**** BEGIN LOGGING AT Thu Mar 05 02:59:56 2009 Mar 05 03:39:40 raster. Mar 05 03:50:21 paroli :p Mar 05 03:50:54 mwester! Mar 05 03:51:19 Dave: boo Mar 05 03:51:35 boo cellphones Mar 05 03:56:59 mobile phones suck Mar 05 03:57:07 lets go back to rotary phones Mar 05 03:57:12 with really long cords back home Mar 05 03:57:14 haha oh god no Mar 05 03:57:23 those black curly ones Mar 05 03:57:25 ack Mar 05 03:57:30 always getting tangled and shit Mar 05 03:57:38 and your pets chewing through them Mar 05 03:59:03 Dave: I heard that ;) Mar 05 03:59:20 raster: is there a way to switch off the scrollbars in elm ? Mar 05 04:00:03 mirko-paroli: theme. Mar 05 04:00:16 scrollbars only appear if needed Mar 05 04:00:19 ie when scrolling Mar 05 04:00:24 or if contents change size/pos Mar 05 04:00:39 of course entries in a scroller are subject to some nastiness here Mar 05 04:00:45 and there needs to be refinement Mar 05 04:00:57 but for now it is functional enough until i get back to it Mar 05 04:01:06 there is a todo item for it Mar 05 04:02:11 ok so in theory I might be able to, if I change the theme a bit Mar 05 04:07:49 yes Mar 05 04:08:06 tho how a user will knwo where in a long list of things they are when scrolling.... beats me Mar 05 04:10:47 MRMOKU :o Mar 05 04:39:03 Hello Sargun. Mar 05 04:40:01 Hello Dave. Mar 05 04:47:08 raster: that is a good question, maybe I can simply change it a bit Mar 05 04:47:25 my concern was mainly the horizontal scrollbar Mar 05 04:48:03 thats just me not having differentiated h and v scrolling Mar 05 04:48:16 ah ok Mar 05 04:48:22 and display of the h/v bars Mar 05 04:48:29 i should - just havent done it Mar 05 04:57:24 ah ok Mar 05 05:36:15 lol Mar 05 05:36:18 DIPLSYA=:0 ./paroli-launcher Mar 05 05:36:21 obviously not going to work Mar 05 06:02:05 vininim: got something to tell us? Mar 05 06:03:21 it runs Mar 05 06:03:33 not sure it's properly accessing framework though Mar 05 06:03:58 tried sending a message to myself, didn't work Mar 05 06:04:14 r u running from source directory? Mar 05 06:04:30 yes Mar 05 06:04:37 check the paroli.cfg file Mar 05 06:05:17 if #defaults = GSM:Test, SIM:Test, Audio:Test, SMS:Test Mar 05 06:05:19 exists Mar 05 06:06:04 it has to be commented out Mar 05 06:06:13 otherwise paroli will not connect to the framework Mar 05 06:06:30 isn't # a comment itself? Mar 05 06:06:47 if there is a # in front of that line Mar 05 06:06:48 than yes Mar 05 06:06:53 http://pastebin.com/m579861c7 Mar 05 06:07:10 yapp that should do Mar 05 06:07:24 can you relaunch it and paste the log to a pastebin ? Mar 05 06:08:41 btw you are looking at the wrong file Mar 05 06:08:51 see line 6 Mar 05 06:09:02 in the folder where the executable is Mar 05 06:09:09 there is also a config Mar 05 06:09:13 check that one Mar 05 06:09:25 line 6 should read # This config is used when paroli is run from within the source folder Mar 05 06:09:34 oh Mar 05 06:18:38 k, it works but I sent two messages to myself >_< Mar 05 06:20:08 mmm.. also difficult to know what is sent and received, but it works :) Mar 05 06:21:41 People has the same scrolling problem of the others contacts apps: it's easily to aciddentaly call someone while scrolling Mar 05 06:21:53 if there is a '>' in front of the message than it is sent Mar 05 06:22:03 if there is nothing it was an incoming messages Mar 05 06:22:09 at least that's the way it should be Mar 05 06:22:21 the list dragging thingy I agree that is a problem Mar 05 06:23:04 want the log of a successfull operation? =) Mar 05 06:23:10 sure Mar 05 06:23:35 so it works on shr ? Mar 05 06:24:28 yes, except for dbus Mar 05 06:26:22 http://omploader.org/vMWJ1bw/paroli-withoutdbus.log Mar 05 06:30:10 http://omploader.org/vMWJ1cA/paroli-dbus.log Mar 05 06:40:11 ah Mar 05 06:40:19 vininim: there is a data folder Mar 05 06:40:28 in it there is a dbus folder Mar 05 06:41:31 you need to copy the two files to its destination folders, not sure if they are the same on SHR but I guess so Mar 05 06:41:32 k Mar 05 06:43:18 tichy.conf goe to /etc/dbus-1/system.d/ Mar 05 06:44:31 i think ... Mar 05 06:44:46 but I guess it could lie in SHR's settings somewhere as well Mar 05 06:44:47 not sure Mar 05 06:45:00 ah Mar 05 06:45:01 wait Mar 05 06:46:11 paroli shouldn't try to create a dbus thing Mar 05 06:46:12 odd Mar 05 06:52:11 copying the files work Mar 05 07:17:53 mickeyl: pong Mar 05 07:22:27 heyho Mar 05 07:26:28 ah morphis Mar 05 07:26:31 :) Mar 05 07:26:48 where abouts in GER are you from btw? Mar 05 07:32:39 mirko-paroli: currently I am living in the near of Bremen Mar 05 07:33:06 hmm Mar 05 07:33:11 I am in germany in 1.5 weeks Mar 05 07:33:29 where do you live? Mar 05 07:33:33 Sweden Mar 05 07:33:37 oh Mar 05 07:33:38 but I'll be in Chemnitz for CLT Mar 05 07:33:45 CLT? Mar 05 07:33:51 and I'll be in Berlin for a few days Mar 05 07:33:53 Chemitz Linux Tage Mar 05 07:34:04 ah ok Mar 05 07:34:21 and I'll be back for Re:publica Mar 05 07:34:31 hehe Mar 05 07:34:33 april 1st Mar 05 07:34:42 Bremen is not that far ;) Mar 05 07:34:49 republic is on april 1st? Mar 05 07:34:57 I think so Mar 05 07:35:01 it starts then Mar 05 07:35:01 yeah Mar 05 07:35:06 2 or 3 days I think Mar 05 07:36:48 sadly I have no time this year to visit the republic or the cdl Mar 05 07:43:08 mirko-paroli: but If you will be in the near of bremen, tell me :) Mar 05 07:44:29 morning Mar 05 07:50:19 hehe morphis I will :) Mar 05 07:57:33 btw was the patch not good yesterday? Mar 05 07:58:57 why? Mar 05 07:59:04 is it in the repo? Mar 05 07:59:07 I commited it donight Mar 05 07:59:16 s/donight/tonight Mar 05 07:59:37 so it should be in already Mar 05 07:59:43 or you will commit it? Mar 05 07:59:43 jepp Mar 05 07:59:46 hmm ok Mar 05 07:59:50 no, it is commited Mar 05 07:59:54 will check again Mar 05 08:04:38 rev 39372 Mar 05 08:04:59 No changeset 39372 in the repository Mar 05 08:05:08 am I checking at the wrong place? Mar 05 08:05:16 in e-svn? Mar 05 08:05:16 svn.enlightenment.org ? Mar 05 08:05:28 humpf Mar 05 08:05:44 k I'll try sth else Mar 05 08:05:48 takes a couple of min Mar 05 08:05:52 I commited it yesterday from my git-svn-repo and after that I have done an update on my svn-only repo Mar 05 08:07:56 k locally I get it Mar 05 08:08:09 only the page seems to be not up to date Mar 05 08:08:11 superb Mar 05 08:08:12 thx Mar 05 08:08:16 I will get our rev bumped Mar 05 08:09:23 hm Mar 05 08:09:33 hm ? Mar 05 08:11:09 it didnt know that trac caches the svn status Mar 05 08:15:28 morphis: morning Mar 05 08:15:36 spaetz: morning Mar 05 08:15:50 Do I see it correctly that neither lists, nor Sliders are available in python-elementary yet? Mar 05 08:16:08 no Mar 05 08:16:12 both are available Mar 05 08:16:14 there are some thing referring to them in the c_elementary version but the are not imported into the elementary module Mar 05 08:16:32 from c_elementary import Object, Window, Background, Icon, \ Box, Button, Scroller, Label, Toggle, Frame, Table, \ Clock, Layout, Hover, Entry, AnchorView, AnchorBlock, Bubble, \ Photo, Hoversel, Toolbar, init, shutdown, run, exit Mar 05 08:16:34 but they are not exported Mar 05 08:16:41 jepp Mar 05 08:16:53 because I dont tested them until nwo Mar 05 08:16:58 s/nwo/now Mar 05 08:17:15 ahh, so how would it be possible to use them? I tried a c_elementary.List and it complained that it wants some evas.Canvas Mar 05 08:17:30 and I was handing it an elementary.Window in __init__ Mar 05 08:17:43 can you give me the output? Mar 05 08:17:55 hold on, need to log into my FR Mar 05 08:18:33 k Mar 05 08:21:02 bah, what's going on? Now I only have to do "import elementary, elementary.init() to have python segfaulting" Mar 05 08:21:05 weird Mar 05 08:21:13 rebooting Mar 05 08:24:27 morphis: http://dpaste.com/6834/ Mar 05 08:24:55 simplest test case I could find Mar 05 08:28:17 hm Mar 05 08:28:31 I dont get this error when constructing the list Mar 05 08:28:57 * spaetz checks revision number in unstable Mar 05 08:29:28 python-elementary - 0.1+svnr38899-ml0 Mar 05 08:30:20 ok, I tested in locally on my pc because I dont have my fr available at the moment Mar 05 08:30:21 weird Mar 05 08:30:38 the weird thing is, that I never call __init__ from evas.Object Mar 05 08:30:53 It's reproducable Mar 05 08:31:12 actually the same when I try a slider Mar 05 08:31:28 even slider works for me Mar 05 08:31:36 it current unstable? Mar 05 08:31:39 ahh, no Mar 05 08:31:42 s/it/its Mar 05 08:31:46 AttributeError: 'module' object has no attribute 'Slider' Mar 05 08:31:58 when I do s = e.c_elementary.Slider(w) Mar 05 08:32:51 mmh, so it can't be the latest revision, I gues Mar 05 08:32:52 s Mar 05 08:33:37 no there is no version in svn where Slider an List are exported Mar 05 08:33:43 I will commit this the evening Mar 05 08:34:27 OK, will retry with your changes. But this is extremly strage. I only use the stock packages on shr-unstable and I can reproduce the error every time Mar 05 08:35:04 hm, maybe someone other available who can test this? Mar 05 08:36:00 * spaetz pokes mrmoku :-) Mar 05 08:36:23 can you reproduce some python-elementary weirdness for us? Mar 05 08:36:46 it's as simple as entering "python" and type the 4 lines in http://dpaste.com/6834/ Mar 05 08:37:01 (get out of python with Ctrl-D) :) Mar 05 08:37:14 :) Mar 05 08:38:30 and good morning by the way ;-) Mar 05 09:17:20 dos1 Mar 05 09:17:27 I miss you Mar 05 09:17:28 :) Mar 05 09:39:38 morgen Mar 05 09:40:08 mickeyl: still it's