**** BEGIN LOGGING AT Fri Mar 26 02:59:57 2010 Mar 26 03:26:45 no Mar 26 03:26:51 You just buffer the new command Mar 26 03:27:13 and if too many commands come in you have to throttle Mar 26 03:27:24 Or close the socket Mar 26 03:28:01 right, i need just buffer the new command Mar 26 03:28:20 without letting server_feed to fetch it, right? Mar 26 03:28:47 i just wonder if it could happen. i need test Mar 26 03:29:25 Do note that it isnt really valid V250 protocol Mar 26 03:29:39 Anything that comes over the socket when a command is pending is a cancel request Mar 26 03:30:11 Well behaved clients should not do it Mar 26 03:30:39 okay, so how to cancel it? Mar 26 03:31:55 do we need to handle it right now? Mar 26 03:31:56 cancelation is tricky Mar 26 03:32:14 i want to put it as low priority Mar 26 03:32:46 thats fine Mar 26 03:33:01 just buffer it up for now **** BEGIN LOGGING AT Fri Mar 26 13:06:14 2010 **** ENDING LOGGING AT Fri Mar 26 13:09:37 2010 **** BEGIN LOGGING AT Fri Mar 26 13:12:15 2010 Mar 26 14:37:21 Hi, i try to run oFone examples on scratchbox but there are some exception Mar 26 14:37:24 [sbox-FREMANTLE_X86: ~/ofono-0.15/test] > run-standalone.sh ./test-phonebook Mar 26 14:37:24 Traceback (most recent call last): Mar 26 14:37:24 File "./test-phonebook", line 3, in ? Mar 26 14:37:24 import dbus Mar 26 14:37:24 ImportError: No module named dbus Mar 26 14:37:39 Anybody help... Mar 26 14:38:20 install the python dbus module Mar 26 14:38:55 PtDBus package ? Mar 26 14:39:20 no idea, depends on your distro I guess Mar 26 14:40:19 thanks, i'll try Mar 26 14:43:20 It should be PyDbus or python-dbus Mar 26 14:43:27 my sistem is ubuntu 8.04 , i am developing in scratchbox, i have following packages to install Mar 26 14:43:28 python2.5-dbus - Python wrappers around D-Bus. Mar 26 14:43:28 python2.5-qt4-dbus - Python bindings for Qt dbus mainloop. Mar 26 14:43:28 python-dbus - simple interprocess messaging system (Python interface) Mar 26 14:43:28 python2.5-edbus - DBus python integration for ecore. Mar 26 14:43:28 python-dbus-dbg - Debug build of the D-Bus Python interface Mar 26 14:43:52 python-dbus :) Mar 26 14:44:17 thank you :) Mar 26 14:44:25 you're welcome :) Mar 26 14:45:06 Btw, I tried Neophysis and it worked well Mar 26 14:45:32 I assume you guys solved the issues you were having? Mar 26 14:45:42 Djdas, python-dbus is already the newest version. Mar 26 14:45:42 0 upgraded, 0 newly installed, 0 to remove and 552 not upgraded. Mar 26 14:47:33 denkenz, not yet, the oFono version inside the 20100306 image is still 0.16 Mar 26 14:47:49 we are working on sms now so didn't try Mar 26 14:47:55 your suggestions Mar 26 14:48:07 I plan to try this weekend Mar 26 14:48:23 Strange, the image I tried had 0.17 Mar 26 14:49:05 with the kernel provided in /boot? :O Mar 26 14:49:33 bbg, this is stange, maybe python2.5-dbus? Mar 26 14:49:52 I believe so, but I flashed directly, didn't go through SD Card Mar 26 14:49:52 bbg, which version of Python are you using Mar 26 14:50:40 denkenz, this is interesting as I test it on SD only (in NADN I have my daily distro) Mar 26 14:51:08 denkenz, can you try a more recent version? 0.18 or 0.19 Mar 26 14:51:23 I need a toolchain, the OpenMoko one doesn't have signalfd Mar 26 14:51:27 if you're not able to cross compile I can send you the packages to install Mar 26 14:51:52 ok, may I send you the ipk by email? Mar 26 14:52:15 hmm, ipk means gcc on Freerunner? Mar 26 14:52:18 Djdas, 2.5.2-3maemo2 Mar 26 14:52:25 You guys don't cross compile? :) Mar 26 14:52:58 eheh denkenz, we have a buildhost with OE toolchain and BitBake ;) Mar 26 14:53:32 the host liks to our Git Server, downloads the sources and automatically builds the packages ;) Mar 26 14:53:37 If you have a 64-bit toolchain give it to me Mar 26 14:53:54 Otherwise a vmware image or virtual box image Mar 26 14:54:09 I'll ask to the sysadmin I don't know the platform, will let you know Mar 26 14:54:49 well I have a toolchain I use on different PCs but I didn't try to compile newest oFono versions Mar 26 14:55:34 I can cross-compile in an eeepc with ArchLinux, a desktop with Slackware13 (32bit) and a laptop (64bit) with Ubuntu9.10 Mar 26 14:55:56 using the same toolchain, I can provide you a link to test it Mar 26 14:56:06 Put it up somewhere so I can give it a whirl Mar 26 14:56:10 I'm too lazy to build my own Mar 26 14:56:24 bbg, so maybe you need python2.5-dbus Mar 26 14:56:33 ok, please wait Mar 26 14:57:57 no hurry, sending me a link is fine too Mar 26 14:58:58 Djdas, certainly, i tried but ... Mar 26 14:59:02 The following packages have unmet dependencies: Mar 26 14:59:02 python2.5-dbus: Depends: python-dbus (= 0.83.0-1maemo1) but it is not going to be installed Mar 26 14:59:32 in maemo repository only python-dbus (= 0.83.0-1maemo3) version Mar 26 15:00:11 bbg, sorry but I'm not very in confidence with ubuntu Mar 26 15:00:30 denkenz, you can try this toolchain: http://sourceforge.net/projects/neophysis/files/developer-tools/djdas-toolchain-20091029.zip/download Mar 26 15:01:37 denkenz, all the packages