**** BEGIN LOGGING AT Fri Apr 29 02:59:58 2011
Apr 29 06:39:31 <morphis> heyho
Apr 29 07:09:46 <ChristW> Moin
Apr 29 07:11:25 <JaMa> moin
Apr 29 07:12:11 <ChristW> First time that I have code I touched running on FR. That's the first time in, like, 3 years or so...
Apr 29 07:17:39 <Azog> methril: i had the same problem, running shr on the micro sd solved it
Apr 29 07:44:01 <ChristW> I have code like this:                 GHashTable *properties = (GHashTable *) elm_genlist_item_data_get(it);
Apr 29 07:44:03 <ChristW>                 if (properties) {
Apr 29 07:44:05 <ChristW> ... }
Apr 29 07:44:28 <ChristW> I can then lookup data by using "tmp = g_hash_table_lookup(properties, ..."
Apr 29 07:44:42 <ChristW> What properties does the hash table have for a contact?
Apr 29 08:41:10 <n3v3rmind> hi guys!
Apr 29 08:44:23 <n3v3rmind> i've a problem with my python application...i've created a .py file that is in /usr/bin and a .desktop file that is in /usr/share/application and a .png file that is in /usr/share/pixmaps
Apr 29 08:44:47 <n3v3rmind> i've the application icon on the desktop of my openmoko
Apr 29 08:45:06 <n3v3rmind> if i press it...it run...
Apr 29 08:45:27 <n3v3rmind> but it run only one time! i've to restart the phone to reuse the icon
Apr 29 08:49:04 <n3v3rmind> http://pastebin.com/kydCWWC4 this is .desktop file maybe i've made some errors..
Apr 29 08:51:56 <n3v3rmind> some ideas?
Apr 29 09:11:40 <DocScrutinizer> openmokoDFU.zip anybody? where has that file been?
Apr 29 09:21:41 <morphis> mickey|office: good morning!
Apr 29 09:22:35 <mickey|office> morning morphis 
Apr 29 09:24:05 <DocScrutinizer> moo
Apr 29 09:27:05 <mrmoku> ChristW: depends on what fields you contact has :)
Apr 29 09:27:10 <morphis> mickey|office: I started now with writing the aurora stuff in python with pyside
Apr 29 09:27:48 <ChristW> mrmoku: A telephone number, at least, I guess...
Apr 29 09:28:01 <ChristW> (And now my audio is messed up :-( )
Apr 29 09:28:50 <ChristW> Kill Ubuntu on Parallels!
Apr 29 09:31:43 <ChristW> I guess this calls for a reboot of my computer after installing some updates... And I had it set up 'just right'!
Apr 29 09:31:47 * ChristW grumbles.
Apr 29 09:31:51 <ChristW> I'll be back.
Apr 29 09:40:44 <morphis> mickey|office: but I will be a kind of interesting how to sort all the things into the right boxes 
Apr 29 09:49:22 <CIA-10> SHR: 03mok 07isi-wireshark-plugin * r346da3b30159 10/ (Makefile src/isi-pipe.c src/isi-pipe.h src/packet-isi.c): add a dissector for the pipe resource
Apr 29 09:49:34 <CIA-10> SHR: 03mok 07isi-wireshark-plugin * rd38655c23733 10/src/packet-isi.c: remove some unneeded includes from packet-isi.c
Apr 29 09:49:48 <DocScrutinizer> \o/
Apr 29 09:50:36 * DocScrutinizer just notes this is quite probably a wireshark-isi-plugin
Apr 29 09:51:53 <morphis> mickey|office: but the whole qml extension stuff in pyside seems to be not very good supported
Apr 29 09:53:31 <mickey|office> no? 
Apr 29 09:55:11 <morphis> # This isn't supported by PySide yet.
Apr 29 09:55:11 <morphis> # Q_EXPORT_PLUGIN2(qmlqtimeexampleplugin, QExampleQmlPlugin);
Apr 29 09:55:36 <morphis> than the whole documentation of pyside seems the be copied from the C++ one with some adjustments
Apr 29 09:55:45 <morphis> http://www.pyside.org/docs/pyside/PySide/QtDeclarative/QDeclarativeExtensionPlugin.html
Apr 29 09:57:21 <morphis> even the examples are very incomplete
Apr 29 09:57:56 <mickey|office> the documentation copy is not a bad sign
Apr 29 09:58:18 <mickey|office> pyqt did that as well, it means it can be literally transferred without gross exceptions
Apr 29 09:58:29 <mickey|office> anyways, if we can't get it into OE, we're lost anyways :)
Apr 29 09:58:39 <mickey|office> i will continue to work on that over the weekend
Apr 29 09:58:47 <mickey|office> we need to change qt4-tools-native
Apr 29 09:58:56 <mickey|office> from static to dynamic linking
Apr 29 09:58:58 <mickey|office> enabling exceptions
Apr 29 09:58:59 <mickey|office> etc. etc.
Apr 29 09:59:09 <mickey|office> apiextractor is almost built
Apr 29 09:59:17 <mickey|office> it just fails during linking probably because of static qt
Apr 29 09:59:43 <morphis> ah ok
Apr 29 09:59:51 <morphis> but you think it will work?
Apr 29 10:00:27 <mrmoku> DocScrutinizer: indeed it is... but will help a lot to implement PDP hopefully :)
Apr 29 10:01:07 <DocScrutinizer> whatever PDP is
Apr 29 10:01:28 <morphis> mickey|office: yes bad documentation is not a problem, but we have to make somethink different then the "default" way is
Apr 29 10:01:28 <DocScrutinizer> recall my PDP11
Apr 29 10:03:48 <mrmoku> DocScrutinizer: GPRS
Apr 29 10:04:12 <darkh> Does anyone know a way to findout in a recipe or bbclass if we are currently building for the host (native) or the target?
Apr 29 10:05:11 <mickey|office> morphis: if we can't get it integrated, we need to use PyQt. I managed to get this working before, although it's also quite a lot of work :)
Apr 29 10:05:22 <mickey|office> luckily, it does hardly matter
Apr 29 10:05:27 <mickey|office> the APIs are 95% compatible
Apr 29 10:05:36 <mickey|office> and PyQt is even still faster than PySide
Apr 29 10:05:46 <mickey|office> although just a bit
Apr 29 10:06:11 <mrmoku> darkh: for what to do exactly?
Apr 29 10:06:36 <mrmoku> darkh: there is _virtclass-native override... like in:
Apr 29 10:06:39 <mrmoku> recipes/bcel/bcel_5.2.bb:DEPENDS_virtclass-native = "fastjar-native xerces-j-native regexp-native"
Apr 29 10:07:10 <darkh> i need to execute function foo if native and foo-native if native.
Apr 29 10:07:31 <darkh> i meant foo on target and foo-native on native
Apr 29 10:08:59 <mrmoku> darkh: something like this works too:
Apr 29 10:08:59 <mrmoku> recipes/glib-2.0/glib-2.0_2.23.6.bb:do_configure_prepend_virtclass-native() {
Apr 29 10:09:22 <darkh> ah that could be what im looking for
Apr 29 10:24:39 * mrmoku off to a client :/
Apr 29 11:23:45 <CIA-10> freesmartphone.org: 03morphis 07aurora * r2b771321cfe8 10/aurora/aurora-components/src/components/qml/ (6 files): aurora-components: remove components we will redesign in our own way
Apr 29 11:23:46 <CIA-10> freesmartphone.org: 03morphis 07aurora * rde7518eef2b3 10/aurora/ (37 files in 8 dirs): Remove aurora cpp stuff, new one will be written in Python
Apr 29 11:23:49 <CIA-10> freesmartphone.org: 03morphis 07aurora * rb385ac4654b5 10/aurora/ (6 files in 4 dirs): aurora: first draft of the new version written in python
Apr 29 11:23:51 <CIA-10> freesmartphone.org: 03morphis 07aurora * r0cd58616acec 10/aurora/.gitignore: aurora: add missing gitignore
Apr 29 11:23:58 <CIA-10> freesmartphone.org: 03morphis 07aurora * re7156462a2f5 10/aurora/ (9 files in 4 dirs): aurora: some restructuring for use of Qt's QML
Apr 29 11:24:06 <morphis> mickey|office: I pushed now the first bits of the python variant
Apr 29 11:24:25 <GNUtoo> so, no vala?
Apr 29 11:24:43 <GNUtoo> anyway python has some advantages too
Apr 29 11:27:18 <mickey|office> yes, vala
Apr 29 11:27:20 <mickey|office> but later
Apr 29 11:27:22 <GNUtoo> ah ok
Apr 29 11:27:26 <mickey|office> we need to solve some technical hurdles first
Apr 29 11:27:31 <GNUtoo> ok
Apr 29 11:27:31 <mickey|office> and don't want to block development for that
Apr 29 11:27:34 <GNUtoo> ok
Apr 29 11:27:56 <mickey|office> since the QML components will be the same, no matter what language the business logic is written in
Apr 29 11:28:05 <GNUtoo> ahh ok
Apr 29 11:28:08 <GNUtoo> nice
Apr 29 11:35:37 <mickey|office> so mrmoku works on umts now, do you have time to work on audio?
Apr 29 11:35:51 <mickey|office> or are you busy with other things?
Apr 29 11:39:26 <morphis> mickey|office: do you are experienced with sip?
Apr 29 11:39:37 <mickey|office> no, mrmoku is
Apr 29 11:39:42 <mickey|office> err
Apr 29 11:39:44 <mickey|office> which sip do you mean?
Apr 29 11:39:48 <mickey|office> session initiation protocol?
Apr 29 11:39:55 <mickey|office> or riverbankcomputing sip?
Apr 29 11:40:04 <morphis> riverbankcomputing sip
Apr 29 11:40:23 <mickey|office> ah. well, i used to bang on it to get it into OE some years ago...
Apr 29 11:40:33 <mickey|office> i also wrote the sip.bbclass etc.
Apr 29 11:40:48 <mickey|office> but it's long ago
Apr 29 11:40:50 <morphis> ok
Apr 29 11:41:05 <morphis> as I thought about doing the binding for qmlRegisterType on our own
Apr 29 11:41:10 <morphis> but first I will ask on the mailinglist
Apr 29 11:49:04 <mickey|office> ok, i'm definitely not _that_ familiar with it to give you a hint
Apr 29 11:49:14 <mickey|office> Phil is usually quite responsive
Apr 29 11:49:23 <mickey|office> so mailing list is a good idea
Apr 29 11:52:08 <morphis> ok
Apr 29 11:52:17 <morphis> will do that when I get access to it
Apr 29 11:52:35 <morphis> this http://lists.qt.nokia.com/pipermail/qt-qml/2010-October/001466.html is even interesting
Apr 29 11:52:49 <morphis> it's about the mixd.tv project, they are using python and qml together
Apr 29 11:53:03 <morphis> I wrote them a mail to request the code as they outline it's open source
Apr 29 12:00:48 <GNUtoo> mickey|office, where should I start with audio and what is required?
Apr 29 12:07:44 <GNUtoo> as required I means do I need 2 sim cards
Apr 29 12:07:48 <GNUtoo> and similar things
Apr 29 12:08:10 <GNUtoo> else I could call the free operator number
Apr 29 12:08:13 <GNUtoo> 4242
Apr 29 12:08:18 <GNUtoo> press 1 for foo
Apr 29 12:08:21 <GNUtoo> press 2 for bar
Apr 29 12:08:23 <GNUtoo> etc...
Apr 29 12:08:52 <GNUtoo> the ideal would be an echo test
Apr 29 12:09:15 <GNUtoo> but I've no USRP nor 2 sim+phone
Apr 29 12:09:24 <GNUtoo> that are capable of doing echo test
Apr 29 12:09:39 <GNUtoo> I think the openmoko segfault when you try to record a call with a kenrel problem
Apr 29 12:09:44 <GNUtoo> was that bug fixed
Apr 29 12:09:54 <GNUtoo> the htcdream, I don't think it's possible
Apr 29 12:09:59 <GNUtoo> the nexusone either
Apr 29 12:14:22 <GNUtoo> JaMa|Wrk, btw libnl 3.0 exists
Apr 29 12:14:27 <GNUtoo> hi
Apr 29 12:14:41 <GNUtoo> s/btw/hi,/
Apr 29 12:15:05 <JaMa|Wrk> let's hope it's compatible with libnl2 :)
Apr 29 12:17:45 <JaMa|Wrk> hmm latest glib-2.0 breaks current gdbus-binding-tool-native
Apr 29 12:34:59 <GNUtoo>  * pkg_parse_from_stream_nomalloc: Excessively long line at 13. Corrupt file?