not quite right. Even with prefixes caching i get enourmous delay (like 7 seconds) after picking up and before the other side hears mic signal. Here's the log illustrating that it's not a debugging enabled issue: http://pastebin.com/f1561ddf2 Mar 05 09:40:31 What happens if I try to read from unaligned address (e.g. word from odd address) in userspace on the FR? Do I get a signal? Mar 05 09:43:00 PaulFert`: can you try with another rigntone? Mar 05 09:43:26 the prefix cache got it down to 2 seconds here Mar 05 09:44:31 mickeyl: Sure :) But you see the log, yes? If you think it can be related to the ringtone, i'm promptly changing it back :) Mar 05 09:44:36 dent: depends on /proc/cpu/alignment, here it seems to be fixup+warn Mar 05 09:45:45 i see the log, yes Mar 05 09:46:15 it doesn't look like the new timeout handling is responsible Mar 05 09:46:38 since there's no modem activity between ATA and the set scenario Mar 05 09:46:39 pabs3: fixup+warn means it will just proceed fine from the POV of the app? Mar 05 09:47:21 dent: I'm guessing there would be a performance penalty, but everything would be fine Mar 05 09:48:11 DocScrutinizer2: just did some tests on echo with a fully disassembled FR. Tried to hold speaker with hand with various force and didn't get any considerable echo (calling from landline on FR with %N0001 everything disabled). Also tried to hold PCB to the front case with varying force. Looks like echo is a bit louder. Mar 05 09:48:39 mickeyl: sure, but picking up was slow before as well :) Mar 05 09:51:40 DocScrutinizer2: assembled it again, got a considerable echo. (i don't use the screws for some time, but if i press firmly to hold the case together, echo seem to increase. And yes, looks like %N0187 eliminates it fully. Mar 05 09:53:07 pabs3: thanks Mar 05 09:57:51 mickeyl: hey... did ptitjes get commit access to OE? :-) Mar 05 09:58:06 no Mar 05 09:58:09 did he ask for it? Mar 05 09:58:14 i thought so... Mar 05 09:58:16 no Mar 05 09:58:19 ptitjes: ping Mar 05 09:58:20 quickdev and Ainulindale have Mar 05 09:58:31 since months Mar 05 09:58:33 both are quite inactive this period... Mar 05 10:02:12 ok, new patch is ready for shr-settings Mar 05 10:02:36 we don't import *all* possible plugins modules before drawing the main GUI Mar 05 10:02:42 which takes 2 seconds on cold start Mar 05 10:02:49 (and is ikely to increase over time) Mar 05 10:03:02 but only import those plugins dynamically that we actually need Mar 05 10:03:16 * spaetz invokes the git gods Mar 05 10:04:15 nah Mar 05 10:04:17 we're not inactive Mar 05 10:04:19 just, well Mar 05 10:04:20 not there Mar 05 10:04:27 let me try to find a new girlfriend, guys Mar 05 10:04:32 :-) Mar 05 10:04:32 (almost there) Mar 05 10:05:04 girl friends are bad for OSS productivity :) Mar 05 10:05:04 plus giving ptitjes access to OE would be a bad idea IMHO Mar 05 10:05:12 next thing you know he'll redevelop everything Mar 05 10:05:19 in cobol Mar 05 10:05:21 sorry Ainulindale , couldn't resist. I know you don't find it funny Mar 05 10:05:23 :-) Mar 05 10:05:37 spaetz: well don't worry about that I feel good these days Mar 05 10:05:53 (Mostly because I met a really nice girl) Mar 05 10:06:02 cool, congrats Mar 05 10:06:31 (plus she's studying philosophy for her PhD) Mar 05 10:06:41 ELM_ENGINE=x11-16 looks crappy with current theme, but with some simplifications it would really make a speed difference Mar 05 10:06:49 (so she's very interesting to talk with) Mar 05 10:07:03 wasn't there a shr effort to build a new theme for elementary some time ago Mar 05 10:07:10 yep there was Mar 05 10:07:12 no idea, Mar 05 10:07:37 ok, pushed my shr-settings patch Mar 05 10:07:43 We'll have to talk a bit guys, you'll have to brief me on the latest changes, I'm outdated Mar 05 10:07:52 Plus I'm seeing Angus next week Mar 05 10:08:04 oohhh, interesting Mar 05 10:08:07 (he's staying at my place) Mar 05 10:08:12 ptitjes will probably be there too Mar 05 10:08:57 mrmoku: is it tricky to update shr-settings on the unstable repo? If not, care to trigger one? Mar 05 10:09:06 Ainulindale: no more effort on that so it is dead? Mar 05 10:09:26 shaving 1-2 sec off shr-settings start time (till main GUI shows) is not too bad Mar 05 10:09:31 morphis: more a WIP than dead I think Mar 05 10:09:33 ainulindale is alive Mar 05 10:10:16 spaetz: would it be save for testing also? Mar 05 10:10:24 Hire: yeah it has been known to happen Mar 05 10:10:29 Ainulindale: ok Mar 05 10:10:39 sure :) Mar 05 10:12:25 mrmoku: I'd say yes, but then who knows what unintended side effects there might be. Mar 05 10:12:29 :-) Mar 05 10:12:38 * mrmoku has triggered shr-settings build for unstable and is moving to the homeoffice Mar 05 10:12:43 but it *should* be safe (famous last words) Mar 05 10:12:48 spaetz: at least it will break shr-settings only :-) Mar 05 10:12:54 right :) Mar 05 10:12:54 s/will/would/ Mar 05 10:12:55 mrmoku meant: spaetz: at least it would break shr-settings only :-) Mar 05 10:13:08 * mrmoku does not assume spaetz broke something :P Mar 05 10:13:11 cu Mar 05 10:13:17 cu Mar 05 10:15:21 spaetz, I agree... Mar 05 10:15:37 socat? Mar 05 10:30:44 DocScrutinizer2: OTOH i'm not sure that mic gain with the stock state file (which i tested) is enough to really make use of AGC. Mar 05 10:38:16 shr-unstable would switch to fso/milestone5.5 as base branch? Mar 05 10:38:24 or is that not usable at all Mar 05 10:45:43 and how can I find which pipeline gstreamers would use for a certain file type? Mar 05 10:45:54 gstreamer is a modular mystery to me Mar 05 10:46:20 check gst-inspect Mar 05 10:46:46 andthat's what I did and now it output tons of stuff without possibility to CTRL-C it :-( Mar 05 10:47:00 should have read manuals beforehand Mar 05 10:47:16 but thanks for the hint, will dig deeper Mar 05 10:48:21 issueing gst-instpect made the FR freeze. Not even the screen backlight will come back Mar 05 10:48:29 * spaetz takes out battery :-( Mar 05 10:51:09 hi mrmoku|homeoffice :-). Can you also bitbake package-index, please? Mar 05 10:51:45 will try ;) Mar 05 10:53:21 spaetz: built Mar 05 10:53:25 and mickeyl, I'm all for pushing things upstream. less for us to maintain :-) Mar 05 10:53:26 (for unstable) Mar 05 10:53:29 thanks Mar 05 10:54:21 good Mar 05 10:54:26 just needs someone doing Mar 05 10:54:30 i have been preaching for months now Mar 05 10:54:51 any specific packages you have in mind, or all shr apps in general? Mar 05 10:55:24 for a start, everything original to the project Mar 05 10:55:37 the more controversial things after review Mar 05 10:55:50 so I'd appreciate you opening a branch Mar 05 10:55:54 then we can easily merge back Mar 05 10:56:31 or not -- if nothing is controversial Mar 05 10:57:04 well, I doubt that OE would be interested in frameworkd-shr-settings Mar 05 10:57:18 but for things like ophonekitd, I see why it could go up Mar 05 10:57:20 uhm Mar 05 10:57:36 how is anyone supposed to build your apps then? Mar 05 10:58:20 the whole idea of OE is that anyone can build anything Mar 05 10:58:24 no matter for which machine Mar 05 10:58:26 or distro Mar 05 10:58:31 ok, good to know Mar 05 10:59:28 spaetz: if you confirm me that shr-settings works as expected I will build it for testing too Mar 05 10:59:53 mrmoku: hold on, I just opkg upgraded. trying now Mar 05 11:02:56 works fine for me un shr-unstable Mar 05 11:03:15 just bluez4 doesn't want to upgrade... (file conflicts) Mar 05 11:03:22 nifty new logos. Mar 05 11:03:31 err, icons Mar 05 11:05:56 spaetz: that bluez4 problem is because you have an older bluez installed in unstable Mar 05 11:06:00 and they conflict Mar 05 11:06:08 * spaetz gives green light to move current shr-settings to -testing Mar 05 11:06:16 works like a charm Mar 05 11:06:29 so I remove bluez-utils first? Mar 05 11:06:43 ok, will do (don't really use bluetooth anyway) Mar 05 11:21:41 what the heck is this: EDJE ERROR: file /usr/share/elementary/themes/default.edj, group elm/frame/base/default has a non-fixed part. add fixed: 1 1; ??? Problem part is: elm.swallow.content Will recalc min size not allowing broken parts to affect the result. Mar 05 11:27:58 spaetz: in x.log? Mar 05 11:28:15 * mrmoku has that one too since a loooong time Mar 05 11:28:42 it complains when it literally cannot properly calculate the min size of something due to its geometry setup Mar 05 11:28:43 no, STDERR when calling some python-elementary Mar 05 11:28:50 so it tries a slower uglier algorithm instead Mar 05 11:28:51 and bitches Mar 05 11:29:17 might that be a reason why I can't resize some windows when displaying them via ssh -X on my desktop? Mar 05 11:29:32 then it seems like our theme is broken Mar 05 11:29:36 or something Mar 05 11:29:49 no Mar 05 11:29:54 unlikely Mar 05 11:30:02 more likely just layout Mar 05 11:30:19 mmh, hard to debug Mar 05 11:30:28 from python-elementary Mar 05 11:30:39 * spaetz won't bother raster to ask for help with this :-) Mar 05 11:31:13 i know some theme elements in upsteram elementary are not min-sizecalc clean Mar 05 11:31:15 i fixed it tho Mar 05 11:31:18 already Mar 05 11:31:26 ok Mar 05 11:31:43 I don't know whether we are now using the latest and greates in shr-unstable Mar 05 11:31:57 unlikely Mar 05 11:32:00 u will always be behind Mar 05 11:32:02 spaetz: well... not yet Mar 05 11:32:11 hence the discussion about how to do it on the ML :-) Mar 05 11:32:41 * spaetz shouts branches. Mar 05 11:32:56 raster: I build a local image