**** BEGIN LOGGING AT Mon Sep 15 02:59:59 2014 Sep 15 03:59:41 evening richip Sep 15 04:01:02 Good evening Sep 15 04:02:16 Wasn't able to catch Morphis all weekend Sep 15 04:04:22 probably 2hrs from now :) Sep 15 04:04:25 or tomorrow. Sep 15 04:06:24 sokay. Plenty of time Sep 15 04:06:31 How do you recommend a dev environment be set up? Sep 15 04:06:46 Not just for app dev but system level dev Sep 15 04:06:49 is there a doc? Sep 15 04:07:21 there is a porting guide that is under development Sep 15 04:07:43 but first thing is setting up for openembedded builds. Sep 15 04:09:24 Sure. I remember long time ago setting one up. Even back then, it was a pain (for lack of instruction) for someone on Fedora Linux Sep 15 04:09:40 are you still on Fedora? Sep 15 04:09:47 Is there any particular setup for openembedded builds that we'd like to follow Sep 15 04:09:49 Yes :) Sep 15 04:13:01 For now, I'll just try http://www.openembedded.org/wiki/OEandYourDistro#Fedora_13_.2C_14_.2C_15_and_16 Sep 15 04:13:22 kk Sep 15 04:13:28 also look at this: http://webos-ports.org/wiki/Porting_Guide Sep 15 04:13:38 which has setups for webOS ports repos Sep 15 04:14:28 Whoops ... that was outdated. This instead: http://www.openembedded.org/wiki/Getting_started Sep 15 04:14:33 Okay. Thanks Sep 15 04:30:08 richip, usually it takes about 4GB of RAM, Core2Duo or better, 60GB of HD for builds Sep 15 04:31:48 check. Clearing up some space now Sep 15 05:05:45 the 4GB is for linking Qt Sep 15 05:05:47 and webkit Sep 15 05:49:47 richip, how are you doing? Sep 15 06:04:07 morning Sep 15 06:04:36 morning Garfonso Sep 15 06:46:12 morning Sep 15 06:52:44 morning Tofe Sep 15 06:53:11 Time for bed here…nite all Sep 15 07:14:44 morning Sep 15 07:23:08 Tofe: ping Sep 15 07:52:01 Herrie|Veer: pong Sep 15 07:58:39 Managed to "break" your stacked cards LOL :P Sep 15 08:01:13 Let me see if I can reproduce it and send you details Sep 15 09:30:13 Herrie: the new one ? Sep 15 09:31:05 I can also break it, sometimes you get a stack of black shadows instead of the cards Sep 15 09:31:48 It's my algorithm for inserting/moving/removing a "ghost card" when doing the drag'n'drop that isn't bug-free yet Sep 15 09:32:15 The problem mostly happens when going in-between two stacks, and then back to a stack Sep 15 09:32:33 Debugging that is pretty awful Sep 15 09:35:46 (because as soon as you hit a breakpoint, the mouse pointer is released, and triggers a card drop) Sep 15 11:32:09 Tofe: Yes the new one... Could get the shadows indeed and also a stuck card Sep 15 11:34:59 Where's morphis when you need him? He's still enjoying the Oktoberfest beers :P ? Sep 15 11:36:37 Tofe: Can you check the PR's for browser? 1 from me for code cleanup and 1 for launch parameters. I'm not very familiar with how QML deals with this. Sep 15 11:47:10 Herrie|Veer: merged the cleanup, and commented the launch parameters Sep 15 12:00:44 Thnx Sep 15 12:10:18 Tofe: For dropdowns in QML seems like we can use ComboBox and ComboBoxStyle? But some bits from ComboBoxStyle don't seem to be available in 5.2 only in 5.3. You have any other suggestions besides nasty stuff with rectangles :P ? Sep 15 13:16:49 Herrie|Veer: which bits ? Sep 15 13:18:31 Herrie|Veer: I don't see anything like that in the Qt 5.3 documentation Sep 15 13:18:32 Specifically the width Sep 15 13:19:18 dropDownButtonWidth is in the documentation for ComboBoxStyle in 5.3 but doesn't seem to work on 5.2 :P Sep 15 13:20:20 http://doc.qt.digia.com/qt-5.2/qml-qtquick-controls-styles-comboboxstyle.html