IRC log for #maemo-ssu on 20130104

02:03.45*** join/#maemo-ssu jonwil (~jonwil@27-33-137-199.static.tpgi.com.au)
02:15.37jonwilok, I think Gentoo box has decided to come back to life. And I have a solution for the problem being discussed last night by gregoa
02:15.41jonwilAt least I think I do
02:16.11gregoagreat!
02:27.18jonwilnow I just need to remember how to use git :)
02:43.36*** join/#maemo-ssu mickname_ (luodemm1@lyta.org.aalto.fi)
02:44.45*** join/#maemo-ssu ThreeM_ (~DonJohnso@cassiopeia.uberspace.de)
02:45.59jonwilok, gregoa, I should have a .deb file for you to test soon :)
02:46.48*** join/#maemo-ssu ruskie (ruskie@sourcemage/mage/ruskie)
02:46.51gregoajonwil: that sounds very promising, and I'm happy to test
02:51.27jonwilok, firstly I will test it myself and see what I get sent by the code in various ways and see if the fix does what I think it should on my end
02:51.32jonwilthen I will give you a .deb to test
02:51.51jonwilthen if it fixes your end, I will update the changelog and stuff and push to git (and once someone with access shows up) to cssu-devel :)
03:00.16jonwilgregoa, try this please
03:00.17jonwilhttp://www.cncmods.net/files/connui-home-cellular_3.2.1_armel.deb
03:00.23jonwilinstall that .deb file with dpkg -i
03:00.32jonwilthen reboot the phone
03:00.38jonwilthen send or pastebin opername.log
03:01.01jonwiloh wait delete opername.log before you reboot
03:01.07gregoajonwil: installing ...
03:01.10jonwilok
03:02.27jonwilif it works, great, I will figure out how to update Git :)
03:02.46jonwiland the changelog
03:02.49jonwiland stuff :)
03:02.53jonwiland make a 3.2.2 version :)
03:03.07jonwildoes it work?
03:03.21gregoa... rebooting ...
03:03.30jonwilok
03:05.25gregoastill 'A1' (on an almost empty screen). let's wait
03:05.35jonwilok, get me the opername.log file then
03:06.02*** join/#maemo-ssu user_ (~user@1.38.24.96)
03:06.07*** part/#maemo-ssu user_ (~user@1.38.24.96)
03:07.32gregoajonwil: http://paste.debian.net/221529/ ← here your are. sorry for not reporting better news, and thanks for all your efforts!
03:08.55jonwilok, the good news is that from that log, it IS picking up the name "bob"
03:09.05jonwilbut for some reason its then choosing to use the name "A1" anyway
03:09.15jonwilBack to the stock widget for a bit more reverse engineering I go :)
03:09.38gregoajonwil: :) thanks again!
03:15.08jonwilok, I am building another build, will test it myself then hand it to you if it works for me
03:32.28jonwilok, download this and try it please
03:32.30jonwilhttp://www.cncmods.net/files/connui-home-cellular_3.2.1_armel.deb
03:32.36jonwilsame thing abefore
03:32.38jonwilas before
03:32.43jonwilinstall it, remove opername.log, reboot
03:33.38jonwiland pastebin the opername.log you get so I can verify that its doing what I think it should be
03:34.41jonwilaqnything?
03:34.46jonwilanything?
03:40.09gregoajonwil: ... rebooting ...
03:40.20jonwilok
03:40.44jonwilso yeah assuming this does what I think it will, there should be one more build for you to install
03:41.48gregoasounds fine but I guess we have to defer to next round to tomorrow :)
03:42.00jonwildoes it work?
03:43.11gregoaATM I see no operator at all -- so yes there's some change :)
03:43.49gregoajonwil: yeah! 'bob' appeared!
03:43.52jonwilok, send me the opername.log
03:45.23gregoajonwil: https://paste.debian.net/221533/ ; looks like it works now! cool & thanks!
03:46.08jonwilok, so thats good
03:46.26jonwilI will update Git and make a 3.2.2 build for people to play with
03:47.18gregoajonwil: sounds good. and thanks again, I really appreciate your efforts in improving this stuff!
