**** BEGIN LOGGING AT Wed Mar 02 02:59:57 2011
Mar 02 03:00:44 <denkenz> that's not per spec
Mar 02 03:01:02 <denkenz> 22.101 says to use those numbers as additional ones
Mar 02 03:03:38 <Jeevaka> so, you are referring to "Additional emergency call numbers that may have been downloaded by the serving network when the SIM/USIM is present."
Mar 02 03:04:44 <Jeevaka> in addition to this, we should have the default_en_list which wil be 112 an 911
Mar 02 03:05:39 <Jeevaka> 112, 911 + EFecc + Emergency number list downloaded by serving network
Mar 02 03:06:35 <denkenz> yep
Mar 02 03:13:35 <Jeevaka> as per spec, 000, 08, 110, 999, 118 and 119 when a SIM/USIM is not present. 
Mar 02 03:14:18 <Jeevaka> why are we not removing it from the list when we are having the sim?
Mar 02 03:16:02 <denkenz> we are
Mar 02 03:16:09 <Jeevaka> my mistake, it is removed when the EFecc is read
Mar 02 03:16:51 <denkenz> last I checked that logic is pretty solid and even handles empty EFecc
Mar 02 03:20:29 <Jeevaka> nothing has changed, still its in the same state ;)
Mar 02 03:21:51 <denkenz> good
Mar 02 03:22:23 <denkenz> depending on what IFX does with additional ECCs in the end you will either have to union the ECC list or just append it
Mar 02 03:23:34 <Jeevaka> is it possible to have duplicates in the emergency number list downloaded by the serving network
Mar 02 03:24:12 <Jeevaka> i mean 110 is present twice in the network emergency number list
Mar 02 03:26:50 <Jeevaka> basically due to the category association
Mar 02 03:27:04 <denkenz> Yeah, the category will get funny
Mar 02 03:27:39 <denkenz> But then whoever designed the additional ECC algorithm was nuts anyway
Mar 02 03:27:48 <denkenz> Not my problem ;)
Mar 02 03:28:12 <denkenz> Guess you'll just have to union the numbers list
Mar 02 03:32:34 <Jeevaka> ok will check
Mar 02 13:36:00 <demarchi> denkenz: humn... you broke gps on mbm
Mar 02 13:36:22 <demarchi> 	registered = ofono_modem_get_integer(modem, "Registered");
Mar 02 13:36:23 <demarchi> 	if (registered != 0)
Mar 02 13:36:24 <demarchi> 		return;
Mar 02 13:37:00 <demarchi> since you don't check for gps != NULL, now you are settings Registered=1 before the gps device is added
Mar 02 13:37:22 <demarchi> hence "ofono_modem_set_string(modem, GPS_DEVICE, devnode)" is never executed
Mar 02 14:22:57 <denkenz> demarchi: My MD300 does not have a GPS
Mar 02 14:23:47 <denkenz> So we need a better solution
Mar 02 14:25:18 <demarchi> yeah... 
Mar 02 14:47:58 <Jeevaka> denkenz: balrog-k1n: ping
Mar 02 15:03:08 <denkenz> Jeevaka: pong
Mar 02 15:04:05 <Jeevaka> incase of setup call handled in ofono side, alphaid and icon for call setup is part of the voicecall
Mar 02 15:04:36 <Jeevaka> i'm afraid we wont be able to do the same incase of setupcall handledi in the modem side
Mar 02 15:05:24 <denkenz> Probably not
Mar 02 15:05:29 <denkenz> Does setup call even send us an STKI?
Mar 02 15:07:47 <Jeevaka> incase of ifx, we will get SATN for user confirmation
Mar 02 15:08:38 <denkenz> Is there a case where no user confirmation is required?