every one or two days with current e Mar 05 11:32:57 easy said when I am not the one capable of doing the work :) Mar 05 11:33:01 just to see what happens :-) Mar 05 11:33:06 branches wont help u Mar 05 11:33:16 mrmoku: thats good Mar 05 11:33:22 imho generally speaking Mar 05 11:33:29 u will be better off following head Mar 05 11:33:35 just maybe be behind a day or 2 Mar 05 11:33:47 as u will deal with fewer problems overall Mar 05 11:33:52 raster: spaetz meant branches in the shr-overlay me thinks... not e Mar 05 11:33:55 ie problems u have will get fixed Mar 05 11:34:04 or problems u dont even know u have yet - get fixed before u hit them Mar 05 11:34:09 mrmoku: yep. shr-overlay branching for shr-testing Mar 05 11:34:11 but u will get some new ones crop up too Mar 05 11:35:18 raster: as long as the balance goes in the right direction ... ;) Mar 05 11:36:03 chances are the balance is "u'll be better off" Mar 05 11:36:17 of course no one listens to me and screams to stabilise Mar 05 11:36:21 and stick ot a specific release Mar 05 11:36:53 freesmartphone.org: 03mickey 07framework * rd7dca1626ae6 10/framework/subsystems/ogsmd/gsm/parser.py: Mar 05 11:36:53 freesmartphone.org: ogsmd: parser.py: flag +CGMR as solicited message that has a PDU. Closes FSO #369, which was actually a valid Mar 05 11:36:53 freesmartphone.org: bug. Mar 05 11:36:58 but then u just wall yourself off in a box of "sorry - you live with the bugs and features you have. you get nothing new or fixed" Mar 05 11:37:54 raster: the reason we are still sticking to a very old EFL_SRCREV for shr is that we have problems with segfaults Mar 05 11:38:06 (not e segfaults but shr segfaults) Mar 05 11:38:43 anything using entries is segfaulting... Mar 05 11:39:40 well send bt's Mar 05 11:39:43 deal with the problem Mar 05 11:39:47 dont run away from it! Mar 05 11:40:30 raster: we don't run away (and I think I already posted a segfault in here) Mar 05 11:40:51 and if I remember correctly you told me it might have to do with threading... Mar 05 11:41:07 threading coudl cause any number of bugs Mar 05 11:41:53 yep... and I have no experience in debugging that :( Mar 05 11:42:15 unless u really know your code Mar 05 11:42:15 I get backtraces leading somewhere into eina_lists stuff... Mar 05 11:42:18 and i mean all of it Mar 05 11:42:23 dont bother Mar 05 11:42:33 if u have more than 1 thread accessesin any efl calls Mar 05 11:42:36 directly or indirectly Mar 05 11:42:38 u are in trouble Mar 05 11:42:43 just remove the code Mar 05 11:42:48 well its quickdev's code... and he is busy with other stuff... so were doomed Mar 05 11:43:17 thats what I intend to do for unstable... lots needs to get rewritten to use opimd anyway Mar 05 11:43:34 (exception - ecore_pipe calls are safe or more specifically ecore_pipe_write() is intended to be used by threads other than the efl main thread) Mar 05 11:44:45 though Ainulindale says we need the multithreaded nature... Mar 05 11:45:16 * mrmoku never understood why... Mar 05 11:45:33 anyway have to eat now... bbiab Mar 05 11:46:11 what for? Mar 05 11:46:29 Ah, lfp-efl Mar 05 11:46:37 Yes it's necessary as ophonekitd is tied to the glib mainloop Mar 05 11:46:56 but none of the calls to e stuff are made outside of the e specific thread Mar 05 11:46:57 then u need it Mar 05 11:47:11 but do not use efl calls from the glib loop thread Mar 05 11:47:16 yep Mar 05 11:47:17 not even indirectly Mar 05 11:47:20 eg via callbacks? Mar 05 11:47:31 PaulFert`: yup, mic gain seems low. If echo needs fully assembled case, it's a good indication it's probably transmitted "ota" inside case Mar 05 11:47:42 raster: Pipes Mar 05 11:47:46 (quickdev did that) Mar 05 11:47:58 It'll be rewritten to use dbus Mar 05 11:49:01 ok Mar 05 11:49:07 then as long as no one did anything silly Mar 05 11:49:09 it should be fine Mar 05 11:50:25 raster: yep but let us not exclude that possibility right away :-) Mar 05 11:52:21 sure Mar 05 11:52:22 :) Mar 05 11:56:45 what the heck can I use as a "Hover" style? Mar 05 11:56:51 it seems to require a string Mar 05 11:58:31 Elementary.h lists the styles Mar 05 11:59:12 styles can be extended by a theme extension Mar 05 11:59:17 see elm_theme_extension_add() Mar 05 11:59:31 an app can overlay (override) a theme with elm_theme_overlay_add() Mar 05 12:00:36 k Mar 05 12:05:31 interesting "default","popout" and "vertical_hoversel" all look the same :-) Mar 05 12:05:37 Ainulindale: I agree for the windows ophonekitd launches itself... but contacts and messages would not need threads, no? Mar 05 12:06:44 Ainulindale: ahh, no forget it... they do because of the possibility to configure the type of frontend... Mar 05 12:06:49 :-) Mar 05 12:07:20 jOERG_rw: Has anyone tried to attach a large C to to the other side of R4408 (MICBIAS wolfson pin)? If RF comes in the wolfson chip via this line and audio ripple comes back via the same line, why doesn't it help? Mar 05 12:07:50 well... I could bump the EFL_SRCREV to current for unstable... then more sophisticated people could take a look ;) Mar 05 12:08:02 jOERG_rw: the RF will still be there, but ripple will be shorted by the cap, no? Mar 05 12:08:10 if that's jk4401 end, it will block audio as well as ripple Mar 05 12:08:32 jOERG_rw: no, i mean the other side of R4408, attached to the wolfson pin. Mar 05 12:08:35 but that would need fiddling on the buildhost... which I did not want to do... hmm Mar 05 12:09:20 mrmoku: I'm all for breaking unstable and have people fix it the proper way :) Mar 05 12:09:29 wolfson output is low source impedance Mar 05 12:09:36 mrmoku: agreed Mar 05 12:09:52 jOERG_rw: that means _very_ powerful ripple is coming from it? Mar 05 12:10:04 Ainulindale: agreed for not fiddling on the buildhost? Mar 05 12:10:11 yup, to put it like this Mar 05 12:10:12 agreed for bumping EFL_SRCREV Mar 05 12:10:25 and as long as you upgrade it you don't have to fiddle Mar 05 12:10:28 jOERG_rw: Hey there! Mar 05 12:10:30 How's life? Mar 05 12:10:42 jOERG_rw: why not split R4408 in two R in series and attach the cap inbetween? Mar 05 12:11:01 Ainulindale: current efl wont compile with the fso/ms5 branch... would have to fiddle (a little ;) Mar 05 12:11:13 spaetz: they are very different. look at elementary_test Mar 05 12:11:21 hover and hover 2 use efault and popout Mar 05 12:11:26 PaulFert`: that's exactly what I mentioned yesterday. could do that, but no way to do it MP Mar 05 12:11:29 then hoversel uses ... vertical_hoversel Mar 05 12:12:19 (and remember that I fiddled when building the next testing image... which would be the more difficult part) Mar 05 12:12:24 [2009-03-04 20:55:38] but you had to place a R *before* C, for the LPF, and another one *behind* C for getting some signal at all (source impedance) Mar 05 12:13:00 I guess it is cleaner if I teach the makefile to have different OE trees for unstable / testing... Mar 05 12:13:07 fso does it and we should me thinks Mar 05 12:14:14 jOERG_rw: yep. Didn't understand that yesterday. Speaking about MP, in A8 the problems are solved by stopping RF from coming to wolfson, so that's probably a non-issue and those who want to use handset can probably perform manual rework? Mar 05 12:14:33 exactly Mar 05 12:16:17 * jOERG_rw sheers for some coffee Mar 05 12:20:30 Ainulindale: mooo :) Mar 05 12:29:28 raster: I just tried both "default" and "poput" from python and they both look the same (as in no decorations at all). I'll have a look at elementary-test Mar 05 12:30:02 mrmoku: does it makes sense to follow fso/milestone5.5 for unstable or is that not ready yet? Mar 05 12:30:13 * spaetz grabs lunch Mar 05 12:41:03 spaetz: well that is the thing I have to teach the Makefile... follow different OE branches for unstable / testing... Mar 05 12:41:11 how ready it is? don't care for unstable ;) Mar 05 12:47:31 hmm... need help on gta01 kernel... Mar 05 12:50:13 k :-) Mar 05 12:53:14 hmm, some with xchat around? Mar 05 12:53:22 someone Mar 05 12:54:14 how to set up connection port for xchat IRC on N810? Mar 05 12:54:36 (sorry for somewhat OT) Mar 05 12:55:20 raster: ping Mar 05 12:55:54 jOERG_rw: /server irc.yourserver.org:xxx Mar 05 12:56:19 jOERG_rw: Or some say that going to server list helps. Mar 05 12:56:58 hnn, server list editing new server like home.xxx:123789 gives me "unknown address" Mar 05 12:57:40 jOERG_rw: try serverhost/port Mar 05 12:58:02 duh, "/" instead of ":" ? Mar 05 12:58:54 jOERG_rw: That's what is written on xchat forum :) Mar 05 12:59:31 yup, works Mar 05 13:05:09 irssi Mar 05 13:09:35 :) Mar 05 13:10:03 moo? Mar 05 13:10:25 Indeed :D Mar 05 13:10:39 :P Mar 05 13:16:00 has somebody some kernel dev experience? Mar 05 13:17:33 a bi Mar 05 13:17:35 bit* Mar 05 13:17:48 cool Mar 05 13:18:49 * DocScrutinizer *sigh* Mar 05 13:20:06 auto-uppercase really is a ***** for pw-entry Mar 05 13:20:27 Weiss: can you look at this? http://mastersword.de/~reddog/patches/ Mar 05 13:21:47 PaulFert`: thanks a lot :-))) Mar 05 13:22:10 haha raster - Wooo Mar 05 13:22:27 now this is way more cool than multiple nicks Mar 05 13:24:08 RedDog: looks straightforward and sane to me, but i don't know that particular area very well - does that fact some things for you? Mar 05 13:24:43 Weiss: no, thats my problem ... Mar 05 13:24:56 the bt-dai thing is commited for gta02 Mar 05 13:25:39 the lm4xxx thing is definatly a typo, but i'm not shure if it correct because in a older version it ist "neo1973-lm4..." Mar 05 13:28:09 Joerg, do