**** BEGIN LOGGING AT Thu May 26 02:59:58 2016
May 26 03:27:48 <DougReeder> GoGinrai, under webOS, this dials a number: luna-send -n 1 palm://com.palm.telephony/dial '{"number":"8001234567"}'
May 26 03:28:19 <DougReeder> Under LuneOS, I get the error: {"returnValue":false, "errorText":"Malformed json."}
May 26 03:29:15 <DougReeder> I’m looking through https://github.com/webOS-ports/webos-telephonyd but I don’t see where it defines the API
May 26 03:32:24 <DougReeder> Or is it in https://github.com/webOS-ports/org.webosports.app.phone ?
May 26 03:43:48 <DougReeder> Is that functionality implemented yet, GodGinrai?
May 26 03:52:01 <DougReeder> IIRC, “Malformed json” means the request didn’t match the JSON schema specified for the service - e.g. some addtional parameter must be given.
May 26 03:54:18 <DougReeder> Or, Tofe, do you know?
May 26 05:27:46 <Herrie|Pre3> DougReeder: We use Mer's VoiceCall implementation which basically a plugin in QML and talks to oFono.
May 26 05:28:23 <Herrie|Pre3> So I guess we didn't implement a luna-send for it yet. Someone would need to write it I guess.
May 26 05:28:42 <Herrie|Pre3> But we don't really need it, unless you want to do command line phone calls :p
May 26 05:28:56 <Herrie|Pre3> But I might be wrong here. Sure Tofe can elaborate
May 26 05:29:27 <Herrie|Pre3> Our phone stack differs quite a bit from legacy, since we went from closed Palm TIL to fully open source bits :)
May 26 05:36:03 <Herrie|Laptop> DougReeder: I might actually be wrong, seems we do have quite some methods defined: https://github.com/webOS-ports/webos-telephonyd/blob/d55e40e83c01c725eff64e0ea105eb277a91df69/src/telephonyservice.c#l71
May 26 05:39:39 <Herrie|Laptop> Seems it expects a number and blockId as params though: https://github.com/webOS-ports/webos-telephonyd/blob/a0b292f0b851a54933678c50fe2d56804abcdbcb/src/telephonyservice_call.c#L62
May 26 05:39:44 <Herrie|Laptop> Not sure what blockId is :P
May 26 07:11:34 <Tofe> Morning
May 26 07:12:57 <Tofe> DougReeder: I think it should be webos-telephonyd, however I'm not sure we implemented that API in LuneOS
May 26 07:17:59 <Tofe> The block_id boolean is to enable the CLIR flag in oFono, whatever that means
May 26 07:19:46 <Tofe> https://www.nfon.com/gb/solutions/resources/glossary/clir/
May 26 07:20:33 <Tofe> So, it's just to hide the number, or said differently, block the caller's id :)
May 26 08:53:23 <morphis> Tofe: ping
May 26 09:00:51 <Herrie|Pre3> morphis: Anything I can answer?
May 26 09:03:06 <morphis> Herrie|Pre3: if you remember about hybris_check_android_shared_mutex in libhybris, then yes :-)
May 26 09:03:17 <morphis> trying to fix a crash I get on Android 6
May 26 09:03:25 <morphis> Tofe wrote that code initially
May 26 09:03:53 <Herrie|Pre3> morphis: Ah no :P I guess you really need Tofe for that :)
May 26 09:04:17 <morphis> hehe :-)
May 26 09:04:21 <morphis> Herrie|Pre3: how are things going
May 26 09:04:22 <morphis> ?
May 26 09:06:37 <Herrie|Pre3> Pretty good, but we can always use some hands :P
May 26 09:07:25 <Tofe> morphis: pong
May 26 09:08:00 <Herrie|Pre3> Did 2 Yocto upgrades and at Krogoth now + QT 5.6 :) Still have some small bugs but nothing too major ;)
May 26 09:08:10 <morphis> Herrie|Pre3: sounds very good :-)
May 26 09:10:54 <morphis> Tofe: have question for you about the code you wrote some time ago in hybris for handling pthread mutex
May 26 09:11:34 <Tofe> shoot :)
May 26 09:13:32 <morphis> Tofe: basically hitting https://paste.ubuntu.com/16702938/
May 26 09:14:32 <Tofe> address 0x2 looks pretty invalid
May 26 09:14:33 <morphis> if I remember well we're not doing a real conversion of the mutex but rather creating a new one for one initialized statically on the Android side
May 26 09:15:24 <Tofe> yes, something like that; iirc, the "pointer" is in that case an array index to fetch the android one
May 26 09:16:26 <Tofe> Maybe there's a hook missing on Android 6, in which case we wouldn't be able to convert the pointer to the correct one
May 26 09:17:14 <morphis> hm
May 26 09:17:43 <Tofe> too bad you don't have the full stack :p
May 26 09:18:24 <morphis> I have the full stack to some degree
May 26 09:19:34 <Tofe> knowing where in the stack that pointer comes from would help identify the cause
May 26 09:19:36 <morphis> let me add some more debugging to our conversion
May 26 09:19:42 <morphis> yes
May 26 09:19:50 <morphis> its coming from the private libGLES binary
May 26 09:19:58 <Tofe> ok
May 26 09:20:19 <morphis> there might be some constant changed or so
May 26 09:20:44 <Tofe> ah, this is bringing back quite old memories :)
May 26 12:12:28 <DougReeder> Herrie, Tofe, I’m thinking in the dial service call, blockId should be optional.  But that’s not critical. Is blockId a boolean?
May 26 12:16:24 <Tofe> DougReeder: yes
May 26 12:16:36 <Tofe> For now you can just put false, I think
May 26 12:23:14 * DougReeder nods
May 26 12:25:23 <DougReeder> When I try that, I dont’ get the Malformed json error, but the returnValue is still false.
May 26 12:26:48 <Tofe> mmh. That doesn't look very helpful
May 26 12:27:55 <Tofe> What kind of string did you send?
May 26 12:28:11 <Tofe> Could you try with a really wrong number?
May 26 13:01:34 <DougReeder> Sorry, I missed anything recent
May 26 13:04:45 <Tofe> What kind of string did you send?
May 26 13:04:49 <Tofe> Could you try with a really wrong number?
May 26 14:20:12 <DougReeder> I tried with an 800 number, IIRC "number": "8001234567"
May 26 14:21:17 <DougReeder> I don't have the N4 here at work, so I can't test anything new till late tonight.
**** ENDING LOGGING AT Fri May 27 02:59:58 2016