**** BEGIN LOGGING AT Mon Jan 05 02:59:58 2009 Jan 05 05:10:30 mickey_away: fso-testing feeds and images are up to date for gta01, gta02 and a780 Jan 05 05:54:22 I see bug, I don't know what it's caused by, yay! Jan 05 06:32:03 anybody here know how to talk to frameworkd from something other than python? Like perl or c? I want an app to be able to grab the current cell id Jan 05 06:32:22 is mdbus the best way? Jan 05 06:40:14 budfive, why would you want to write an app in perl? Jan 05 06:40:22 budfive, have you heard of this thing called dbus? Jan 05 06:53:41 clearly not. Jan 05 06:53:52 i thought dbus was a communication framework Jan 05 06:54:07 are there bindings to other languages? or is it python only? Jan 05 06:55:12 aaah. i just found what i need, i think Jan 05 06:56:53 slightly unrelated question. If I call org.freesmartphone.GSM.Network.GetStatus() , I do not get the cell tower id, even though this information exists and appears in the FSO log. anybody know if FSO can give me this info? Jan 05 06:57:16 Right now I'll write something grepping the log, but this obviously isn't the right way to do it Jan 05 07:15:02 budfive, update your fso Jan 05 07:15:14 budfive, there are bindings in other languages Jan 05 07:20:19 ok. thanks Jan 05 08:19:59 Hey Jan 05 08:20:19 hey raster Jan 05 08:20:52 * raster hides Jan 05 08:25:09 * raster pokes his head out to see if the coast is clear Jan 05 08:30:26 raster, I'd like to have a min height of a scrollbar. Is elm_scroller_content_min_limit useful? Jan 05 08:30:58 hmmm Jan 05 08:31:03 no Jan 05 08:31:19 that limits the h or v axis to the min size of the contents of the scroller Jan 05 08:31:21 eg Jan 05 08:31:27 if i have a list of labels Jan 05 08:31:44 i can make the scroller have its min size horizontally limited to the min size of the contents Jan 05 08:31:47 but not vertically Jan 05 08:32:02 there isnt anything to say "be just bing enouhg to show a scrollbar" Jan 05 08:32:22 i do need to work a bit on all the controls you might want to use for a scroller Jan 05 08:33:40 I just want to make a workaround for this probhlem: When I'm placing an elm_entry into a scroller, clicking on the entry does not work immediately Jan 05 08:35:05 hmmm Jan 05 08:35:10 um Jan 05 08:35:14 in what way? Jan 05 08:36:31 the scroller is very small in height...10-15 pixel.. clicking in that area does not work. But if I press the mouse..hold it..and move it slightly..the bars are displayed and the entry has focus Jan 05 08:37:02 hmmm Jan 05 08:37:04 it should work Jan 05 08:37:49 works in elm_test Jan 05 08:40:35 raster, sc.content_min_limit(0, 1); is doing it :) Jan 05 08:41:02 well its forcing the scroller to tbe the height of the text Jan 05 08:41:03 :) Jan 05 08:41:45 (i need to make a scroller style that doesnt actually allow u to drag and scroll or show scrollbars for single line entries) Jan 05 08:42:37 while i'm at it... i need to say Jan 05 08:42:40 neon is nice Jan 05 08:43:01 what's neon? Jan 05 08:43:02 Neon is an inert gas, which is commonly used for electronic signs Jan 05 08:43:31 it's also armv7's new simd stuff Jan 05 08:43:57 unfortunately - useless for anyone with a freerunner Jan 05 08:44:38 raster, do you own a nokia n8xx? Jan 05 08:44:49 n800- yes Jan 05 08:45:28 e works on it? Jan 05 08:45:29 When you pass charges through it (and something else), it converts the energy into light energy, or photons, which can vary depending on the colour you can see. Neon signs are seen in various places, including crappy diners, "Open" signs, and nasty, old "Disco" clubs. Jan 05 08:46:20 hey AndreasD Jan 05 08:46:20 quickdev: on my n800 - yes Jan 05 08:46:27 in runs on every arm device i have Jan 05 08:46:35 hey quickdev Jan 05 08:46:43 how are things? Jan 05 08:46:55 http://www.rasterman.com/files/the-gang-03.jpg Jan 05 08:47:04 i have a new member of the gang Jan 05 08:47:09 but he only comes as a board Jan 05 08:47:16 AndreasD, things are fine - had 2 weeks holidays. Wednesday school is beginning again :) and for you? Jan 05 08:47:17 no screen on him - screen is attached to dvi Jan 05 08:47:55 beagle board? ;) Jan 05 08:48:13 yup Jan 05 08:48:19 thus the neon Jan 05 08:48:20 :) Jan 05 08:48:27 man its fast Jan 05 08:48:39 clock for clock it leaves the 2442 in the dust Jan 05 08:49:01 and thats without even using neon Jan 05 08:49:03 or the dsp Jan 05 08:49:16 http://www.rasterman.com/ Jan 05 08:49:22 i did a performance comparison Jan 05 08:49:40 Anyone know of a good guide on 3GPP/SS7 standards which is compact, and makes sense? Jan 05 08:49:51 it averages 4x the speed of the gta02 Jan 05 08:49:57 and only @ 500mhz Jan 05 08:49:58 quickdev: been sick since 21. dec but are starting to