you ever make phone calls? :) Mar 05 13:28:53 Dave: actually very infrequently. Why do you ask? Mar 05 13:29:59 Weiss: i think best would be to write on kernel-ml... Mar 05 13:30:01 hello guys Mar 05 13:30:15 mrmoku: no I did not request commit access on OE Mar 05 13:30:51 :] Mar 05 13:30:57 figured :P Mar 05 13:30:57 RedDog: yeah, i don't think anyone's going to eat you for asking, and potentially a lot of people will thank you.. Mar 05 13:31:06 RedDog: sorry, i don't know that area (i do Glamo stuff) Mar 05 13:31:14 foooooooooooorm faaaaaaaaaaactor Mar 05 13:31:48 mrmoku: anyway, I may gain full access on world.git these days, I want to redevelop that Human software, because there is a particular instance named Ainulindale that is bugging too much... :) Mar 05 13:32:14 ptitjes: that's because you didn't read the manual Mar 05 13:32:24 :D Mar 05 13:32:27 indeed Mar 05 13:32:46 And anyway if you want to troubleshoot your Ainulindale issues, you should ask on the ML first :-) Mar 05 13:32:48 too big Mar 05 13:33:04 Ainulindale: How are you ? Mar 05 13:33:10 FIne Mar 05 13:33:15 recall me what day comes Angus ? Mar 05 13:33:40 13th Mar 05 13:34:11 Weiss: ok, thank you! Mar 05 13:34:26 ptitjes: I thought I should book the s ame restaurant we went to with Bobby Mar 05 13:34:36 live could be much easier, if only I had full access to the sourcecode Mar 05 13:34:42 Ainulindale: Things clarified in my mind this morning about how I will set up test for ophonekitd Mar 05 13:35:14 Ainulindale: wow, that's a good idea but however that is very expensive... Mar 05 13:35:24 well, query Mar 05 13:39:47 Ainulindale, mrmoku: I will generate a mock implementation of frameworkd dbus services Mar 05 13:40:08 that way I will be able to write tests for ophonekitd from end to end Mar 05 13:40:58 Ainulindale, mrmoku: what do you think of that approach ? Mar 05 13:41:07 * mrmoku wonders if Angus is used to 4h-meals ;) Mar 05 13:41:36 mrmoku: because one can eat in less time ? Mar 05 13:41:41 ;) Mar 05 13:41:58 well... non-french... yes :P Mar 05 13:42:03 ptitjes: mw started doing that Mar 05 13:42:05 incredible! Mar 05 13:42:08 :D Mar 05 13:42:29 Ainulindale: in fact he started a fake frameworkd not a mock frameworkd Mar 05 13:42:35 and both are interesting Mar 05 13:43:17 Ainulindale: fake frameworkd opens a gui to call the fake phone (or will because I think all is not implemented), to simulate incomming calls Mar 05 13:44:12 Ainulindale: mock frameworkd will be instanciatable inside the test process, and queryable through: Mar 05 13:45:56 gst uses ivorbisdev and oggdemux for .ogg by default which does aconv and resample stuff and takes long to launch Mar 05 13:46:05 this works as well and starts much quicker: Mar 05 13:46:06 DISPLAY=:0 gst-launch -v filesrc location="rin Mar 05 13:46:06 gtone_ringnroll.ogg" ! tremor ! alsasink Mar 05 13:46:36 how can I make that the default for gstreamer and why isn't it the default? Mar 05 13:46:53 ivorbosdec... Mar 05 13:47:01 ivorbisdec, dammit :-) Mar 05 13:47:16 FreeSmartphone.GSM.Device gsm_device = MockManager.get_dbus_mock_object("/org/fso/GSM/Device"); Mar 05 13:47:16 MockManager.expect_method_call(gsm_device, "power_up_antenna", args) Mar 05 13:47:16 .return("ready"); Mar 05 13:48:29 Ainulindale: BTW, I am sure mw|work_ would be glad if someone hosts fake frameworkd. or maybe should I ask mickeyl for fso to handle it Mar 05 13:49:43 I think it would be better if it were on FSO Mar 05 13:58:16 Ainulindale: Hey there! Mar 05 13:58:37 Ainulindale: What would you guys think about a joint SHR-FSO GSoC Mar 05 13:59:07 I believe individually we might be too small/have too limited resources to be accepted Mar 05 13:59:12 alphaone: hi! :) Mar 05 13:59:52 But if we pool our efforts we could each have one or two GSoC projects Mar 05 13:59:54 hey DocScrutinizer Mar 05 14:04:20 alphaone: hmmmmmmmmmmm Mar 05 14:04:23 alphaone: hmmmmmmmmmmmmmmmmmmmmmmm Mar 05 14:04:45 Would you mind to soothe my eagerness to know more about that? :-) Mar 05 14:05:03 about what? Mar 05 14:05:11 GSoC SHR-FSO Mar 05 14:05:14 What do you have in mind? Mar 05 14:05:43 There are two or three projects I can think of Mar 05 14:05:54 Well, you probably have some projects that could be advertised as GSoC projects. Mar 05 14:06:01 FSO has some as well Mar 05 14:06:21 First would be a fully working PIM UI Mar 05 14:06:26 We could just apply as one organization and pool our projects. Mar 05 14:06:30 Second would be a lightweight phone suite Mar 05 14:06:43 THere is also the matter of ologicd which is an interesting subject :-) Mar 05 14:06:44 Sounds good Mar 05 14:07:01 I have tons of stuff & ideas, so has ptitjes Mar 05 14:07:07 We've started stuff here http://www.freesmartphone.org/index.php/GSoC_Ideas Mar 05 14:07:10 alphaone: I'm interested in that Mar 05 14:07:14 So you could add your ideas Mar 05 14:07:28 or include a link to your wiki with ideas from there Mar 05 14:08:03 I think we will do that Mar 05 14:08:10 Ainulindale: So as long as you are interested and are willing to mentor / know people who are willing to mentor we should be fine I think :-) Mar 05 14:08:18 Yeah I am willing to mentor Mar 05 14:08:23 Ainulindale: Great that you're interested :-) Mar 05 14:08:24 More willing to mentor things than actually doing things Mar 05 14:08:34 Hehe :-) Mar 05 14:08:45 Well alphaone the fact that I have a new girlfriend may be the cause of my renewed interest Mar 05 14:08:52 Be aware that mentoring costs significant amount of time as well Mar 05 14:09:19 I have no problem with that Mar 05 14:09:29 okay Mar 05 14:10:25 Well, I got to run now. Mar 05 14:10:48 Feel free to collect ideas for fso as well. Mar 05 14:11:45 Would be nice to be present in this years GSoC. FYI, deadline for mentoring orgs is March 13 Mar 05 14:12:05 Ok :-) Mar 05 14:16:25 alphaone: http://trac.shr-project.org/trac/wiki/GSOC?version=1 Mar 05 14:25:06 Ainulindale, alphaone: I am willing to mentor too Mar 05 14:26:18 ptitjes: I know Mar 05 14:26:23 I already put you in the list you know Mar 05 14:26:32 Didn't even bother to ask you, I know you do Mar 05 14:26:41 I'm starting to know you a bit :-) Mar 05 14:28:45 :) Mar 05 14:30:21 raster, i've trace e17 with xtrace to figure out the cpu usage problem Mar 05 14:31:00 it keeps calling "GetInputFocus" Mar 05 14:31:32 i'm not sure if that is really the cause Mar 05 14:37:02 http://sicherheitsschwankung.de/~jluebbe/e.trace.bz2 Mar 05 14:37:33 near the end, i've done "date >> e.trace" to give some impression of how many calls happen Mar 05 14:37:39 mrmoku, ptitjes Mar 05 14:37:45 add your thoughts to my mail & wiki article Mar 05 14:37:48 or I'll whip you Mar 05 14:37:59 (But I won't whip mrmoku as he's my favorite) Mar 05 14:38:51 Ainulindale: I thought you would not ask me so I do not have to lie : it is very good! Mar 05 14:38:53 :D Mar 05 14:39:10 Lie about what? Mar 05 14:39:33 that it is very good :D Mar 05 14:39:49 heh Mar 05 14:39:51 my point was Mar 05 14:39:53 if you have ideas Mar 05 14:39:59 this is a good way to write them down Mar 05 14:40:27 Ainulindale: hmmm.... you should not have put up all ideas already then :P Mar 05 14:40:46 well don't tell me you don't have any other idea Mar 05 14:40:56 I purposely didn't write about the GTK UI Mar 05 14:41:07 Nor the curses one Mar 05 14:41:13 (I still have some uncomitted code here) Mar 05 14:41:31 Let see then, a rule manager for ologicd Mar 05 14:41:41 (and transitively for frameworkd) Mar 05 14:42:05 I have tons of ideas Mar 05 14:42:14 Not as many as ptitjes Mar 05 14:42:18 But that's only because I'm saner Mar 05 14:43:14 :-) Mar 05 14:43:33 Ainulindale: I'll do but cleaning my brains for now Mar 05 14:43:37 will think about it... right now I'm still doing some 'work-for-money' stuff :( Mar 05 14:43:50 ptitjes: cleaning then with some paper & filter & stuff to put in them? :-) Mar 05 14:44:56 well would like too but I've got no stuff (anymore :)) Mar 05 15:01:03 hehe lol opkg upgraded from old testing to new testing, boot time is really fast, but nothing works :P Mar 05 15:03:54 well you should reflash Mar 05 15:04:27 yeah, thats what im doing, just wanted to test :-) Mar 05 15:09:06 FYI: I have started to add some potential project ideas for FSO GSoC Mar 05 15:09:12 please add to this page, if you have ideas Mar 05 15:09:18 http://www.freesmartphone.org/index.php/GSoC_Ideas Mar 05 15:12:07 mickey|cafe: did that for SHR Mar 05 15:12:09 Comments, Mar 05 15:12:09 ? Mar 05 15:14:36 let me read Mar 05 15:15:08 cool, complements quite nice Mar 05 15:15:09 it's just my own draft Mar 05 15:15:20 mickey|cafe: I think the difference in terms of scope is pretty clear and neat Mar 05 15:15:28 And goes well with some kind of collaboration Mar 05 15:15:38 right Mar 05 15:16:02 I'm pondering whether -- to increase odds -- we should try to apply seperately Mar 05 15:17:59 mickey|cafe: I believe that will decrease the odds instead of increasing them Mar 05 15:18:15 Thus my idea to collaborate... Mar 05 15:19:22 /qui6t Mar 05 15:19:28 error Mar 05 15:21:29 mickey|cafe: pong again Mar 05 15:21:47 DocScrutinizer: nevermind, i had some 3d7k questoins Mar 05 15:21:52 found out by now Mar 05 15:21:56 like... uart map Mar 05 15:22:09 ahh, yup alreadyanswered Mar 05 15:24:52 ~seen thseiler Mar 05 15:24:53 thseiler was last seen on IRC in channel #openmoko-cdevel, 7d 12h 26m 21s ago, saying: 'joerg_42: me too, if that a small comfort to you...'