**** BEGIN LOGGING AT Thu Sep 22 02:59:59 2016 Sep 22 03:50:19 Thanks, Herrie, that will save many app authors (and users trying unmodified webOS apps) many hours of puzzlement. Sep 22 03:53:27 I installed yesterday’s testing build, and http://issues.webos-ports.org/issues/1177 was still there. Has the fix not been merged yet? Sep 22 04:15:19 DougReeder: Will be in the next nightly. It was merged just after previous nightly was run Sep 22 04:15:45 * DougReeder nods Sep 22 04:43:56 DougReeder: https://github.com/webOS-ports/luna-appmanager/pull/14 Sep 22 04:44:04 Waiting for Tofe or nizovn to review Sep 22 04:44:10 But tested it and was working Sep 22 04:44:23 Just my C++ is no good, so I prefer them to check it before I cause memory leaks etc :P Sep 22 04:56:07 * DougReeder nods Sep 22 05:11:23 Though it should be straight forward for this one ;) Sep 22 05:11:35 * DougReeder nods Sep 22 05:34:41 Herrie: merged, no leaks ;) Sep 22 05:42:41 Tofe: That's what I thought, but I prefer not to touch C++ :P Sep 22 05:42:49 But this was straight forward enough :P **** BEGIN LOGGING AT Thu Sep 22 10:10:23 2016 Sep 22 11:20:11 JaMa: ping Sep 22 12:39:19 Andolamin: ping Sep 22 13:49:33 Tofe: hey, I saw you guys imported some stuff from the Ubuntu UITK :) Sep 22 13:49:50 great to see! Sep 22 13:55:02 faenil: yes, we took some pieces, when it corresponded to our needs, and when there wasn't too many dependencies on other Ubuntu components Sep 22 13:55:29 Tofe: cool, glad to hear. I don't know if we chatted since then, but I joined Canonical more than a year ago Sep 22 13:55:50 so if you guys want to cooperate or help out or join, feel free to talk to me :) Sep 22 13:56:06 faenil: oh, no, I didn't know, I just saw you were still around a lot :) That's great! Sep 22 13:56:13 we're trying to cooperate with Kirigami guys as well (the KDE plasma QML tookit) Sep 22 13:56:40 so it would be great if could all eventually end up contributing to QtQuickControls2 upstream Sep 22 13:56:47 and all base on that (+ platform-specific components) Sep 22 13:56:54 that's the future of Ubuntu UITK, at least Sep 22 13:57:08 (base on QQC2 to cooperate with upstream and other toolkits) Sep 22 13:57:29 Yes, that would be a good thing; today we didn't really use QtQuickControls1 heavily, because we needed each time to style it a lot Sep 22 13:59:11 Also for us a big part of our apps is HTML5+CSS, only some core apps are QML (phone, compositor, keyboard, browser) Sep 22 14:00:45 Tofe: yeah and QQC1 is quite heavy, performance wise Sep 22 14:01:10 Tofe: I see, I thought your goal was to use QML as much as possible, I think I misunderstood :) Sep 22 14:02:30 Well, QML is nice, and we can use QML apps without issue; but LuneOS (as was webOS) is based on HTML5/CSS/JS apps, fundamentaly Sep 22 14:04:09 yeah I am aware Sep 22 14:04:09 Ideally, we could re-package and re-use QML apps from other sources (Ubuntu, Plasma...) theme it for LuneOS, and use it as-is Sep 22 14:04:20 I thought one of the goals was to change that :) but I clearly misunderstood Sep 22 14:04:37 Tofe: yes that would be great as well Sep 22 14:04:47 Tofe: nemo would need the same (if it ever comes back to life) Sep 22 14:06:37 No, currently we don't have a special focus on QML apps, though they are welcome too. Simply, apps devs around here are used to Enyo JS, so QML doesn't come to the minds :) Sep 22 14:07:17 (For me it's a bit special, as I don't have any special webOS background, so I'm far more at ease with QML than with Enyo) Sep 22 14:12:54 Tofe: I understand how faenil could think that, given the preference for QML I have seen in this channel Sep 22 14:16:58 GodGinrai: yes, I can understand it too; but in the end, apart for the phone app, the only parts that are in QML today are the ones that were in Qt C++ in webOS Sep 22 14:18:26 I don't deny I prefer writing in QML than in Enyo, but that can be explained by the fact that I don't know how to write an Enyo app :) Sep 22 14:19:43 Actually, I prefer Enyo to QML :P Sep 22 14:21:12 :) Sep 22 14:21:31 it would be interesting to see a performance comparison Sep 22 14:22:25 Oh I'm sure that QML is more performant Sep 22 14:22:37 JS is slow Sep 22 14:23:06 to be fair, one should also compare the resulting ergonomics, which is certainly more consistent with Enyo, and less dependent on the dev's skills Sep 22 14:23:07 but the thing that's nice about Enyo and web apps is the ease of coding the UI Sep 22 14:36:02 GodGinrai: QML uses JS as well :p Sep 22 14:37:53 only if you choose