feel fine again Jan 05 08:49:59 already read about it ;) Jan 05 08:50:05 (as opposed to 400 for gta02) Jan 05 08:50:13 AndreasD, good to hear.. Jan 05 08:52:44 quickdev: btw. for the paste days I get http://pastebin.ca/1299990 Jan 05 08:52:54 when trying to make a shr image Jan 05 08:53:19 AndreasD, then you didn't patch the openembedded tree Jan 05 08:53:58 make update-openembedded Jan 05 08:54:00 make updated-shr Jan 05 08:54:06 make patch-openembedded Jan 05 08:54:08 should do it Jan 05 08:54:27 I've already done that, however I've updated the makefile Jan 05 08:54:54 yes..you should use the newest makefile :) Jan 05 08:55:10 when I run: make patch-openembedded I get cat: .patched: No such file or directory Jan 05 08:55:52 is that normal or could that be the problem Jan 05 08:55:57 ? Jan 05 08:56:06 raster, different microarch IIRC. Jan 05 08:56:13 so no comparison Jan 05 08:56:13 i know Jan 05 08:56:18 with clock speed. Jan 05 08:56:21 well there is a comparison Jan 05 08:56:27 i just compared Jan 05 08:56:28 :) Jan 05 08:56:36 it is ineed much more modern Jan 05 08:56:57 and is brimming with features and silicon i have yet to even begin to do work for me Jan 05 08:58:08 raster, I don't blame OM for not getting a better processor Jan 05 08:58:31 for gta02 - they could have given the release date Jan 05 08:58:41 but release date was delayed in the order of 6-9 months or more Jan 05 08:58:52 if it was not delayed - it'd have made sense Jan 05 08:59:00 but hey - water under the bridge Jan 05 09:07:23 quickdev: just tried updating and patching again but did not resolve the issue Jan 05 09:07:39 AndreasD, does it patch sucessfully? Jan 05 09:07:45 yes Jan 05 09:10:18 quickdev any other idea? Jan 05 09:11:25 AndreasD, yes..probably your directory structure is changed - could you give me an ls -alF? ;) Jan 05 09:11:52 quickdev: in the root? Jan 05 09:12:00 yes Jan 05 09:12:48 quickdev: http://pastebin.ca/1299995 Jan 05 09:15:34 AndreasD, did you want to compile shr-unstable? Is in shr-unstable a shr symlink pointing to ../shr? Does the conf/local.conf contain BBFILES += "${TOPDIR}/shr/openembedded/packages/*/*.bb" ? What does conf/topdir.conf contain? Jan 05 09:18:17 quickdev: yes, yes, yes. It contains http://pastebin.ca/1299998 Jan 05 09:20:01 AndreasD, in shr/, is there a openembedded/packages/images/shr-lite-image.bb file? Jan 05 09:20:38 quickdev: yes Jan 05 09:21:05 quickdev: could it be the bitbake cache that should be cleaned? Jan 05 09:21:31 quickdev: and do you have any idea on how to do that? Jan 05 09:21:46 AndreasD, mv tmp tmp_renamed Jan 05 09:22:26 if you want to clean the cache....try to build everything from scratch...following wiki:"Building SHR" Jan 05 09:22:27 quickdev: nope did not help :-( Jan 05 09:23:08 quickdev: okay, I'll probably end up doing that, anyway thanks for your help Jan 05 09:23:15 no prob.. Jan 05 09:23:34 have to go, see you around Jan 05 09:31:14 raster, have you already implemented the window in a window? Jan 05 09:32:10 yup Jan 05 09:32:14 inwin in e_win Jan 05 09:32:26 its just a feature of win Jan 05 09:35:14 i do get to things Jan 05 09:35:17 and list is next Jan 05 09:35:26 though i'm tossing up what precisely to do with list Jan 05 09:35:57 though for now i will implement the list thats not intended for LOOOONG lists of items Jan 05 09:36:07 just short lists Jan 05 09:38:18 raster, is 250 entries short? Jan 05 09:38:25 hmm Jan 05 09:38:32 (contact list) Jan 05 09:38:34 i'd consider it a boundary Jan 05 09:38:45 borderline Jan 05 09:38:57 it'll work Jan 05 09:39:08 just be slower that an optimised list Jan 05 09:39:08 ok :) Jan 05 09:39:20 i dont set hard limits on stuff :) Jan 05 09:39:32 you just pay a price for abuse with speed/memory :) Jan 05 09:39:47 the advantage of the "slow list" Jan 05 09:39:55 is that u'll be able to put in all sorts of widgets into the list Jan 05 09:40:05 like icons or entries or anything else u like Jan 05 09:40:12 it will be able to pack lots of children Jan 05 09:40:17 plus have a label of its own Jan 05 09:40:45 an optimised list will be more restricted Jan 05 09:40:55 eg only suport label + icon Jan 05 09:41:01 yeah Jan 05 09:41:03 and icon is declared by a filename Jan 05 09:41:20 either that or i will need a "listitemcreate" callback Jan 05 09:41:22 and a destroy Jan 05 09:41:33 that will create/destroy specific item entries Jan 05 09:41:42 and this may be called any timew during the life of the list Jan 05 09:41:52 but to avoid this complexity of create/destroy on the fly Jan 05 09:42:07 i'm going to implement the "u create it all at the start and then just deal with it" list Jan 05 09:42:18 thus the inten of it being for smaller number of items