Apr 29 12:35:03 <GNUtoo> I've that with opkg
Apr 29 12:35:06 <GNUtoo> on om-gta02
Apr 29 12:35:13 <GNUtoo> and no trace of strace on the phone
Apr 29 12:38:16 <GNUtoo> ok deleting /var/lib/opkg/* make it work again
Apr 29 12:55:47 <DocScrutinizer> who's interested in a more durable replacement for the backup RTC battery (either in FR or N900, they are the same) - I'm considering to get me some at farnell, but shipping to Germany is 20EUR, while the battery is .68 .. 1.18 / unit
Apr 29 12:56:55 <DocScrutinizer> so if I'd order 20, it's like 2Eur plus 55ct postage per unit
Apr 29 12:57:01 <DocScrutinizer> in europe
Apr 29 12:58:41 <DocScrutinizer> http://de.farnell.com/taiyo-yuden/pas414hr-va5r/cap-0-06f-3-3v-80ohm-4-8mm-coin/dp/1853000?Ntt=pas414hr-va5r
Apr 29 13:00:02 <DocScrutinizer> (just in case you don't know: this is the component that's broken on almost all FR and N900 and causes RTC to lose time on every swap of main battery)
Apr 29 13:07:53 * pabs3 would be if he were in the EU
Apr 29 13:08:18 <pabs3> N900 has that problem too?!
Apr 29 13:09:53 <GNUtoo> hi, how to record calls on the freerunner?
Apr 29 13:09:57 <GNUtoo> I tried and failed
Apr 29 13:10:02 <GNUtoo> the wiki is not up to date
Apr 29 13:10:19 <GNUtoo> I'm using that kernel:
Apr 29 13:10:24 <GNUtoo> Linux om-gta02 2.6.34.8 #1 Mon Feb 21 04:29:33 CET 2011 armv4tl GNU/Linux
Apr 29 13:12:05 <JaMa|Wrk> hmm code generated by gdbus-codegen changed a lot in last version and there is no gdbusproxymanager..
Apr 29 13:12:08 <DocScrutinizer> pabs3: yes, same component, same problem
Apr 29 13:13:38 <DocScrutinizer> pabs3: check postage from Germany to wherever you live, I don't really care where to address that simple letter to
Apr 29 13:14:32 <pabs3> opposite side of the world (Australia), probably more than the component
Apr 29 13:14:50 <DocScrutinizer> hmm, lemme check
Apr 29 13:14:54 <JaMa|Wrk> DocScrutinizer: how hard is it to replace it @home?
Apr 29 13:15:05 <DocScrutinizer> average soldering skills
Apr 29 13:15:26 <DocScrutinizer> the component is visible without microscope ;-D
Apr 29 13:16:12 <DocScrutinizer> just don't overheat the cell
Apr 29 13:17:07 <DocScrutinizer> on FR is's really easy. On N900 disassembling is a bit eeew
Apr 29 13:17:56 <JaMa|Wrk> I don't mind loosing time on fr at all, and only slightly annoyed on N900, so don't know if it's worth fixing for me :)
Apr 29 13:18:10 * JaMa|Wrk prefers not to pull battery out :)
Apr 29 13:20:40 <DocScrutinizer> pabs3: standard letter international is 75ct while eu letter is 55ct
Apr 29 13:21:09 <GNUtoo> is it possible to record and play during calls with .34?
Apr 29 13:21:15 <GNUtoo> on om-gta02
Apr 29 13:21:24 <DocScrutinizer> yes
Apr 29 13:21:29 <DocScrutinizer> hw supports that
Apr 29 13:21:30 <GNUtoo> ok
Apr 29 13:21:45 <GNUtoo> how do I do that, I tried it reading that:
Apr 29 13:22:15 <GNUtoo> Capture Left Mux: Line or RXP-RXN 
Apr 29 13:22:19 <GNUtoo> I put line
Apr 29 13:22:20 <JaMa|Wrk> has n900 with  https://bugs.maemo.org/show_bug.cgi?id=6063
Apr 29 13:22:31 <GNUtoo> Capture Right Mux: PGA  there was no PGA
Apr 29 13:22:45 <GNUtoo> so I end up with a wav file that didn't contain any sound
Apr 29 13:22:50 <GNUtoo> *ended
Apr 29 13:23:09 <DocScrutinizer> JaMa|Wrk: weird
Apr 29 13:23:22 <GNUtoo> I'll retry
Apr 29 13:23:48 <GNUtoo> ...without touching alsamixer
Apr 29 13:24:54 <DocScrutinizer> you need to keep the gsmhandset.state relevant settings, plus add in the settings needed for recording/playback
Apr 29 13:25:10 <DocScrutinizer> they are NOT mutually exclusive
Apr 29 13:25:17 <GNUtoo> ok
Apr 29 13:25:27 <DocScrutinizer> you just need to combine them into one statefile
Apr 29 13:25:28 <GNUtoo> but what are the settings needed for recording/playback
Apr 29 13:25:34 <GNUtoo> that changed trough kenrels
Apr 29 13:26:00 <DocScrutinizer> no idea what's alsa foo in fso/shr today
Apr 29 13:26:17 <DocScrutinizer> I just can tell from hardware PoV
Apr 29 13:27:01 <GNUtoo> http://pastie.org/1847209
Apr 29 13:27:21 <mickey|office> i wonder whether there is a dedicated output for voice data
Apr 29 13:27:32 <mickey|office> or are we supposed to dump it to the PCM out?
Apr 29 13:27:37 <mickey|office> and then route to earpiece?
Apr 29 13:28:14 <GNUtoo> ah I found capture left/right mux
Apr 29 13:28:56 <GNUtoo> sorry for the nose
Apr 29 13:29:02 <GNUtoo> *noise
Apr 29 13:37:17 <mickey|office> DocScrutinizer: where is voice data coming from under maemo?
Apr 29 13:37:24 <mickey|office> (i.e. which alsa output)
Apr 29 13:37:38 <DocScrutinizer> good question
Apr 29 13:38:02 <DocScrutinizer> microphone is digitally coupled to a IIS input
Apr 29 13:38:22 <DocScrutinizer> so I'd guess this is a separate alsa device
Apr 29 13:38:28 <lindi-> GNUtoo: what are you trying to do?
Apr 29 13:38:36 <DocScrutinizer> interface to cmt is another IIS
Apr 29 13:38:41 <DocScrutinizer> bidir iirc
Apr 29 13:39:14 <GNUtoo> lindi-, something like the echo test
Apr 29 13:39:34 <lindi-> GNUtoo: so just for debugging?
Apr 29 13:39:42 <GNUtoo> just for the n900
Apr 29 13:39:46 <lindi-> ah
Apr 29 13:39:47 <GNUtoo> like that:
Apr 29 13:39:49 <lindi-> wrong hw then :)
Apr 29 13:39:55 <GNUtoo> n900<->freerunner
Apr 29 13:39:59 <lindi-> ah
Apr 29 13:40:10 <GNUtoo> basically we do not have voice yet with freerunner
Apr 29 13:40:14 <GNUtoo> oops
Apr 29 13:40:18 <GNUtoo> basically we do not have voice yet with n900
Apr 29 13:40:26 <GNUtoo> so I need something to test
Apr 29 13:40:30 <GNUtoo> like an echo test
Apr 29 13:40:39 <GNUtoo> and what better thing than a freerunner to do that
Apr 29 13:41:00 <DocScrutinizer> SSI, not IIS
Apr 29 13:41:04 <DocScrutinizer> ?
Apr 29 13:41:07 <GNUtoo> I will need another sim tough
Apr 29 13:41:13 <GNUtoo> ah ssi
Apr 29 13:41:24 <GNUtoo> IIS was some unknown acronym for me
Apr 29 13:42:14 <DocScrutinizer> see http://linuxplumbersconf.org/2009/slides/Jyri-Sarha-audio_miniconf_slides.pdf p.15 diagram
Apr 29 13:43:17 <DocScrutinizer> voice-source/sink plus the SSI-up/downlink left bottom corner is the cmt interface aiui
Apr 29 13:43:32 <mickey|office> hmm
Apr 29 13:43:35 <mickey|office> a seperate alsa device?
Apr 29 13:43:40 <DocScrutinizer> ALSA hw:0.0 is mic-in and *-out
Apr 29 13:43:40 <mickey|office> or just a seperate input?
Apr 29 13:44:08 <mickey|office> ah, interesting. are all alsa devices appearing already on our kernel?
Apr 29 13:44:08 <DocScrutinizer> on maemo some PA-plugin is interfacing to cmt-SSI
Apr 29 13:44:29 <DocScrutinizer> iirc
Apr 29 13:44:50 <DocScrutinizer> you probably can't access cmt SSI via standard ALSA
Apr 29 13:45:33 <DocScrutinizer> that's where cmtspeech comes in I guess
Apr 29 13:46:09 <mickey|office> hmm
Apr 29 13:46:15 <mickey|office> possible
Apr 29 13:46:27 <mickey|office> although i only understand yet that cmtspeech is the way to and from the modem
Apr 29 13:46:34 <mickey|office> s/that/how/
Apr 29 13:46:34 <apt> mickey|office meant: although i only understand yet how cmtspeech is the way to and from the modem
Apr 29 13:46:53 <mickey|office> which works fine, but is not enough ;)
Apr 29 13:47:27 <mickey|office> guess someone really needs to read the source of the PA plugin
Apr 29 13:47:41 <DocScrutinizer> cmtspeech should be a 'janus' between cmt SSI and usual playback and recording using ALSA
Apr 29 13:49:05 <DocScrutinizer> basically a pipe
Apr 29 13:49:17 <DocScrutinizer> bidir
Apr 29 13:49:21 <GNUtoo> hmmm still blank wav
Apr 29 13:50:47 <DocScrutinizer> 1) record from mic  2) pipe that data to cmt SSI (using cmtspeech) 3) playback some sound to speaker/earpiece via ALSA 4) pipe audio *from* cmt SSI via cmtspeech to that ALSA output
Apr 29 13:51:29 <DocScrutinizer> that's my humble take on it
Apr 29 13:56:03 <mickey|office> hmm
Apr 29 13:56:07 <mickey|office> 4) i'm not sure
Apr 29 13:56:35 <mickey|office> or do you mean via standard alsa means on 4)
Apr 29 13:56:43 <mickey|office> to/from SSI is cmtspeech, no doubt
Apr 29 13:56:55 <mickey|office> to/from earpiece is alsa, i guess
Apr 29 13:58:48 <DocScrutinizer> yes
Apr 29 13:59:28 <DocScrutinizer> speakers/earpiece/headset and handset/wired-headset-mic are plain ALSA
Apr 29 13:59:50 <DocScrutinizer> cmt is no ALSA domain at all aiui
Apr 29 14:00:15 <mickey|office> right
Apr 29 14:00:28 <mickey|office> ok, so we just have to decide/find the right device/channel
Apr 29 14:00:39 <DocScrutinizer> so cmtspeech should act like a mp3-player + recorder
Apr 29 14:00:40 <mickey|office> and the rest should be guessing the actual raw data format
Apr 29 14:00:48 <mickey|office> that comes from / goes to cmtspeech
Apr 29 14:00:56 <DocScrutinizer> just it doesn't play mp3 but cmt
Apr 29 14:00:56 <mickey|office> so that we can properly configure the device
Apr 29 14:01:00 <mickey|office> *nod*
Apr 29 14:01:06 <DocScrutinizer> and it doesn't store to file but to cmt
Apr 29 14:01:26 <mickey|office> pretty cool, really
Apr 29 14:01:43 <mickey|office> can't wait to overlay the gsm audio with some things
Apr 29 14:02:13 <mickey|office> like speech synthesis indicating how long the call is
Apr 29 14:02:20 <DocScrutinizer> np
Apr 29 14:02:22 <mickey|office> or whatever crazy ideas
Apr 29 14:02:45 <mickey|office> pitch shifting heh
Apr 29 14:02:47 <mickey|office> robot voice
Apr 29 14:02:47 <DocScrutinizer> N900 design is AP centric. No audio without CPU
Apr 29 14:02:53 <GNUtoo> lol
Apr 29 14:03:48 <DocScrutinizer> FR *can* do this as well, but standard setup on FR  is direct analog feedthru from mic to modem, and from modem to audio-out
Apr 29 14:03:59 <DocScrutinizer> you could even power down the CPU
Apr 29 14:04:22 <DocScrutinizer> reboot the whole system while continuing your phonecall
Apr 29 14:04:31 <GNUtoo> lol 
Apr 29 14:04:35 <DocScrutinizer> you can't on N900
Apr 29 14:04:48 <DocScrutinizer> the design is very different
Apr 29 14:05:52 <DocScrutinizer> cmt audio interface is plain digital SSI
Apr 29 14:06:01 <DocScrutinizer> mic is digital SSI
Apr 29 14:06:12 <DocScrutinizer> codec is digital
Apr 29 14:06:18 <DocScrutinizer> all routed thru CPU
Apr 29 14:07:28 <DocScrutinizer> any more questions?
