**** BEGIN LOGGING AT Fri Sep 11 02:59:58 2015
Sep 11 03:28:07 <DougReeder> Hey, do we have any sound files that we’re already using in LuneOS, that I could use for a a custom alert sound in Testr?
Sep 11 05:44:10 <Herrie> DougReeder: See Settings. THere are a few there
Sep 11 05:44:54 <Herrie> YOu can find some in /usr/palm/sounds too
Sep 11 07:43:21 <Herrie|Veer> morning
Sep 11 07:45:12 <Tofe> Morning!
Sep 11 08:21:59 <Herrie|Veer> Was your Fido build successful?
Sep 11 08:24:12 <Tofe> Herrie|Veer: I stopped it for the night, and it's back again this morning... qtwebkit in progress...
Sep 11 08:24:33 <Tofe> So far, so good.
Sep 11 08:27:32 <Herrie|Veer> Which reminds me I should update the wiki from dizzy to fido :P
Sep 11 08:41:43 <Herrie|Veer> morphis: ping
Sep 11 08:41:53 <morphis> Herrie|Veer: pong
Sep 11 08:45:29 <Herrie|Veer> Let's say hypothetically we'd like to port LuneOS to a new device, what is the max version of Android we can use in combination with libhybris? 5.1 already?
Sep 11 08:46:14 <Tofe> I don't think hybris can be used with >=5.0 yet
Sep 11 08:46:42 <Herrie|Veer> I saw morphis was quite active there recently so I guess he might know :P
Sep 11 08:46:59 <Tofe> oh, I know I should never have left that channel :p
Sep 11 08:47:10 <morphis> Tofe: hybris can be used with >= 5.0 already
Sep 11 08:47:17 <morphis> we merged support for that in recent upstream
Sep 11 08:47:39 <Tofe> That's great!
Sep 11 08:47:49 <morphis> Herrie|Veer: theoretically 5.1 is possible but it wouldn't be straightforward
Sep 11 08:48:50 <Tofe> Would there be any interest in trying 5.0 on N4 ?
Sep 11 08:49:08 <morphis> for sure but we still have problems with even getting 4.4 working on N4
Sep 11 08:49:28 <morphis> running out of file descriptors as somehow they are not closed
Sep 11 08:49:49 <Herrie|Veer> Jolla and Ubuntu are already on 4.4+ for N4 aren't they?
Sep 11 08:50:24 <Tofe> Jolla community on N4 doesn't seem that active, I've more seen N5 work lately
Sep 11 08:50:28 <morphis> Herrie|Veer: no
Sep 11 08:50:42 <morphis> Herrie|Veer: however Ubuntu is but its a different story
Sep 11 08:50:47 <morphis> mir vs. wayland
Sep 11 08:50:54 <morphis> and the problem is in our hwcomposer drive
Sep 11 08:51:04 <morphis> theirs is completely different and not compatible
Sep 11 08:51:13 <Tofe> "of course."
Sep 11 08:51:54 <Herrie|Veer> Ah OK :P
Sep 11 08:52:24 <Herrie|Veer> Yeah Ubuntu likes to be different. Can be a good and bad thing ;) Good things come from it too :)
Sep 11 08:52:51 <morphis> yeah
Sep 11 08:53:09 <morphis> so that is for sure a big problem to be fixed before being able to move on
Sep 11 08:53:43 <Herrie|Veer> I have no problem with 4.2.2 but some guys using multi-rom have it seems....
Sep 11 08:53:59 <Herrie|Veer> Or maybe that's their user error :P
Sep 11 08:54:11 <Tofe> morphis: do you think I might be of some help there?
Sep 11 08:54:29 <morphis> for sure :)
Sep 11 08:54:30 <Herrie|Veer> I only use N4 for LuneOS so I don't care too much which Android version is under the hood ;)
Sep 11 08:55:41 <Tofe> morphis: ok fine; I'll first finish my fido build, then I'll ask you what current status/version you use
Sep 11 08:55:54 <Tofe> I'll have some time tomorrow to look into it
Sep 11 08:56:29 <morphis> Tofe: see http://git.shr-project.org/git/?p=meta-smartphone.git;a=commit;h=35ed44de2b82aff7c54e2e8011a846a3288d6da3
Sep 11 08:57:57 <Tofe> ah great
Sep 11 09:05:30 <Herrie|Veer> morphis: Shouldn't we pick some commits from upstream libhybris as well? I see some that might be of interest for us like "Prevent resizing problem on nexus 4"?