Jan 05 09:42:32 as the icons need to all be created at the start Jan 05 09:42:38 and then packed and calculated Jan 05 09:42:47 and will stay around always Jan 05 10:01:25 <\marco> hi to all Jan 05 10:02:35 <\marco> is there any elementary tutorial? I can't find anything :/ Jan 05 10:04:02 download the sources Jan 05 10:04:13 i need to write one Jan 05 10:04:21 its on my todo fror the immediate future Jan 05 10:04:26 but i'll be a tut on using it in c Jan 05 10:04:34 until then there'es test.c in elementayr's src Jan 05 10:04:36 its commented Jan 05 10:04:46 <\marco> raster: thanks a lot :) Jan 05 10:04:46 also elementary-alarm Jan 05 10:04:50 elementary-sms too Jan 05 10:04:52 they are simple Jan 05 10:04:54 well alarm is Jan 05 10:04:58 sms is more complex by now Jan 05 10:05:10 dont be frightened by test.c Jan 05 10:05:14 <\marco> eheh ok Jan 05 10:05:17 its just one long file that opens up a lot of windows Jan 05 10:05:23 each window is just a function in there Jan 05 10:05:30 tghats just whay its big Jan 05 10:05:32 err Jan 05 10:05:34 thats just why its big Jan 05 10:05:42 typo-heaven Jan 05 10:05:59 <\marco> yep, I've seen it Jan 05 10:06:16 <\marco> *saw Jan 05 10:06:41 <\marco> doh Jan 05 10:06:43 <\marco> *seen Jan 05 10:06:44 <\marco> XD Jan 05 10:07:19 http://www.rasterman.com/files/the-gang-03.jpg <- what? No 1973? :) Jan 05 10:08:27 polz: i lost mine somewhere Jan 05 10:08:31 *shrug* Jan 05 10:08:35 who knows where it went to Jan 05 10:08:57 my ipaq3660 has also gone walkabout Jan 05 10:09:06 as has my zaurus sl5000 Jan 05 10:09:14 and my clie nx80v Jan 05 10:09:31 i had 2 zaurus a320's too at one point Jan 05 10:09:52 raster: ask the underpants gnomes, maybe they've changed their business model and might know where all your toys are Jan 05 10:09:54 a neuros OSD Jan 05 10:09:56 as well Jan 05 10:10:54 i've messed with a lot of embedded systems over the years Jan 05 10:11:05 i just seem to have a few casualties every time i move Jan 05 10:11:28 i suspect they are happily contributing to landfill somewhere Jan 05 10:57:54 raster, I've rows in a scroller. A row consists of label + button. When the text of the label is too big, it makes the scroller scroll horizontally. Is there a way to forbid that? I'd like the text of the label to be shrinked in that case Jan 05 11:01:17 hey ho Jan 05 11:40:59 morning Jan 05 11:50:35 quickdev: as in label... Jan 05 11:50:35 ? Jan 05 11:55:23 raster, let me ask slightly different. Is it possible to disable horizontal scrolling? Jan 05 11:55:52 can the parent widget limit the children's widget sizes? Jan 05 11:56:30 no Jan 05 11:56:37 child widget min size is the controller Jan 05 11:56:41 there is no way to go in reverse Jan 05 11:56:45 it can be added Jan 05 11:56:51 but you cant right now Jan 05 12:07:16 raster, if I want a multiline label, I have to use an elm_entry, right? Jan 05 12:08:10 hmm no Jan 05 12:08:16 actually labels are textblocks Jan 05 12:08:22 tho in butotn widgets ther are just text Jan 05 12:08:24 so 1 line Jan 05 12:08:29 but standar dlabel is multi-line Jan 05 12:08:36 see elementary_test L( Jan 05 12:08:37 :) Jan 05 12:08:45 the paragraph at the top is a label in a frame Jan 05 12:11:52 raster, there it is wrapped manually through "
". Is it possible to wrap it automatically? :) Jan 05 12:12:10 I've got some sentences which should be wrapped according to screen width Jan 05 12:12:17 hmm no Jan 05 12:12:20 label doesnt auto-wrap Jan 05 12:12:30 otherwise its kind of hard to calc a min WxH size Jan 05 12:12:39 as its not fixed Jan 05 12:12:39 entry is auto-wrap? Jan 05 12:12:44 as W gets less H gets more Jan 05 12:12:46 yes Jan 05 12:12:51 entry can be non-editable Jan 05 12:13:03 the idea is so u can display text blocks that u can select from and copy from Jan 05 12:13:05 but not edit Jan 05 12:25:07 raster, entry works fine, but I've got a problem with an entry in a box, which is in a scroller: http://rafb.net/p/qF4tPG64.html Jan 05 12:25:16 the text is not displayed in that case Jan 05 12:25:56 if I add a button instead of the entry to the box...which is the child of the scrolller...the button is displayed Jan 05 12:26:46 raster: do you plan to implement a dropdown list (like the keyboardlayout/dictionary list) a filechooser widget in elementary soon Jan 05 12:27:01 s/a /and a / Jan 05 12:27:02 bumbl meant: raster: do you plan to implement and a dropdown list (like the keyboardlayout/dictionary list) a filechooser widget in elementary soon Jan 05 12:27:27 hmm not the right s/// Jan 05 12:31:48 raster, it seems the problem is unrelated to the scroller - just to an entry in a box..hm Jan 05 12:34:36 raster, bumbl, have I forgotten