03:48.02jonwilok
03:48.06*** join/#maemo-ssu amiconn_ (amiconn@rockbox/developer/amiconn)
03:48.27jonwillets hope others using the widget get the intended results too
03:49.30gregoayup; kerio is a good candidate for testing, he previously also had problems with the op-name (on roaming)
03:51.10jonwilkerio: ping
04:03.45*** join/#maemo-ssu DocScrutinizer05 (~HaleBopp@openmoko/engineers/joerg)
04:43.31jonwilWho has access to upload new things to community-devel?
04:43.42jonwilI have a new version for http://maemo.merlin1991.at/cssu/community-devel/pool/free/c/connui-home-cellular/
04:43.44jonwil3.2.2
05:53.53*** join/#maemo-ssu luf (luf@nat/ibm/x-eoenslrjjkedpclg)
05:55.41lufjonwil: as I wrote I need also _armel.changes file.
05:58.24jonwilI posted all files here
05:58.25jonwilhttp://www.cncmods.net/files/connui-home-cellular/
05:58.38jonwilthat should be what is needed to handle upload to cssu-devel
06:00.25jonwilif there are any missing files, tell me and I will grab them off my box for you
06:01.02lufjonwil: now I see all files I need. I'll upload in 10 minutes
06:01.09jonwilok, thanks
06:11.45lufjonwil: new connui-home-cellular is now in cssu-devel
06:14.31*** join/#maemo-ssu LaoLang_cool (~LaoLang_c@221.226.175.141)
07:39.30kerioluf: will test
07:39.51freemangordonhmm, what is the big idea? priv->operator_state_name = g_strdup("NULL");
07:40.18freemangordonwhat kind of ugly hack is that?
07:41.47freemangordoninsttead of improving, it seems to me that code quality of cbs-widget(or whatever the name) is getting worse :(
07:46.57merlin1991I go to sleep early once and suddenly everyone needs somehting from me
07:47.12freemangordonmerlin1991: you sleep?
07:47.18merlin1991at times
07:47.28freemangordon:D
07:47.52freemangordonmerlin1991: zlib got stuck in cssu-devel
07:47.58freemangordonincoming folder
07:48.38merlin1991let me see
07:50.00merlin1991hates UDebs
07:50.00keriofreemangordon: so fic it!
07:50.02keriofix
07:50.16freemangordonkerio: no, i'll ask jonwil
07:50.46freemangordoni have tklock bug to fix
07:51.24freemangordonkerio: BTW I still cannot get it how you managed to do what you've reportd for tklock
07:51.39freemangordondid you check syslog for weird stuff?
07:52.01merlin1991freemangordon: repo is fixed
07:52.14freemangordonzlib is in or I have to reupload it?
07:52.39freemangordonmerlin1991: ^^^
07:52.44merlin1991in
07:52.51freemangordonok, thanks
07:53.03merlin1991btw if the package is right all you have todo is to touch the .changes file
07:53.11freemangordonI know
07:53.35merlin1991the error was, that the repo was not configured to include udebs
07:53.53freemangordonmerlin1991: BTW, in case you've missed it, I compiled microb with system zlib and libpng, works like a charm :). latest zlib and libpng ofc
07:53.54lufmerlin1991: sleep ... is it some new app for N900 from you? :D
07:54.25freemangordonluf: do we have BT ftp browsing already? :P
07:55.04luffreemangordon: sure all the time :D
07:55.09freemangordongreat :D
07:56.53merlin1991I still heaven't gotten a vmware image though
07:57.16merlin1991need to find out WHY microb fails to compile under my "normal" setup
07:57.36freemangordonmerlin1991: so if everything remains ok, we'll be able to close all bugs against libpng :)
07:57.49freemangordonmerlin1991: going to upload it, just tell me where
07:57.51merlin1991only that we can't close them ;)
07:58.01merlin1991freemangordon: drop it in your home
07:58.06freemangordonmerlin1991: (once the new update is out)
07:58.08freemangordonok
07:58.27merlin1991also it would be great if you could do a build of the version for stable for chemist
07:59.15freemangordonof what?
07:59.25merlin1991microb-engine ofc :)
07:59.37freemangordonhmm...
