**** BEGIN LOGGING AT Wed Oct 07 02:59:58 2015 Oct 07 04:28:36 nizovn: morning Oct 07 04:29:14 nizovn: you're absolutely right. good catch. Oct 07 05:16:27 JaMa: Ping Oct 07 05:37:05 iscgar: I just wondering why you did such a massive code-style change in pmcertificatemgr Oct 07 05:39:50 morphis: There wasn't a consistent code style and it was written in a way the let bugs slip in easily, so I thought that it'd be better to choose a good enough code style and write everything in this style. Oct 07 05:42:12 morphis: but it's still work in progress, so I'd love to hear if there are alternatives. Oct 07 05:46:10 iscgar: I normally try to stick to what is available Oct 07 05:46:22 and not to change the code style of existing projects (even if it is mixed) Oct 07 05:48:45 morphis: Me too, but I figured that it'd be easier to change it if we're going to maintain it from now on and add new functionality. Oct 07 05:48:54 good Oct 07 05:49:02 it's you turn to decide on that Oct 07 05:49:17 I doubt that upstream will do any further work in the open so why not Oct 07 05:49:26 morphis: That's my thought as well Oct 07 05:49:41 So we can just as well make it more logical and less prone for wrong interpretation etc Oct 07 05:50:00 Since we don't expect anything from upstream in the open anytime soon Oct 07 05:50:14 right Oct 07 05:50:45 my thoughts exactly Oct 07 05:52:47 If it would be an actively maintained project I'm still for a uniform code style but that would need to be in cooperation with the upstream devs Oct 07 05:52:58 In this case I think we're free to do as we see fit. Oct 07 05:53:25 Any security fixes we always welcome :) And new features too :D Oct 07 05:53:40 Herrie: uniform code style is a great thing however when dealing with a large project as we do a uniform code style per single component is really enough Oct 07 05:53:44 and even that is hard Oct 07 05:56:31 morphis: Yeah ;) Oct 07 05:58:05 morphis: Pushed release with Fido, QT 5.5 & QtWebEngine for Browser out this morning ;) Oct 07 05:58:42 Herrie: wow! Oct 07 05:58:45 congratulations! Oct 07 05:59:34 Herrie: I do hope they release more code, but this is unlikely as morphis said. Oct 07 06:01:21 Herrie: and congratulations on the new release. my N7 finally booted up this morning so I'll test it tonight. is there any reason for it to be broken? Oct 07 06:29:50 iscgar|Pre: We updated Yocto from Dizzy to Fido. When we upgraded to Dizzy before both Nexus GSM (maguro) and Nexus 7 (Grouper) stopped working Oct 07 06:30:07 morphis: Heavy lifting was done by JaMa and Tofe ;) Oct 07 06:30:15 I just put some things together :P Oct 07 06:31:15 iscgar: mostly because nobody maintained it Oct 07 06:38:05 I see. I hope I'll have the time to test it and fix things if they break. shrinking the ports list with new releases is probably not what we want. Oct 07 06:53:35 Morning! Oct 07 07:15:29 Tofe: morning. Oct 07 07:15:59 iscgar|pre: Indeed. Just almost all devs have a N4 and almost nobody has a N7 to test, that's the problem ;) Oct 07 09:15:46 JaMa: ping Oct 07 18:09:33 Herrie: rebuilding qtwebengine didn't help, so I'm rebuilding the whole thing, and this time I'll try to be as close as possible from testing Oct 07 18:09:51 Maybe a regression was introduced between 5.5.0 and 5.5.1 Oct 07 18:11:33 (I realized a bit late that I could have avoided rebuilding the whole thing, but now it's too late anyway) Oct 07 18:18:09 Tofe: Ah OK Oct 07 18:18:20 ANyway the N4 audio not working is solved thnx to hint from morphis ;) Oct 07 18:19:02 Herrie: nice Oct 07 18:20:33 morphis: Any reason why we don't use latest upstream pulseaudio-modules-droid and just add our patches on top? Oct 07 18:25:05 Ah seems we use some files from Ubuntu for headphone detection Oct 07 18:30:39 Herrie: right Oct 07 18:31:35 Herrie: we shoudl be able to rebase that patch on top of upstream Oct 07 18:34:11 Seems this is the largest chunk https://github.com/webOS-ports/pulseaudio-modules-droid/commit/a5359e1c6796e455222af4acf5db42cb129c28ba and then just the minor change for PA5+: https://github.com/webOS-ports/pulseaudio-modules-droid/commit/f75332b539a438c1ab98b4ba0cefabf23d85480c Oct 07 18:34:21 That should be possible as patch right? Oct 07 18:36:19 Anyway it's not a lot of work to cherrypick these 2 commits Oct 07 19:23:50 Herrie: pong/2 Oct 07 19:25:33 JaMa: PulseAudio issue is solved on N4 Oct 07 19:26:06 Seems we no longer need: https://github.com/webOS-ports/pulseaudio-modules-droid/commit/daa1e2502a3efa6f62fb1fbce13b07752e781796 Oct 07 19:27:29 this is the local patch which breaks it for N4? Oct 07 19:33:05 evening Oct 07 19:35:48 Herrie: going to flash my N7 in a minute (I won't be able to test audio because the sound card is missing a resistor and I'm too lazy to fix it, but everything else should be testable) Oct 07 19:51:41 Herrie: so far so good. The keyboard is breaking the view when shown, though. Is it a known issue or an N7 specific? Oct 07 20:07:18 Herrie: it also seem to sporadically ignore hardware buttons and draw the touch ripple at some random places instead of under the touch zone. Was it like that before? Oct 07 20:28:12 JaMa: Yes, when we don't apply this local patch, all is OK for N4 Oct 07 20:28:27 iscgar: Yeah I believe these are known issues with N7 Oct 07 20:28:40 Good to know it still boots Oct 07 20:28:58 The VKB I know was there, not sure about the tap ripple Oct 07 20:32:20 Herrie: well that's good. Unfortunately the VKB issue prevents me from testing my patch for #633, so I'll be building a qemux86 image tonight and see if it's fixed. Oct 07 20:37:29 iscgar: OK Oct 07 22:46:53 i noticed that issue with tap ripple as well. It's because we "rotate" it to current orientation, but forgot that certain orientation can be locked. Oct 08 00:32:00 Hey, QTWebEngine guys: does it create separate layers for the GPU. If so, when? Oct 08 00:33:03 Does it create a new layer for the GPU if a web app specifies a transform property? An opacity property? Oct 08 00:34:02 http://www.html5rocks.com/en/tutorials/speed/high-performance-animations/ discusses what is performant in mainstream browsers. Oct 08 00:34:37 I’d like to add some animations to the GUI, where they can be smooth. Oct 08 00:36:54 http://alistapart.com/article/ui-animation-and-ux-a-not-so-secret-friendship discusses why we’d want to do this. Oct 08 00:47:45 Oh, and likewise does declaring a filter get you a new GPU layer? Oct 08 00:48:13 And how many layers is too many for devices like the N4? **** ENDING LOGGING AT Thu Oct 08 02:59:58 2015