something? http://rafb.net/p/Tkk8Qb79.html Jan 05 12:40:02 however..I found a different solution Jan 05 13:15:58 quickdev: Trying to follow http://shr.bearstech.com/trac/wiki/Building%20SHR i get http://pastebin.com/m47cc6a17 Jan 05 13:17:39 AndreasD, never had that. Are the permissions of openembedded/git/* files ok? Jan 05 13:19:05 quickdev: locally? I don't have openembedded folder Jan 05 13:20:24 Initialized empty Git repository in /media/sda6/freerunner/shr-makefile/openembedded/.git/ Jan 05 13:20:24 fatal: read error (Connection reset by peer) Jan 05 13:20:24 cd: 4: can't cd to openembedded Jan 05 13:20:24 error: could not lock config file .git/config Jan 05 13:20:32 AndreasD, have you tried it again? Jan 05 13:20:46 AndreasD: seems like you are not able to clone from git.openembedded.net Jan 05 13:20:54 yes.. Jan 05 13:21:21 quickdev: yes tried twice Jan 05 13:22:17 AndreasD: host seems up but I am unable to clone either. Jan 05 13:22:34 AndreasD: you should probably ask in #oe if something is wrong with that host. Jan 05 13:22:58 Kensan: thanks Jan 05 13:22:58 AndreasD: yep, me too can't clone now Jan 05 13:29:16 hi Jan 05 13:30:02 dcordes: hi there Jan 05 13:30:32 is there a bugtracker for fso related issues? Jan 05 13:31:14 mickey_away: qualcomm_msm seems to work well on the new generation htcs Jan 05 13:31:21 dcordes: http://trac.freesmartphone.org Jan 05 13:31:54 PaulFertser_ thanks. I prefer it over ml Jan 05 13:34:20 hey PaulFertser_ Jan 05 13:34:28 quickdev: Hi quickdev :) Jan 05 13:35:33 dcordes: any new htc? Jan 05 13:35:53 why there isn't a fre-scaling driver on the neo freerunner? oO Jan 05 13:36:31 sylar91: what's a free-scaling driver? Jan 05 13:36:55 Has anyone ever used BlueLee's keyboards? Jan 05 13:37:28 PaulFertser_: a driver by which you can change dinamically the frequency of a cpu? Jan 05 13:38:25 sylar91: because nobody has implemented that yet Jan 05 13:38:26 sylar91: what about http://wiki.openmoko.org/wiki/User:CesarB/cpufreq ? Jan 05 13:38:35 PaulFertser_: that's not finished Jan 05 13:38:37 unfortunately Jan 05 13:39:05 ah, I can't see anybody in ML talking about that so I questioned if there is any progress.... Jan 05 13:39:06 Kensan: but it's a good hint to why there's no finished driver yet, and whom to ask about it. ;) Jan 05 13:42:05 sylar91: you might want to talk to Jonas Bonn see http://lists.openmoko.org/pipermail/openmoko-kernel/2008-October/006075.html Jan 05 13:42:08 mrmoku nope Jan 05 13:42:20 mrmoku only kaiser and blueangel Jan 05 13:42:26 sylar91: he picked up some work of cesarb and has something (tm) working ;) Jan 05 13:42:36 Kensan: ah ok Jan 05 13:42:37 thanks Jan 05 13:42:38 dcordes: and how well do they work? Jan 05 13:43:39 sylar91: I don't know if Jonas is still working on it since he never posted his patchset to the list... Jan 05 13:43:53 blueangel has good kernel support but no frameworkd Jan 05 13:44:12 kaiser both so-so Jan 05 13:44:43 we just tested frameworkd on the htcraphael (100) and it seems to work well Jan 05 13:45:00 dcordes: thanks... my collegue is thinking about a new phone... will tell him to wait ;) Jan 05 13:46:50 ps: if anywan whats to try ecalc just opkg install http://sylar-stuff.googlecode.com/files/ecalc_0.1-r0_armv4t.ipk (I posted this yesterday but anybody answered...) Jan 05 13:46:59 s/anywan/anyone/ Jan 05 13:47:00 sylar91 meant: ps: if anyone whats to try ecalc just opkg install http://sylar-stuff.googlecode.com/files/ecalc_0.1-r0_armv4t.ipk (I posted this yesterday but anybody answered...) Jan 05 13:49:36 mrmoku he can get a raphael Jan 05 13:49:57 dcordes: looking on htc.com right now... which one is it? Jan 05 13:50:25 Touch Pro Jan 05 13:50:53 mrmoku: heh, fine that one he would have selected me thinks :-) Jan 05 13:51:05 it's available with cdma (msm7501a) and gsm (msm7201a) Jan 05 13:51:27 anyone knows how to use qwo on shr?= Jan 05 13:51:53 mrmoku the keyboard and displa resolution are rich Jan 05 13:51:56 just install it and select the qwo keyboard from illume-setting Jan 05 13:53:27 ah.. thanks will try Jan 05 13:53:36 tifereth what's qwo keyboard? Jan 05 13:54:02 http://wiki.openmoko.org/wiki/Qwo Jan 05 13:58:55 sounds interesting Jan 05 13:59:56 what's the shr status? when is there a release? Jan 05 14:02:06 dcordes: good question... next question ;) Jan 05 14:02:25 anybody heard something from Ainulindale? Jan 05 14:02:30 dcordes, with some experience typing with qwo really makes fun Jan 05 14:05:07 how can I disable illume keyboard dictionary? Jan 05 14:06:36 just select another layout Jan 05 14:07:09 tifereth: I create my own layout but dictionary in turning on by it self Jan 05 14:08:02 dcordes: how does installing an image onto the raphael