**** BEGIN LOGGING AT Sun Jun 26 02:59:58 2016 Jun 26 06:01:10 morning Jun 26 06:01:25 morning Jun 26 06:01:45 nizovn: You don't happen to have a clue how I can use bitbake to patch source files with Windows line endings? Jun 26 06:02:41 no, never tried that Jun 26 06:02:55 Lastest freeimage has LibWebP which seems to have it's source code with Windows line endings... And I need to patch it... Because it doesn't work with GCC 5.1+ without a patch :S Jun 26 06:05:14 Herrie: have you solved error: invalid character ' ' in raw string delimiter ? Jun 26 06:06:15 nizovn: Yeah Jun 26 06:06:41 I need to use a slightly modified version of https://chromium-review.googlesource.com/#/c/297211 Jun 26 06:06:52 But I cannot get the patch to apply in BitBake Jun 26 06:06:56 It works fine on my GH Jun 26 06:16:06 nizovn: This is the patch: https://github.com/Herrie82/FreeImage/commit/b4e1d12de6111f0bef1eda5c0e7cd5636621df57 or https://github.com/Herrie82/FreeImage/commit/b4e1d12de6111f0bef1eda5c0e7cd5636621df57.patch Jun 26 06:17:43 Herrie: yes, i get the same error Jun 26 06:21:57 I guess I could convert them to linux line endings first with another patch but that's nasty :P Jun 26 07:26:37 I tried locally on my git to recreate, apply the patch then make a new .patch from it but also no luck :S Jun 26 07:26:41 this worked for me Jun 26 07:27:09 applied original patch to sources, then generated new patch Jun 26 07:34:10 nizovn: I tried that on Windows but didn't work. i guess on *nix it will Jun 26 07:34:38 If you coudl share your patch that would be great Jun 26 07:42:15 Herrie: https://gist.github.com/nizovn/6a1d8023276bfe5aecae7203c1a14a59 Jun 26 07:53:37 Thxn Jun 26 08:47:03 OK PR updated, let's see what JaMa says :) Jun 26 08:59:51 Tofe: ping Jun 26 09:00:28 Herrie: pon Jun 26 09:00:29 g Jun 26 09:02:30 Got my Mako now as well Jun 26 09:02:47 Volume down is a bit flaky & screen has some minor scratches but seems OK for the rest Jun 26 09:02:53 \o/ Jun 26 09:03:10 Volume button doesn't seem too hard to fix, just need to get a Torx 5, only got a 6 here :P Jun 26 09:03:32 I noticed a small bug in the App Menu ;) Jun 26 09:03:46 While you're buying that kind of screwdriver, make sure you also get a T4 Jun 26 09:04:03 it's what's needed to open a N4 and replace battery Jun 26 09:04:36 Oh, there is a bug in app menu ? Jun 26 09:04:39 1. Open Calendar, go to app menu, got to Accounts & Preferences, Close it with Done. Close the Calendar App. Relaunch the Calendar app. You'll see that the App Menu is now called App Menu instead of Calendar and it doesn't work anymore Jun 26 09:05:19 interesting Jun 26 09:05:44 could be the "Current app", as seen by LNC, isn't correct Jun 26 09:06:24 That could be not sure Jun 26 09:06:56 Also I noticed that the input focus bug seems only present when VKB is show. When you hide VKB and press the input, you can type Jun 26 09:07:10 Might be maliit or compositor, not sure Jun 26 09:11:03 Yes, I'll have to squish those little annoying bugs Jun 26 10:17:19 nizovn: ping Jun 26 10:17:26 Tofe: pong Jun 26 10:17:38 nizovn: do you have a scenario where a popup window appears? Jun 26 10:18:00 yes, long tap on any link in browser Jun 26 10:18:08 actually it almost work Jun 26 10:18:23 ok Jun 26 10:19:33 it tries to set surface as popup but fails with setPopup called without input device Jun 26 10:20:48 https://github.com/RSATom/Qt/blob/master/qtwayland/src/client/qwaylandwlshellsurface.cpp#L188 Jun 26 10:20:56 probably something small Jun 26 10:23:12 mmh, ok Jun 26 10:47:52 nizovn: Does your PR also fix http://issues.webos-ports.org/issues/1127 ? It seems it might be related? Jun 26 10:49:20 no, 1127 seems like focusing bug Jun 26 10:53:24 nizovn: OK Jun 26 16:20:52 Tofe: so with a dirty fix i'm getting surfaces with popup windowtype. what should i test next, seems your recent commits need further work? Jun 26 16:23:33 Well, I've been trying to see why QtWayland fails to call setPopup Jun 26 16:23:59 it looks like the lastInputDevice isn't set Jun 26 16:24:39 I've tried forcing it a bit, but it doesn't really solve the issue and introduces weird other bugs Jun 26 16:24:48 yes, i just replaced it with defaultInputDevice :) Jun 26 16:24:54 my commits are not complete yet, indeed Jun 26 16:28:21 nizovn: with my last commit in luna-next, it should be complete. Or, at least, consistent. Jun 26 16:28:43 ok, will test Jun 26 16:29:50 nizovn: how do you know you get surfaces with popup window type ? you've put traces in the cardshell? Jun 26 16:30:21 i put qwarning in luna-next in surfaceCreated Jun 26 16:30:41 ok Jun 26 16:31:18 I'll try your dirty fix approach; it's not worse than mine anyway :) Jun 26 16:31:59 actually, it's almost good, it returns default one only when last is null :) Jun 26 16:33:14 my approach was to set lastInputDevice in QWaylandTouchExtension::touch_extension_touch Jun 26 16:34:58 ok, i don't even know what you are talking about, seems it's too complex for my head :) Jun 26 16:43:42 with my approach I get QWaylandSurface::None for the window-type, so I guess your approach works better :) Jun 26 16:49:09 ah, I'm doing the test too soon... Jun 26 16:53:13 seems i was wrong, i'm getting popup windowtype only in surfaceMapped, not surfaceCreated Jun 26 17:13:52 Tofe: so i guess we should create windows in surfaceMapped, not in createView? Jun 26 17:14:34 I would rather not modify our current window creating order :) Jun 26 17:14:45 I'm going to put a connect on windowTypeChanged Jun 26 17:15:29 ok, interesting Jun 26 17:32:58 mmh it is killed right away; maybe because of my dirty fix Jun 26 17:33:11 can you show me you modification? Jun 26 17:33:14 your* Jun 26 17:34:27 https://bpaste.net/show/d58c69cee872 Jun 26 17:34:59 let me try Jun 26 17:35:00 but i also get segfaults when second and further browser cards are closed Jun 26 17:35:23 yes, maybe because it should be NULL in that case Jun 26 17:37:00 seems our contextmenu works sometimes.. Jun 26 17:39:25 ok, I got it to work with a variation of your patch Jun 26 17:41:04 Problem is, it's wonderfully small Jun 26 17:42:05 popup? Jun 26 17:42:16 yes, and the font it's using Jun 26 17:42:21 because of the dpi Jun 26 17:42:32 and because we don't set the high dpi scaling for qt Jun 26 17:42:44 morphis: ping Jun 26 17:44:02 ah, ok; on emulator it looks good.. Jun 26 17:44:38 Forcing the centering of the popup doesn't work that well for submenus, I see Jun 26 17:55:24 yes, for qml combobox maybe it doesn't either Jun 27 02:34:31 * DougReeder checks in **** ENDING LOGGING AT Mon Jun 27 02:59:58 2016