Sep 15 13:21:33 It could be a bug in Qt 5.2 though, ofc Sep 15 13:22:52 I see drow instead of drop, could be somebody had fat fingers there :P Sep 15 13:22:58 Let me try that Sep 15 13:29:25 Yeah somebody fatfingered that and managed to get it into the release :P Sep 15 13:30:06 Anyway the implementation sucks, seems to open the options in a new screen instead of on the same screen... Will look for an alternative... Sep 15 13:34:58 morning Sep 15 13:38:53 morphis: morning Sep 15 13:39:05 Survived the Oktoberfest beer :P ? Sep 15 13:39:43 Herrie|Veer: wasn't really there, living more in the north :) Sep 15 13:40:24 Ah :P Still a good excuse for beer though LOL :P Sep 15 13:40:29 :D Sep 15 13:43:21 Got a PR pending for the https issue for LunaUniversalSearch Sep 15 13:44:17 Also how to debug Enyo apps in latest nightlies? webapp-launcher doesn't work anymore? Sep 15 13:44:19 Herrie|Veer: yeah saw it Sep 15 13:44:27 Or I need to run it from a specific path? Sep 15 13:44:31 no Sep 15 13:44:41 you can just connect to the inspector Sep 15 13:44:49 adb forward tcp:1122 tcp:1122 Sep 15 13:45:03 currently no need to start the webappmanager in debug mode Sep 15 13:45:13 will have the inspector always enabled Sep 15 13:46:16 Yeah but that doesn't work with Enyo apps that don't launch :P Sep 15 13:46:34 Nothing appears in my list when I browse to localhost:1122 for thos Sep 15 13:46:45 Nothing in journalctl for these either :'( Sep 15 13:46:48 hm Sep 15 13:47:51 Herrie|Veer: change /lib/systemd/system/LunaAppManager.service to start LunaAppManager with -t -c Sep 15 13:48:11 and /lib/systemd/system/LunaWebAppManager.service to start LunaWebAppManager with --verbose Sep 15 13:48:54 will add a switch for that soon Sep 15 13:49:08 so it gets automatically enabled with switching the devmode on Sep 15 13:49:09 does LunaWebAppManager cache app code? Sep 15 13:49:18 that would be great. :) Sep 15 13:49:21 it does Sep 15 13:49:39 which some times makes app development harder Sep 15 13:49:47 does it reload them on applicationManager/rescan call? Sep 15 13:49:59 no Sep 15 13:50:08 that only reloads the internal app database Sep 15 13:50:17 but doesn't forward that to the webappmanager Sep 15 13:50:34 ok.. hm. But a luna-next restart does help Sep 15 13:50:47 Garfonso: yeah as that restart LunaWebAppManager too :) Sep 15 13:51:33 mcgeek87: ping Sep 15 13:51:53 is it possible to restart LunaWebAppManager only? Sep 15 13:53:57 yeah Sep 15 13:54:18 but possible that then some of the boot time apps are not started Sep 15 13:56:38 morphis: morning ! Sep 15 14:54:53 Tofe: so card stacks are working better now? Sep 15 14:59:15 morphis: Theoretically yes, it's more intuitive and offers more flexibility for other features but it's still quite buggy Sep 15 14:59:29 According to Tofe debugging is a bit harder Sep 15 15:03:53 searching for the developer page on how to build webos for non supported devices ... Sep 15 15:04:04 could someone give me a link? :S Sep 15 15:04:38 OldDroid: you mean luneOS? Sep 15 15:04:46 yep Sep 15 15:05:15 can't find it on the website ... Sep 15 15:05:17 :/ Sep 15 15:05:50 btw do you guys need a fast and relieable vps for a nice website? Sep 15 15:06:33 OldDroid: is that what you are looking for http://webos-ports.org/wiki/Porting_Guide ? Sep 15 15:06:35 OldDroid: http://webos-ports.org/wiki/Porting_Guide Sep 15 15:06:40 thanks Sep 15 15:08:31 It's a work in progress document so feel free to add and correct while