work? complicated? Jan 05 14:12:43 yoyo_: hm.. not sure, perhaps you should look at other customised layouts Jan 05 14:17:47 heyho Jan 05 14:26:42 hey morphis Jan 05 14:26:45 some minutes for me? Jan 05 14:27:00 one minute Jan 05 14:27:08 morphis, found two bugs in the bindings Jan 05 14:27:13 must first log off and log in :) Jan 05 14:30:32 quickdev: so be back Jan 05 14:31:08 morphis_, I found the cause of the segfault and the cause of the black backgrounds: http://rafb.net/p/nvIUy911.html Jan 05 14:32:50 quickdev: oh great Jan 05 14:32:56 I will apply and checkin Jan 05 14:33:20 yeah, fine :) Jan 05 14:33:42 I am currently implementing the toolbar widget Jan 05 14:33:50 the basics works Jan 05 14:36:43 quickdev: the segfault was the processing of the command line arguments? Jan 05 14:37:58 morphis, no. the segfault was, because the argc/argv was free'd ;) Jan 05 14:38:11 ah ok Jan 05 14:38:12 I removed the argv_copy, because I didn't see any purpose Jan 05 14:38:30 ok, I just copied the code from the other e-python bindings Jan 05 14:38:37 maybe the segfault is there too Jan 05 14:38:59 maybe etk is copying argc/argv and thus that problem does not exist Jan 05 14:44:03 maybe Jan 05 14:44:05 I will check Jan 05 14:49:56 quickdev: why you removed the line with PyMem_Free(argv)? Jan 05 14:50:12 morphis_, because that free'd argv, which is used by elementary Jan 05 14:50:13 argv is still used and if we removes it we will have a mem leak Jan 05 14:50:21 ah ok Jan 05 14:51:09 but elementary will release the memory when elm_shutdown is called? Jan 05 14:52:04 morphis_, aren't that the argv parameters that should be there for the whole app execution? Have a look at elementary's source :) Jan 05 14:52:50 in a normal c program that are the vars from the main function Jan 05 14:53:25 yeah..do you free them in a normal c program? Jan 05 14:53:34 but in our case we allocate extra memory for that (the args are already in sys.argv) Jan 05 14:53:35 no Jan 05 15:02:55 ainu: when can we expect your fancy news post? Jan 05 15:20:25 bumbl: when I won't have so much work Jan 05 15:32:21 Hi. Jan 05 15:39:44 Ainulindale: hey, how are you? very busy, I see Jan 05 15:40:41 <\marco> there's a way to know which revision(A5,A6 etc..) is my FR? Jan 05 15:42:00 mrmoku: yeah :-) Jan 05 15:42:28 \marco: look in the battery compartment if it's not running, or cat /proc/cpuinfo if it is. Jan 05 15:42:45 <\marco> mwester: thanks a lot! :) Jan 05 15:44:53 <\marco> mwester: mhm.. "Revision : 0360" is equals to...A5? Jan 05 15:50:23 \marco: A6 iirc Jan 05 15:50:37 <\marco> sicu: thanks :) Jan 05 15:51:09 \marco: http://wiki.openmoko.org/wiki/Finding_hardware_revision Jan 05 15:51:13 =] Jan 05 15:52:19 <\marco> sicu: doh.. :D sorry.. I'm blind Jan 05 15:54:47 mwester: g'day sir Jan 05 15:57:17 raster, still here? Jan 05 15:59:33 G'day Bruce, say g'day to Bruce, Bruce Jan 05 16:01:50 mwester: When would you have time to discuss the userspace-reset idea? I assume you are busy now *heh* Jan 05 16:07:19 Kensan: can you recommend a working gta02 shr build? Jan 05 16:13:49 dcordes_: sorry, I haven't been using SHR all that much lately so I wouldn't know... Jan 05 16:13:58 dcordes_: maybe quickdev knows a stable version. Jan 05 16:14:06 quickdev: ping Jan 05 16:14:09 ehm Jan 05 16:14:19 dcordes_: happy 09 to you btw ;) Jan 05 16:14:42 dcordes_, if you want some features..take unstable....testing is sonly slightly more stable - if at all Jan 05 16:15:23 Kensan: thanks. also wish you a happy new year Jan 05 16:15:34 * mrmoku would strongly recommend the unstable xmas edition :-) Jan 05 16:15:37 quickdev: are all the images in testing and stable tested? Jan 05 16:15:48 mrmoku: is it tested? got a link? Jan 05 16:16:08 dcordes_, we've got only testing and unstable - they're partly tested ;) Jan 05 16:16:17 http://shr.bearstech.com/shr-unstable/images/om-gta02/openmoko-shr-lite-image-glibc-ipk--20081224-om-gta02.rootfs.jffs2 Jan 05 16:16:33 at the moment I'm trying to compile the most recent E Jan 05 16:28:34 is the lack of ts calibration specific to the shr images or is it normal in all builds for the om-gta02 machine? Jan 05 16:29:00 dcordes_, I never needed that Jan 05 16:29:16 because you use the gta02. we try the xmas image on htc universal Jan 05 16:31:05 dcordes_, yes, I use gta02 Jan 05 16:31:07 the openmoko-dialer2 works ootb with htc universal Jan 05 16:31:15 so I hope frameworkd will as well Jan 05 16:31:17 ootb? Jan 05 16:31:29 openmoko-dialer2 is from the 2007.2 days? Jan 05 16:31:49 yes oldskool Jan 05 16:32:22 how can one rotate the screen orientation by 90° / 270° ? Jan 05 16:32:24 in e Jan 05 16:39:31 take some lsd Jan 05 16:41:42 dcordes_: xrandr --rotate Jan 05 16:42:00 no fancy buttons? Jan 05 16:42:15 dcordes_: hm, don't know. Jan 05 16:42:39 dcordes_: for openmoko we have rotate daemons Jan 05 16:42:53 which rotate according to the data of the accels Jan 05 16:43:34 think out of the one-device-box Jan 05 16:43:51 also for the tscal Jan 05 16:45:38 dcordes_: there must be some simple app which rotates the screen... Jan 05 16:47:44 wasn't there one in the old images? Jan 05 16:48:28 http://wiki.openmoko.org/wiki/Rotate Jan 05 16:48:36 opkg install http://projects.openmoko.org/frs/download.php/455/accel-rotate_0.41_armv4t.ipk Jan 05 16:48:38 works great for me Jan 05 16:51:57 quickdev: we have no accerlometer in htcuniversal Jan 05 16:52:14 dcordes_, try e settings panel? Jan 05 16:52:17 *tried Jan 05 16:52:50 that made e segfault Jan 05 16:53:21 ping raster ;) Jan 05 16:53:45 well i would say Jan 05 16:53:58 raster isn't available at the moment Jan 05 16:54:44 xrandr -o right works Jan 05 16:55:02 you guys could include a script that executes that on the shr desktop Jan 05 16:55:21 for those who don't like the accel-rotate or have no accerlometer Jan 05 16:58:19 dcordes_, could be built into the settings app Jan 05 16:59:17 actually it should work with e - so the best solution would be to solve the segfault Jan 05 16:59:25 true Jan 05 17:03:04 dcordes_: http://projects.openmoko.org/plugins/scmsvn/viewcvs.php/tapUntap/?root=fdom Jan 05 17:03:19 should solve your problem Jan 05 17:03:48 quickdev: who is responsible for libframeworkd-glib? Jan 05 17:04:49 bumbl: thanks Jan 05 17:05:10 mrmoku, mainly Ainulindale - but I made some additions / modifications, too Jan 05 17:05:19 mrmoku, what do you want to know? :) Jan 05 17:05:32 quickdev: there was some api breakage for sms sending... remember? Jan 05 17:05:37 I'm there for questions if necessary Jan 05 17:05:48 mrmoku, no I don't Jan 05 17:06:07 sending a sms gives back the timestamp when the message center received the sms Jan 05 17:06:09 ainu: enough work for today? Jan 05 17:06:14 No, still at work Jan 05 17:06:31 that is in frameworkd-devel, so if I would like to try that sms sending would not work I think Jan 05 17:06:59 Well we should branch lfg Jan 05 17:07:43 http://git.freesmartphone.org/?p=framework.git;a=commit;h=898dff053f32a0a78a93b17b79514693e904391f Jan 05 17:07:54 yep, branching a devel branch for lfg would be good I think :-) Jan 05 17:08:50 I'd love to use frameworkd-devel... if just for having a working clock ;) Jan 05 17:32:28 where in the SHR rootfs is defined which channel is used for the AT gta02 channel in frameworkd? Jan 05 17:33:37 dcordes, which modem channel? I think gsm0710muxd is doing that? Jan 05 17:34:20 usr/lib/python2.5/site-packages/framework/subsystems/ogsmd/modems/singleline/modem.py: return "/dev/ttySAC0" Jan 05 17:34:28 ah ok Jan 05 17:34:31 ah..which device ;) Jan 05 17:34:42 well we are fuxing with the uni Jan 05 17:35:02 as I mentioned earlier, in the old images with openmoko-dialer2 phone works ootb Jan 05 17:35:24 gta02 used no muxing at that time? Jan 05 17:35:52 htcuniversal Jan 05 17:38:42 dcordes: no back then there was no muxer afaik Jan 05 17:39:06 or at least gsmd was used for that which was/is rather unstable Jan 05 17:39:14 ok I guess we can assume htcuniversal modem has no muxing support Jan 05 17:40:06 we'll try qualcomm_msm Jan 05 17:40:12 or singleline modem Jan 05 17:52:41 hi Jan 05 17:54:33 hi Jan 05 18:01:57 is there any precompiled mailclient for shr available? i found nothing Jan 05 18:03:46 hi all: about parsing gps data Jan 05 18:04:17 the UBX protocol, do this kind of message works? $PUBX,40,GLL,1,0,0,0*5D Jan 05 18:05:35 The problem is I want to test UBX protocol, ubx NEMA or ubx binary Jan 05 18:29:48 framework return me 103% capacity battery :) Jan 05 18:29:50 hahah Jan 05 18:35:10 yoyo_: enjoy the extra 3% - the freerunner will drain the battery quite fast :p Jan 05 18:35:24 :/ Jan 05 18:39:32 hyyy state charge / discharge is constans :P charge all the time :) Jan 05 18:39:36 cool Jan 05 18:45:52 you have a super charger Jan 05 18:46:03 the frameworkd makes the magic Jan 05 18:46:36 yyy in frame work is logick error Jan 05 18:46:59 during talk "call" frame work send ower dbus signal idle, idle_dim Jan 05 18:47:31 and if you have set suspend it will put phone to suspend during call Jan 05 18:47:34 :/ beeed Jan 05 18:48:04 you are trying to say Jan 05 18:48:55 frameworkd doens't love when you awake him when is suspended? Jan 05 18:48:55 frame work is able to pus phone to sleep during talk Jan 05 18:50:04 there's a rule in that requests the CPU resource during a call Jan 05 18:51:09 I think frame work don't shoud send suspend state on IdleNotifier.State