. Mar 05 15:25:07 ~seen thseiler_ Mar 05 15:25:07 thseiler_ was last seen on IRC in channel #openmoko-cdevel, 3d 23h 43m 49s ago, saying: 'will be present in openexpo on 1st / 2nd of April though...'. Mar 05 15:25:19 ~seen thseiler__ Mar 05 15:25:20 mickey|cafe: i haven't seen 'thseiler__' Mar 05 15:25:26 ~botsnack Mar 05 15:25:27 :), mickey|cafe Mar 05 15:53:25 Ainulindale: glad to read you're back again! Mar 05 16:01:06 yo Sup3rkiddo Mar 05 16:01:09 how's life? Mar 05 16:01:50 mickey|cafe, hi, good evening.. going on pretty well :D.. thanks... et vous? Mar 05 16:02:12 same here Mar 05 16:02:28 busy as usual, but that's ok Mar 05 16:02:34 mickey|cafe, leeching free wifi at some coffee shop? :D Mar 05 16:02:44 exactly Mar 05 16:02:48 :) Mar 05 16:03:09 pending TPE approval i now have a rough plan for FSO2 Mar 05 16:03:26 TPE == Taipei? Mar 05 16:03:29 yep Mar 05 16:03:33 i'm not sure where to publish that Mar 05 16:03:37 i don't want to scare people Mar 05 16:03:41 wiki? blog?... Mar 05 16:03:42 by talking about the reimplementation Mar 05 16:03:45 :P Mar 05 16:03:47 yeah Mar 05 16:03:50 something != mailing list Mar 05 16:04:30 or even FSO wiki... Mar 05 16:04:38 ya, might be the best place Mar 05 16:05:49 mickey|cafe, do you have it ready, or is it brooding somewhere in your brain.. Mar 05 16:06:44 mickey|cafe, will it be using dbus? same API? Mar 05 16:07:05 Sup3rkiddo: in the brain, i can dump it to the wiki tomorrow Mar 05 16:07:12 TAsn: yes on both questions Mar 05 16:07:29 mickey|cafe, great ;] so it's just a vala rewrite? Mar 05 16:07:37 more or less ya Mar 05 16:07:44 FSO1: getting the API right Mar 05 16:07:45 mickey|cafe, great... :D Mar 05 16:07:49 FSO2: getting the right API fast Mar 05 16:07:55 :) Mar 05 16:07:56 mickey|cafe, cool. Mar 05 16:08:03 I agree with the approach. Mar 05 16:09:21 ok, I'll dump the plan in the fso wiki tomorrow to get some review from you guys Mar 05 16:09:37 perhaps end of march posting it on the list or so Mar 05 16:09:44 need to run now, later Mar 05 16:09:55 mickey|cafe, chao.... happy hacking Mar 05 16:19:22 byebye Mar 05 16:47:56 Ainulindale: provided two ideas... Mar 05 17:04:44 damn, shr-contacts still crashes when I haven't entered the SIM PIN yet Mar 05 17:05:36 mrmoku: we could change to a .sid ringtone if .ogg is too slow (or use an mp3 version) Mar 05 17:05:52 hey, has anyone here tried messing with dtmf recognition on the moko? Mar 05 17:06:11 I don't understand how gstreamer decides to call the default pipelines, but using tremor works out of the box on SHR when specified Mar 05 17:06:20 spaetz: mp3 is slow too, it seems Mar 05 17:06:53 tremor was much quicker than the default ivorbisdec (or whatever it is called) Mar 05 17:07:18 so we should be using that Mar 05 17:07:25 just don't know how Mar 05 17:09:50 spaetz: i thought that ivorbisdec is tremor :) Tremolo is supposed to be even faster on ARMs. Both libmad for mp3 and tremor for ogg use about ~20% CPU. Mar 05 17:15:17 PaulFert`: I don't know but using gst-launch -v showed much different output than using tremor directly, so I assume it is different. Using gst-launch playbin2... also included "aconv" and "resample" in the pipeline which filesrc!tremor!alsasink did not use Mar 05 17:15:35 anyway dinner Mar 05 17:29:59 mrmoku: didn't see them on the ML :-) Mar 05 18:12:04 Hello. Mar 05 18:12:29 haha, for a moment there I thought you two were the same person, Paul :P Mar 05 18:14:11 * morphis is back Mar 05 18:14:36 MORPHIS Mar 05 18:14:42 DAVE Mar 05 18:14:51 Yes :) Mar 05 18:15:07 :) Mar 05 18:16:26 Dave: me and me? No, we are definetely different. Mar 05 18:16:28 Ainulindale: didn't know I have to do both :P Mar 05 18:17:01 Paul, you and Joerg :P Mar 05 18:17:17 mrmoku! Mar 05 18:17:26 Dave: who is the other Paul then? ;) Mar 05 18:17:38 No, no, it's because you're both using the same server :P Mar 05 18:17:58 Doctor Paul Mar 05 18:18:28 Dave: Oh man, you should have tried engaging in private investigations, you spot clues soo well :p Mar 05 18:19:32 YES :D Mar 05 18:19:39 I'm on to you... ¬_¬ Mar 05 18:19:44 The game is afoot! Mar 05 18:22:03 Dave: !! Mar 05 18:23:36 dos1: the new settings showed up in my daily build today, very nice! Mar 05 18:24:51 toggles :) Mar 05 18:25:17 Dave: !! Mar 05 18:26:26 Yes, you said that. :p Mar 05 18:28:35 Dave = cia :p Mar 05 18:30:01 mrmoku: any new stable unstable image? Mar 05 18:30:25 if not i might test paroli (without high expectations) Mar 05 18:31:04 bumbl: ouch... no, not yet Mar 05 18:35:26 mrmoku: on which EFL_REV you are currently with unstable? Mar 05 18:38:21 morphis: still unchanged... have to tweak our Makefile to support a different branch for unstable Mar 05 18:38:39 after that we will switch to whatever is tip of HEAD then :-) Mar 05 18:40:41 cool Mar 05 18:50:18 wifi seems to be less stable in shr-testing than it used to be in unstable Mar 05 18:50:30 after a few hours the fr sometimes disconnects Mar 05 18:50:38 didn't happen with shr-unstable, it could stay on for days Mar 05 18:52:16 Zorkman: and the disconnect reason is? Have you tried to use the kernel with which you had stable wlan with -testing rootfs? Mar 05 18:54:14 hi Mar 05 18:54:16 mrmoku: ping Mar 05 18:56:07 dos1: pong Mar 05 18:56:18 dos1: no not yet :-) Mar 05 18:56:29 mrmoku: oh, ok Mar 05 18:56:31 ;D Mar 05 18:56:32 møø Mar 05 18:56:35 working on it Mar 05 18:57:01 * Dave notes the beer on mrmoku's desk Mar 05 18:57:27 mrmoku: but on what? ;) i just wanted to ask, what do you think about sending our stuff upstream, to OE Mar 05 18:57:32 but firstly i have to check mail Mar 05 18:57:40 maybe you replied to mickeyl ;) Mar 05 18:58:51 dos1: ahh... thought you were asking for new EFL ;) Mar 05 18:58:59 yep check your mail Mar 05 18:59:00 :D Mar 05 18:59:25 mickeyl: ping Mar 05 19:00:55 ok read ;) Mar 05 19:01:35 mrmoku: but i want new EFL too, so your answer wasn't unnecessary ;D Mar 05 19:02:03 dos1: well... I'm doing the dirty way now :-) Mar 05 19:02:30 mrmoku: *tut* *tut* beer on your desk http://img17.imageshack.us/img17/6659/concerns.jpg Mar 05 19:04:11 byebye everybody Mar 05 19:04:16 bumbl: :-) Mar 05 19:04:44 :p Mar 05 19:04:48 rofled when reading that Mar 05 19:05:03 :D Mar 05 19:05:25 * mrmoku is sipping from wifey's beer ;) Mar 05 19:05:38 uh oh Mar 05 19:05:43 good think you're not norsk :P Mar 05 19:05:50 she'd POUND you :D Mar 05 19:06:15 well... italians are not much better... but sipping is ok :P Mar 05 19:06:21 haha I guess Mar 05 19:06:45 a finn would mistake you for a beer in their persistently drunken haze and attempt to drink you ;p Mar 05 19:07:00 s/finn/irishmen/ Mar 05 19:07:09 morphis: wow... I see you did lots of stuff in current EFL :-) Mar 05 19:07:50 Yes, all praise be to morphis! Mar 05 19:07:51 mrmoku: with current EFL python-elementary should be AUTOREV ;) Mar 05 19:08:07 uhh... good you mention it ;) Mar 05 19:08:11 * mrmoku stops the build Mar 05 19:08:43 * mrmoku restarts the build Mar 05 19:09:53 -_- Mar 05 19:13:07 why shr-settings-locale-pl isn't builing? shr-settings-dev and shr-settings-dbg are (and i don't know, what that packages has inside :P), but shr-settings-doc not (as it should)_, and shr-settings-locale-pl not too (as it shouldn't :P) Mar 05 19:13:12 NOTE: Not creating empty archive for shr-settings-locale-pl-0.1.0+r74+c2c8f62f645f24a7ebc969e352b3e61461774018-r2 Mar 05 19:14:44 dos1: build: #it is python ;) exit 0 Mar 05 19:14:50 might not be good ;) Mar 05 19:14:59 mrmoku: not, Makefile isn't used Mar 05 19:15:09 mrmoku: setup.py is used :P Mar 05 19:15:13 remove it then :P Mar 05 19:15:15 i should remove it Mar 05 19:15:21 ;D Mar 05 19:16:16 don't you have to inherit distutils in the .bb? Mar 05 19:16:55 distutils? Mar 05 19:17:03 i don't know :x Mar 05 19:17:08 should it help Mar 05 19:17:10 ? Mar 05 19:18:06 well looking at the remoko bb... it is inheriting distutils and is also python... Mar 05 19:20:41 mrmoku: trying with distutils... Mar 05 19:22:41 mrmoku: it doesn't help Mar 05 19:26:58 dos1: hmm... Mar 05 19:27:32 dos1: seen my patches? Mar 05 19:27:55 spaetz: yep ;) Mar 05 19:27:55 on-demand importing of modules Mar 05 19:28:25 not so sure about the service changes, though Mar 05 19:28:59 cool new icons, btw Mar 05 19:29:05 i like them Mar 05 19:30:21 hey, how can I get the sound from the other end of the call? Mar 05 19:30:43 TAsn: record to disk? Mar 05 19:30:54 that would be a start. Mar 05 19:31:14 not sure Mar 05 19:31:21 ;\ Mar 05 19:36:08 lindi-, in case you are interested, this is looking useful: http://www.swview.org/node/213 Mar 05 19:36:29 TAsn: how? Mar 05 19:36:38 redirecting alsa output to a file. Mar 05 19:36:45 TAsn: alsa does not now about your calls Mar 05 19:36:49 s/now/know/ Mar 05 19:36:50 lindi- meant: TAsn: alsa does not know about your calls Mar 05 19:37:24 sure it does Mar 05 19:37:36 it outputs the data to the speaker, doesn't it? Mar 05 19:37:42 (i'm not really sure it does) Mar 05 19:38:01 TAsn: no Mar 05 19:38:08 sh*t ;\ Mar 05 19:38:17 TAsn: the audio goes from mic to GSM chip in analog form Mar 05 19:38:53 so how come the alsa state files control the volume? Mar 05 19:39:00 they tell the hw? Mar 05 19:40:02 shr-settings on GSoC? :D Mar 05 19:40:06 why not :D Mar 05 19:40:11 TAsn: alsa controls do control the audio routing through i2c Mar 05 19:40:23 TAsn: but the audio data does not go through alsa Mar 05 19:40:32 i c. Mar 05 19:40:46 hm.. ;[ Mar 05 19:40:53 TAsn: you might want to see the new