Apr 29 14:09:07 <mickey|office> who's going to implement the missing bits? :)
Apr 29 14:09:21 <mickey|office> [you don't have to answer that]
Apr 29 14:10:41 <GNUtoo> hmmm
Apr 29 14:10:59 <GNUtoo> I asked where to start and what was required earlier
Apr 29 14:11:21 <DocScrutinizer> I can give another hint, regarding >>the rest should be guessing the actual raw data format<< see "8 to 48 hhz SRC" in that diagram
Apr 29 14:11:44 <DocScrutinizer> obviously the codec alsways runninh 48kHz, and cmt always running 8kHz
Apr 29 14:13:01 <DocScrutinizer> you shouldn't worry about that, alsa plug plugin autoconverts, so you can open plughw:0.0 at 8kHz and up/downsampling is done transparently by ALSA
Apr 29 14:13:56 <DocScrutinizer> simple enough, huh? ;-D
Apr 29 14:14:20 <mickey|office> GNUtoo: so now you should be able to start. 1) finding the right alsa devices, input/output, 2) finding the right raw format, 3) opening/configuring the alsa device, 4) feeding data into/from
Apr 29 14:14:46 <mickey|office> i'd say lets expand cmtspeech with different modes
Apr 29 14:14:51 <mickey|office> one is loopback (which is already in)
Apr 29 14:15:00 <mickey|office> 2nd real one
Apr 29 14:15:49 <GNUtoo> I should retry with alsa?
Apr 29 14:15:59 <GNUtoo> I've states that *should* work
Apr 29 14:16:10 <GNUtoo> I took them from a running meego
Apr 29 14:16:13 <GNUtoo> with the same kernel
Apr 29 14:16:53 <DocScrutinizer> btw meego-arm must have sth working along that line
Apr 29 14:17:07 <DocScrutinizer> I heard they did calls
Apr 29 14:17:08 <GNUtoo> calls work in meego-arm
Apr 29 14:17:10 <mickey|office> ya, first thing is finding out how to put something to earpiece and record from mic
Apr 29 14:17:10 <GNUtoo> with sounds
Apr 29 14:17:15 <GNUtoo> I tested
Apr 29 14:22:43 * DocScrutinizer mumbles loopback for cmtspeech could mean 'normal operation' loopback from cmt-SSI to ALSA and vice versa. Much like a loopback mounted device for storage
Apr 29 14:23:33 <DocScrutinizer> should be obvious from source
Apr 29 14:33:39 <DocScrutinizer> lsmod:
Apr 29 14:33:42 <DocScrutinizer> omap_ssi               18116  2 ssi_mcsaab_imp,cmt_speech
Apr 29 14:34:46 <DocScrutinizer> IroN900:~# modinfo cmt_speech
Apr 29 14:34:47 <DocScrutinizer> filename:       /lib/modules/2.6.28-omap1/cmt_speech.ko
Apr 29 14:34:49 <DocScrutinizer> license:        GPL
Apr 29 14:34:50 <DocScrutinizer> description:    CMT speech driver
Apr 29 14:34:52 <DocScrutinizer> author:         Kai Vehmanen <kai.vehmanen@nokia.com>
Apr 29 14:34:53 <DocScrutinizer> author:         Peter Ujfalusi <peter.ujfalusi@nokia.com>
Apr 29 14:34:55 <DocScrutinizer> srcversion:     A15FDEFF8FF827E0794E6FF
Apr 29 14:34:56 <DocScrutinizer> depends:        omap_ssi
Apr 29 14:34:58 <DocScrutinizer> vermagic:       2.6.28-omap1 preempt mod_unload modversions ARMv7
Apr 29 14:35:32 <DocScrutinizer> HTH
Apr 29 14:36:09 <DocScrutinizer> btw mcsaab is the codec
Apr 29 14:36:15 <DocScrutinizer> or related to codec
Apr 29 14:36:45 <DocScrutinizer> multi-channel serial audio fooo bar
Apr 29 14:37:23 <DocScrutinizer> wonder which /dev or /sys files cmt_speech.ko exports
Apr 29 14:41:35 <DocScrutinizer> actually it could be as simple as ~#cat /dev/cmtaudio/in | alsaplay&; alsarecord > /dev/cmtaudio/out
Apr 29 14:42:49 <DocScrutinizer> or use socat? ;-D
Apr 29 14:45:27 <DocScrutinizer> IroN900:~# ls -l /dev/cm*
Apr 29 14:45:28 <DocScrutinizer> crw-rw---- 1 root pulse 10, 58 1970-01-01 01:00 /dev/cmt_speech
Apr 29 14:45:57 <DocScrutinizer> eeew group puke err pulse
Apr 29 14:46:05 <DocScrutinizer> fair enough, for a hint
Apr 29 14:47:03 <DocScrutinizer> I bet this is the direct phy if to cmt
Apr 29 14:47:43 <DocScrutinizer> open for read and you will get 8kHz 16bit (LE?) audio samples
Apr 29 14:48:39 <DocScrutinizer> write arbitrary crap to it, to hear noise at far end
Apr 29 14:49:25 <DocScrutinizer> so make that...:
Apr 29 14:49:50 <DocScrutinizer> actually it could be as simple as ~#cat /dev/cmt_speech | alsaplay&; alsarecord > /dev/cmt_speech
Apr 29 14:50:54 <GNUtoo> mrmoku, !!! it seem that the the modem_nokiaisi doesn't register anymore
Apr 29 14:50:59 <GNUtoo> and I need that fior cmt!!!
Apr 29 14:51:12 <GNUtoo> 1970-01-01T00:04:34.937075Z [DEBUG] fsogsmd : GLib <fsogsm.modem_nokia_isi>: isitransport.vala:63: ISI PROBLEM in PHASE 2, FAIL
Apr 29 14:51:20 <GNUtoo> 1970-01-01T00:04:34.937441Z [ERROR] NokiaIsiModem <phonet:phonet0>: Can't open main channel; open returned false
Apr 29 14:51:20 <DocScrutinizer> NB you can't feed those samples directly to mcsaab as that one is working @48kHz
Apr 29 14:51:22 <GNUtoo> hmmm
Apr 29 14:51:39 * GNUtoo wants to try cmt_speech cat
Apr 29 14:52:43 <DocScrutinizer> hey, I lured you in? :-D
Apr 29 14:53:34 <DocScrutinizer> IroN900:~# lsof |grep cmt
Apr 29 14:53:35 <DocScrutinizer> pulseaudi   780      pulse  mem    CHR      10,58                3131 /dev/cmt_speech
Apr 29 14:53:37 <DocScrutinizer> pulseaudi   780      pulse   35u   CHR      10,58                3131 /dev/cmt_speech
Apr 29 14:55:03 <GNUtoo> yes I know that device node
Apr 29 14:56:30 <DocScrutinizer> just wanted to explain why I am reluctant to try and cat that file on maemo
Apr 29 14:57:01 <DocScrutinizer> I'd have to shutdown pukeaudio to free the /dev/cmt_speech
Apr 29 14:57:13 <GNUtoo> I'll try
Apr 29 14:57:17 <GNUtoo> on meego or SHR
Apr 29 14:57:18 <DocScrutinizer> and I'm not sure I could initiate any phonecall then
Apr 29 14:58:10 <DocScrutinizer> suggestion: cat >testfile.raw
Apr 29 14:58:30 <DocScrutinizer> then you have all the time n the world to feed testfile.
Apr 29 14:58:54 <DocScrutinizer> then you have all the time n the world to feed testfile.raw to aplay with vrious format parameters
Apr 29 14:59:10 <GNUtoo> yes that's what I tought
Apr 29 14:59:14 <GNUtoo> cating to a file
Apr 29 14:59:18 <GNUtoo> and then trying to read it
Apr 29 14:59:26 <DocScrutinizer> :nod:
Apr 29 14:59:44 <DocScrutinizer> I predict it will need a -r 8000
Apr 29 15:00:18 <DocScrutinizer> and probably a -f s16le
Apr 29 15:00:30 <DocScrutinizer> or any other 16bit format
Apr 29 15:00:34 <GNUtoo> I predict...an empty file
Apr 29 15:00:44 <DocScrutinizer> hehe, that as well
Apr 29 15:01:10 <GNUtoo> I mean....I have an empty file
Apr 29 15:01:29 <DocScrutinizer> umm, you tried to record an ongoing call?
Apr 29 15:01:36 <GNUtoo> yes
Apr 29 15:01:43 <DocScrutinizer> meh
Apr 29 15:01:51 <DocScrutinizer> weird shit
Apr 29 15:02:10 <DocScrutinizer> read source of  /lib/modules/2.6.28-omap1/cmt_speech.ko
Apr 29 15:04:06 <GNUtoo> ok
Apr 29 15:05:07 <DocScrutinizer> maybe cat isn't exactly the right tool here
Apr 29 15:05:17 <DocScrutinizer> more sth like dd or socat
Apr 29 15:05:29 <GNUtoo>         .mmap           = cs_char_mmap,
Apr 29 15:05:35 <GNUtoo>         .read           = cs_char_read,
Apr 29 15:05:35 <GNUtoo>         .write          = cs_char_write,
Apr 29 15:07:06 <DocScrutinizer> yeah, pukeaudio also used mem what I gather is mmap
Apr 29 15:08:12 <DocScrutinizer> no idea, actually. Just handwaving
Apr 29 15:14:03 <GNUtoo> maybe we could take from nitdroid if they have something
Apr 29 15:14:13 <GNUtoo> there is a cmt_audio.cpp in the ril I think
Apr 29 15:19:57 <GNUtoo> ah they use it as a loopback too
Apr 29 15:20:02 <DocScrutinizer> I think the modem needs some special kick in the ass to switch to SSI audio. It has standard speaker and mic connection/interface as well
Apr 29 15:20:03 <GNUtoo> according to the git commits
Apr 29 15:20:29 <GNUtoo> I really wonder
Apr 29 15:20:39 <GNUtoo> are there any mic wired to the modem?
Apr 29 15:20:49 <GNUtoo> or does the CPU need to do the wiring
Apr 29 15:20:50 <DocScrutinizer> nope, not in N900
Apr 29 15:20:57 <GNUtoo> nice
Apr 29 15:21:14 <DocScrutinizer> CPU must forward the SSI *from* mic to SSI *to* cmt
Apr 29 15:21:15 <GNUtoo> because I wondered if it was better privacy wise than some other phones 
Apr 29 15:21:24 <GNUtoo> ok
Apr 29 15:22:58 <DocScrutinizer> I'd look for magic secret cmds sent to ISI, in meego, to start SSI audio
Apr 29 15:23:14 <GNUtoo> ok
Apr 29 15:23:16 <GNUtoo> thanks a lot
Apr 29 15:23:46 <GNUtoo> I have it in the ril I think
Apr 29 15:23:51 <DocScrutinizer> also some debug output in  /lib/modules/2.6.28-omap1/cmt_speech.ko alike the arecord -vv VU-meter would help a lot I guess
Apr 29 15:23:53 <GNUtoo> +            cmtspeech_state_change_call_status(cmtspeech, newState);
Apr 29 15:24:25 <GNUtoo> there are debug environment flags
Apr 29 15:24:26 <GNUtoo> I think
Apr 29 15:24:29 <GNUtoo> if I remember well
Apr 29 15:25:15 <DocScrutinizer> bbl, RL calling
Apr 29 15:25:23 <GNUtoo> ok
Apr 29 15:25:26 <GNUtoo> RL?
Apr 29 15:25:45 <dcordes> hi
Apr 29 15:25:57 <dcordes> GNUtoo: you ping'd recently iirc
Apr 29 15:26:41 <GNUtoo> yes
Apr 29 15:27:19 <GNUtoo> cmtspeech.h:int cmtspeech_state_change_call_status(cmtspeech_t *context, bool state);
Apr 29 15:28:41 <GNUtoo> maybe it was again for news about cotulla
Apr 29 15:28:48 <GNUtoo> I was in need of news
Apr 29 15:29:58 <dcordes> yes it really sucks with the qsd8* situation
Apr 29 15:30:45 <dcordes> they are so lifeless without the audio
Apr 29 15:31:39 <GNUtoo> indeed
Apr 29 15:31:47 <GNUtoo> even if you can call without it....
