**** BEGIN LOGGING AT Sat Jul 13 02:59:58 2013 Jul 13 10:38:46 morning Jul 13 13:08:35 morphis: looks like direct rendering will be fairly easy to do with QtWayland, if this code is working: https://github.com/xknight/demo-compositor Jul 13 13:11:00 morphis: I realize I'm using Qt 5.1, and you are using 5.0.2, which maybe explains the differences of behavior Jul 13 13:11:37 Tofe: hm Jul 13 13:11:58 I thought already that really have to move on to 5.1 for better qtwayland support Jul 13 13:12:35 Tofe: are you here now for a little bit longer? Jul 13 13:12:50 I would like to do some planing for next week Jul 13 13:12:56 one hour or so Jul 13 13:13:24 Sure, we can do that Jul 13 13:15:31 ok, give me 15min Jul 13 13:15:35 ok Jul 13 13:58:57 morphis: I may have to leave soon Jul 13 14:01:47 Tofe: ok, I am back Jul 13 14:01:59 sorry, took a little bit longer Jul 13 14:02:17 morphis: that's fine, I still have a couple of minues Jul 13 14:02:21 minutes Jul 13 14:02:26 ok Jul 13 14:02:32 so what do we want to do next week? Jul 13 14:02:58 I would say, beginning to put some pieces together Jul 13 14:03:22 i.e. show a client wayland app in a card Jul 13 14:03:43 and getting things smooth Jul 13 14:03:55 yes Jul 13 14:04:20 I would like to have the card view with wayland clients mostly finished at the end of next week Jul 13 14:04:44 that would be good, yes Jul 13 14:05:04 so we need to break this down into small work portions Jul 13 14:05:41 Ok. currently, on the device, QtWayland works, and with the default compositor we can show a client app ? Jul 13 14:05:48 Tofe: I created already a "Sprint 2 - planing" list on our board Jul 13 14:05:54 somehow Jul 13 14:05:59 it's not really working Jul 13 14:06:05 test_webapp doesn't work Jul 13 14:06:15 snowshoe gets displayed within a card Jul 13 14:06:29 it's not really usable Jul 13 14:06:41 so I would say we have the following tasks: Jul 13 14:07:05 - get card view look like with LSM (correct card size, etc.) Jul 13 14:07:38 - implement native window class and window model Jul 13 14:08:00 - get wayland clients displayed correctly within a card Jul 13 14:09:15 the first task is for me of course Jul 13 14:09:35 I can take the second on Jul 13 14:09:36 e Jul 13 14:09:44 the last one I don't know: what doesn't work with test_webapp ? Jul 13 14:10:29 I think it's currently related to not setting a correct size for the window Jul 13 14:10:47 as I am using QApplication::primaryScreen()->size() for this atm Jul 13 14:10:57 and that seems to return 0x0 on the device Jul 13 14:11:17 so we need to decide how the clients should handle their screen size Jul 13 14:12:02 well, they can't retrieve it directly, because it can change if we rotate the device Jul 13 14:12:10 yes Jul 13 14:12:27 so we have to find a proper way to handle this Jul 13 14:12:37 maybe through a QML component ? Jul 13 14:12:51 or something exposed in QML Jul 13 14:13:11 (I mean a QML plugin, not a "Component" object) Jul 13 14:13:24 yes Jul 13 14:13:53 one for getting the correct screen size and one to detect screen orientation Jul 13 14:14:00 second one should be already there afaik Jul 13 14:14:06 I think so yes Jul 13 14:14:12 there should be a sensor for that Jul 13 14:14:14 (and there is a HTML5 API for it too) Jul 13 14:14:24 so we just need the first one Jul 13 14:14:43 yes Jul 13 14:14:53 QtWayland doesn't provide something ? Jul 13 14:15:02 and some way to define the actual screen size for a client window Jul 13 14:15:08 (as it can't use whole screen Jul 13 14:15:55 good question Jul 13 14:16:25 ah, yes, that's right, the screen size should be the size available for a fullscreen client in LSM Jul 13 14:17:48 yes, but we don't have fullscreen for a maximized window Jul 13 14:18:02 btw. we should decide which card states we actually have Jul 13 14:18:12 (for the screen size we can make so: we have the full screen size, adapted w.r.t. the rotation, and LSM can ask that "screen size manager" to reserve some regions of the screen) Jul 13 14:18:17 carded, maximized, fullscreen Jul 13 14:18:47 I'm not sure I really get the difference between maximized and fullscreen Jul 13 14:18:53 is it only the direct access ? Jul 13 14:18:57 