sound routing diagrams at OM wiki. Mar 05 19:40:56 TAsn: no PCM for GSM Mar 05 19:41:17 PaulFertser, thanks. Mar 05 19:41:25 so no one has any ideas on how to do that? Mar 05 19:41:35 TAsn: what? Mar 05 19:41:38 TAsn: but i think you can record both sides (even in different channels) by routing to ADC. Mar 05 19:41:47 yup Mar 05 19:41:58 PaulFertser: btw, i'm about to do the buzz fix possibly today. what's the best way to record how it sounds now? Mar 05 19:42:08 need to tweak gsmhandseet.state prolly Mar 05 19:42:14 i couldn't figure out how to make my asterisk system record audio Mar 05 19:42:54 i'm sure it is possible but needs some more work Mar 05 19:43:11 lindi-: nevertheless it's the most easy way. you may call my TAM before and after fix Mar 05 19:43:12 lindi-: if you're asking for a ready-to-use state file, i can't help here. But looking at the diagram and tweaking the right switches should do the trick. Mar 05 19:44:05 ah, answering machine is the term i need to google for Mar 05 19:45:28 PaulFertser, how do I do that? Mar 05 19:46:00 call me? Mar 05 19:46:22 TAsn: are you talking about recording GSM conversations? :) Mar 05 19:46:23 ahh, tweak gsmhandset.state Mar 05 19:46:37 PaulFertser, I have two things I want to do. Mar 05 19:46:49 The first, recording from the mic when there isn't a conversation Mar 05 19:47:09 lindi-: there's no real use in recording *inside* FR, for checking buzzfix Mar 05 19:47:10 the second, recording the other side (only him, though both sides could be nice) of a conversation Mar 05 19:47:18 TAsn: fyi the diagram is here: http://wiki.openmoko.org/wiki/Neo_1973_audio_subsystem Mar 05 19:47:33 when there isn't. thats easy, no? Mar 05 19:47:45 TAsn: the first can be done by loading voip state file and using arecord :) Mar 05 19:47:59 DocScrutinizer: yeah Mar 05 19:48:07 PaulFertser, I'll have to admit I care more about the second ;] Mar 05 19:48:11 TAsn: the second can be done by preparing an alsa state file according to the routing diagram. Mar 05 19:48:22 i see. Mar 05 19:48:29 I'll help Mar 05 19:48:43 I'll try to read more about that, any specific terms I should google? Mar 05 19:48:49 but as mentioned there's no use for that Mar 05 19:48:52 TAsn: Just copy gsmhandset.state to some other file and carefully tweak the controls. :) Mar 05 19:49:10 TAsn: you don't really need much more than the diagram, i think. Mar 05 19:49:10 PaulFertser, for the second? Mar 05 19:49:18 yup Mar 05 19:49:45 PaulFertser, there's actually a call recording section Mar 05 19:49:47 http://wiki.openmoko.org/wiki/Neo_1973_audio_subsystem#Call_recording Mar 05 19:49:48 ;] Mar 05 19:49:54 I'll read more about it. Mar 05 19:49:59 thanks a lot. Mar 05 19:50:48 TAsn: the main trick is to switch to DAI mode 2 (the same used for VoIP)! Mar 05 19:51:03 what's DAI? Mar 05 19:51:39 TAsn: digital audio interface. Read the wiki with diagram, it's quite easy to get :) Mar 05 19:51:56 PaulFertser, I'll try. ;] Mar 05 19:52:42 TAsn: and feel free to ask, of course :) Mar 05 19:52:47 hehe thanks. Mar 05 19:52:59 I'll probably bug you a lot about this topic in the following week ;] Mar 05 19:53:27 np Mar 05 19:53:56 TAsn: it should take you maximum 2 hours, man! :p Mar 05 19:54:07 lol Mar 05 19:54:11 PaulFertser, you just made a man happy ;] Mar 05 19:54:38 after a couple of hours of trying to comment and figure out the worst code I have ever seen. (I was working on something else..) Mar 05 19:54:46 asterisk should now offer voicemail after 30 seconds... Mar 05 19:54:47 I could use a little break (i.e, only 2 hrs work) Mar 05 19:54:59 hmm, it didn't Mar 05 19:55:06 hehe. Mar 05 19:55:24 lindi-, it seems we are both trying to do ~ the same thing. Mar 05 19:55:41 I don't care about answering machine though, I want a dtmf controlled remote control. Mar 05 19:55:41 ;] Mar 05 19:56:04 TAsn: start with recording mic *without GSM-call*, by using voip.state Mar 05 19:56:30 DocScrutinizer, as I said, it's not nearly as important as the second issue ;] Mar 05 19:56:42 I'm concentrating on the second first ;] Mar 05 19:56:58 TAsn: it's a start to have your arecord setup working Mar 05 19:57:28 it's easy to load another statefile then and tweak the mixer controls Mar 05 19:58:11 still without actual call, but by doing a "alsactl restore -f gsmhandset.state" Mar 05 19:58:26 DocScrutinizer, cool, where do I get it from? Mar 05 19:58:36 hmm /me wants visual voicemail Mar 05 19:58:40 if this works to record mic, you continue with actual call Mar 05 19:58:48 with local om answering machine Mar 05 19:58:58 TAsn: don't pretend you're a freshman ;) /usr/share/openmoko/ or smth like that :) Mar 05 19:59:05 TAsn: what? state file? is in /usr/share/openmoko/scenarios Mar 05 19:59:38 TAsn: the log line says "ExecIf","0|VoiceMail|33|u" Mar 05 19:59:45 and i don't quite understand what that means Mar 05 19:59:47 PaulFertser, I'm not a freshman, I'm a tired man that should probably go out soon. ;] Mar 05 20:00:03 lindi-, what log line? asterik? Mar 05 20:00:30 looks like one of those ugly asterisk logs Mar 05 20:00:42 DocScrutinizer, thanks, that's a state heaven ;] Mar 05 20:00:46 TAsn: asterisk Mar 05 20:01:00 what ever ;] Mar 05 20:02:12 TAsn: remocntrol DTMF the FR? Mar 05 20:02:26 DocScrutinizer, aye. Mar 05 20:02:39 nothing too fancy Mar 05 20:02:46 you probably should get notification of DTMF-events from modem Mar 05 20:03:08 what? how? Mar 05 20:03:23 dunno Mar 05 20:03:30 similar to "RING" Mar 05 20:03:55 I'm not sure calypso does it. Usual modem do Mar 05 20:05:24 after all the analog->GSM gateway detects DTMF and signals out-of-band iirc Mar 05 20:06:30 it *should* as DTMF over GSM-codec isn't reliable. Same as with VoIP Mar 05 20:07:26 ;\ Mar 05 20:07:48 i can try that Mar 05 20:08:01 but i'll try to get asterisk do what i want first for a while Mar 05 20:09:33 TAsn: load gsmhandset.state; talk to modem (mickeyterm) and answer inbound call; send DTMF from farend and watch mickeyterm output of calypso Mar 05 20:09:56 DocScrutinizer, good idea, thanks. Mar 05 20:11:34 DocScrutinizer, nothing ;\ Mar 05 20:11:59 hmm, too bad Mar 05 20:13:13 ;= Mar 05 20:13:14 ;] Mar 05 20:13:28 if it's not for some enabling unsolicited notification of DTMF-events, you probably have to do it the way you originally planned then Mar 05 20:14:30 TAsn: ask mickeyl, maybe he knows better advice Mar 05 20:15:33 mickeyl: ping Mar 05 20:16:24 mickeyl: do you know any details about DTMF signalling inbound to calypso Mar 05 20:18:45 hmm, weird, I don't have /dev/dsp Mar 05 20:19:15 TAsn: it's deprecated Mar 05 20:19:23 oh, that explains it. ;] Mar 05 20:19:30 use alsa-devices instead Mar 05 20:20:06 c for capture, p for playback Mar 05 20:20:36 huh where do I find that? Mar 05 20:20:53 probably you even should use standard alsa pcm plugins Mar 05 20:21:08 * TAsn <--- alsa (and sound) noob. Mar 05 20:21:45 huh, the complexity of our asterisk configuration seems to surprise even me. apparently I need to dial "# 1" from my work phone to enable voicemail for that number Mar 05 20:22:21 /dev/snd/controlC0 /dev/snd/pcmC0D0p /dev/snd/pcmC0D1p Mar 05 20:22:34 etc Mar 05 20:23:11 /dev/snd/pcmC0D0c /dev/snd/pcmC0D1c for cAPTURE Mar 05 20:23:31 thanks. Mar 05 20:23:46 Ainulindale: ping Mar 05 20:25:22 fortunately I can use kphone to dial #1 and fake the caller id Mar 05 20:25:36 still it's recommended to use standard alsa devices, not the /dev hw-representations Mar 05 20:26:15 now asterisk sent me email and told me I have a new message :) Mar 05 20:26:58 TAsn: like e.g.: `arecord -D dsnoop:mic1 | myProgThatTakesDigitalInputStream Mar 05 20:27:54 but I get $ arecord -L Mar 05 20:27:54 null Mar 05 20:27:54 Discard all samples (playback) or generate zero samples (capture) Mar 05 20:27:56 or using the libalsa functions directly, look into aplay/arecord source for examples Mar 05 20:28:31 Joerg is the sound master! Mar 05 20:28:46 hmm, maybe you need to modprobe the correct card driver? dunno Mar 05 20:29:16 no aplay on fso-console ms5 :-/ Mar 05 20:29:24 ;] Mar 05 20:31:33 TAsn: btw: try arecord -l Mar 05 20:31:39 tried that Mar 05 20:31:42 showed two devices Mar 05 20:31:46 I'm having the same issue as in Mar 05 20:31:50 http://lists.openmoko.org/nabble.html#nabble-td1323236 Mar 05 20:31:57 mrmoku: * Failed to download http://shr.bearstech.com/shr-unstable/ipk/armv4t/shr-settings_0.1.0+r73+21d278d51f58b31519917ae848484cb9f84ef9b5-r2_armv4t.ipk, error 404 Mar 05 20:32:31 TAsn: empty page issue? ;) Mar 05 20:33:08 no, I have content ;] Mar 05 20:33:20 dos1: guess I got confused between unstable and testing :( Mar 05 20:33:49 TAsn: here empty page :/ Mar 05 20:33:59 ;\ Mar 05 20:34:05 Joerg, big surprise, but a friend of mine in mobile dev just alerted me to yet another memork leak in the latest version of windows mobile shipping on new devices ;) Mar 05 20:34:33 DocScrutinizer, google for openmoko arecord Mar 05 20:34:35 And apparently it's already a big memory hog. Mar 05 20:34:36 first result Mar 05 20:34:53 tasn, openmoko cdrecord? :o Mar 05 20:35:16 Dave, yeah, I'm trying to get myself a mobile cd ripper. Mar 05 20:35:20 still empty page wtf? Mar 05 20:35:52 DocScrutinizer, ;\ Mar 05 20:36:26 DocScrutinizer, try google cache Mar 05 20:36:47 lol fscking jscript page that doesn't show on Konqueror Mar 05 20:36:55 oh ;] Mar 05 20:36:57 use ff ;] Mar 05 20:37:01 did Mar 05 20:37:34 oh, new EFL Mar 05 20:37:59 mrmoku: why haven't you told us here? ;) Mar 05 