Apr 29 15:45:12 <dcordes> GNUtoo: I am intersted in the 'calling without it' thing
Apr 29 15:45:41 <dcordes> GNUtoo: audio routing - hot to do it? can you operate it on nexus outside of android ?
Apr 29 15:45:49 <dcordes> or in replicant ?
Apr 29 15:46:02 <GNUtoo> I wrote code for that
Apr 29 15:46:04 <GNUtoo> I proposed it
Apr 29 15:46:08 <GNUtoo> nobody wanted it
Apr 29 15:46:14 <GNUtoo> let me find it
Apr 29 15:47:19 <dcordes> well hello I want it :)
Apr 29 15:47:39 <GNUtoo> http://gnutoo.homelinux.org/downloads/dcordes/audio_test_programs_qsd.tar.bz2
Apr 29 15:47:41 <GNUtoo> try that
Apr 29 15:48:02 <dcordes> the good thing is: I run android on the phone. there is no wince init in the way. so maybe nexus android audio routing mechanism also works on leo now
Apr 29 15:48:08 <dcordes> thanks !
Apr 29 15:48:34 <GNUtoo> routing for GNU/Linux without alsa is in that tarball
Apr 29 15:48:45 <GNUtoo> I bet you will find how to compile it
Apr 29 15:48:53 <GNUtoo> if you have any questions ask me
Apr 29 15:48:55 <dcordes> sounds good. how to compile it? does it have any notable dependencies? 
Apr 29 15:49:03 <dcordes> hehe :D
Apr 29 15:49:16 <GNUtoo> I can try again to compile it
Apr 29 15:49:20 <GNUtoo> bitbake devshell
Apr 29 15:49:22 <GNUtoo> source devshell
Apr 29 15:49:23 <GNUtoo> and try
Apr 29 15:49:39 <dcordes> ok
Apr 29 15:50:01 <GNUtoo> arm-oe-linux-gnueabi-gcc routing.c -o routing
Apr 29 15:50:04 <GNUtoo> was easy
Apr 29 15:50:27 <GNUtoo> note that I think that volume don't work
Apr 29 15:50:33 <GNUtoo> as android handle volume in software
Apr 29 15:50:41 <GNUtoo> and that hardware volume was disabled in their kernels
Apr 29 15:50:56 <dcordes> confusing
Apr 29 15:51:08 <GNUtoo> try my programs
Apr 29 15:51:16 <GNUtoo> and then if they work fine
Apr 29 15:51:21 <GNUtoo> advertize them to captainigloo 
Apr 29 15:51:27 <GNUtoo> note that I don't have an HD2
Apr 29 15:51:29 <dcordes> captainigloo: ping
Apr 29 15:51:37 <GNUtoo> so they are untested on leo
Apr 29 15:51:39 <dcordes> captainigloo: GNUtoo might have something interesting for our devices here
Apr 29 15:51:51 * dcordes bitbakes devshell
Apr 29 15:51:51 <GNUtoo> I told him also several times
Apr 29 15:52:06 <GNUtoo> but maybe he just didn't realize
Apr 29 15:52:15 <GNUtoo> or thought that he didn't need it yet
Apr 29 15:52:20 <GNUtoo> at theses early stages
Apr 29 15:53:04 <dcordes> maybe...
Apr 29 15:53:17 <dcordes> ouf parsing takes ages
Apr 29 15:53:34 <GNUtoo> parsing is now obtimized for multicores
Apr 29 15:53:34 <dcordes> I might consider a hdd with larget spin
Apr 29 15:53:45 <GNUtoo> no, consider multicore
Apr 29 15:53:52 <dcordes> 4BitBake Build Tool Core version 1.12.0, bitbake version 1.12.0
Apr 29 15:54:03 <dcordes> model name	: Intel(R) Core(TM)2 Duo CPU     P8600  @ 2.40GHz
Apr 29 15:54:27 <GNUtoo> I've core i7, but because of theses **** bios I've some speed issues with it
Apr 29 15:54:36 <DocScrutinizer> ~rl
Apr 29 15:54:37 <apt> i heard rl is Real Life (the bad world out there), or known as the Big Blue Box (tm).  Conways game using only rational numbers
Apr 29 15:54:48 <GNUtoo> ok
Apr 29 15:55:19 <GNUtoo> so in libcmtspeech there are lot of interesting stuff including:
Apr 29 15:55:27 <GNUtoo>  * A Doc directory
Apr 29 15:56:03 <GNUtoo>  * DUMMYSRC <string>:
Apr 29 15:56:31 <GNUtoo>  Select the source for generated speech frames. Supported options    are "sine" (default, generate a sine tone), or a file name (path   must start with either "/" or ".")
Apr 29 15:56:35 <dcordes> DocScrutinizer: you are going to play it ?
Apr 29 15:57:25 <GNUtoo> The file    must be a raw 8/16bit, mono, PCM audio file (with little-endian byte    order).
Apr 29 15:57:50 <GNUtoo> so it should be "easy" hmmm(won't be easy at the end but anyway)
Apr 29 15:58:05 <dcordes> libcmtspeech what is that ?
Apr 29 15:58:15 <GNUtoo> something for n900
Apr 29 15:58:18 <dcordes> http://meego.gitorious.org/meego-middleware/libcmtspeech-ifx
Apr 29 15:58:29 <GNUtoo> basically the library that handles the audio in calls
Apr 29 15:59:41 <dcordes> ah so totally on topic
Apr 29 16:00:03 * dcordes slaps laptop
Apr 29 16:00:05 <dcordes> faster faster !
Apr 29 16:00:11 <GNUtoo> lol
Apr 29 16:00:22 <GNUtoo> that's not the good way....try another way:
Apr 29 16:00:30 <GNUtoo> take a big fan....
Apr 29 16:00:32 <dcordes> $ source devshell
Apr 29 16:00:32 <dcordes> bash: devshell: Datei oder Verzeichnis nicht gefunden
Apr 29 16:00:37 <GNUtoo> and put fresh air on it
Apr 29 16:00:43 <GNUtoo> dcordes, devshell in in:
Apr 29 16:00:57 <GNUtoo> deploy/addons/arm-oe-linux-gnueabi-shr-nokia900-devshell
Apr 29 16:01:00 <GNUtoo> for the n900
Apr 29 16:01:06 <dcordes> aight
Apr 29 16:01:08 <dcordes> thx
Apr 29 16:01:19 <GNUtoo> so for leo it would be sed s#nokia900#leo#
Apr 29 16:01:41 <dcordes> can I change wd ?
Apr 29 16:01:49 <GNUtoo> wd?
Apr 29 16:01:54 <dcordes> pwd
Apr 29 16:01:59 <GNUtoo> yes
Apr 29 16:02:00 <dcordes> or must I work in oe
Apr 29 16:02:01 <dcordes> dir
Apr 29 16:02:04 <dcordes> ok cool
Apr 29 16:02:05 <GNUtoo> source that
Apr 29 16:02:11 <GNUtoo> it'll error on ./configure
Apr 29 16:02:23 <GNUtoo> go in the dir where there is the .c files to compile
Apr 29 16:02:28 <GNUtoo> that can be outside of oe
Apr 29 16:02:33 <GNUtoo> or anywhere
Apr 29 16:02:47 <GNUtoo> and then compile them with the command I gave
Apr 29 16:03:35 <dcordes> routing:         ELF 32-bit LSB relocatable, ARM, version 1 (SYSV), not stripped
Apr 29 16:03:38 <dcordes> volume2:         ELF 32-bit LSB executable, ARM, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.16, not stripped
Apr 29 16:03:48 <GNUtoo> ok
Apr 29 16:03:51 <dcordes> tada. thanks!
Apr 29 16:03:52 <GNUtoo> scp routing
Apr 29 16:03:55 <GNUtoo> and try it
Apr 29 16:03:56 <dcordes> yep ..
Apr 29 16:04:03 <GNUtoo> it's not very robust
Apr 29 16:04:09 <GNUtoo> with the command line arguments
Apr 29 16:04:14 <dcordes> doesn't matter will do for poc
Apr 29 16:04:17 <GNUtoo> so read the source or ask me before
Apr 29 16:05:03 <GNUtoo>         printf("Usage: ./route type path_name\n");
Apr 29 16:05:03 <GNUtoo>         printf("Use \"./route help routes\" to get all the routes\n");
Apr 29 16:05:03 <GNUtoo>         printf("Use \"./route help types\" to get all types\n");
Apr 29 16:05:51 <dcordes> that's good
Apr 29 16:05:58 <DocScrutinizer> Triggering protocol message is: CMTSPEECH_SPEECH_CONFIG_REQ   This event is used by the cellular modem to initiate transfer of downlink speech frames (from network).
Apr 29 16:06:32 <captainigloo> dcordes: pong
Apr 29 16:06:49 <DocScrutinizer> After this message, application should be prepared to receive CMTSPEECH_EVENT_DATA events. 
Apr 29 16:06:51 <dcordes> go go go !
Apr 29 16:07:26 <GNUtoo> you got audio?
Apr 29 16:07:33 <GNUtoo> (for dcordes )
Apr 29 16:07:39 <captainigloo> oh ?
Apr 29 16:07:43 <captainigloo> with alsa ?
Apr 29 16:07:46 <GNUtoo> captainigloo, no
Apr 29 16:07:51 <captainigloo> ah :)
Apr 29 16:07:54 <GNUtoo> captainigloo, alsa must be tested and published
Apr 29 16:07:59 <captainigloo> oh ?
Apr 29 16:07:59 <GNUtoo> altough it's written
Apr 29 16:08:13 <GNUtoo> so we need to wait more for alsa
Apr 29 16:08:29 <GNUtoo> but basically nothing new
Apr 29 16:08:34 <captainigloo> ah ok
Apr 29 16:08:35 <GNUtoo> only some old stuff such as:
Apr 29 16:08:47 <GNUtoo> audio during calls with a routing program I wrote 
Apr 29 16:08:53 <GNUtoo> and that I even told you about
Apr 29 16:09:05 <captainigloo> you i remember
Apr 29 16:09:13 <GNUtoo> yes
Apr 29 16:09:14 <captainigloo> s/you/yes
Apr 29 16:09:28 <GNUtoo> do you want it?
Apr 29 16:09:34 <captainigloo> i'm a bit busy
Apr 29 16:09:38 <GNUtoo> ok
Apr 29 16:09:40 <GNUtoo> lol
Apr 29 16:09:47 <captainigloo> but if you could post the link somewhere
Apr 29 16:09:49 <captainigloo> i will check later
Apr 29 16:10:04 <GNUtoo> http://gnutoo.homelinux.org/downloads/dcordes/audio_test_programs_qsd.tar.bz2
Apr 29 16:10:11 <GNUtoo> it
Apr 29 16:10:28 <GNUtoo> it's just a test program, not very robust with the command line arguments
Apr 29 16:10:45 <GNUtoo> but it worked on nexusone
Apr 29 16:11:44 <captainigloo> bookmarked
Apr 29 16:11:48 <captainigloo> thanks
Apr 29 16:12:02 <captainigloo> bbl
Apr 29 16:15:01 <GNUtoo> DocScrutinizer:
Apr 29 16:15:08 <GNUtoo> int cmtspeech_state_change_call_status(cmtspeech_t *context, bool status){
Apr 29 16:15:11 <GNUtoo> [...]
Apr 29 16:15:41 <GNUtoo>   if (self->s.call_server_status != status) {
Apr 29 16:15:49 <GNUtoo>     if (status) {
Apr 29 16:15:50 <GNUtoo> }
Apr 29 16:15:54 <GNUtoo> else {
Apr 29 16:15:58 <GNUtoo>         priv_emit_speech_config_req(self, 0, 0);
Apr 29 16:16:07 <GNUtoo> strange....
Apr 29 16:16:28 <GNUtoo> basically
Apr 29 16:16:59 <GNUtoo> change_status is what is called by the applications(the vala test, the nitdroid test etc...)