Mar 02 15:11:31 <Jeevaka> im not aware of any such case
Mar 02 15:17:37 <denkenz> so we always get a SATN on setup call
Mar 02 15:18:24 <Jeevaka> yep
Mar 02 15:19:48 <Jeevaka> I have done the changes for the stk and ifx driver side to send user confirmation
Mar 02 15:20:05 <denkenz> So user-confirmation is easy I guess
Mar 02 15:20:12 <Jeevaka> yes
Mar 02 15:20:29 <denkenz> I'm assuming the modem takes a positive terminal response as a cue to continue with setup call
Mar 02 15:20:40 <denkenz> The 2nd alpha id is tricky
Mar 02 15:21:01 <Jeevaka> new driver function introduced to inform about the user confirmation
Mar 02 15:21:22 <denkenz> driver function? err
Mar 02 15:21:47 <Jeevaka> positive terminal response not used to continue with the setup call instead SATD is used to inform about the user confirmation
Mar 02 15:23:08 <denkenz> Ah ok, that weird case
Mar 02 15:23:45 <Jeevaka> issue is how to use the 2nd alphd identifier
Mar 02 15:24:29 <denkenz> I don't think you can
Mar 02 15:24:55 <denkenz> If we had a guarantee that the call will always be setup once the SATD has been sent, then we can do some magic inside voicecall
Mar 02 15:25:03 <denkenz> But I'm not really sure you can guarantee that
Mar 02 15:25:36 <Jeevaka> cant be guaranted
Mar 02 15:26:22 <Jeevaka> as you know, we should display the second alpha identifier as per the spec
Mar 02 15:26:41 <denkenz> yes, but its not like we can do it here
Mar 02 15:27:04 <denkenz> Do we at least get a SATF once the setup call has been performed?
Mar 02 15:27:08 <denkenz> Or failed?
Mar 02 15:28:13 <Jeevaka> if we use the stkagent dbus property it wil be unset when the terminal response is sent by modem
Mar 02 15:28:50 <Jeevaka> SATF not yet been verified
Mar 02 15:29:41 <denkenz> How do we know when the modem sends a terminal response?
Mar 02 15:31:44 <Jeevaka> atleast the core has been implemented with the assumption that if the received pdu has tag 0x81, then we are unsetting the alpha id
Mar 02 15:33:06 <denkenz> file / line #?
Mar 02 15:33:33 <denkenz> Ah I see
Mar 02 15:33:35 <Jeevaka> stk.c function: ofono_stk_proactive_command_handled_notify
Mar 02 15:33:45 <denkenz> So SATN is giving us terminal responses as well
Mar 02 15:34:15 <denkenz> Yeah OK, then we can do this
Mar 02 15:34:49 <Jeevaka> thats also a question mark as I havent tested myself hope balrog-k1n has tested it
Mar 02 15:34:53 <denkenz> Just set some flag inside voicecall atom that it should use this particular string on the next created outgoing call
Mar 02 15:35:09 <denkenz> He might have on the Freerunner
Mar 02 15:35:14 <denkenz> Or the MBM
Mar 02 15:35:53 <denkenz> One way to find out on IFX ;)
Mar 02 15:37:02 <Jeevaka> nod
Mar 02 17:00:16 <demarchi> denkenz: how can i use g_at_chat_send() or an equivalen function if i don't want the chat to read anything?
Mar 02 17:00:46 <demarchi> i.e. write the command and call the callback as soon as all bytes were written?
Mar 02 17:01:22 <demarchi> doing what you said about AT*E2PGSNPD, the callback is never called
Mar 02 17:06:06 <holtmann> demarchi: Do we actually get an OK back from the modem?
Mar 02 17:07:47 <demarchi> holtmann: no
Mar 02 17:12:38 <denkenz> demarchi: Sounds like you need to just write the command to the channel
Mar 02 17:12:45 <denkenz> Make sure the channel is non-block
Mar 02 17:13:13 <demarchi> denkenz: yes... but after i need to unref the channel, no?
Mar 02 17:13:28 <demarchi> and this unref would be done in a callback
Mar 02 17:14:08 <demarchi> or you are saying to write directly to channel without setting up a GAtChat... ?
Mar 02 17:19:50 <denkenz> demarchi: Exactly, don't bother creating the AT Chat 
Mar 02 17:19:59 <denkenz> if the callback never arrives anyway
Mar 02 18:16:36 <balrog-k2n> Jeevaka: yes, i only tested this on the freerunner and phonesim
Mar 02 18:17:00 <balrog-k2n> and Call Setup does not work on the Freerunner (yet)
Mar 03 02:40:45 <zgli> denkenz: Could I get the NITZ value with example nettime plugin enable on IFX modem by default? I did not see any relevant At command for it.  
**** ENDING LOGGING AT Thu Mar 03 02:59:57 2011