you're working through it Sep 15 15:09:12 OldDroid: really need some extensions Sep 15 15:09:20 I may have some time to rework it this week Sep 15 15:09:30 OldDroid: the real first step is getting cm build Sep 15 15:09:42 well, thats not the problem ^^ Sep 15 15:09:54 just wanted to look into the sysmgr Sep 15 15:09:56 stuff Sep 15 15:11:29 OldDroid: You're looking into N5 right? Sep 15 15:11:39 yup Sep 15 15:11:43 OldDroid: it build it already successfully? Sep 15 15:11:55 building atm Sep 15 15:12:25 ok Sep 15 15:12:36 OldDroid: next step is create the meta-layer Sep 15 15:12:45 hmm, tell me more ^^ Sep 15 15:13:34 the bsp stuff for openembedded? Sep 15 15:15:21 pewww, looks quite heavy :/ Sep 15 15:15:49 right Sep 15 15:16:23 OldDroid: look at https://github.com/webOS-ports/meta-smartphone/tree/jansa/daisy/meta-lg Sep 15 15:16:40 take https://github.com/webOS-ports/meta-smartphone/blob/jansa/daisy/meta-lg/conf/machine/mako.conf as base and copy it to hammerhead.conf Sep 15 15:17:43 there are really some things I need to document Sep 15 15:18:47 OldDroid: let me better do that before you start with it Sep 15 15:19:17 i know, some stuff for sdcrad, system part, kernel headers, its features Sep 15 15:19:39 ok Sep 15 15:19:44 morphis: yes and no -- the infrastructure is better, but there are still some bugs here and there that are annoying. But nothing technically challenging. Sep 15 15:19:48 some device related infos like screen size Sep 15 15:21:03 wait, is mako kernel built using official lg kernel source? Sep 15 15:21:08 or from cm ones? Sep 15 15:22:40 OldDroid: we're using a modifed cm tree Sep 15 15:23:08 OldDroid: https://github.com/shr-distribution/linux/tree/mako/3.4/master Sep 15 15:23:32 ^yeah, ok, because in the one layer config there is a kernel description line which says that its built using official lg sources Sep 15 15:23:39 just thought ^^ Sep 15 15:24:03 possible that this comment is out dated Sep 15 15:26:52 ok ive got an idea moorphis Sep 15 15:27:13 if you have time at the weekend? Sep 15 15:27:21 mostly not Sep 15 15:27:34 damn :/ Sep 15 15:28:04 so you want to create the layer? Sep 15 15:28:15 or i missunderstood? Sep 15 15:30:44 no, I will extend the documentation Sep 15 15:30:53 and you will help me to find things I missed :) Sep 15 15:31:11 okay ^^ Sep 15 15:31:14 thats a word :D Sep 15 15:31:40 i will extend the android source based stuff Sep 15 15:32:37 because if you build cm, you can't use the tutorial about initializing the build env from aosp Sep 15 18:08:14 Roy__, any update on the OpenSource website? Sep 15 18:09:16 ka6sox: well, it works, it's just dog slow. Sep 15 18:09:27 Hopefully Arthur can fix it this week. Sep 15 18:09:33 Roy__, thanks! Sep 15 18:50:20 Tofe: for a simple are you sure, yes/no dialog I can use something liek this right? https://qt.gitorious.org/qt/qtwebkit/source/0cba511b6bf9cebbf325b3cfebeee7e5f85effe9:Tools/MiniBrowser/qt/qml/ConfirmDialog.qml Sep 15 18:52:22 Well, right, but it's not as simple as it seems: it depends on Dialog.qml and DialogButton.qml in the same directory Sep 15 18:53:03 Yeah I know this ;) Sep 15 18:53:05 That's OK Sep 15 18:53:47 apart from that, the code seems quite ok yes Sep 15 18:54:52 OK will reuse that somehow Sep 15 18:55:12 Just need a simple confirmation dialog to user to confirm the deletion of all bookmarks/history Sep 15 18:55:46 UI wise settings starts to look OK, just need to add some backend stuff Sep 15 18:56:00 And figure out what we do and