Apr 29 16:17:36 <GNUtoo> if status is true, it does nothing ( priv_change_state(self, STATE_INITIALIZED); )
Apr 29 16:17:43 <GNUtoo> if it's false it does that:
Apr 29 16:17:54 <GNUtoo> it terminate everything and calls:
Apr 29 16:17:59 <GNUtoo>         priv_emit_speech_config_req(self, 0, 0);
Apr 29 16:18:07 <GNUtoo> which does that:
Apr 29 16:18:38 <GNUtoo> (all that was in dummy-backend/cmtspeech_dummy.c in libcmtspeechdata)
Apr 29 16:19:16 <GNUtoo> priv_emit_speech_config_req does that :
Apr 29 16:19:18 <GNUtoo>  * Emits a SPEECH_CONFIG_REQ towards APE:
Apr 29 16:21:00 <GNUtoo> bbl
Apr 29 16:21:11 <dcordes> # routing
Apr 29 16:21:11 <dcordes> /usr/sbin/routing: line 1: syntax error: unexpected word (expecting ")")
Apr 29 16:22:28 <DocScrutinizer> you should read http://meego.gitorious.org/meego-middleware/libcmtspeech-ifx/blobs/wsiv_swdev-integ/doc/libcmtspeechdata_api_docs_main.txt
Apr 29 16:22:58 <DocScrutinizer> it's quite enlightening about what weird things they expect your app to do
Apr 29 16:23:23 <dcordes> captainigloo: hmm maybe I did some mistake
Apr 29 16:23:57 <dcordes> captainigloo: compile time
Apr 29 16:26:08 <dcordes> ~/routing # ls /dev/ | grep msm
Apr 29 16:26:08 <dcordes> msm_aac_in
Apr 29 16:26:08 <dcordes> msm_audio_ctl
Apr 29 16:26:08 <dcordes> msm_audio_route
Apr 29 16:26:08 <dcordes> msm_mp3
Apr 29 16:26:11 <dcordes> msm_pcm_in
Apr 29 16:26:13 <dcordes> msm_pcm_out
Apr 29 16:26:16 <dcordes> msm_qcelp_in
Apr 29 16:26:32 * dcordes needs gnutoo :(
Apr 29 16:26:49 <dcordes> I compiled it like he said
Apr 29 16:26:53 <dcordes> no idea why it won't work
Apr 29 16:27:43 <dcordes> ~/routing # ./volume2 20
Apr 29 16:27:43 <dcordes> Setting in-call volume to 400
Apr 29 16:27:53 <dcordes> then it seems to freeze
Apr 29 17:07:13 <dcordes> argh
Apr 29 17:07:15 <dcordes> bbl
Apr 29 17:18:05 <GNUtoo> back
Apr 29 17:20:22 <dcordes> GNUtoo: un success ful
Apr 29 17:20:28 <GNUtoo> ah?
Apr 29 17:20:30 <GNUtoo> I'll retry
Apr 29 17:20:36 <dcordes> /usr/sbin/routing: line 1: syntax error: unexpected word (expecting ")")
Apr 29 17:20:43 <dcordes> volume2 works
Apr 29 17:20:46 <GNUtoo> ah
Apr 29 17:20:49 <GNUtoo> let me look
Apr 29 17:20:52 <dcordes> well it hangs when I send stuff
Apr 29 17:21:02 <dcordes> but routing doesn't even show the help. ito nly give this error. I compiled as you told me
Apr 29 17:21:12 <GNUtoo> ok 
Apr 29 17:21:15 <GNUtoo> I'll try
Apr 29 17:35:49 <GNUtoo> dcordes, sorry I gave you an old version of the progrtams
Apr 29 17:38:59 <mrmoku> GNUtoo: registration problem solved?
Apr 29 17:39:05 <GNUtoo> yes
Apr 29 17:39:10 <GNUtoo> using the GUI
Apr 29 17:39:23 <mrmoku> ok
Apr 29 17:40:18 <GNUtoo> ok I'll push to gitorious
Apr 29 17:43:23 <GNUtoo> dcordes, https://gitorious.org/htc-msm-2-6-32/uttilities
Apr 29 17:43:29 <GNUtoo> howto do a call:
Apr 29 17:44:07 <GNUtoo> *)Plug headphones inside the phone and put them on your ears or near them, beware of the possible high volume
Apr 29 17:44:22 <GNUtoo> *) mdbus2 -s org.freesmartphone.ogsmd  /org/freesmartphone/GSM/Device org.freesmartphone.GSM.Call.Initiate 4242 voice
Apr 29 17:44:29 <GNUtoo> replace 4242 by the number to call
Apr 29 17:44:41 <GNUtoo> ./route CALL_START HEADSET_SPKR_STEREO
Apr 29 17:44:49 <GNUtoo> that will route to headphones
Apr 29 17:44:52 <GNUtoo> despite of the name
Apr 29 17:46:36 <dcordes> GNUtoo: cool thanks man
Apr 29 17:46:52 <playya__> dcordes, remove -c switch from the gcc
Apr 29 17:47:00 <playya__> call
Apr 29 17:47:14 <dcordes> ~lart dcordes 
Apr 29 17:47:16 * apt takes a big bite out of dcordes's jugular vein
Apr 29 17:47:21 <dcordes> oh !
Apr 29 17:47:26 <playya__> this always happends if you want to exec an object file
Apr 29 17:47:34 <dcordes> you can take a bite of my sbuclavian artery
Apr 29 17:47:41 <GNUtoo> arm-oe-linux-gnueabi-gcc routing.c -o routing
Apr 29 17:47:50 <dcordes> playya__: thanks
Apr 29 17:47:58 <playya__> you're welcome
Apr 29 17:48:04 <GNUtoo> that's what I pasted before....thanks playya__ nice catch
Apr 29 17:48:12 <DocScrutinizer> where do I find #unix-gurus ?
Apr 29 17:48:41 <GNUtoo> now back to cmt_speech
Apr 29 17:48:44 <DocScrutinizer> or meybe here are a few who don't even know or are too humble :-)
Apr 29 17:48:52 <GNUtoo> dcordes, hurry to get it working
Apr 29 17:49:08 <DocScrutinizer> how would I feed kbd-lookalike input to an arbitrary app?
Apr 29 17:49:23 <GNUtoo> hmmm
Apr 29 17:49:25 <GNUtoo> no idea
Apr 29 17:49:32 <GNUtoo> but maybe captainigloo knows
Apr 29 17:49:33 <dcordes> GNUtoo: ok
Apr 29 17:49:40 <GNUtoo> or #e people
Apr 29 17:49:42 <playya__> DocScrutinizer, libfakekey
Apr 29 17:49:50 <DocScrutinizer> o.O
Apr 29 17:49:50 <playya__> or use /proc ;)
Apr 29 17:50:22 <dcordes> git clone git://gitorious.org/htc-msm-2-6-32/uttilities.git
Apr 29 17:50:22 <DocScrutinizer> yeah, what I been thinking, sth like echo "P" >/proc/<process>/stdin
Apr 29 17:50:48 <DocScrutinizer> don't think it's that easy
Apr 29 17:51:07 <dcordes> GNUtoo: should I also compile the other binaries ?
Apr 29 17:51:17 <GNUtoo> no
Apr 29 17:51:20 <GNUtoo> only the route thing
Apr 29 17:51:28 <GNUtoo> I don't remember the use of the other stuff
Apr 29 17:51:40 <dcordes> okay
Apr 29 17:52:10 <DocScrutinizer> playya__: any details about "use /proc"?
Apr 29 17:52:22 <playya__> it's a X app?
Apr 29 17:52:31 <DocScrutinizer> usually yes
Apr 29 17:52:51 <DocScrutinizer> playya__: it's about speech input
Apr 29 17:52:58 <DocScrutinizer> generic
Apr 29 17:53:10 <playya__> hmm
Apr 29 17:53:24 <playya__> maybe register a uinput device from a daemon
Apr 29 17:53:30 <lindi-> DocScrutinizer: are you talking about applications connected to a terminal emulator?
Apr 29 17:53:39 <DocScrutinizer> I thought about the speech engine creating /dev/input99
Apr 29 17:53:46 <playya__> yes
Apr 29 17:54:26 <dcordes> ~/routing/uttilities # ./routing
Apr 29 17:54:29 <dcordes> Usage: ./route type path_name~
Apr 29 17:54:32 <dcordes> GNUtoo: wokrs!
Apr 29 17:54:33 <DocScrutinizer> lindi-: sending >CR> to an arbitrary app's stdin when I say 'enter'
Apr 29 17:54:53 <GNUtoo> dcordes, try to make a call
Apr 29 17:55:21 <DocScrutinizer> playya__: what does libfakekey?
Apr 29 17:55:26 <dcordes> GNUtoo: I didmn't run any command yet
Apr 29 17:55:39 <GNUtoo> dcordes, indeed that's why I said to try to make a call
Apr 29 17:55:50 <GNUtoo> I already pasted the commands upper
Apr 29 17:55:54 <dcordes> ok
Apr 29 17:56:01 <GNUtoo> <GNUtoo> howto do a call:
Apr 29 17:56:06 <lindi-> DocScrutinizer: ok so would what virtual keyboards do be ok for you?
Apr 29 17:56:19 <playya__> afaik maemo-keyboard and e-keyboard are using it
Apr 29 17:56:23 <DocScrutinizer> yes, I think it would
Apr 29 17:56:36 <dcordes> ./route CALL_START HEADSET_SPKR_STEREO
Apr 29 17:56:42 <dcordes> GNUtoo: don'T have headset
Apr 29 17:56:48 <dcordes> GNUtoo: I have headphones
Apr 29 17:56:54 <GNUtoo> ah? even headphones?
Apr 29 17:56:57 <playya__> DocScrutinizer, eschreibung: library for converting characters to X key-presses [runtime]
Apr 29 17:56:58 <playya__>  libfakekey is a simple library for converting UTF-8 characters into 'fake' X key-presses. 
Apr 29 17:56:59 <GNUtoo> try with headphones
Apr 29 17:57:03 <dcordes> ok!
Apr 29 17:57:03 <GNUtoo> call an operator number
Apr 29 17:57:07 <dcordes> connected
Apr 29 17:57:08 <dcordes> ok
Apr 29 17:57:10 <GNUtoo> like one that is free as in cost
Apr 29 17:57:14 <dcordes> 1212
Apr 29 17:57:15 <GNUtoo> and that says press one for 
Apr 29 17:57:19 <GNUtoo> press two for bar
Apr 29 17:57:21 <GNUtoo> etc...
Apr 29 17:57:27 <DocScrutinizer> playya__: thanks
Apr 29 17:57:28 <playya__> i know dcordes PIN :P
Apr 29 17:57:51 <dcordes> how can I close the contacts window ?
Apr 29 17:57:54 <GNUtoo> playya__, isn't 1212 an operator number instead
Apr 29 17:58:17 <dcordes> playya__: dcordes has no pin
Apr 29 17:58:18 <GNUtoo> dcordes, with mdbus2 or as usual with the top bar + close button
Apr 29 17:58:43 <dcordes> my top bar is gone
Apr 29 17:58:44 <lindi-> DocScrutinizer: xte "str Return"  emulates hitting enter
Apr 29 17:58:47 <GNUtoo> I know leviathan's pin
Apr 29 17:58:52 <GNUtoo> I bet it's 1337
Apr 29 17:58:55 <dcordes> not the frist time my top bar disappeared for no reason
Apr 29 17:59:02 <dcordes> lol leviathan 
Apr 29 17:59:10 <DocScrutinizer> lindi-: \o/
Apr 29 17:59:18 <dcordes> calling
Apr 29 17:59:22 <dcordes> looolllll
Apr 29 17:59:24 <dcordes> sepaker !
Apr 29 17:59:33 <dcordes> aaahahahhahahaha !
Apr 29 17:59:40 <GNUtoo> so it works
Apr 29 17:59:49 <DocScrutinizer> sebaka the wookie
Apr 29 17:59:50 <GNUtoo> maybe that's because I routed to headphones first
Apr 29 18:00:05 <dcordes> :D
Apr 29 18:00:14 * dcordes lights some fireworks
Apr 29 18:00:30 <dcordes> the funny part is: the operator lady has robotic voice
Apr 29 18:00:34 <GNUtoo> lol
Apr 29 18:00:39 <dcordes> it is a known problem: the DSP needs some init sequence
Apr 29 18:00:42 <dcordes> I need to find it out
Apr 29 18:00:44 <GNUtoo> ok
Apr 29 18:00:45 <dcordes> cotulla knows
Apr 29 18:00:52 <dcordes> haha :D
Apr 29 18:01:05 <GNUtoo> maybe you prefer the robotic voice?
Apr 29 18:01:14 <GNUtoo> (since it's funny)
Apr 29 18:01:15 <antrik> DocScrutinizer: hm... perhaps you could get GoldenDelicious or so to import the caps? surely easier to handle than doing it privately :-)
Apr 29 18:01:24 <dcordes> yes it gives me lot of joy
Apr 29 18:01:38 <DocScrutinizer> antrik: thought about that, yes
Apr 29 18:01:40 <antrik> (don't they need them anyways for GTA04?...)