to, correct? Can't you use C w/ it? Sep 22 14:43:38 GodGinrai: you can also use C++ with it (through Qt APIs) Sep 22 14:47:01 ^ and that's what I meant about QML being more performant Sep 22 14:50:14 I think Enyo is quite OK performance wise. Sep 22 14:50:40 With 2.7 packaging only the needed stuff instead of whole lib should make leaps of difference I guess Sep 22 14:50:41 Herrie|Pre3: sure. But it's simple fact that C is more performant than JS Sep 22 14:50:47 (and C++) Sep 22 14:51:02 C(++) in general yes. QML not sure. Sep 22 14:51:16 I like QML because it's easier to fiddle UI for me. Sep 22 14:51:29 But then again never gave Enyo a fair chance myself. Sep 22 14:51:36 lol Sep 22 14:51:55 I can do a little in everything but don't excel anywhere LOL Sep 22 14:52:37 I like I can use QtCreator and Tofe created quite some magic so we can do stub calls etc Sep 22 14:52:52 So development and testing of things is really easy Sep 22 14:53:11 For me, I can fiddle UI in Enyo very easily Sep 22 14:53:14 I know Enyo has this too. I've seen it, just documentation is really not up to par. Sep 22 14:53:17 (or in any webpage, honestly) Sep 22 14:53:37 I agree that Enyo needs better documentation Sep 22 14:53:52 Not for the basics, honestly, but for the more complex things they never wrote docs for Sep 22 14:53:55 (like enyo.Router) Sep 22 14:53:58 GodGinrai: I see you're volunteering to solve #1021 :P Sep 22 14:54:11 Herrie|Pre3: link? Sep 22 14:55:03 issues.webos-ports.org ;) Sep 22 14:55:06 Then 1021 there Sep 22 14:55:35 Source code can be found at http://github.com/webos-ports/core-apps/com.palm.app.clock Sep 22 14:55:40 It has hardcoded pixels :P Sep 22 14:55:53 I tried to get it running in Chrome but failed miserably :P Sep 22 14:56:06 In the CSS. That should become % I guess. Sep 22 15:01:43 It's still Enyo 1 but I got other apps to run in Chrome with the correct switches just this one not... Sep 22 15:02:10 And without it it's a bit hard to sort... I could do the webinspector in VirtualBox maybe :P Sep 22 15:02:13 But well... Sep 22 16:18:05 Herrie: not sure, you left some failed ApplicationManager::instance()->launch() calls without setting errorText intentionally? Sep 22 16:19:52 imho, the idea to prefer QML and to use enyo as less as possible is good.. Sep 22 16:42:15 nizovn: I fixed the ones I saw, could be there are more, will check Sep 22 17:04:47 nizovn: You mean on places like https://github.com/webOS-ports/luna-appmanager/blob/master/Src/base/application/ApplicationManagerService.cpp#L1571 ? Sep 22 17:07:02 no, L1328 L1357 L1421 L1438 Sep 22 17:15:11 Tofe: seems currently NotificationArea is a little broken e.g. i can't close some notifications, sometimes they appear as black Sep 22 17:15:51 to reproduce, create dashboard in Testr, create notification using org.webosports.notifications, and try to close the second one Sep 22 17:22:15 nizovn: ah, ok, not good Sep 22 17:52:45 nizovn: Black dashboard we have for a while already Sep 22 17:52:51 Tofe already looked into it but seems a bit random Sep 22 17:53:07 The other one I've seen as well I think, not sure though Sep 22 17:57:25 nizovn: Those didn't have an errorText on 3.0.5 either: https://github.com/woce/LunaSysMgr/blob/beta/Src/base/application/ApplicationManagerService.cpp#L727, https://github.com/woce/LunaSysMgr/blob/beta/Src/base/application/ApplicationManagerService.cpp#L757, https://github.com/woce/LunaSysMgr/blob/beta/Src/base/application/ApplicationManagerService.cpp#L820 and https://github.com/woce/L Sep 22 17:57:26 unaSysMgr/blob/beta/Src/base/application/ApplicationManagerService.cpp#L838 Sep 22 17:57:33 So from a dev's perspective this wouldn't change anything Sep 22 17:58:03 DougReeder had issues with the other 2 because he (and other devs) relied on having a message content returned Sep 22 18:00:21 And we didn't.... Sep 22 18:00:39 Somehow HP rewrote a lot of the lunasysmgr code between WOCE open sourced bits and Open webOS. Sep 22 18:00:55 WOCE was 3.0.5 1:1 almost, while the OWO bits were changed quite a bit it seems Sep 22 18:05:43 Herrie: notice that WOCE code doesn't have errMsg's you introduced either. seems they are set by WebAppMgrProxy::instance()->appLaunch() calls, which have much more parameters then our launch() Sep 22 18:06:12 i guess it will return some errMsg on legacy Sep 22 18:12:40 I'd need to check it Sep 22 19:25:36 Herrie: I'll take a look at that clock code when I get a chance. Sep 22 19:42:38 GodGinrai: Thnx **** ENDING LOGGING AT Fri Sep 23 02:59:58 2016