Sep 11 09:05:54 <morphis> we have that one already
Sep 11 09:06:23 <Herrie|Veer> "Make sure to release all buffers"
Sep 11 09:06:56 <morphis> Herrie|Veer: feel free to bump libhybris and see what recent upstream changes
Sep 11 09:07:53 <Herrie|Veer> I'll probably leave that to Tofe because I guess we cannot use upstream directly but we'd need to cherry pick?
Sep 11 09:08:47 <Tofe> We used to be pretty close to upstream iirc
Sep 11 09:09:25 <morphis> Herrie|Veer: we can use upstream directly
Sep 11 09:09:32 <morphis> that is what we do today
Sep 11 09:09:40 <morphis> see https://github.com/webOS-ports/meta-webos-ports/blob/dizzy/meta-luneos/recipes-core/libhybris/libhybris_git.bbappend
Sep 11 09:09:46 <morphis> we just bump what meta-android gives us a bit
Sep 11 09:10:06 <Herrie|Veer> morphis: ah
Sep 11 09:10:18 <Herrie|Veer> So we don't actually need our own branch anymore?
Sep 11 09:10:24 <morphis> right
Sep 11 09:10:33 <Herrie|Veer> I got confused by the one we have because it's quite outdated :P
Sep 11 09:10:51 <morphis> always look at the recipes first
Sep 11 09:10:59 <Herrie|Veer> Yeah I should :P
Sep 11 09:11:09 <Herrie|Veer> Anyway JaMa switched us to Fido now :)
Sep 11 09:11:21 <Herrie|Veer> Have a minor issue with new PulseAudio on N4
Sep 11 09:11:29 <Herrie|Veer> And emu is broken.
Sep 11 09:11:44 <Herrie|Veer> JaMa's looking into emu
Sep 11 09:12:15 <morphis> Herrie|Veer: also worth to look what the recent changes https://github.com/mer-hybris/qt5-qpa-hwcomposer-plugin/commits/master fixes
Sep 11 09:12:28 <morphis> in terms of 4.4 upgrade
Sep 11 09:13:03 <morphis> especially https://github.com/mer-hybris/qt5-qpa-hwcomposer-plugin/commit/c24ad7fea662d94b1a03f5c40b2532071b3e7cc1
Sep 11 09:13:17 <Tofe> morphis: a priori, the problem for the fds is with hwcomper ?
Sep 11 09:13:24 <morphis> feel free to ask sleges over in #sailfish-porters as he had the same from with the fds on n4
Sep 11 09:13:29 <morphis> Tofe: yes
Sep 11 09:13:49 <morphis> we miss to close some of the fds we get and therefor run out of them after some time
Sep 11 09:13:58 <Herrie|Veer> They have some "fix fence leak on mako" in there
Sep 11 09:14:12 <morphis> Herrie|Veer: where?
Sep 11 09:14:41 <morphis> https://github.com/mer-hybris/qt5-qpa-hwcomposer-plugin/commits/mako-fence-leak-fix?
Sep 11 09:14:45 <Herrie|Veer> in their mako-fence-leak-fix branch, 3 commits
Sep 11 09:14:47 <Herrie|Veer> Yeah
Sep 11 09:14:56 <morphis> not sure of those help but worth a try
Sep 11 09:15:17 <morphis> but there must be a reason why they are not on master yet :)
Sep 11 09:16:33 <Herrie|Veer> Yeah probably might be worth checking with Sledges
Sep 11 09:19:42 <Herrie|Veer> For PulseAudio we get droid-util.c: Assertion 'pa_droid_input_port_name(cur_devicem &name) failed. Looks familiar?
Sep 11 09:20:01 <morphis> no
Sep 11 09:27:30 <morphis> Herrie|Veer: but we should still use https://github.com/mer-hybris/pulseaudio-modules-droid, right?
Sep 11 09:29:55 <Herrie|Veer> morphis: Yeah but JaMa bumped from PulseAudio 5 to 6
Sep 11 09:30:07 <Herrie|Veer> And that broke something on Mako, seems OK on TP
Sep 11 09:38:13 <Herrie|Veer> morphis: JaMa is using master-next38 branch which uses the Mer master + 3 patches it seems.