Apr 29 18:02:05 <dcordes> it sounds like mickey mouse after inhaling He
Apr 29 18:02:09 <dcordes> with a machine touch
Apr 29 18:02:13 <DocScrutinizer> *should* - I sincerly hope they didn't opt for another go with those crappy LiIon
Apr 29 18:02:34 <dcordes> GNUtoo: ok to stop joking for a while... your problem is gold !
Apr 29 18:02:47 <dcordes> GNUtoo: now we must implement it in shr/fso
Apr 29 18:03:20 <dcordes> s/problem/program/
Apr 29 18:03:51 <DocScrutinizer> fools, use sox for funny effects!
Apr 29 18:04:02 <GNUtoo> ok
Apr 29 18:04:26 <GNUtoo> dcordes, why not wait for cotulla's alsa
Apr 29 18:04:37 <GNUtoo> ?
Apr 29 18:04:39 <DocScrutinizer> maybe even via laspa
Apr 29 18:04:44 <DocScrutinizer> ladspa
Apr 29 18:04:47 <GNUtoo> maybe he has routing inside it too
Apr 29 18:05:11 <dcordes> GNUtoo: does it cover audio routing?
Apr 29 18:05:32 <GNUtoo> alsa should cover it, but I've no idea if he implemented that too
Apr 29 18:06:18 <DocScrutinizer> playya__: I gather X doesn't by default use ALL /dev/input*, and also doesn't recognize when a new input device appears there after X been started?
Apr 29 18:06:43 <lindi-> DocScrutinizer: hijacking such low level stuff is probably not a good idea
Apr 29 18:06:55 <DocScrutinizer> otoh new kbds get noticed immediately
Apr 29 18:07:21 * DocScrutinizer scratches head
Apr 29 18:07:38 <DocScrutinizer> unix and kbd input - always a sure headache
Apr 29 18:09:31 <dcordes> GNUtoo: if we can make calls happen on nexus, desire, leo right now-  why wait ?
Apr 29 18:11:42 <GNUtoo> because I'm ultra busy
Apr 29 18:11:50 <GNUtoo> if it's something quick why not
Apr 29 18:11:55 <GNUtoo> but if it's not quick...
Apr 29 18:12:17 <GNUtoo> maybe let's do a separate program
Apr 29 18:12:23 <GNUtoo> that listen for some calls
Apr 29 18:12:26 <GNUtoo> if there is a call
Apr 29 18:12:30 <GNUtoo> it switch audio for calls
Apr 29 18:12:44 <GNUtoo> and if the call ends
Apr 29 18:12:54 <GNUtoo> it switches audio back to normal
Apr 29 18:13:04 <antrik> DocScrutinizer: X on Linux uses udev nowadays to register new inputs
Apr 29 18:13:22 <DocScrutinizer> antrik: yeah
Apr 29 18:13:30 <DocScrutinizer> oh, X
Apr 29 18:13:40 <DocScrutinizer> o.O
Apr 29 18:13:56 <GNUtoo> dcordes, personally my priority is the N900 with regard to FSO+SHR
Apr 29 18:14:21 <GNUtoo> (since alsa is not there yet at the moment)
Apr 29 18:14:40 <GNUtoo> then if alsa comes I'll still have to finish n900
Apr 29 18:14:50 <GNUtoo> since I started
Apr 29 18:14:56 <GNUtoo> finish means help in the crucial part
Apr 29 18:15:00 <antrik> uinput simulates input devices AIUI? so I guess if you add one that pretends to be a keyboard, X will automatically pick it up
Apr 29 18:15:01 <GNUtoo> not do everything
Apr 29 18:15:05 <GNUtoo> but unblock people
Apr 29 18:15:13 <DocScrutinizer> lindi-: no xte on any of my machines
Apr 29 18:15:16 <GNUtoo> or work on things nobody is working on
Apr 29 18:15:22 <GNUtoo> for instance I can let mrmoku work on modem
Apr 29 18:15:43 <GNUtoo> but since nobody wants to work on cmt_speech it seem that I have to do it
Apr 29 18:16:02 <DocScrutinizer> antrik: I think that's the way to go for a universal speech input daemon
Apr 29 18:17:10 <antrik> I'm not sure. #xorg-devel would be a better place to ask
Apr 29 18:17:19 <DocScrutinizer> :nod:
Apr 29 18:17:29 <CIA-10> freesmartphone.org: 03morphis 07aurora * r095d7ad04b9b 10/aurora/AUTHORS: aurora: add AUTHORS file
Apr 29 18:17:32 <CIA-10> freesmartphone.org: 03morphis 07aurora * r109c4a4a3ef9 10/aurora/Aurora/Components/Theme.qml: aurora: adjust font size of text shown in the statusbar
Apr 29 18:17:34 <CIA-10> freesmartphone.org: 03morphis 07aurora * r17fe5baefc1c 10/aurora/ (4 files in 3 dirs): aurora: add a dummy implementation of the necessary indicators
Apr 29 18:17:36 <CIA-10> freesmartphone.org: 03morphis 07aurora * raf9f7be81b00 10/aurora/ (Aurora/Components/Window.qml aurora main.qml): aurora: make window show the statusbar on top of the screen
Apr 29 18:18:14 <DocScrutinizer> I tried a few freaky things aling the line "echo test >/proc/4554/fd/*" but the xxx'th time with no success
Apr 29 18:19:24 <antrik> nah, that can't work. if an application writes to its own stdin, it won't get a keypress either :-)
Apr 29 18:19:52 <DocScrutinizer> you can make text appear on arbitrary apps' *output* no matter which fd you echo it to, but obviously nothing will be seen by app's stdin and get parsed
Apr 29 18:19:56 <antrik> if stdin is connected to a terminal, it will probably produce screen output; but that's it
Apr 29 18:20:05 <antrik> exactly
Apr 29 18:21:18 <DocScrutinizer> would have to wrap my head further around pty's and that other associated thing
Apr 29 18:21:43 <antrik> if you want to push something to an application's stdin, you have to run that application on a pty slave, and write to the pty master
Apr 29 18:21:47 <GNUtoo> [Rui], hi
Apr 29 18:21:52 <antrik> but that won't do much good for most X apps...
Apr 29 18:21:58 <DocScrutinizer> antrik: exactly :-)
Apr 29 18:22:43 <antrik> only X app I know that actually reads from stdin is MagickPoint...
Apr 29 18:22:57 <antrik> err magicpoint
Apr 29 18:23:00 <DocScrutinizer> yeah, and as we got not even consoles on maemo or - for that behalf - SHR, I'm not much concerned to find a way beyond "faking" X input
Apr 29 18:23:50 <DocScrutinizer> so /dev/input99 is it then
Apr 29 18:24:16 <antrik> I know that X has some extension(s) that allow faking input events (that's probably what this libfakekey uses)
Apr 29 18:24:28 <antrik> that's what you would want to use to send to specific applications
Apr 29 18:24:34 <DocScrutinizer> and xte which lindi- mentioned
Apr 29 18:24:50 <antrik> the uinput way would send it to the X server itself, which would than use the usual method for distributing it to clients...
Apr 29 18:25:01 <DocScrutinizer> yep
Apr 29 18:25:02 <antrik> not sure which is more useful for you
Apr 29 18:25:15 <DocScrutinizer> fine for top level - I.E. window manager
Apr 29 18:25:30 <DocScrutinizer> aka hildon-desktop
Apr 29 18:25:37 <antrik> I never heard of an xte. I know there is an xtest extension, but I have no idea how it is used. it *might* be what libfakekey uses
Apr 29 18:26:02 <DocScrutinizer> antrik: many thanks for the sparring
Apr 29 18:28:04 <mrmoku> DocScrutinizer: if my soldering skills wouldn't be negative... I'd consider taking two of those bats :)
Apr 29 18:28:25 <mrmoku> DocScrutinizer: btw. any idea what 'required_fc_tx' might be?
Apr 29 18:28:31 <GNUtoo> [Rui], you have a galaxy S right?
Apr 29 18:28:50 <GNUtoo> [Rui], there is an user named igor- in #replicant that has one too
Apr 29 18:29:00 <GNUtoo> maybe you could unite to reverse engineer that modem
Apr 29 18:29:03 <DocScrutinizer> mrmoku: no idea. Which context?
Apr 29 18:29:06 <GNUtoo> *modem protocol
Apr 29 18:29:46 <DocScrutinizer> tx sounds like RF
Apr 29 18:29:50 <DocScrutinizer> or serial
Apr 29 18:29:55 <mrmoku> DocScrutinizer: PEP pipe
Apr 29 18:30:03 <mrmoku> broader context is GPRS
Apr 29 18:30:04 <DocScrutinizer> whatever that is
Apr 29 18:30:08 <DocScrutinizer> aaah
Apr 29 18:30:22 <DocScrutinizer> frequency correction?
Apr 29 18:30:32 <GNUtoo> phonet_pipe?
Apr 29 18:30:34 <DocScrutinizer> no idea
Apr 29 18:30:36 <GNUtoo> like the kernel module?
Apr 29 18:30:37 <mrmoku> GNUtoo: yup
Apr 29 18:30:39 <mrmoku> no
Apr 29 18:30:40 <GNUtoo> ok
Apr 29 18:30:44 <GNUtoo> ah?
Apr 29 18:30:54 <mrmoku> not the kernel module... the pipe resource of isi
Apr 29 18:31:01 <GNUtoo> ah ok
Apr 29 18:31:03 * mrmoku got those two confused too :)
Apr 29 18:31:17 <mrmoku> the kernel module is for different phones me thinks
Apr 29 18:31:23 <GNUtoo> ok
Apr 29 18:31:28 <DocScrutinizer> grep thru wireless modem api specs?
Apr 29 18:31:36 <mrmoku> ok, good idea
Apr 29 18:31:43 <GNUtoo> like features phones running symbian
Apr 29 18:31:45 <GNUtoo> ok
Apr 29 18:33:56 <mrmoku> DocScrutinizer: flow control
Apr 29 18:34:14 <DocScrutinizer> meh, chk
Apr 29 18:34:27 <mrmoku> GNUtoo: I think I read some SE phone using a slim version of isi needs that IIRC
Apr 29 18:34:29 <DocScrutinizer> too simple ;-P
Apr 29 18:34:31 <mrmoku> :P
Apr 29 18:34:47 <DocScrutinizer> I wasn't bad with serial
Apr 29 18:34:54 <GNUtoo> ok
Apr 29 18:42:58 <DocScrutinizer> anybody experienced with concept of commercial speech input like dragon dictate or whatever they're called?
Apr 29 18:44:45 <dcordes> ok robot voice fixed.
Apr 29 18:45:03 <dcordes> GNUtoo: how to route to handset ?
Apr 29 18:45:33 <GNUtoo> I don't remember but I can look
Apr 29 18:45:42 <GNUtoo> maybe not immediately tough
Apr 29 18:45:57 <GNUtoo> handset beeing less volume?
Apr 29 18:46:20 <GNUtoo> if so it's complicated
Apr 29 18:46:28 <[Rui]> GNUtoo: back, nice news. will look into it.
Apr 29 18:46:34 <GNUtoo> I started it for dream and didn't finish it yet
Apr 29 18:46:40 <GNUtoo> [Rui], maybe come in #replicant
Apr 29 18:48:59 <nschle85> hello :-)
Apr 29 18:51:24 <nschle85> I have a question, independent on om dev, Ill move into an another flat, which cheap usb dongle (umts/hsdpa) with linux support should i buy ? (i dont hav a N900 :-) ) 
Apr 29 18:57:02 <CIA-10> freesmartphone.org: 03mickey 07cornucopia * r4f9e7c492055 10/fsogsmd/src/plugins/dbus_service/plugin.vala: fsogsmd: dbus_service: set proper status check for every dbus method
Apr 29 18:57:12 <antrik> nschle85: I know some people who successfully use the O2 ones -- but there is no guarantee that they won't suddently change the hardware. only safe thing is to go to the shop, and say that you only buy it if they let you test it beforehand
Apr 29 18:59:46 <nschle85> antrik: thank you, iam using simyo, they have one for 39 euros but i need it for round abot 14 days so the hardware should be cheap and working
Apr 29 19:03:58 <antrik> nschle85: you probably don't want to use the same SIM as you use for phoning; so you can pretty much pick any provider...
