**** BEGIN LOGGING AT Fri Apr 01 02:59:58 2011
Apr 01 14:24:52 <Jeevaka> denkenz: pin
Apr 01 14:24:59 <Jeevaka> denkenz: ping
Apr 01 14:25:43 <denkenz> Jeevaka: pong
Apr 01 14:26:49 <Jeevaka> denkenz: is typing casting a constant pointer to non-constant pointer accepted?
Apr 01 14:27:12 <denkenz> sometimes
Apr 01 14:28:30 <Jeevaka> what about the value constant string case?
Apr 01 14:29:06 <denkenz> I dunno what you're trying to accomplish
Apr 01 14:29:28 <denkenz> it is safe to cast as long as you have no alternative and you know the contents won't be modified
Apr 01 14:29:36 <Jeevaka> ok
Apr 01 14:42:33 <denkenz> Jeevaka: yikes
Apr 01 14:42:35 <denkenz> +	list = g_new0(char *, g_hash_table_size(vc->en_list) + 1);
Apr 01 14:42:35 <denkenz> +
Apr 01 14:42:35 <denkenz> +	g_hash_table_iter_init (&iter, vc->en_list);
Apr 01 14:42:35 <denkenz> +
Apr 01 14:42:36 <denkenz> +	while (g_hash_table_iter_next(&iter, &key, &value))
Apr 01 14:42:36 <denkenz> +		list[i++] = g_strdup(key);
Apr 01 14:42:48 <denkenz> Do you really need to g_strdup here? 
Apr 01 14:43:19 <denkenz> And where's my g_free? ;)
Apr 01 15:58:53 <Jeevaka> denkenz: ping
Apr 01 16:01:18 <Jeevaka> denkenz: g_strfreev is just 2lines below the memory allocation
Apr 01 16:05:56 <Jeevaka> holtmann: ping
Apr 01 16:06:24 <holtmann> Jeevaka: pong
Apr 01 16:06:46 <Jeevaka> holtmann: XLEMA URC is triggered only when the trigger conditions are met. AT+XLEMA=1 is not one of that trigger
Apr 01 16:08:30 <Jeevaka> holtmann: thats the main reason behind XLEMA read command
Apr 01 16:13:39 <holtmann> I think that should be changed in the IFX firmware.
Apr 01 16:13:55 <holtmann> The read command is potentially racy.
Apr 01 16:16:56 <Jeevaka> hmm, what makes you think so it is racy?
Apr 01 16:27:43 <fdalleau> denkenz: ping
Apr 01 16:34:42 <denkenz> Jeevaka: Yes you're right about g_strfreev, was too lazy to look up the code
Apr 01 16:34:49 <denkenz> fdalleau: pong
Apr 01 16:35:14 <fdalleau> hi
Apr 01 16:36:11 <fdalleau> i'm still working on HSP, after discussing with Vudentz, i removed the audio part of it
Apr 01 16:37:43 <fdalleau> Instead of doing sco in ofono, Vudentz suggested to implement RegisterAgent and UnregisterAgent the same way that hfp_hf is doing it.
Apr 01 16:40:17 <denkenz> sounds reasonable, honestly I have not thought of HSP at all
Apr 01 16:40:18 <fdalleau> i wrote a prototype for that but I don't know if this is really what you guys expect
Apr 01 16:40:29 <denkenz> That is holtmann's wild idea ;)
Apr 01 16:41:02 <denkenz> I'm reasonably OK with leaving HSP inside BlueZ
Apr 01 16:41:02 <fdalleau> well, the main reason i'm working on that, is because i thought it would be quick :D and to try to foresee some tasks for HFP
Apr 01 16:41:21 <denkenz> Hah, well now you know ;)
Apr 01 16:44:15 <fdalleau> if hsp is implemented like hfp_hf (using HandsfreeAgent) then it seems very easy to connect the sco, just call the Play method of org.bluez.Headset
Apr 01 16:44:58 <fdalleau> but is it the way to go? the other fred tells me it is too early, what do you think?
