**** BEGIN LOGGING AT Wed Apr 08 02:59:59 2015
Apr 08 12:44:45 * DougReeder waves good morning
Apr 08 14:27:29 <fprimex> hey guys, I got dragged into the deep end by life. was intending to work on a crash reporter for luneos.
Apr 08 14:27:44 <fprimex> don't know if you still need that, but I'm still open to it
Apr 08 14:30:17 <morphis> fprimex: hey
Apr 08 14:30:26 <morphis> fprimex: sure, just continue where you stopped last time :)
Apr 08 14:31:41 <Herrie|Veer> DougReeder: morning
Apr 08 14:31:45 <Herrie|Veer> What's up?
Apr 08 14:32:10 <DougReeder> I’ve added HTML5 geolocation to Testr
Apr 08 14:36:26 <Herrie|Veer> :)
Apr 08 14:37:33 <Herrie|Veer> Is it more reliable compared to the geoclue stuff we have :P ?
Apr 08 14:42:30 <DougReeder> Oh, it’s just a different API.  It would presumably pull from the same underlying source.
Apr 08 14:44:17 <DougReeder> We want to implement that API (like webOS does), so apps can use the same API as on other platforms.
Apr 08 14:46:30 <Herrie|Veer> :) Curious to test :)
Apr 08 14:47:29 <morphis> DougReeder: I doubt it will work out of the box as we don't have anyone from the webkit side using geoclue2
Apr 08 14:47:48 <DougReeder> It doesn’t work, at present. :-(
Apr 08 14:47:53 <morphis> qtlocation would be the candidate to wrap that up but still doesn't have geoclue2 support
Apr 08 14:54:08 <Tofe> morphis: I didn't look, will it support it for Qt 5.5 ?
Apr 08 14:59:56 <morphis> Tofe: not sure, just heard someone worked on it
Apr 08 15:02:53 <Tofe> ok
Apr 08 15:03:46 <Herrie|Veer> Were we able to track the source of the 5.4.1 regression bug?
Apr 08 15:33:46 <morphis> no
Apr 08 15:34:07 <morphis> however I would more or less skip 5.4.1 now and directly go to 5.5 later
Apr 08 15:39:43 <JaMa> morphis: just curious, why skip 5.4*?
Apr 08 15:39:53 <JaMa> just not worth spending time to fix it?
Apr 08 15:41:58 <morphis> right
Apr 08 15:42:12 <morphis> everything seems to work correctly, just the keyboard not
Apr 08 15:42:28 <morphis> even if that problem appears on 5.5 too I would more spend the time there to fix it
Apr 08 15:46:16 <nizovn> Herrie|2: re: yes, it should
Apr 08 15:50:30 <JaMa> ok, fair enough
Apr 08 15:51:05 <JaMa> fyi: I got delayed a bit with other stuff, so not sure if I'll start with testing-> stable today
Apr 08 15:51:40 <morphis> np
Apr 08 15:56:04 <Herrie|Veer> nizovn: Great will test when home :)
Apr 08 15:59:53 <Herrie|Veer> nizovn: Is 817 something you could look into?
Apr 08 16:04:36 <nizovn> Herrie|Veer: yes, will look after led support
Apr 08 16:05:12 <Herrie|Veer> OK :) Well maybe I can give it a try myself first :)
Apr 08 16:05:41 <nizovn> ok
Apr 08 16:05:42 <Herrie|Veer> Shouldn't be that hard just don't know much c but seems fairly straight forward based on example actually
Apr 08 16:05:53 <Herrie|Veer> You can check it then :)
Apr 08 16:06:42 <nizovn> ok :)
Apr 08 16:51:23 <Herrie|2> nizovn: Flashing N4 with latest image now to test the fix for luna-sysmgr-common
Apr 08 17:16:48 <Herrie> nivzovn: I still see some "Unknown keyword: optional" but seems to be less
Apr 08 17:19:09 <nizovn> need to think
Apr 08 17:34:38 <Herrie> nizovn: Mar 27 20:59:42 mako LunaSysService[806]: Unknown keyword: optional
Apr 08 17:36:27 <nizovn> Herrie: indeed, seems LunaSysService uses JsonMessageParser directly, so it's better to apply convert function in that class
Apr 08 17:51:46 <nizovn> Herrie: thanks for testing; do you want to fix it?
Apr 08 17:55:52 <Herrie> I'm no good with C ;)
Apr 08 17:55:59 <Herrie> So if you could do a PR ;)
Apr 08 17:56:07 <Herrie> Or it's just C&P from luna-sysmgr-common?
Apr 08 17:57:12 <Herrie> And change a few things?
