**** BEGIN LOGGING AT Sun Nov 16 03:00:00 2014 Nov 16 04:41:49 I was briefly excited by the possibility of using an inpute of type “search”, but our current version of webkit is too old to support it. How old is the version of webkit we’re using? Nov 16 10:05:22 DougReeder: We were using the one of QT 5.2.1 but upgrading to QT 5.4 now Nov 16 10:51:58 DougReedr: If I recall correctly we have 538.1 or something in Black Eye Nov 16 11:02:54 nizovn: Not sure what was the problem, figured it would be quickly to just delete the folder and redo the build from scratch ;) Nov 16 11:03:22 ok Nov 16 12:01:29 DougReeder: you can activate upsync, if you want. Otherwise it will be overwritten if a change to the contact happens on the server. Nov 16 13:03:29 Herrie: still have problems building ? Nov 16 13:04:35 Herrie: for a build from scratch, you can eventually delete tmp-glibc and sstate-cache. Deleting the meta-* directories often isn't helping. Nov 16 13:22:44 About the high CPU load at boot time with LunaWebAppManager: here is a paste of the journal, showing that there is definitely a problem with the enyo build we ship with the testing release: https://bpaste.net/show/fff55e972e33 Nov 16 13:23:16 I'm not good at analyzing this: does someone know how to interpret that ? Nov 16 14:30:47 Tofe: The Enyo 1.0 build that is? Nov 16 14:31:08 Tofe: Just deleted the whole webos-ports-env folder :P Nov 16 14:31:15 And started from scratch, building now, but takes long like you know Nov 16 14:35:09 Shouldn't our framework be in 1.0 and not 0.1? Nov 16 14:59:06 Herrie|2: I just don't know :) Nov 16 16:12:12 It looks like 538.1 is from April of 2013, which is not bad. Hmm, Safari had input of type search by then, so evidently it hadn’t made it upstream. Nov 16 16:13:05 The important point is whether modern websites display correctly, and that does happen, mostly. Nov 16 16:14:41 Enyo 0.10 is identical to Enyo 1.0 - it was basically a name change. Nov 16 16:15:06 Internally, Enyo 1.0 is a link to Enyo 0.10. Nov 16 16:15:12 DougReeder: With QT 5.4 upgrade we move to QtWebEngine instead of QtWebKit Nov 16 16:15:43 Ah, is there a list of differences? Nov 16 16:16:52 It's using Chromium fork of WebKit because it develops a lot quicker compared to WebKit itself :P http://blog.qt.digia.com/blog/2013/06/25/experimenting-with-chromium-and-qt/ Nov 16 16:17:10 Once we have a build with Tofe's fix I can give it a go to see what I get Nov 16 16:17:38 Maybe Tofe has time to go to http://whatsmyuseragent.com/ on his fixed N4 :P Nov 16 16:18:13 Since I cannot connect to WiFi currently without VKB (well I could with command line connman but don't have much time to look into the exact command again :P) Nov 16 16:21:58 But can try the VBOX to see what I get there :P Nov 16 16:23:59 So, I thank you all for working to upgrade the HTML renderer - a modern web browser will be critical going forward. Nov 16 16:25:38 DougReeder_: The one in Black Eye release is already a LOT better compared to legacy webOS ;) Nov 16 16:25:52 * DougReeder_ nods Nov 16 16:32:53 But using Chromium/Blink instead of WebKit means we'll have the best of the best and quick :P Nov 16 16:47:22 We'll need to call out the shift when it happens - apps devs will want to test using Chrome rather than Safari. Nov 16 16:51:19 Seems that we're still on 538.1 somehow in the browser :S Nov 16 16:51:38 Maybe I need to update the browser somehow, not sure what UA manipulation I put in, need to double check Nov 16 16:51:44 Could be I had it hardcoded somehow :S Nov 16 17:07:46 Ah seems it's hardcoded indeed: https://github.com/webOS-ports/org.webosports.app.browser/blob/master/qml/UserAgent.qml#L42 Nov 16 17:08:46 Let me do some work on this... Nov 16 17:09:08 Also need to use "import QtWebEngine 1.0" now ;) **** ENDING LOGGING AT Mon Nov 17 03:00:00 2014