07:59.46freemangordonI need to setup SB first
08:00.23merlin1991err how did you build without having sb setup ?
08:00.23merlin1991o_O
08:00.59freemangordonI have packages from -testing installed in armel target ;)
08:01.11freemangordonwill have to setup a new target for stable
08:01.56merlin1991still wants to know WHY fmg can build and he can't
08:03.01keriomerlin1991: because he's better than you!
08:07.40freemangordonmerlin1991: SB image is copying in ~fmg/SB, will take a while though ;)
08:08.14freemangordonseems my upload speed sux
08:19.27*** join/#maemo-ssu luf (luf@nat/ibm/x-fsuzakaeickxkbgm)
08:29.34*** join/#maemo-ssu jonwil (~jonwil@27-33-137-199.static.tpgi.com.au)
08:35.01*** join/#maemo-ssu Martix (~martix@eduroam-244.fi.muni.cz)
08:36.50lufmerlin1991: what about the new git repo for osso-gnomevfs-extra ?
08:38.27*** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali)
08:46.44*** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali)
08:54.24merlin1991luf: coming right up
08:55.14jonwilI think with the changes I made earlier today, all the issues with connui-home-cellular should be resolved
08:55.41merlin1991did you push them to git? I'm interested what it was :D
08:55.48merlin1991luf: https://gitorious.org/community-ssu/osso-gnomevfs-extra
08:57.29jonwilyes all the changes I made to connui-home-cellular are in both git and the community-devel repo
08:59.17jonwilLets hope everyone gets the correct operator name now and there is no need for a 1.2.3 :)
08:59.33jonwilI mean 3.2.3
08:59.44merlin19911 2 3 would be way more awesome ;)
09:00.26jonwil:P
09:11.20*** join/#maemo-ssu dhbiker (~dhbiker@95.87.145.172)
09:13.11jonwilok, so now that I finished work on connui-home-cellular, what should I target next? :)
09:13.31jonwilMaybe I will go back to the investigation of clockd that I started yesterday
09:16.54jonwilor go back to looking at icd...
09:18.03*** join/#maemo-ssu zogg_ (~zoggrules@109.64.160.7)
09:22.43luffreemangordon: I pushed osso-gnomevfs-extra into cssu gitorious
09:41.08lufAgain N900 crash/reboot (this time during ssh login)
09:46.51lufyes the same reason as the last one ...
09:46.55lufpali: ping
09:47.51Paliluf, pong
09:48.21lufpali: kp52 kernel oops. how cen I report it (what is your preferred wy)?
09:48.43Paliluf, can you send me oops log?
09:48.57Palior do you know when it oopsing?
09:49.30Palijonwil, something usefull for RE could be USB widget (switch between PCSuite and Mass storage)
09:49.54PaliI really hate that idea that you need to unplug usb cable and plug it again for switching mode
09:50.02lufright now and few days ago (I upgraded kp to the latest from tmo)
09:50.40lufI used PCSuite and ssh from notebook to N900
09:50.49luf(pwd request)
09:51.37Palisame here, but for some reason I needed to switch to usb mass storage (ssh via usb network is not enought...)
09:52.01Paliand I see that it is possible to add mass storage mode to g_nokia driver
09:52.12Paliso no switch
09:52.34Palibut that change must be adopted to that stupid closed nokia widget...
09:52.49lufHowever http://pastebin.com/52Zg0GeZ
09:54.36Paliluf, use this tool for extracting oops from mtd partition: http://maemo.org/packages/view/sp-oops-extract/
09:54.53kerioit's preinstalled, isn't it?
09:54.56keriohm, maybe not
09:55.10lufpali: how to use?
09:55.40lufI see.
09:55.40kerioluf: sp-oops-extract /dev/mtd2ro
09:56.07Paliyou can also create dump of mtd, transfer to pc and extract via that tool on pc
09:56.19Pali(if you do not want to install it to n900 rootfs)
09:56.40lufhttp://pastebin.com/d3yExyfk
09:56.52jonwilI think the details of how the usb widget works are understood, its just a matter of someone with the UI skills cloning it
09:56.58jonwilI mean writing a new one/replacement
09:57.15jonwilIf there are unknown questions about how the widget works, those I can reverse engineer I guess :)
09:57.26lufYou can see all four or five crashes (the same backtrace for all)
09:57.31lufpali: ^^^
09:57.58jon_yPali: about your kp garage git repo, do you just dump the "debian" dir into the vanilla kernel source tree?