Sep 11 09:39:52 <morphis> then this needs some deeper debugging
Sep 11 09:55:03 <morphis> Herrie|Veer, Tofe: Jolla guys don't have the fd leak problem fixed yet
Sep 11 09:59:04 <Herrie|Veer> Ah OK so some possibilities for Tofe to fix and upstream :P
Sep 11 10:43:04 <Tofe> Herrie|Veer: and save the world
Sep 11 10:46:51 <Tofe> Herrie|Veer: my build is doing the rootfs now, so it's nearly finished and all went well apparently
Sep 11 10:49:17 <Herrie|Veer> :)
Sep 11 15:19:41 <Herrie> nizovn: morning
Sep 11 15:19:57 <Herrie> Arresel might drop by shortly, he has some questions on the BT stuff ;)
Sep 11 15:19:57 <nizovn> morning
Sep 11 15:20:13 <nizovn> ok
Sep 11 15:20:22 <Herrie> s/Arresel/Aressel ;)
Sep 11 15:20:30 <Herrie> He's not that good with C++ like me :P
Sep 11 15:20:43 <Herrie> So he needs some pointers on how to integrate stuff on Enyo side
Sep 11 15:22:12 <morphis> nizovn: pretty nice work on the BT API side in webappmanager :)
Sep 11 15:25:01 <nizovn> morphis: thanks :), but it could be better imho, as bluetooth pairing for now need setting app to be open
Sep 11 15:25:10 <morphis> right
Sep 11 15:25:29 <morphis> however that is not that bad as not being able to pair at all :)
Sep 11 15:25:38 <morphis> it should be the same on Android
Sep 11 15:25:47 <morphis> as there you only get discoverable when the settings app is open
Sep 11 15:27:03 <Herrie> There he is :P
Sep 11 15:27:07 <Herrie> Aressel: morning ;)
Sep 11 15:27:20 <Aressel> Good morning :)
Sep 11 15:27:27 <Aressel> You guys are just getting off work, right?
Sep 11 15:28:32 <Herrie> Yeah more or less ;) I had the afternoon off more or less ;)
Sep 11 15:28:42 <Herrie> But normally I'd be done in 30 mins or so ;)
Sep 11 15:28:45 <Aressel> Lucky!
Sep 11 15:29:51 <Aressel> nizovn, you around? Herrie tells me that you’re the one who integrated the bluetooth manager.
Sep 11 15:30:06 <nizovn> Aressel: yes
Sep 11 15:30:23 <Aressel> I’m working on tieing the settings app into it, and had a couple of questions
Sep 11 15:34:26 <nizovn> Aressel: you basically need to implement these functions: https://github.com/webOS-ports/luna-webappmanager/blob/f83938e3f9da1edac0cccb9f0af1f661664ebdf0/src/extensions/BluetoothManager.js#L6-L17
Sep 11 15:34:28 <Aressel> the first two things that I need to figure out are: 1. Does the bluetooth manager have a stored list of “known” and/or “connected” devices, or should that be stored by the settings app, and 2. how exactly does navigator.BluetoothManager.discover work? The only parameter is “value”. I’m not sure what to pass into that.
Sep 11 15:38:51 <nizovn> 1. no, list should be created in settings app, ondevice* will be called on status change
Sep 11 15:40:07 <nizovn> 2. "discover" is just making device discoverable and starts discovery process, so it takes boolean to start/stop
Sep 11 15:42:04 <Aressel> nizovn: Thanks! That should be enough to get me going. I assume that after I start the discovery process, I just catch the events as they come?
Sep 11 15:42:22 <nizovn> yes
Sep 11 15:43:16 <Aressel> Awesome, thanks for your help!
Sep 11 15:45:21 <nizovn> theoretically i can provide separate lists, but not sure how then to update them, updating whole list seems not best to me
Sep 11 15:46:54 <Aressel> Just need to figure out something that can be shared between settings app and device menu, I think.
Sep 11 15:50:07 <Herrie> Aressel: Not 100% sure, but device menu would normally only do power on/off and select a device from list
Sep 11 15:50:16 <Herrie> Rest should be done by settings app (see WiFi)
Sep 11 15:50:26 <Herrie> How to populate the list I need to figure our exactly ;)
Sep 11 15:50:30 <Aressel> Ya, that’s what I was thinking too
Sep 11 15:50:32 <Herrie> Since Device Menu is QML no JS ;)
Sep 11 15:50:42 <Herrie> WiFi follows that principle as well
Sep 11 15:50:45 <nizovn> Aressel: when i debugged plugin, i used this diff to your modification of settings app(with bt): https://bpaste.net/show/86078640240e , thanks a lot for it btw, it was very helpful :)
Sep 11 15:50:46 <Aressel> hmm, that makes it a little trickier…
Sep 11 15:51:41 <Aressel> nizovn: Thanks! Glad it helped :)
Sep 11 15:52:31 <Herrie> Aressel: This is what we do in DEvice Menu for WiFi
Sep 11 15:52:32 <Aressel> Herrie: is there some globally-accessable data store that we can use?