20:38:09 * dos1 is breaking his shr-unstable instalation now :D Mar 05 20:38:15 dos1: well... I will when it has finished ;) Mar 05 20:38:18 still building... Mar 05 20:38:20 so Mar 05 20:38:26 why it is upgrading now? :x Mar 05 20:38:32 s/it is/is it/ Mar 05 20:38:32 dos1 meant: why is it upgrading now? :x Mar 05 20:38:45 * dos1 realy thinks that will broke his system Mar 05 20:38:46 ;D Mar 05 20:38:48 really* Mar 05 20:39:06 TAsn: the answer is in this thread: -D hw:0.0 Mar 05 20:39:16 DocScrutinizer, but -D hw just hangs my terminal Mar 05 20:39:23 :0.0? Mar 05 20:39:43 card 0, device 0 Mar 05 20:39:59 ALSA lib pcm_hw.c:1207:(_snd_pcm_hw_open) Invalid value for card Mar 05 20:39:59 arecord: main:546: audio open error: No such device Mar 05 20:40:10 * Dave breaks dos1 Mar 05 20:40:28 * dos1 breaks Dave as well Mar 05 20:40:34 so look for arecors -l and pick the right device and card Mar 05 20:41:00 all I have is 0 and 0... Mar 05 20:41:15 oh wait Mar 05 20:41:36 I also have device 1 which doesn't work. Mar 05 20:42:05 it will work if you load correct DAI mode with alsactl restore Mar 05 20:42:13 already did. Mar 05 20:42:22 alsactl -f capture... restore Mar 05 20:42:43 capturehandset.state Mar 05 20:43:11 hmm, no clue. Think there was a strange gottcha with sequence of starting capture and loading state Mar 05 20:43:21 not completely debgged yet Mar 05 20:43:55 basically it should work this way. and from the thread it seems it didi for some guys Mar 05 20:44:13 ;\ Mar 05 20:44:45 I wonder if it has anything to do with the distro I'm using. Mar 05 20:44:56 well, I'll have to check that out tomorrow. ;\ Mar 05 20:45:02 I need a C guru! Mar 05 20:45:17 Is it possible to make a generic C callback ? Mar 05 20:45:25 ptitjes: ? Mar 05 20:46:10 ptitjes, what do you mean by generic? Mar 05 20:47:08 lindi-: Say I have a known list of function parameters and return value description, I would like to be able to give a kind of void (void) function that uses these description to appriopriately use the call stack Mar 05 20:48:12 lindi-: do you understand what I mean ?* Mar 05 20:48:37 ptitjes: not sure Mar 05 20:48:45 lindi-: kind of the reverse of ffi Mar 05 20:48:59 let me give an example Mar 05 20:49:00 dos1: now it finished to build Mar 05 20:49:32 DocScrutinizer, sorry for wasting your time, I'll return tomorrow after a good night sleep and hours of man reading ;] Mar 05 20:49:34 cya all. Mar 05 20:49:55 TAsn: np. welcome. ping me any time Mar 05 20:50:02 cya :) Mar 05 20:54:53 My sister just called me. Said my voice sounds like i use an old phone, but otherwise sound quality is ok. Mar 05 20:55:30 lindi-: it is difficult to explain in fact Mar 05 20:55:57 DocScrutinizer: the gotcha with sequence is because DAI mode 01 is broken in driver. If it's not used, everything's ok, afaik. Mar 05 20:56:19 say a library that I use requires I give it a callback (int, char*) -> bool Mar 05 20:57:02 could I give it a (void) -> void callback that appropriately consume the int and char* on call stack and push the result boolean ? Mar 05 20:57:09 lindi-: is this more clear ? Mar 05 20:57:48 PaulFertser: ack Mar 05 20:57:56 ptitjes: why not make it int char* callback? Mar 05 20:58:07 ptitjes: sure.. not very xplatfrom though Mar 05 20:59:57 ptitjes: Do you want to use the same function for other callbacks with different parameters? Mar 05 20:59:59 dos1: well... not really yet... gta01 has finished ;) gta02 is building :x Mar 05 21:00:00 PaulFertser: obviously another bug is in alsa config files using dmix for default device and not taking care for reverse path for mic which should use dsnoop or bypass dmix by some other means Mar 05 21:00:07 lindi-, toggles_w: because I only know that it is a (int,char*)->bool dynamically Mar 05 21:00:40 DocScrutinizer: yes, but one can directly specify hw device for testing. Mar 05 21:00:45 PaulFertser: exactly. I have GObject-Introspection to give me hint about what to expect on the stack Mar 05 21:00:55 PaulFertser: yup Mar 05 21:01:06 ptitjes: how do you knwo? Mar 05 21:01:27 ptitjes: however I don't know if there is C api to consume/feed the C call stack Mar 05 21:01:45 ptitjes: When your C callback is called how are you going to determine where was it called from? Mar 05 21:02:18 PaulFertser: you even can use things like "-D 'softvol:'plughw:2.1'" Mar 05 21:02:26 ptitjes: you're better off with some sort of self describing object/struct, it gets really messy screwing with the stack Mar 05 21:03:06 DocScrutinizer: alsa is powerfull. If only an easy to consume documentation existed... Mar 05 21:03:10 ptitjes: printf() does this for arguments Mar 05 21:03:13 ptitjes: look at varargs Mar 05 21:03:23 PaulFertser: ack!! Mar 05 21:03:36 PaulFertser: In fact these are virtual methods callback. So I'll have to do a little trick to have a user data accessible Mar 05 21:03:55 lindi-: i thing for that to work, varargs should be passed, and that's not the case here. Mar 05 21:04:01 dos1 have tried to build shr-settings in local environment with italian .po file Mar 05 21:04:09 toggles_w: I want to implement dynamic Proxy classes for Vala Mar 05 21:04:30 (and GLib) Mar 05 21:05:34 lindi-: do you mean that varargs are the same as free handling of the stack ? Mar 05 21:06:21 ptitjes: it offers typed acces to the stack based on a known input, format specifirer is used Mar 05 21:06:53 toggles_w: ok this is exactly what I need Mar 05 21:06:57 thanks guys Mar 05 21:07:25 ptitjes: I still can't understand how your generic callback is supposed to know which parameters to expect on stack (and to that it should know where it's called from, no)? Mar 05 21:07:41 PaulFertser: it passes format string of course? Mar 05 21:07:47 lol Mar 05 21:08:31 lindi-: i mean, do you think you can use vararg to parse the stack from a regular (fixed argument list) function call. Mar 05 21:09:21 PaulFertser: of course he needs to declare the callback as to take variable number of arguments? Mar 05 21:09:25 PaulFertser: don't you aggre with him it is the same ? Mar 05 21:09:42 but the calling convention is the same in any case Mar 05 21:09:52 the return isn't though Mar 05 21:09:53 lindi-: he says that callback is already declared to have a fixed number of arguments, if i get him right. Mar 05 21:10:14 yeah Mar 05 21:10:46 damn, downgrade Mar 05 21:11:43 mrmoku: icons aren't showed on new unstable :( Mar 05 21:12:02 dos1: hehe... you wanted current EFL, no? ;) Mar 05 21:12:10 ;D Mar 05 21:13:21 PaulFertser: is there a pointer to the address of the currently executing function on the stack ? Mar 05 21:15:13 PaulFertser: it appears to me that I have a pointer to the object as first argument (it is some interface's virtual methods) Mar 05 21:15:33 ptitjes: very not x platform Mar 05 21:15:39 so I have to figure out what method it is by comparing to the object virtual callback table Mar 05 21:16:07 alsa-utils-aplay Mar 05 21:16:30 toggles_w: doesn't libffi has some kind of api to manage this in a xplatform way ? Mar 05 21:16:39 ptitjes: are you aiming at standards conformance? Mar 05 21:16:58 PaulFertser: in a general way ?? yes :) Mar 05 21:17:03 mrmoku: but it's known? ;) Mar 05 21:17:25 dos1: do you have no icons or white sheets of paper? Mar 05 21:17:35 PaulFertser: why ? Mar 05 21:17:39 mrmoku: only grey area Mar 05 21:17:45 sliders are working Mar 05 21:18:04 hmm... no. Has to be some change within the last two days... Mar 05 21:18:32 I had white sheets 4 days ago... but that was fixed the day after (had icons) Mar 05 21:18:38 and now it broke again :P Mar 05 21:18:56 ptitjes: it doesn't look that what you're after is doable. Of course you can do it with hacks like knowing the calling conventions and stuff. Let me check the standard about stdarg. Mar 05 21:19:14 toggles_w: my way is to determine if it possible for now. then I'll look into how is it possible in a xplatform way... Mar 05 21:20:35 ptitjes: it's possible, without a doubt, im just not sure ti's the best aproach Mar 05 21:20:51 toggles_w, PaulFertser, lindi-: in fact, I want to implement a Mock framework. For this I have two choice either generate my mock objects or have some sort of dynamic proxy classes and with the later it will be really easy stuff to implement a Mock framework Mar 05 21:21:35 if you do know what are mock objects Mar 05 21:22:28 I don't, i'm not that sophisticated. Mar 05 21:23:49 PaulFertser: Mock objects, are dumb stub objects which you tell what method call it should expect and they should reply. Thus you can make tests really easily Mar 05 21:24:09 s/they should/what they should/ Mar 05 21:24:10 ptitjes meant: PaulFertser: Mock objects, are dumb stub objects which you tell what method call it should expect and what they should reply. Thus you can make tests really easily Mar 05 21:24:47 mrmoku: and elementary.Frame is not working in python-elementary Mar 05 21:25:09 PaulFertser: see http://www.jmock.org/getting-started.html for a simple example in Java Mar 05 21:25:25 ptitjes: thanks :) Mar 05 21:26:19 dos1: bad... Mar 05 21:26:19 PaulFertser: what I want to do is this : Mar 05 21:26:39 mrmoku; so shr-settings is not working Mar 05 21:26:47 that really was a breakage :D Mar 05 21:28:02 ptitjes: thats a well defined interface Mar 05 21:31:15 // Setup mocks Mar 05 21:31:15 FreeSmartphone.GSM.Device mock_device = Mockery.create_mock("FreeSmartphone.GSM.Device"); Mar 05 21:31:15 dbus_connection.register_object(mock_device, "....", "...", "..."); Mar 