Apr 29 19:05:46 <antrik> BTW, are you sure you won't survive two weeks without internet?... ;-)
Apr 29 19:07:00 <nschle85> antrik: no, a am planning a europe trip withing this time, checking bank accounts, checking mails....
Apr 29 19:07:32 <nschle85> antrik: tracking waht happens on shr :-)
Apr 29 19:11:31 <nschle85> antrik: and last but not least, my phone at home is provided by sipgate ....
Apr 29 19:11:59 <DocScrutinizer> ohohoh, data roaming
Apr 29 19:12:31 <darkh> 1 minute online could be more expensive than the whole stick ;)
Apr 29 19:12:39 <DocScrutinizer> still expensive, while with airbag now
Apr 29 19:13:25 <nschle85> no, simyo: 10 euros for 30 days flat, after 1 GB it has limited speed
Apr 29 19:13:41 <DocScrutinizer> I don't know of any semi decent data rate for international roaming
Apr 29 19:14:22 <nschle85> i dont need roaming (would be nice to have )
Apr 29 19:14:29 <DocScrutinizer> and simyo best thing that you can get is no data traffic at all abroad
Apr 29 19:15:05 <DocScrutinizer> worst is your account has -80EUR  in a few seconds
Apr 29 19:15:12 <DocScrutinizer> the gets cut
Apr 29 19:15:29 <nschle85> DocScrutinizer:  i need the stick only in germany
Apr 29 19:15:40 <DocScrutinizer> ooh, what abut the EU trip then?
Apr 29 19:16:07 <nschle85> DocScrutinizer: i talked about planning that
Apr 29 19:16:11 <DocScrutinizer> seems I been mistaken
Apr 29 19:16:18 <DocScrutinizer> ok
Apr 29 19:16:28 <DocScrutinizer> then go for simyo or whatever you like
Apr 29 19:16:44 <DocScrutinizer> I think O2 is really decent
Apr 29 19:17:02 <DocScrutinizer> though a itsy tad more expensive than e.g simyo
Apr 29 19:17:12 <DocScrutinizer> but then has better bandwidth
Apr 29 19:17:14 <nschle85> DocScrutinizer: we booked some ferries but detailed hotel booking ill do after voving
Apr 29 19:17:18 <nschle85> moving
Apr 29 19:17:44 <DocScrutinizer> way better
Apr 29 19:17:58 <nschle85> DocScrutinizer: i am looking for some cheap hardware
Apr 29 19:18:05 <DocScrutinizer> O2
Apr 29 19:18:16 <DocScrutinizer> surfstick
Apr 29 19:18:27 <DocScrutinizer> like 25EUR irrc
Apr 29 19:18:39 <nschle85> DocScrutinizer: is it simlock and netlock free ?
Apr 29 19:18:41 <DocScrutinizer> dunno about linux
Apr 29 19:18:53 <DocScrutinizer> nfc
Apr 29 19:19:13 <nschle85> DocScrutinizer: linux is a must have ! 
Apr 29 19:19:40 <DocScrutinizer> sure, see what antrik said
Apr 29 19:19:57 <DocScrutinizer> there are O2 shops all over the town
Apr 29 19:20:34 <DocScrutinizer> (in Germany)
Apr 29 19:23:46 <nschle85> DocScrutinizer: thank you, you have no alternative to buy the 39€ stick from simyo. 
Apr 29 19:24:22 <nschle85> DocScrutinizer: this is the argument to buy the N900 now and not to throw away 39€
Apr 29 19:24:50 <nschle85> DocScrutinizer: N900 can thetering ? i hope so
Apr 29 19:25:00 <DocScrutinizer> basically yes
Apr 29 19:25:22 <nschle85> DocScrutinizer: what does basically mean ?
Apr 29 19:25:28 <DocScrutinizer> works like a charm with windows, on linux results are ambivalent
Apr 29 19:25:45 <GNUoo|n900> ok backend of cmtspeech is alreasy nokia....hmmm
Apr 29 19:25:48 <DocScrutinizer> brute force always works though
Apr 29 19:26:08 <antrik> nschle85: the problem with simyo is that you probably can't test the hardware before buying; so it would be pretty risky
Apr 29 19:26:44 <nschle85> antrik: the stick they are selling has 100% linus support
Apr 29 19:27:24 <nschle85> DocScrutinizer: are you talking about shr image for N900 or Maemo ?
Apr 29 19:27:28 <antrik> do you get a guarantee that they won't change the hardware?
Apr 29 19:28:21 <antrik> anyways, I guess you won't get much cheaper than that if you want to buy it without a contract attached
Apr 29 19:30:00 <nschle85> antrik: if i buy a Huawei Data Stick E173 on simyo platform, hat can they change ?
Apr 29 19:30:20 <nschle85> -hat +what
Apr 29 19:31:16 <ArtoX> Hello everyone
Apr 29 19:31:31 <ArtoX> I am having issues building shr-unstable
Apr 29 19:31:40 <ArtoX> for quite a while now
Apr 29 19:32:14 <ArtoX> http://pastebin.com/wVGfjJaA
Apr 29 19:32:27 <ArtoX> somehow it fails there
Apr 29 19:32:30 <nschle85> DocScrutinizer: is thetering working with N900 and installed OS (memo) and linux clients ? 
Apr 29 19:32:37 <ArtoX> even when I start fresh from scratch
Apr 29 19:32:50 <antrik> nschle85: well, with WLAN sticks it often happend in the past that people bought one after reading about good Linux support, but than it turned out the "same" product uses a completely different chipset now... I guess this could happen here too
Apr 29 19:33:20 <nschle85> antrik: this i did not know
Apr 29 19:34:07 <antrik> don't know whether this happend for UMTS sticks too
Apr 29 19:34:26 <antrik> guess the risk is not too big...
Apr 29 19:34:39 <antrik> and you can still send it back if it really doesn't work :-)
Apr 29 19:34:41 <nschle85> ArtoX: have you attached all necessary patches ?
Apr 29 19:35:40 <nschle85> antrik: if N900 would work for me i would buy one now but DocScrutinizer does not answer
Apr 29 19:35:57 <ArtoX> nschle85: I just tried to build shr as state by the wiki
Apr 29 19:36:05 <ArtoX> and I have just run make update
Apr 29 19:38:31 <nschle85> ArtoX: i did not build shr from scratch for a while, please read the troubleshooting section
Apr 29 19:39:48 <nschle85> Artox: section: ffmpeg and libav 
Apr 29 19:40:08 <Artox> found it
Apr 29 19:40:17 <Artox> should have reread the wiki really
Apr 29 19:40:31 <Artox> last time I read it was weeks ago
Apr 29 19:40:34 <Artox> thx
Apr 29 19:41:14 <nschle85> Artox: i have always the same problems like you, so I always escalade every build issue.
Apr 29 19:42:28 <nschle85> Artox: JaMa now documents how to build shr in the wiki, before the bug is fixed.
Apr 29 19:43:59 <nschle85> Artox: does it work ?
Apr 29 19:44:13 <Artox> havent tested yet
Apr 29 19:44:59 <nschle85> Artox: please test it :-)
Apr 29 19:45:33 <nschle85> Artox: if it does not work, we have anoter problem
Apr 29 19:45:50 <Artox> *** Please tell me who you are.
Apr 29 19:45:52 <Artox> what?
Apr 29 19:46:19 <Artox> git wants to know my identity
Apr 29 19:46:20 <nschle85> Artox: who ? what do you mean ?
Apr 29 19:46:39 <nschle85> Artox: ok, wait
Apr 29 19:46:49 <lindi-> Artox: yep you should configure your name and email?
Apr 29 19:46:55 <Artox> k
Apr 29 19:48:18 <Artox> bitbake running again
Apr 29 19:50:02 <nschle85> Artox: you can do this:
Apr 29 19:50:26 <nschle85> wget -nv http://patches.openembedded.org/patch/935/mbox/ -O pw-am-935.patch
Apr 29 19:50:48 <nschle85> patch -p1 < ../../pw-am-935.patch
Apr 29 19:51:36 <Artox> it already works
Apr 29 19:51:38 <Artox> but thx
Apr 29 19:51:46 <Artox> it built libav
Apr 29 19:52:49 <nschle85> Artox: i am one of the documenters, how did you get it working ?
Apr 29 19:53:52 <Artox> as stated in the wiki
Apr 29 19:54:02 <Artox> cd shr-unstable/openembedded
Apr 29 19:54:04 <Artox> contrib/patchwork/pw-am.sh 935
Apr 29 19:54:09 <Artox> those 2 commands did it
Apr 29 19:55:03 <nschle85> Artox: and why did you get some git questions about your name ?
Apr 29 19:55:43 <nschle85> Artox: and how did you solve them ?
Apr 29 19:56:20 <nschle85> Artox: please tell ma all you did, i would like to document it
Apr 29 19:56:33 <nschle85> -ma +me
Apr 29 19:56:53 <Artox> mhm
Apr 29 19:56:59 <Artox> cant scroll back
Apr 29 19:57:04 <Artox> buffer overwritten in terminal
Apr 29 19:57:06 <Artox> but
Apr 29 19:57:17 <Artox> it told mne how to set email and name
Apr 29 19:57:28 <Artox> git config --global user.name "yourname"
Apr 29 19:57:30 <Artox> sthg like that
Apr 29 19:58:20 <Artox> I was justr suprised by this "*** Please tell me who you are."
Apr 29 19:58:24 <Artox> remembered me of google
Apr 29 19:58:27 <Artox> and sony
Apr 29 19:58:28 <nschle85> Artox: did you entered some git commands ?
Apr 29 19:58:45 <Artox> 2 commands to set my name and my email address
Apr 29 19:59:03 <Artox> the error message on the screen told me those
Apr 29 19:59:44 <nschle85> Artox: ups i thought you want to know who I am :-)
Apr 29 19:59:56 <Artox> oh
Apr 29 20:00:22 <Artox> but usually only Microsoft and Google want to know who I am
Apr 29 20:00:32 <Artox> and I was really shocked when I was asked by a shr script
Apr 29 20:01:50 <nschle85> Artox: Git needs at least your nick name and email address, this information is published if you commit some code changes
Apr 29 20:02:44 <nschle85> Artox: I you only compile your shr image its not necessary to give git this kind of information
Apr 29 20:02:48 <Artox> well
Apr 29 20:02:52 <Artox> it wanted them stil
Apr 29 20:02:58 <Artox> for fetching a patch
Apr 29 20:03:56 <nschle85> Artox: hmmm, i gave git this information but my build server uses wget
Apr 29 20:04:21 <nschle85> Artox: wget -nv http://patches.openembedded.org/patch/935/mbox/ -O pw-am-935.patch
Apr 29 20:04:39 <nschle85> Artox: cd shr-unstable/openembedded
Apr 29 20:04:57 <nschle85> Artox: patch -p1 < ../../pw-am-935.patch
Apr 29 20:05:21 <bobo> hi there. I have a palm pre there. make flash-kernel failed, and after I accidently tried to boot shr, the palm screen tells me: http://palm.com/ROM . is there a way to restore bootr and just boot the old webos system? or is this old system gone?
Apr 29 20:05:26 <Artox> git am -s pw-am-$patchnumber.patch
Apr 29 20:05:32 <Artox> thats what the script does after wget
Apr 29 20:05:40 <GNUtoo> mickey|bbl, ping
Apr 29 20:05:44 <Artox> and git wont do this unless it has a name and email
Apr 29 20:06:24 <GNUtoo> mickey|bbl, the code you wrote doens't get the dbus signals
Apr 29 20:06:32 <GNUtoo> for cmt_speech
Apr 29 20:06:34 <GNUtoo> the vala source
Apr 29 20:06:48 <GNUtoo> it should print that:
Apr 29 20:06:51 <GNUtoo>     debug( @"onCallStatusSignal $id w/ status $status" );
Apr 29 20:06:54 <Artox> last issue: |  * opkg_install_cmd: Cannot install package locale-base-en-us.
Apr 29 20:06:56 <GNUtoo> it doesn't
Apr 29 20:06:58 <Artox> that error is clear
Apr 29 20:07:13 <nschle85> Artox: you cat use the script, described in wiki, so you must configure git, or you use the sequence i told you
Apr 29 20:07:39 <Artox> I wonder where its hardcoded that shr-image explicitly wnats locale-base-en-us
Apr 29 20:07:44 <Artox> cause I disabled it in my local.conf
Apr 29 20:07:49 <Artox> I only have de locale
Apr 29 20:10:20 <bobo> how long does it usually take to "make flash-kernel" for the SHR on a palm pre?