durink call Jan 05 18:51:34 that's a layering violation Jan 05 18:51:38 we have fixed it with a rule Jan 05 18:51:50 check the latest rules file Jan 05 18:52:13 :/ Jan 05 18:52:14 ok Jan 05 18:52:41 we have the CPU resource since some months Jan 05 18:52:51 once you request the CPU resource, the suspend state will never be hit Jan 05 18:53:11 the new rule now requests the CPU resource whenever you're on a call Jan 05 18:53:14 that is the most flexible way Jan 05 18:53:17 application listen idlenotification :/ Jan 05 18:54:29 taraaa :) ophonekitd turns of after geting suspend idlenotifier.state Jan 05 18:54:34 it is wrong Jan 05 18:56:43 mickey|ICE870, Are you recoding the entire FSO in Vala? Jan 05 18:58:02 one more time ophonekitd recive id5 of idle notifier (suspend) bach dbus error Jan 05 18:58:24 didn't make up my mind on that. we're heading for some of the subsystems where we expect a substantial performance gain Jan 05 19:04:40 if i were to priorize I'd say Jan 05 19:04:40 odeviced Jan 05 19:04:45 oventsd Jan 05 19:04:47 ogpsd Jan 05 19:04:51 resource handling Jan 05 19:04:53 ogsmd Jan 05 19:09:34 but i won't encourage anyone to work on the vala reimp until we're confident our APIs are ok Jan 05 19:09:47 hum is it possible to change the keybindings in a config-file? Jan 05 19:09:51 http://shr.bearstech.com/git/?p=shr-overlay.git;a=blob;f=openembedded/packages/e-wm-config-illume-shr/e-wm-config-illume-shr/e.src;h=cbe328f644fc16c45aaec6ea00c6eb34d66154f5;hb=HEAD Jan 05 19:09:58 wer can i find this config? Jan 05 19:33:22 does one of you get a Timing Advance different than "0" as answer from "AT%EM=2,1"? Jan 05 19:34:12 hey folks, i'm getting strange errors while using the latest fso-dbus-api-reference-implementation (what a buzzword :)) : http://pastebin.ca/1300302 Jan 05 19:34:42 which part of the result is the timing advance? Jan 05 19:34:59 >>> gsmdebug.DebugCommand("AT%EM=2,1\r\n") Jan 05 19:35:00 11th value Jan 05 19:35:02 ['%EM: 819,13,17,16,31,36335,23,1,0,0,0,0,0,0,0,155,0,0,2,255', 'OK'] Jan 05 19:36:54 and another 0 Jan 05 19:48:54 bad marshal data, strange Jan 05 19:48:54 find . -name "*.pyo"|xargs rm -f Jan 05 19:48:54 then relaunch Jan 05 20:00:15 x Jan 05 20:01:01 y Jan 05 20:01:29 sry :) Jan 05 20:01:38 mickey_away: will try it - thanks Jan 05 20:09:24 mickey_away: still the same when deleting all .pyo-files. But when I delete all the bytecodefiles (*.pyc) I get another error once, further runs return the "bad marshal data" error : http://pastebin.ca/1300342 Jan 05 20:24:18 ah it seems TA is only != 0, if data is exchanged with the BTS (for example during a call) Jan 05 21:13:20 is there an AT commando to force the calypso connecting to a given cell id? Jan 05 21:18:59 where can I see the frameworkd version used in my image? Jan 05 21:21:53 shr uses an obsolete version? Jan 05 21:24:34 can't tell you how it works with opkg, but in debian you would do it like this: apt-cache show fso-frameworkd Jan 05 21:27:31 dcordes, opkg list_installed | grep frameworkd Jan 05 21:28:24 dcordes: shr uses the stabilization branch of frameworkd Jan 05 21:28:34 you would need the master Jan 05 21:29:02 I already installed master. was just wondering if I can easily determine the version/branch from outside of the rootfs Jan 05 21:30:32 mickey|bs: still the same when deleting all .pyo-files. But when I delete all the bytecodefiles (*.pyc) I get another error once, further runs return the "bad marshal data" error : http://pastebin.ca/1300342 Jan 05 21:31:27 dcordes: you could look into /var/lib/opkg/armv4t Jan 05 21:32:32 mrmoku: I think only replacing the frameworkd fodler horribly broke it Jan 05 21:33:46 dcordes: under /usr/lib/python/site-packages? Jan 05 21:34:13 yea Jan 05 21:34:28 I need the config stuff for the devel too Jan 05 21:34:40 mirko: that looks pretty strange. could be a miscompilation. did it just stop working or wasn't working from the beginning? Jan 05 21:34:43 yep, and you have to delete the .pyo files Jan 05 21:35:57 dcordes: and sms sending won't work I think Jan 05 21:37:40 dcordes: if you want I can try to build a frameworkd-devel package for shr unstable... Jan 05 21:38:30 mrmoku: that would be very helpful. or a complete image if you have everything packaged readily Jan 05 21:40:03 mrmoku: why dod you think sms won't work Jan 05 21:40:33 dcordes: because there was an API change to return the timestamp when the sms was sent Jan 05 21:40:41 and libframeworkd-glib was not yet adjusted Jan 05 21:41:00 mickey|bs: hi. I still wasn't able to debug pppd. after running syslogd -C logread -f also gives the error Jan 05 21:42:12 hm that's the reason why I criticized the 'extra layer of abstraction' Jan 05 