09:58.17*** join/#maemo-ssu dhbiker (~dhbiker@95.87.145.172)
09:58.18Palijonwil, I do not know what usb widget doing for switching to mode
09:58.36Paliit calling via sudo some shell script? or setting some HAL property?
09:58.49jonwilok, I will find out what its doing
09:58.52jon_y3g modem switch?
09:59.07jon_ydoes it involve 3G USB dongles?
09:59.38jon_ysorry, I picked up on the "switching to mode" part
09:59.45jonwilit use com.nokia.ke_recv dbus
10:00.07keriothe same as ke-recv-test p/m i think
10:00.13kerioapparently there's also a "charging mode" in ke-recv
10:00.16Palijon_y, see: http://talk.maemo.org/showthread.php?t=78371
10:00.20keriothat could be useful to switch back to
10:00.23Palisteps how to compile
10:00.57jon_yPali: OK
10:01.32jonwilit also uses HAL to watch USB
10:01.33jonwilor so
10:01.52jonwilorg/freedesktop/Hal/devices/usb_device_1d6b_2_musb_hdrc specifically
10:01.55Palijonwil, write that info to some file
10:02.08jonwildon't have anywhere to write it
10:02.13jonwilif wiki was working I would write it there
10:03.37jonwilIts also calling libhal_manager_find_device_string_match(blah,"button.type","usb.cable",blah) or so
10:04.22jonwilso basically its using libhal (open and documented) to monitor/deal with USB stuff
10:05.19jonwilthen ke_recv (also open) to actually switch modes
10:06.23Paliok
10:06.25jonwilits also doing /bin/ls /dev/ttyGS* somewhere for some reason
10:06.41Paliwhat is ttyGS*?
10:06.45jonwilno idea
10:06.47jonwilits also doing /usr/sbin/osso-usb-mass-storage-is-used.sh
10:06.53kerioPali: g_serial
10:07.11Paliah, yes, g_nokia have g_serial included
10:07.33Pali(and export here pnatd daemon for modem)
10:08.48jonwiloh and it does /bin/grep /proc/modules -e %s too
10:08.58jonwilwith g_nokia and g_file_storage as the values passed in
10:09.41jonwilseems like g_file_storage is module for mass storage
10:09.50jonwiland g_nokia is module for pcsuite
10:10.30jonwilalso reference to /tmp/.current_usb_mode
10:10.33kerioPali: is g_nokia open?
10:11.03Palikerio, yes, all kernel drivers are open
10:11.22kerioso why didn't you (or someone else) merge g_nokia and g_file_storage already?
10:11.38Palikerio, I wrote that I see that it is possible
10:11.51Palibut possible on 2.6.3x kernels...
10:11.58Palidriver g_nokia is modular
10:12.07Palibut driver g_filestorage is monolitic
10:12.12kerio:(
10:12.27Palig_filestorage was replaced in new kernel with g_masssttorage which is modular
10:12.31jonwilif it wasn't so hard to reverse engineer gtk apps with no x86 binary I would just reverse engineer entire USB widget
10:12.51jonwilbut I hope the info I have provided helps someone else come up with a new widget
10:12.53Paliso first we need to backport massstorage code to 2.6.28
10:13.00Paliand then add massstorage code to g_nokia
10:13.01kerioPali: i see
10:13.31Palijonwil, I will look at that widget
10:14.01PaliI wrote more 2 buttons widgets :-)
10:14.02jonwilok, so what should I look at next?