don't have with webkit.experimental stuff in QT 5.2 Sep 15 18:57:30 There seem to be a tons of features, but no proper documentation it seems at QT site :S https://rschroll.github.io/beru/2013/08/21/qtwebview.experimental.html Sep 15 18:58:47 Herrie, which TZ are you in just out of curiosity? I think I'll tackle another task on the browser and may need to hit you up with some questions later on Sep 15 18:59:17 Actuall same question for Tofe too.. :) Sep 15 19:01:16 thrrgilag: I'm GMT+2 Sep 15 19:07:32 Tofe: thanks, I'm GMT-7 (at least until November when magic happens to reset us back to standard time) Sep 15 19:12:26 thrrgilag, left coasters... Sep 15 19:13:22 ka6sox: yup (i was a right coast person for a few years but the ocean was on the wrong side so I had to move back Sep 15 19:22:55 thrrgilag, I couldn't get used to seeing it on the "wrong" side too... Sep 15 19:23:15 however for me its due south. Sep 15 19:25:50 thrrgilag, have you tried ToDo on LuneOS? Sep 15 19:28:37 ka6sox, no not familier with it Sep 15 19:28:44 * thrrgilag quickly searches goolgle Sep 15 19:30:03 ka6sox: ToDo? Sep 15 19:32:32 I've been testing todo.txt enyo, the app I wrote a while back for webos Sep 15 19:33:11 I have a minor tweak done, and a related tweak to the browser that was merged earlier today Sep 15 19:33:11 would be interesting to see if it works on LuneOS :) Sep 15 19:33:43 For the most part it works fine, just won't auth against dropbox but I have that fixed now Sep 15 19:34:33 thrrgilag: why would it not auth against dropbox? Sep 15 19:34:47 I am actually going to do some more work to bring it more up to date Sep 15 19:35:05 morphis, it has an option for storing the todo text file on Dropbox for cloud sync Sep 15 19:35:14 by default it stores locally though Sep 15 19:36:14 thrrgilag: yeah, but why would it not work on LuneOS? Sep 15 19:36:46 Because of the way I had coded it, it called out to the browser to bring the user to the dropbox auth page Sep 15 19:36:53 which has hardcoded to the palm browser Sep 15 19:37:22 I made it a generic call, and added support to the luneos browser this morning to take the URL target parameter Sep 15 19:37:58 ah ok Sep 15 19:38:08 thrrgilag: how do you get the oauth call back to your app then? Sep 15 19:38:54 It's doing an old oauth1 spec that dropbox originally had before they got better about it Sep 15 19:39:46 so it does some calls behind the scenes, gets a token, uses that when bringing up the auth page, dropbox autherizes it, etc Sep 15 19:39:55 so nothing has to get passed back from the browser Sep 15 19:39:58 ah Sep 15 19:40:14 I am no that much familar with oauth but that has changed with the new version right? Sep 15 19:40:53 yeah works much differently now, I just haven't updated this app to support it yet but that's on my (todo list). :) Sep 15 19:41:00 thrrgilag, using Onyx or Moochi? Sep 15 19:41:19 It's Enyo 1 actually, that's another thing on my todo list Sep 15 19:41:53 thrrgilag: but I am right that with newer oauth (is it 2.x?) you have to get something back from the webpage, right? Sep 15 19:42:33 yeah, with oauth2 there are some additional validation steps on both sides Sep 15 19:53:20 thrrgilag: ok Sep 15 19:55:50 thrrgilag: CET (GMT/UTC +1) Sep 15 19:55:54 European mainland Sep 15 21:10:38 ok, I've solve one issue. The problem was... a bug in "move" in the list of delegates in Qt... yepee. Don't use "move". Sep 15 21:11:02 Night is calling! bye **** ENDING LOGGING AT Tue Sep 16 02:59:58 2014