built by our server are in a opkg-style repository at http://build.neophysis.bialive.it/neophysis-testing/ipk/ Mar 26 15:03:27 you can find oFono inside armv4t, but you'll find only 0.16 (our currently officially supported) in the repo, but 0.17 and pre-0.16 are available inside armv4t/morgue folder Mar 26 15:04:17 to install packages in Neophysis you can use the ipkmgr.sh script with the following syntax: ipkmgr.sh -x pkgname.ipk Mar 26 15:05:22 if you have filesystem space problems you can modify the script (it's not yet a full package manager it just extracts files in / dir) to save temporary files somewhere else Mar 26 15:05:35 the script is located in /opt/bin dir Mar 26 15:10:32 So what do I do with this vdi file? Mar 26 15:11:18 it's a VBox Disk image Mar 26 15:11:42 Doesn't help me since I'm running Gentoo in vmware :) Mar 26 15:11:56 ops :D Mar 26 15:11:57 sorry Mar 26 15:12:32 uhm, I should have a simple tarred toolchain, let me search :) Mar 26 15:12:44 Yah, that'll be better ;) Mar 26 17:49:57 denkenz, I found the file and uploading but my bandwidth is very poor so please be patient Mar 26 17:50:41 it's a tgz file you can extract in /usr/local (you should have /usr/local/neophysis) Mar 26 17:51:00 inside there is a toolchain and a source dir where you can find sample sources dir for ofono and phonesim Mar 26 17:52:04 good, just send me email once you have it ready and I'll try it out Mar 26 17:52:17 No worries if it is next week ;) Mar 26 17:52:29 ok, and I'll send you infos for building programs Mar 26 17:52:50 sounds good Mar 26 17:52:56 do you prefer a private email or in list? Mar 26 17:53:15 up to you, if the toolchain is good it could be useful / interesting to others Mar 26 17:53:52 well it can build even qt4 programs so I think it could be useful ;) Mar 26 18:13:42 denkenz, I have one question about network registration properties Mar 26 18:14:33 I noticed using GetProperties method that unless properties don't change, they don't appear in list Mar 26 18:15:19 so for example my signal strenght widget doesn't show the Strenght property until it changes Mar 26 18:16:09 but if this doesn't occur it seems I don't have any signal while operator name is valid.... Mar 26 18:16:22 what should be the correct GUI behavior? Mar 26 18:17:36 send me the at trace with the behavior Mar 26 18:18:43 well I simply use a GetProperties at application startup and I get only operator name and registration status Mar 26 18:19:01 the widget recevies the PropertyChanged signal too Mar 26 18:20:03 so at its startup the operator name is correctly filled but I get the "no signal" icon because I don't hav any Strength property in my list Mar 26 18:20:03 and this until the network strenght changes Mar 26 18:20:03 you can test it with phonesim too Mar 26 18:20:54 until you change the network strength slider you won't get any Strength property in your NetworkManager.GetProperties list Mar 26 18:22:43 the same occurs with CellId and LocationAreaCode (IIRC the name), which at the moment I don't care but just to let you know Mar 26 18:34:24 hmmm Mar 26 18:34:43 So in general oFono should take care of the details, if it doesn't then its either a bug in oFono or the driver Mar 26 18:35:45 I don't know what GSM standard says but shoudn't those properties be available at registration time? Mar 26 18:36:08 like operator name and registration status Mar 26 18:36:19 This is why I want the AT trace Mar 26 18:36:30 In general yes Mar 26 18:37:12 However, oFono does pull the Signal Strength at startup Mar 26 18:37:42 ok, then I'll send you the trace to let you investigate, maybe could be my operator mistake Mar 26 18:38:00 Sequence is this: Mar 26 18:38:15 Startup -> Query current registration status -> if registered/roaming query signal strength Mar 26 18:38:22 After that we rely on the driver to tell us Mar 26 18:38:56 uhm... I have one doubt Mar 26 18:39:08 I never had to mind issuing network registration as ofono does registering upon modem power up Mar 26 18:39:24 maybe the sequence you wrote is not sent in that case? Mar 26 18:40:07 We always perform those steps Mar 26 18:40:21 However, if we end up issuing the register ourselves, then the driver is responsible for reporting signal strength Mar 26 18:40:33 It may be that the Modem is stupid and reports strength before CREG Mar 26 18:43:08 Hah, I even have a comment about that in src/network.c:1339 Mar 26 18:45:01 duh! Mar 26 18:45:04 :) Mar 26 18:45:22 so if it's not registered nor roaming it won't get the strength Mar 26 18:45:50 exactly, so if calypso sends SQ right before +CREG then we won't get updated Mar 26 18:45:59 It then needs to be properly quirked in the driver Mar 26 18:46:29 and considering calypso "fantasies" I can bet on it :D Mar 26 18:49:31 ok upload finished you can get the file here: https://sourceforge.net/projects/neophysis/files/toolchains/neophysis_toolchain_20100122.tar.bz2/download Mar 26 18:49:59 I'll send an email to the list when I get home :) Mar 26 18:50:10 thank you very much, bye **** ENDING LOGGING AT Sat Mar 27 02:59:56 2010