Apr 08 18:01:14 <nizovn> just in JSONUtils.h move changed lines from VALIDATE_SCHEMA_AND_RETURN_OPTION to L118
Apr 08 18:02:08 <nizovn> initialize mSchema with converted schema
Apr 08 18:05:48 <nizovn> no, i'm wrong it should be also modified in https://github.com/webOS-ports/luna-sysmgr-common/blob/webOS-ports/master/Src/base/JSONUtils.cpp#L83
Apr 08 18:10:23 <Herrie> nizovn: I'm sure I'll mess this up ;) Maybe better if you do it, since you know what you're doing :P ?
Apr 08 18:11:06 <Herrie> morphis: ping
Apr 08 18:15:22 <Herrie> nizovn: You can probably help me with this too ;)
Apr 08 18:15:44 <Herrie> How would I test my changes to lumberjack code? It has a Makefile but that seems for legacy toolchain
Apr 08 18:21:37 <nizovn> Herrie: so it needs oe recipe? i can compile with legacy toolchain for you, if you want
Apr 08 18:23:46 <Herrie> nizovn: Seems I need webos-ports toolchain
Apr 08 18:23:49 <Herrie> I can isntall that
Apr 08 18:23:57 <Herrie> Then I should be able to just use gcc right?
Apr 08 18:24:05 <Herrie> It seems it doesn't have oe-recipe yet...
Apr 08 18:24:39 <Herrie> Let me push my code :)
Apr 08 18:32:16 <Herrie> nizovn: https://github.com/Herrie82/lumberjack
Apr 08 18:34:06 <nizovn> Herrie: so it needs oe recipe? i can compile with legacy toolchain for you, if you wan
Apr 08 18:34:10 <nizovn> error
Apr 08 18:34:13 <Herrie> Try that first ;)
Apr 08 18:34:27 <nizovn> https://bpaste.net/show/e3a21539abf9
Apr 08 18:35:16 <Herrie> Not sure I needed to update Makefile and not sure I used the right way
Apr 08 18:35:39 <Herrie> I replaced -lmjson with -ljson-c
Apr 08 18:35:43 <Herrie> But that could be wrong?
Apr 08 18:39:12 <Herrie> I guess I might have linked it wrong, see bottom on: https://github.com/json-c/json-c
Apr 08 18:43:34 <nizovn> well, the issue is legacy have <json.h> from mjson, and we need from json-c. he-he, lumberjack by design compiles for webOS, and won't for LuneOS. will make stub recipe
Apr 08 18:44:12 <Herrie> nizovn: Yeah seems so
Apr 08 18:44:19 <Herrie> oe-recipe might be good for future
Apr 08 19:22:50 <nizovn> Herrie: at least you need to include json.h in luna_methods.c
Apr 08 19:33:57 <Herrie> I thought we don't need json.h anymore due to json-c ?
Apr 08 19:35:18 <JaMa> we dont need json/ prefix, but we still need json.h
Apr 08 19:41:10 <Herrie> JaMa: Ah ok
Apr 08 20:11:24 <DougReeder> Is Testr or all org.webosports.* apps specially privileged to send on the private bus?
Apr 08 20:13:09 <DougReeder> The com.palm.location service appears to be available on the private bus but not the public bus.
Apr 08 20:13:23 <nizovn> yes
Apr 08 20:14:05 <DougReeder> Is that Testr or all org.webosports.* apps?
Apr 08 20:14:49 <nizovn> oops, i meant location service is only on private bus
Apr 08 20:16:05 <DougReeder> Would it be a lot of work to make it available on the public bus?
Apr 08 20:17:57 <nizovn> i think no, just add pub roles, but need to check
Apr 08 20:18:53 <DougReeder> Would it be a lot of work to add the permissions dialog also?
Apr 08 20:21:09 <nizovn> you mean dialog to allow application to use location? i think quite a lot work
Apr 08 20:27:05 * DougReeder nods
Apr 08 20:39:03 <DougReeder> We should consider allowing location service on the public bus, to support app porting & development.
Apr 08 20:39:42 <DougReeder> Though without the permission dialog, it does allow user location to be tracked without the user’s knowledge.
Apr 08 21:04:37 <Herrie> nizovn: Thanks for IPK can you send me source too or do PR with changes?
Apr 08 21:26:34 <nizovn> Herrie: as for keywords issue, luna-sysservice has it's own JSONUtils.{h,cpp} and some other files from luna-sysmgr-common, so it needs to be modified to use the common ones.
Apr 09 00:46:19 <DougReeder> The PR for the test of the HTML5 geolocation API is https://github.com/webOS-ports/org.webosports.app.testr/pull/18
**** ENDING LOGGING AT Thu Apr 09 02:59:58 2015