to the interface when the app is suspended, it'll hang unity [14:47] greyback_: so I use AppMan to make sure it is in state running [14:47] mzanetti: it should not hang with an async dbus call [14:47] greyback_: no, but the connect [14:48] hmm [14:48] lemme try something [14:48] ack [14:50] mhall119: anything we can help you with while Saviq is enjoying his holidays? === MacSlow is now known as MacSlow|afk [15:36] mterry: what can we do about https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1372515 ? [15:36] Launchpad bug 1372515 in unity8 (Ubuntu) ""Sorry, incorrect passcode" doesn't fit in spanish translation" [Undecided,New] [15:37] tsdgeos, hrm [15:37] tsdgeos, make it smaller seems least intrusive -- I don't want to drop the whole pad down to fit a second line [15:37] tsdgeos, or have translators test their translations and get creative to let them fit... but we can at least try a technical solution first [15:39] mterry: you're not going to get translators to test the translations [15:39] tsdgeos, :) [15:39] you have to be god to be were a string is [15:39] tsdgeos, I know, they just do it in LP [15:39] and even if they wanted [15:39] tsdgeos, true [15:39] unless you know how to code and check the code and try to infer where the string is ending [15:39] sometimes is a huge amount of work [15:39] not in this case, but [15:40] mterry: i think having that "shrink my size label" in the SDK would be a great idea actually [15:40] for cases where not much can be done [15:40] and you don't really mind if it's 17px or 15px font size [15:41] tsdgeos, yeah... I'm not sure what font metrics info we have available to us to do that in qml, but I'm sure it's possible [15:41] mterry: we have the implicitWidth [15:41] oyu can try playing with it [15:42] tsdgeos, yeah but I don't want to have to set the size, test width, change size, etc. We'd want to assign size once based on the metrics of the string and width available, right? [15:42] mterry: yeap, i am not sure there's something like that available [15:43] :-/ [15:43] tbh i have some code that does that in kde [15:43] tsdgeos, maybe we can change size fast enough that it doesn't matter -- or just keep the label invisible until we settle [15:43] and it actually ends up setting sizes and chaging [15:43] since fonts are not linear [15:43] in point size vs size [15:43] it seems to be the only way to do it [15:46] man qmluitests have gone all crazy :S === MacSlow|afk is now known as MacSlow === alecu is now known as alecu_afk [16:15] * greyback_ annoyed unity8-autopilot brings in lots of qt4 === alecu_afk is now known as alecu === gatox is now known as gatox_lunch === dandrader is now known as dandrader|lunch === dandrader|lunch is now known as dandrader === gatox_lunch is now known as gatox === maxb_ is now known as maxb [18:00] * greyback_ eod [18:04] Is there a quick guide on customizing / making the most use of Unity for new users to Unity but obviously not to Linux [18:04] Coming from the XFCE world and attempting to give Unity a fair shake [18:14] kodiak11, #ubuntu-desktop is the appropriate channel for that. Here we essentially talk about Unity for phone an tablet [18:15] until that new codebase also comes to encompass the desktop as well [18:15] oh snap [18:16] when are you giving me a dang phone??? (US Verizon customer_) [18:36] Hi guys my new hobby is https://bugs.launchpad.net/ubuntu/+source/webbrowser-app/+bug/1328839 [18:36] Launchpad bug 1328839 in QtMir "The top panel is not restored after toggling full screen off" [High,Confirmed] [18:36] so first I am trying to figure out, how a Qt app is supposed to escape this fullscreen state...and it seems like [18:36] platform api is missing a counterpart to [18:36] ua_ui_window_request_fullscreen [18:36] e.g. there is no ua_ui_window_leave_fullscreen or ua_ui_window_show_normal or whatever [18:37] does anyone know otherwise, or know how an app is supposed to leave [18:37] fullscreen [19:48] Hello, can I have a non-breaking space under [Shift [Space]]? [19:51] Can I have a non-breaking space under [Shift [Space]]? === salem_ is now known as _salem === _salem is now known as salem_ === salem_ is now known as _salem === _salem is now known as salem_ === salem_ is now known as _salem