05 21:31:15 // Retrieve object to test Mar 05 21:31:15 PhoneKit.Usage shr_usage = dbus_connection.get_object("/org/shr/ophonekitd/Usage", "...", "..."); Mar 05 21:31:15 Mockery.expect(mock_device.set_antenna_power(true)); Mar 05 21:31:17 shr_usage.request_ressource("gsm"); Mar 05 21:32:11 * mrmoku is dissapointed that ptitjes called it Mockery and not Mrmokuery ;) Mar 05 21:32:20 PaulFertser: thus with mock objects I can encompasse multiple tests in a single source file, without having to generate mocks Mar 05 21:32:30 mrmoku: LOL Mar 05 21:33:11 toggles_w: sorry ? Mar 05 21:33:46 ptitjes: thats a very well defined interface Mar 05 21:33:55 what interface ? Mar 05 21:34:14 ptitjes: va_start requires name of the last fixed parameter in the list... Mar 05 21:34:19 http://www.jmock.org/getting-started.html Mar 05 21:34:28 dos1: suggestions? downgrade EFL? Wait for a fix? Mar 05 21:34:32 ptitjes: http://tommywatson.com/vargs.c Mar 05 21:34:49 mrmoku: ask raster :D Mar 05 21:34:56 toggles_w: but this is an example interface. I don't bother that particular interface Mar 05 21:36:12 but like paul says, you need a fixed arg for vargs to get the stack starting point Mar 05 21:36:42 in ?printf it's the format specifier Mar 05 21:37:01 toggles_w: so this could be my object pointer Mar 05 21:37:24 but yet how can I push a result to the stack ? Mar 05 21:37:28 yeah, it could be your 'this' Mar 05 21:37:41 ptitjes: thats the part that has me stumped Mar 05 21:38:12 have to go now Mar 05 21:38:24 * dos1 thinks he should have shr-testing on sd card ;) Mar 05 21:38:27 bye Mar 05 21:38:32 this is why i think you should do it via an interface, like the hjave example Mar 05 21:38:48 s/hjave/java Mar 05 21:39:23 toggles_w: but the interface to mock is not known in advance Mar 05 21:39:45 it is know by the mock framework user but not by the mock framework developer Mar 05 21:40:20 anyway this restrict my callbacks to as many callbacks as there are base types Mar 05 21:40:50 int int_generic_callback(void *this, ...) Mar 05 21:41:02 char char_generic_callback(void *this, ...) Mar 05 21:41:03 ... Mar 05 21:41:14 you could do that Mar 05 21:41:15 void *ptr_generic_callback(void *this, ...) Mar 05 21:41:39 which is better than an infinite number of callback possibilities :D Mar 05 21:41:40 or 'this' could have a way to stor the result Mar 05 21:41:41 ptitjes: btw, i've looked up in Harbison & Steele. Seems like what you do is not really prohibited by the standard. Mar 05 21:42:17 so your last thing to do would be ((MockInterface *)this)->stor_result(your_result); Mar 05 21:42:22 PaulFertser: nice! could you point me to where you read that ? Mar 05 21:42:45 toggles_w: yes. but this is not thread safe :( Mar 05 21:43:01 ptitjes: why not? Mar 05 21:43:20 dont do it liek ctime() ;-) Mar 05 21:43:21 toggles_w: what if multiple methods are called on the same object at the same time ? Mar 05 21:43:38 ptitjes: I've spent considerable time trying to freely download this book in a machine-readable form and failed. It's probably the most comprehensive book covering all standards as well as pre-standard C and C++ compatibility. Mar 05 21:43:52 stored results will be overwritten Mar 05 21:44:21 PaulFertser: ok! so thanks to have checked this for me. I'm really glad you did. Mar 05 21:44:27 aw, you didn't let me implement the method ;-) Mar 05 21:45:12 ptitjes: but don't count on it. It's not allowed either. The book doesn't say anything about calls like want. Mar 05 21:46:32 PaulFertser: did not understand ? Mar 05 21:46:43 s/?// Mar 05 21:46:44 ptitjes meant: PaulFertser: did not understand Mar 05 21:49:57 ptitjes: The book says that va_start and friends work with functions that are declared to expect ... as the last parameter. Probably you're not supposed to call such functions as a functions with fixed parameter list, but it's not written anywhere and it looks like any implementation should allow that (moreover i have a book translated to russian which may introduce some errors :( ) Mar 05 21:51:59 PaulFertser: ok :) Another question: Do they say if there is an api to get the two last pointer to code of the stack ? Mar 05 21:52:07 PC Mar 05 21:52:08 ptitjes: imagine an implementation that passes the parameters in registers for a fixed list and on stack for a variable list. Mar 05 21:52:29 arf Mar 05 21:52:32 too bad Mar 05 21:52:43 yes I would be fucked :) Mar 05 21:52:48 lol Mar 05 21:52:59 ptitjes: do you agree it's reasonable that compiler will generate registry-passing code for functions that expect only 2 ints, e.g.? Mar 05 21:53:13 yeah you are right Mar 05 21:53:25 but how libffi manages this for the calling side ? Mar 05 21:54:21 ptitjes: magic beans Mar 05 21:54:26 LOL Mar 05 21:55:22 ptitjes: libffi knows the parameter list at runtime, so it acts accordingly. Mar 05 21:55:38 so do I Mar 05 21:56:08 I know the exact function prototype thanks to Gobject-Introspection Mar 05 21:57:27 PaulFertser: it appears to me that using register passing is not portable, is it ? Mar 05 21:57:30 ptitjes: but you don't call the functions Mar 05 21:57:39 nope Mar 05 21:57:56 PaulFertser: no I simulate I am the function :) Mar 05 21:58:11 ptitjes: messing with registers and stack is not portable :) writing programs in C is portable ;) Mar 05 21:58:30 barely portable.. Mar 05 21:59:01 ok thanks for you help men! Mar 05 21:59:26 I will reseach some more about it and try to make an implementation of it Mar 05 21:59:39 I'll let you know if I have any succes Mar 05 21:59:45 ptitjes: but unless you use fastcall calling conventions you're probably safe to assume everything's on stack Mar 05 21:59:47 but for now movie time Mar 05 21:59:48 :) Mar 05 22:00:05 c u all soon Mar 05 22:00:09 ptitjes: looks like gcc on x86 defaults to cdecl, which looks like you want to :) Mar 05 22:00:13 ptitjes: good luck :) Mar 05 22:00:20 thanks Mar 05 22:03:43 And there's also floats that can be passed on a FPU pseudo-stack or smth.. Mar 05 22:14:46 Hi, today i discovered a problem with the gps subsystem. i got a fix but no gps time. anyone can confirm this? time was always 0 long/lat were ok. Mar 05 22:16:45 long/lat can't be fixed without time Mar 05 22:17:08 looks like just Gypsy don't inform your tangoGPS Mar 05 22:17:28 (actuall same for me with shr-20090218) Mar 05 22:18:48 yeah, i know Mar 05 22:18:59 when i get the data directly from fso with dbus its the same Mar 05 22:21:46 shoragan: ping Mar 05 22:22:05 what fso version does shr-testing use? Mar 05 23:13:24 Why do I get this bitbake git error? http://pastebin.com/d7de520b6 "NOTE: Creating tarball of git repository fatal: Not a valid object name 4335" Mar 05 23:17:50 bitbake me a pie Mar 05 23:24:53 (DTMF signalling) i must admit i have no idea. i never even tested this Mar 05 23:24:57 but i will in this very moment Mar 05 23:28:23 nothing Mar 05 23:28:35 i'm afraid you have to detect them by analyzing the frequency :D Mar 05 23:28:37 fun Mar 05 23:29:10 yup Mar 05 23:30:09 Frequency analysis party! Mar 05 23:31:52 nothing a tiny FFT can't fix Mar 05 23:31:58 it's a cpu-time hog :-/ Mar 05 23:32:10 eating power Mar 05 23:32:58 yeah, and IIRC we can only feed audio to loudspeaker _or_ process it, right? Mar 05 23:33:10 no Mar 05 23:33:16 no problem Mar 05 23:33:21 ah, then it's doable Mar 05 23:33:43 we can record both directions of a gsm-audio Mar 05 23:34:13 it's an issue with BT-headset Mar 05 23:35:24 k Mar 05 23:43:19 I just made opkg update and opkg upgrade on SHR unstable and now all application icons are gone Mar 05 23:44:15 bidirectional! Mar 06 00:08:38 freesmartphone.org: 03mickey 07framework * r607459401435 10/framework/subsystems/ogsmd/modems/abstract/mediator.py: ogsmd: override autoprefix for +CUSD, which is kinda special Mar 06 00:08:39 freesmartphone.org: 03mickey 07framework * r9bed3b234f94 10/framework/subsystems/ogsmd/gsm/channel.py: ogsmd: fix SMS sending with new parser. Thanks Daniel for spotting! Mar 06 00:08:40 freesmartphone.org: 03mickey 07framework * r0014180180a3 10/framework/subsystems/ogsmd/modems/abstract/mediator.py: Mar 06 00:08:40 freesmartphone.org: ogsmd: port +CNMA to the new timeout handling Mar 06 00:08:42 freesmartphone.org: (basically the per-command timeout no longer exists) Mar 06 00:55:31 freesmartphone.org: 03alain2210 07framework * rb445a328b91c 10/framework/subsystems/ogsmd/modems/freescale_neptune/mediator.py: Mar 06 00:55:31 freesmartphone.org: ogsmd: [FREESCALE NEPTUNE] add customized versions of DeviceSetAntennaPower and SimGetAuthStatus to cope with Mar 06 00:55:31 freesmartphone.org: the modem's GSM 07.07 violations. Mar 06 01:00:19 :o Mar 06 01:43:15 hm, does someone have moderator access to the openmoko-devel mailing list? I think a post of mine from a few days ago is still waiting Mar 06 01:44:02 devel@lists.openmoko.org, I mean Mar 06 01:45:29 probably too many CCs? Mar 06 01:45:59 roh is moderator of all ML Mar 06 01:46:07 maybe, just the CCs that were on the thread already though Mar 06 01:47:01 hmm, shouldn't block then Mar 06 01:47:24 just CCd to kernel and devel Mar 06 01:48:13 hmm crossposting is a strong indicator for spam Mar 06 01:49:22 yeah, maybe it got canned, it had a fair amount of of snmp log in it too Mar 06 02:04:09 hardware \o/ Mar 06 02:34:05 :O Mar 06 02:34:13 Glofish DX900!? Mar 06 02:35:31 Hmm... E-ten... ¬_¬ Mar 06 02:35:37 * Dave nudges Joerg Mar 06 02:50:20 hahaha WM7... **** ENDING LOGGING AT Fri Mar 06 02:59:57 2009