yes Jul 13 14:19:08 and in maximized you have the statusbar visible Jul 13 14:19:13 + notifications Jul 13 14:19:18 which you don't have in fullscreen Jul 13 14:19:33 and with maximized you can't do direct rendering Jul 13 14:19:48 there's no status bar in LSM in fullscreen ? Jul 13 14:20:02 no Jul 13 14:20:08 thats maximized :) Jul 13 14:20:29 maybe I never got an app in fullscreen then :p Jul 13 14:20:31 http://www.webos-ports.org/images/1/13/Settings-application.png <- maximized Jul 13 14:20:42 Tofe: you never played a 3d game in webos? Jul 13 14:21:32 well, hum, I never had a webos device Jul 13 14:21:43 really? Jul 13 14:21:48 yep! Jul 13 14:21:57 shame on you! :D Jul 13 14:22:15 but I played some 3d games, like Monopoly, on my N900... I guess that doesn't count ! Jul 13 14:23:14 morphis: yes, I rarely speak about it aloud :) Jul 13 14:23:16 should be the same Jul 13 14:23:33 when playing games you shouldn't have the statusbar visible Jul 13 14:23:36 but ok then, I understand fullscreen now. Jul 13 14:23:43 so the game can take the full screen and direct rendering Jul 13 14:24:00 and the gesture area gets you back to card mode I guess Jul 13 14:24:10 yes Jul 13 14:24:15 ok, fine Jul 13 14:24:27 then we have the 3 states Jul 13 14:24:59 yes Jul 13 14:25:08 I will add them the native window implementation Jul 13 14:25:44 and in maximized mode screen height is: screen height - statusbar height - possible notifications height Jul 13 14:25:53 so it can change dynamically Jul 13 14:26:13 so we have to resize the client from the compositor size Jul 13 14:26:14 the notifications aren't composited on top of the window ? Jul 13 14:26:41 http://news.cnet.com/i/bto/20090401/SprintDashboard.jpg Jul 13 14:26:59 oh, beautiful Jul 13 14:27:04 thats the maximized version of the notifcations Jul 13 14:27:24 http://admintell.napco.com/ee/images/uploads/gadgetell/facebook_update_for_webos_640.jpg Jul 13 14:27:29 shows the difference better Jul 13 14:29:35 ok, I'll add that in the main.qml description, I already have a "maximized" qml container, which should adapt to these changes, so we can simply embed the maximized window in this container Jul 13 14:29:51 yes Jul 13 14:29:58 Tofe: I would like to use a enum for this Jul 13 14:30:11 which can be then exported from C++ to QML Jul 13 14:30:11 "this" being ? Jul 13 14:30:17 the staes Jul 13 14:30:21 ok Jul 13 14:31:17 I think the equivalent of the native window implementation would be my current "Cardview/WindowContainer.qml" class, which is still pretty naive right now Jul 13 14:32:00 yes Jul 13 14:32:14 so I'll introduce this enum in the qml description, and rework how it communicates its state and/or resize Jul 13 14:32:29 Tofe: I will put the tasks as cards on our board today or tomorrow Jul 13 14:32:38 ok, thanks Jul 13 14:33:32 I will introduce a fullscreen container at top level, just for the sake of layout right now Jul 13 14:34:01 ok Jul 13 14:34:01 and maybe a button in DummyWindow to switch back and forth to fullscreen Jul 13 14:34:11 that would be great Jul 13 14:34:20 we can also implement a native settings component Jul 13 14:34:24 great and easy, I like these tasks :) Jul 13 14:34:37 so we can just start luna-next with --debug to get those controls Jul 13 14:34:54 yeah, they give you a pretty good overview what we're doing and making the process transparent Jul 13 14:35:30 yep Jul 13 14:35:46 ok, I think I'll have to leave Jul 13 14:36:09 ok Jul 13 14:37:04 so to summarize, I can do the right card sizing + fullscreen layout Jul 13 14:37:31 and empty container for notification area, which I forgot initially Jul 13 14:37:48 all that in next week should be fine Jul 13 14:38:33 great Jul 13 14:38:55 the week after, I'll be abroad, and maybe less available Jul 13 14:39:22 ok Jul 13 14:39:37 but if we have card view mostly ready then it's completely fine :) Jul 13 14:39:44 :) good Jul 13 14:39:55 ok, leaving now, bye ! Jul 13 16:18:13 hey guys. i can't seem to run a build-desktop because github is just dropping my connection to the openwebos repos. any ideas? github broken? Jul 13 16:26:59 https://status.github.com/ looks fine **** ENDING LOGGING AT Sun Jul 14 03:00:00 2013