21:43:07 libframeworkd-glib . of course I think it also eases up some things but it's gonna be a pita keeping up with framework Jan 05 21:43:31 I don't think the situation would be different without it... Then the shr apps would need adjusting Jan 05 21:44:07 mickey|bs: it's my first try with fso-stuff Jan 05 21:44:28 I was talking with Ainulindale and quickdev about it... the solution will probably a devel branch for lfg Jan 05 21:44:30 mrmoku: hm true Jan 05 21:44:32 +b Jan 05 21:46:47 dcordes: tried other ways to read the syslog? Jan 05 21:46:55 mickey|bs: just checked out the latest git Jan 05 21:46:58 dcordes: would ed797b0a4f3af5018a17eebd56c0cdf7db9257eb be current enough (frameworkd)? Jan 05 21:47:01 I think syslog is not working at all Jan 05 21:47:26 hmm in that case it's tough to debug Jan 05 21:47:32 try to get syslog running first Jan 05 21:48:20 anyways, i don't think gprs is important Jan 05 21:48:21 I use the fso-console-image Jan 05 21:48:37 if i were you i'd rather start implementing the htc special commands Jan 05 21:48:42 and come back later to gprs Jan 05 21:48:47 I don't know how to do it Jan 05 21:51:17 start researching syslog options Jan 05 21:51:23 check why it doesn't work with circular buffer Jan 05 21:51:27 debug Jan 05 21:51:36 i can't give you a cooking recipe for debugging Jan 05 21:51:41 you need to construct these skills Jan 05 21:51:44 sorry Jan 05 21:52:12 I hope we get better hardware support soon Jan 05 21:52:35 things like debugging gprs via gprs suck Jan 05 21:53:38 ya, usbnet will come as a relief Jan 05 21:54:07 wifi will be easier Jan 05 21:54:40 the newer models have usbnet Jan 05 21:55:21 all of them. because they have the hsusb as opposed to the kaiser's full speed usb. google kernel has hsusb support only Jan 05 22:01:39 dcordes: I'm trying to build a shr image with MACHINE = htcraphael Jan 05 22:02:37 mrmoku: thanks. raphael won't work. we were trying this on the universal which is pxa. raphael builds armv6-novfp Jan 05 22:02:54 ahh, ok. will stop then :-) Jan 05 22:03:14 you can use htcunivesral := I think the packages will be enough Jan 05 22:03:34 so we can just stick with the om-gta02 image and replace frameworkd with framworkd-devel Jan 05 22:04:49 Error, Your PACKAGE_ARCHS field contains duplicates :( Jan 05 22:05:22 seems like some arch related things are hardcoded in shr Jan 05 22:08:22 dcordes: http://zamora.homelinux.org/frameworkd-devel_0.8.4.9+gitred797b0a4f3af5018a17eebd56c0cdf7db9257eb-r0_armv4t.ipk Jan 05 22:08:34 and frameworkd-devel-config_0.8.4.9+gitred797b0a4f3af5018a17eebd56c0cdf7db9257eb-r0_om-gta02.ipk Jan 05 22:10:49 you have to deinstall frameworkd and frameworkd-config-shr with -force-depends first Jan 05 22:26:54 mrmoku: ok thanks. I hope we can test tomorrow Jan 05 22:28:50 dcordes: ok, I installed them on my FR and at first glance it works nicely.. so good luck :-) Jan 05 22:30:59 TAsn: hi, how is you gtk theming going on? Jan 05 22:31:48 +r Jan 05 22:45:50 Ainulindale: I have 3 new commits for libframeworkd-phonegui-efl waiting. Can I commit already? Jan 05 22:49:03 Anyone else having problems with 'git pull' from git://git.openembedded.net/openembedded ? Explanation/solution ? Jan 05 22:50:04 ingi: yep, oe seems dead... also web is dead Jan 05 22:51:08 It's been like that for 2-3 days..., I've been waiting for resolution on the openmoko-newsgroups but I guess I should subscribe to 'openembedded'-groups instead... Jan 05 22:52:11 mrmoku: what the internet is dead? https://i.broke.the.internet.and.all.i.got.was.this.t-shirt.phreedom.org/ Jan 05 22:53:26 hehe... Jan 05 22:54:02 mickey|bs: any idea? Jan 05 22:54:22 maybe ask in #oe? Jan 05 22:54:39 Kensan: me? Jan 05 22:55:03 mirko: whoever wants to find out why it doesn't work. Jan 05 23:00:30 oe is down, check the topic of the #oe channel Jan 05 23:00:54 there's only emergency developer access right now Jan 05 23:01:58 plus a read-only mirror for the general public Jan 05 23:02:11 that one may be some days old though Jan 05 23:02:19 my question was regarding the fso / "bad marshal data"-error :) Jan 05 23:02:20 These things happen... :( Jan 05 23:03:10 mirko: sorry, no real idea about that one. seems your python is miscompiled or something Jan 05 23:04:03 I was just asking in #oe and they confirm what mickey just said... :-) Jan 05 23:04:16 hmm... Jan 05 23:04:23 how to start debugging? :/ Jan 05 23:05:24 might install some -dbg packages and check the control flow Jan 05 23:05:41 also try to find out what actually is raising this exception Jan 05 23:12:19 need to sleep now, tomorrow starting early *sigh* Jan 05 23:12:20 n8 Jan 05 23:38:02 mwester: hey **** ENDING LOGGING AT Tue Jan 06 02:59:56 2009