Sep 11 15:52:33 <Herrie> https://github.com/webOS-ports/luna-next-cardshell/blob/master/qml/StatusBar/SystemMenu/WiFiElement.qml
Sep 11 15:52:58 <Herrie> It basically does the following call: luna://com.palm.wifi/findnetworks
Sep 11 15:53:20 <Herrie> And will then populate the list of available networks. When the user taps on it it will launch Settings with params
Sep 11 15:54:10 <Aressel> Which isn’t quite what we need, because bluetooth devices are paired and not always discoverable in the same way…
Sep 11 15:55:26 <Herrie> Yeah it needs some tweaking for sure, but the principle should be similar
Sep 11 15:55:36 <Herrie> In Settings this would need ot be extended for BT: https://github.com/webOS-ports/org.webosports.app.settings/blob/506e34e9d8b28cc4122eae6b4232464bb4d2d9b7/source/views/Settings.js#L253
Sep 11 15:57:09 <Herrie> We already have some bits frm legacy: https://github.com/webOS-ports/luna-next-cardshell/blob/master/qml/StatusBar/SystemMenu/BluetoothElement.qml
Sep 11 15:57:37 <Aressel> Assuming that we are trying to handle this in the same way as legacy, we should be able to handle connects and disconnects from the menu, without launching the settings app, no?
Sep 11 15:58:12 <Herrie> Eventually yes
Sep 11 15:58:39 <Aressel> So this is just out of scope for the moment?
Sep 11 15:58:40 <Herrie> Though in 2.2.4 on my Veer it happens quite frequently it launches BT APP when I press an existing paired device
Sep 11 15:59:17 <Herrie> Well for now yes "[17:24] <nizovn> morphis: thanks :), but it could be better imho, as bluetooth pairing for now need setting app to be open"
Sep 11 15:59:22 <Aressel> Huh, interesting. 3.0.5 seems to handle it correctly
Sep 11 15:59:35 <Herrie> I have some LG headset that's funky
Sep 11 15:59:47 <Herrie> It supports too new BT for 2.2.4 and it's acting up sometimes
Sep 11 15:59:51 <morphis> Herrie: pairing != connect
Sep 11 16:00:07 <Herrie> morphis: Yeah true
Sep 11 16:00:16 <Herrie> So you'd reckon we can connect without Settings open?
Sep 11 16:00:19 <Herrie> Just not pair?
Sep 11 16:00:27 <morphis> right
Sep 11 16:00:34 <Herrie> D
Sep 11 16:00:35 <Herrie> :D
Sep 11 16:00:37 <morphis> pairing is a process of getting in trust with something
Sep 11 16:03:37 <Herrie> morphis: Yeah familiar with it ;) Did it a ton of times on Veer, TP with headsets, keyboards etc
Sep 11 16:03:47 <Herrie> Always forget how to get things into pairing mode :P
Sep 11 16:03:59 <morphis> devices normally have a "connect" button for this
Sep 11 16:04:09 <Aressel> Herrie: or some wierd button combo
Sep 11 16:04:40 <Aressel> I have a headset that requires you to hold volume up and next track for 5 seconds or something like that
Sep 11 16:06:18 <Herrie> Aressel: Yeah me too :P
Sep 11 16:06:25 <nizovn> morphis: not sure how to provide devices list to system menu, we need bt service then?
Sep 11 16:06:48 <morphis> nizovn: either that or you put some dbus calls into the QML code
Sep 11 16:07:15 <morphis> nizovn: like https://github.com/nemomobile/nemo-qml-plugin-dbus
Sep 11 16:07:21 <morphis> see https://nemo-qml-plugin-dbus.readthedocs.org/en/latest/
Sep 11 16:08:03 <morphis> would make most sense
Sep 11 16:14:30 <nizovn> morphis: thanks, understood
Sep 11 17:02:45 <DougReeder> Thanks, Herrie, I’m updating the Notifications section of Testr.
Sep 11 17:30:35 <Tofe> Go fido, go!
Sep 11 17:37:03 <Tofe> Herrie: now I do reproduce your issue with cards spawning with html5test.com
Sep 11 19:07:14 <Tofe> morphis: ping
Sep 11 20:22:52 <Tofe> morphis: neve
Sep 11 20:22:58 <Tofe> rmind
Sep 11 20:49:35 <Herrie> Tofe: I didn't see anything obvious in log for html5test.com
Sep 11 20:50:12 <Herrie> For me it doesn't seem to happen EVERY time but like in 90% of cases
Sep 12 02:59:29 <DougReeder> Just created a PR for Testr: https://github.com/webOS-ports/org.webosports.app.testr/pull/23
**** ENDING LOGGING AT Sat Sep 12 02:59:58 2015