Apr 29 20:10:47 <Slyon> bobo, shouldn't take to long
Apr 29 20:13:19 <Slyon> bobo,  you can try to start the palm in recovery mode (USB) and novaterm into bootie
Apr 29 20:14:02 <Slyon> bobo, then type the following commands:
Apr 29 20:14:07 <Slyon> setenv bootfile uImage-2.6.24-palm-joplin-3430
Apr 29 20:14:17 <Slyon> setenv bootargs "root=/dev/mmcblk0p2 rootwait ro init=/sbin/init.backup.bootr"
Apr 29 20:14:29 <Slyon> fsboot
Apr 29 20:14:43 <Slyon> this should boot your palm into webos (if bootr is already installed)
Apr 29 20:14:45 <mickey|bbl> GNUtoo: n900 callhandler does not send any signals yet. use the manual mode, e.g. cmtspeechtest 5
Apr 29 20:14:49 <mickey|bbl> and then setup a call
Apr 29 20:14:52 <mickey|bbl> or vice versa
Apr 29 20:15:35 <GNUtoo> ok
Apr 29 20:15:35 <bobo> Slyon, cool. ok, I will try after "make restore-media" is done. thanks in advance
Apr 29 20:16:27 <GNUtoo> mickey|bbl, where is 5 defined?
Apr 29 20:16:29 <GNUtoo> what is 5?
Apr 29 20:16:59 <mickey|bbl> cmtspeechtest <amount of seconds before synthesizing the 'call in progress' signal>
Apr 29 20:17:12 <GNUtoo> cmtspeech is your vala cmtspeech?
Apr 29 20:17:57 <GNUtoo> ah right
Apr 29 20:18:00 <GNUtoo> I found it
Apr 29 20:19:10 <GNUtoo> when the signals will be there.... it will also benefit dcordes 
Apr 29 20:19:33 <GNUtoo> he needs an activator for audio calls too
Apr 29 20:19:42 <mickey|bbl> it's only the nokia n900 call handler
Apr 29 20:19:47 <mickey|bbl> which does lack signals
Apr 29 20:19:53 <mickey|bbl> all others have them of course
Apr 29 20:19:57 <GNUtoo> ah ok
Apr 29 20:19:59 <GNUtoo> right
Apr 29 20:20:01 <GNUtoo> I understand
Apr 29 20:20:08 <GNUtoo> basiaclly it'
Apr 29 20:20:11 <bobo> Slyon, looks good :)
Apr 29 20:20:13 <GNUtoo> s the unimplemented stuff
Apr 29 20:20:18 <mickey|bbl> yes
Apr 29 20:20:22 <GNUtoo> that makes it behave like that
Apr 29 20:20:22 <GNUtoo> ok
Apr 29 20:21:15 <GNUtoo> I think I should go take the laptop to try that delay
Apr 29 20:21:22 <Slyon> bobo, nice. but this didn't change anything yet
Apr 29 20:21:22 <GNUtoo> because I've only 1 sim 
Apr 29 20:21:35 <bobo> Slyon, if I make saveenv ?
Apr 29 20:22:12 <Slyon> bobo, i don't know. but i guess you should fix the error on the filesystem. i guess you link to the kernel is wrong
Apr 29 20:22:22 <Slyon> or your link to the init file
Apr 29 20:23:07 <bobo> the link to the webos kernel?
Apr 29 20:23:40 <Slyon> bobo, well. bootr changes the link to the apropriate kernel at each boot
Apr 29 20:24:28 <Slyon> so in the default case uImage should point to uImage-2.6.24-palm-joplin-3430 and init should point to init.bootr
Apr 29 20:29:01 <GNUtoo|laptop> ahhh newer cmtspeechtest is there
Apr 29 20:29:40 <bobo> hm. I would like to re-run the Makefile to overwrite the wrong settings with correct ones. I'm not sure which directives I should "make". "recover" looks good. "install-bootr" did not solve the ROM error.
Apr 29 20:31:17 <Slyon> bobo, hmm... i don't know if this works... as the Makefile isn't very error proof yet
Apr 29 20:31:46 <Slyon> i can help you to restore it manually if you want
Apr 29 20:32:02 <Slyon> as a first step, you should paste a ls -la of /boot and /boot/sbin
Apr 29 20:33:24 <DocScrutinizer> nschle85: I answered your question before
Apr 29 20:33:31 <bobo> Slyon, I would like to!
Apr 29 20:33:46 <DocScrutinizer> [2011-04-29 21:24:49] <DocScrutinizer> works like a charm with windows, on linux results are ambivalent   [2011-04-29 21:25:09] <DocScrutinizer> brute force always works though
Apr 29 20:35:22 <GNUtoo|laptop> mickey|bbl, ok the loopback works
Apr 29 20:35:23 <Slyon> bobo, ok so please paste those two things, so we can see which link is wrong. and then fix it. shouldn't be a big thing :)
Apr 29 20:37:07 <DocScrutinizer> GNUtoo|laptop: \o/
Apr 29 20:37:26 <GNUtoo|laptop> I did nothing
Apr 29 20:37:30 <GNUtoo|laptop> only make a call
Apr 29 20:37:47 <GNUtoo|laptop> and launch mickeyl's program
Apr 29 20:37:58 <GNUtoo|laptop> so now I have to extract the "data" out of it
Apr 29 20:38:01 <GNUtoo|laptop> and feed it to alsa
Apr 29 20:38:26 <DocScrutinizer> first dump it to a file, way simpler
Apr 29 20:38:59 <DocScrutinizer> 2nd feed that file back to cmt on next call - assuming the format is identical
Apr 29 20:39:31 <DocScrutinizer> you even may keep that option for a nice "record call" feature ;-D
Apr 29 20:39:34 <bobo> Slyon, http://pastebin.com/YTRNsaMM
Apr 29 20:39:48 <GNUtoo|laptop> yes of course
Apr 29 20:39:57 <GNUtoo|laptop> for getting to a file
Apr 29 20:40:08 <GNUtoo|laptop> *dumping
Apr 29 20:40:15 <DocScrutinizer> got it
Apr 29 20:41:20 <Slyon> bobo, ok here is your error: uImage links to uImage-palmpre.bin but uImage-palmpre.bin isn't there at all
Apr 29 20:41:26 <GNUtoo|laptop> I'll first work on earing calls
Apr 29 20:41:28 <Slyon> to fix it do the following:
Apr 29 20:41:31 <GNUtoo|laptop> since it costs nothing
Apr 29 20:41:47 <GNUtoo|laptop> I just have to ear the 4242 provider number
Apr 29 20:41:47 <DocScrutinizer> dman, missed the fireworks
Apr 29 20:41:55 <Slyon> on the device: mount -o remount,rw /boot
Apr 29 20:42:35 <DocScrutinizer> sure
Apr 29 20:43:07 <DocScrutinizer> so start off with trying to get sound to earpiece with aplay -D whatever
Apr 29 20:43:26 <Slyon> on your computer: wget http://build.shr-project.org/shr-unstable/images/palmpre/uImage-palmpre.bin
Apr 29 20:43:35 <DocScrutinizer> then just pipe the cmt audio to same alsa device
Apr 29 20:43:45 <DocScrutinizer> using same statefile
Apr 29 20:43:46 <Slyon> on your computer: novacom put file://boo/uImage-palmpre.bin < uImage-palmpre.bin
Apr 29 20:44:14 <Slyon> bobo, after this it should work again
Apr 29 20:45:29 <Slyon> bobo, there is a typo in on of the commands. it has to be: file://boot/uImage-palmpre.bin
Apr 29 20:46:49 <bobo> Slyon, yes, noticed. the shr kernel is now on the target and linked.
Apr 29 20:47:20 <Slyon> bobo, nice. now if you reboot. bootr should welcome you 
Apr 29 20:49:32 <bobo> Slyon, yes, the bootr is there. webos starts fine
Apr 29 20:50:55 <bobo> Slyon, shr does not boot.
Apr 29 20:51:29 <Slyon> bobo, is there some error?
Apr 29 20:51:58 <bobo> wait, now there's shr's boot screen
Apr 29 20:53:52 <bobo> wow, there is SHR running. Now I try to find out how to configure. 
Apr 29 20:54:34 <Slyon> bobo, ok. you should report this bug in the makefile as described in the SHR wiki
Apr 29 21:13:57 <GNUtoo|laptop> mickey|bbl, how can I write to a file in vala? is the default java-like api ok?
Apr 29 21:14:07 <GNUtoo|laptop> with the streaminput, streamoutput etc...
Apr 29 21:14:12 <GNUtoo|laptop> I don't find a posix like api
Apr 29 21:21:12 <GNUtoo|laptop> I'll ask in #vala
Apr 29 21:21:55 <antrik> Artox: git records the committer when applying a patch, that's why it needs your name and email. of course this is rather irrelevant if you don't intend to do anything further with this commit...
Apr 29 21:23:13 <GNUtoo|laptop> ah the vapi is in gio not glib-2.0 or posix
Apr 29 21:23:19 <GNUtoo|laptop> I'll look
Apr 29 21:23:32 <GNUtoo|laptop> (thanks to #vala)
Apr 29 21:31:09 <mrmoku> GNUtoo|laptop: there is a posix vapi with file handling too
Apr 29 21:31:17 <GNUtoo|laptop> ok
Apr 29 21:31:37 <mrmoku> the easiest way is Framework.FileHandling
Apr 29 21:31:43 <GNUtoo|laptop> ah ok
Apr 29 21:31:47 <mrmoku> though I don't know if you have that available in the cmtspeach test
Apr 29 21:31:48 <GNUtoo|laptop> I must write a buffer tough
Apr 29 21:31:54 <GNUtoo|laptop> like the posix write in C
Apr 29 21:32:03 <mrmoku> ok
Apr 29 21:32:08 <mrmoku> then not FileHandling
Apr 29 21:32:50 <GNUtoo|laptop> is there valadoc of Framework.FileHandling ?
Apr 29 21:33:09 <mrmoku> don't think so
Apr 29 21:33:16 <GNUtoo|laptop> ok
Apr 29 21:34:15 <mrmoku> http://valadoc.org/references/posix/0.11.5/Posix.html
Apr 29 21:34:49 <mrmoku> http://valadoc.org/references/posix/0.11.5/Posix.write.html
Apr 29 21:35:28 <GNUtoo|laptop> ahhh posix.write and not posix.file.write
Apr 29 22:11:31 <GNUtoo|laptop> aplay -r 8000 voice_data.raw  -f S16_LE seem to work
Apr 29 22:11:42 <GNUtoo|laptop> I've recorded the 4242....lol
Apr 29 22:11:54 <mrmoku> wow :)
Apr 29 22:12:57 <GNUtoo|laptop> http://pastie.org/private/dbugvicytmocjexgnehw
Apr 29 22:13:03 <GNUtoo|laptop> diff from the cmt program
Apr 29 22:14:55 <mrmoku> GNUtoo|laptop: not a big diff :)
Apr 29 22:16:01 <GNUtoo|laptop> indeed it was easy
Apr 29 22:16:13 <GNUtoo|laptop> now we must think of a strategy to integrate it
Apr 29 22:16:17 <GNUtoo|laptop> make an alsa player?
Apr 29 22:16:23 <GNUtoo|laptop> use fso for doing that?
Apr 29 22:16:29 <GNUtoo|laptop> (like for ringtones)
Apr 29 22:17:40 <CIA-10> freesmartphone.org: 03mok 07libgisi * r520b443cdff1 10/ (data/libgisi.vapi gisicomm/gisicomm.vala): 
Apr 29 22:17:40 <CIA-10> freesmartphone.org: add a stub for the GPDS client
Apr 29 22:17:40 <CIA-10> freesmartphone.org: Signed-off-by: Klaus Kurzmann <mok@fluxnetz.de>
Apr 29 22:18:08 <mrmoku> GNUtoo|laptop: IMO this would be a candidate for a fsoaudiod plugin
Apr 29 22:18:15 <GNUtoo|laptop> ok
Apr 29 22:18:56 <GNUtoo|laptop> we should also look on how to integrate call recordings
Apr 29 22:19:01 <GNUtoo|laptop> with om-gta02 too
Apr 29 22:20:03 <mrmoku> ok
Apr 29 22:21:04 <mrmoku> now time to sleep though
Apr 29 22:21:21 <GNUtoo|laptop> ok
Apr 29 22:24:15 <mrmoku> gnight
Apr 29 22:25:29 * GNUtoo|laptop looks forward to use the nokia900 as a phone
**** ENDING LOGGING AT Sat Apr 30 02:59:58 2011