Apr 01 16:45:29 <denkenz> There's nothing wrong with using the agent approach to pass the socket around
Apr 01 16:46:00 <denkenz> However, who's going to call the BlueZ Play method?
Apr 01 16:46:11 <denkenz> And isn't that deprecated?
Apr 01 16:47:03 <fdalleau> i don't know if it si deprecated, i'll check that
Apr 01 16:47:35 <denkenz> I got lost with all the cases, since they're different between HCI and PCM routing I think
Apr 01 16:49:02 <fdalleau> i'm unsure whether ofono should care about wich routing is used
Apr 01 16:50:03 <fdalleau> btw i moved the CKPD into voice call atom to answer and release calls, is that ok for you?
Apr 01 16:50:27 <denkenz> So what is the idea behind HSP in oFono? To send ATA/ATH when CKPD is called?
Apr 01 16:50:37 <fdalleau> basicaly yes :/
Apr 01 16:51:18 <fdalleau> more like a proof of concept
Apr 01 16:51:24 <denkenz> and yes, CKPD should essentially be in voicecall.c
Apr 01 16:51:54 <denkenz> However, as I said, I'm still unsure this is all worth it, especially with HSP VoIP extensions
Apr 01 16:52:11 <denkenz> Since in theory you can use it from skype, etc
Apr 01 16:52:31 <fdalleau> i see what you mean
Apr 01 16:54:25 <fdalleau> but if ofono takes over HFP, how can skype use HSP?
Apr 01 16:57:38 <denkenz> HFP is really meant for cellular
Apr 01 16:57:56 <denkenz> Johan keeps talking of HFP 2.0 and VoIP extensions for it, but they have not materialized yet
Apr 01 16:58:12 <denkenz> So we cross that bridge when we come to it
Apr 01 16:58:20 <denkenz> But HSP... that can be used by anything
Apr 01 16:59:21 <denkenz> Don't necessarily think that TODO tasks are a must do
Apr 01 16:59:38 <denkenz> Sometimes things just don't turn out the way we think
Apr 01 17:00:53 <fdalleau> i understand that's not my point, i want to foresee how the audio will work in HFP, and HSP is a good candidate for prototyping
Apr 01 17:01:09 <denkenz> ah
Apr 01 17:02:20 <fdalleau> stuff like inband ringtone and call volume  will be exactly the same than HFP
Apr 01 17:03:04 <denkenz> no one has even attempted in-band rings I think
Apr 01 17:03:11 <denkenz> that one will be fun ;)
Apr 01 17:04:00 <fdalleau> btw i have one question, how does ofono ring?
Apr 01 17:04:21 <denkenz> oFono doesn't deal with audio
Apr 01 17:05:05 <denkenz> For Meego there was/going to be an audio policy framework that was supposed to do parts of this
Apr 01 17:06:45 <fdalleau> do you know where I could find more info about this? 
Apr 01 17:07:24 <denkenz> I think it was called Ohm-ng
Apr 01 17:07:30 <denkenz> But I never looked that hard
Apr 01 17:08:30 <fdalleau> oh great, that's a good starting point
Apr 01 17:09:00 <fdalleau> ohm-ng gives more result on google than audio policy framework
Apr 01 17:13:16 <fdalleau> denkenz, thank you for your patience, have a nice weekend
Apr 01 19:58:35 <Jeevaka> denkenz: ping
Apr 01 19:59:38 <denkenz> ?
Apr 01 20:01:13 <Jeevaka> has the emergency mode/call patch inline with what you have been expecting?
Apr 01 20:02:00 <denkenz> yes, I've yet to grok patch 6, but overall that is what I wanted
Apr 01 20:03:27 <Jeevaka> great
**** ENDING LOGGING AT Sat Apr 02 02:59:58 2011