10:14.10Paliclockd :-)
10:14.15Paliand libtime
10:14.27jonwilok, maybe I will go back to clockd or icd stuff then :)
10:15.00Palijonwil, and I have git tree for mce where I tried to patch upstream meego mce for fremantle
10:15.00jonwilif I clone clockd, is it ok to completly ignore all the debugging and syslog stuff when cloning it? Would make things easier :)
10:15.18Paliif you want I can give you my tree and you can try mce :-)
10:15.23jonwili.e. pretend that the -d option to clockd doesn't exist
10:15.58Palijonwil, you can ignore debugging & syslog, but add some log to code
10:16.08Pali(e.g time is changing, ...)
10:16.26jonwilhow far did you get with mce?
10:16.41PaliI copied code from diablo mce to harmattan
10:16.45Palicode which was removed
10:16.59Palibut it is still in uncompilable state
10:17.08jonwilI think open MCE things are too far changed from Fremantle MCE to get something usable on Fremantle
10:17.11Palimissing systemui dev packages
10:17.37jonwilI tried before with various of the MCE plugins and got nowhere
10:17.47Palibut freemangordon RE some systemui parts, so then it I try it again
10:17.49jonwilreally need RE guru to handle MCE
10:17.53jonwilI am not RE gutu
10:17.55jonwilguru
10:17.57jonwilso I cant do MCE
10:18.00jonwilclockd I can do
10:18.11Paliok, look at clockd :-)
10:18.20jonwilicd (figuring out externals needed to replace it) I can also probably do
10:18.24jonwilbut yeah I will look into those later :)
10:18.50Palikerio, I found usb_eject function in ke-recv
10:19.09Paliso that function umounting mounted usb device (in host mode)
10:19.34Paliso need to add some button somewhere for calling it
10:23.23jonwilok, so I will see what I can do about clockd (I dont think we need libtime, just clockd)
10:23.31jonwiland also the icd2 stuff
10:24.22Palijonwil, libtime seems to be only dbus handler
10:24.37Palithis can be generated :-)
10:26.06*** join/#maemo-ssu lizardo (lizardo@nat/indt/x-ywdnpikaddlqislk)
10:26.37jonwilno libtime does more than just dbus
10:28.01Palijonwil, look also on: https://wiki.maemo.org/Fremantle_closed_packages/clockd https://wiki.maemo.org/Fremantle_closed_packages/libtime0
10:32.28jonwilthere is old maemo bug report asking for open libtime which was (of course) rejected
10:32.57jonwilbut yeah if I am cloning clockd, cloning libtime looks doable too
10:34.18jonwilhardest part with both is making build system/debian package info/etc, not actual code
10:35.57PaliI can create debian package without problem
10:36.29Palijonwil, just add files which needs to be installed to git
10:36.39jonwilwhat about builf system?
10:36.44Paliand create proper debian directory
10:36.48Palisimple Makefile
10:37.09Palior do you really need autotools?
10:37.28Palior write shell script which compile your binaries
10:38.39jonwilyeah makefie will do or now
10:38.42jonwilfor
10:41.23Palijonwil, clock synchronization from GPS, see https://bugs.maemo.org/show_bug.cgi?id=2750
10:41.25povbotBug 2750: Support clock synchronization by GPS
10:41.38Palifunny that it was fixed 2008-11-26
10:41.47Palibut for harmattan...
10:42.41Paliusign NTP is here: https://bugs.maemo.org/show_bug.cgi?id=472
10:42.43povbotBug 472: Support clock synchronization by NTP
10:42.56jonwilI will simply clone clockd as-is and get someone else to figure it out
10:43.05jonwili.e. do the rest of the work
10:43.24freemangordonjonwil: please, cleanup the cbs-widget code first
10:43.26Palijonwil, on my todo list is NTP support for clockd :-)
10:43.30freemangordonit is a mess now
10:44.15jonwilcleaning up code is not something I am any good at... :P
10:44.15freemangordonjonwil: also, why is priv->operator_state_name = g_strdup("NULL"); needed?
10:44.35freemangordonwhy not simply NULL?
10:44.50jonwilbecause you cant pass NULL to free or strcmp...
10:45.07jonwilif you think you can make cbs-widget code better please do
10:45.22freemangordonjonwil: I am sure I can make it better, but that is not the point
10:45.22jonwilbecause I dont think my skillset will allow me to do it :P
10:45.24tadzikis there a reason why NULL and not just ""?
10:45.57freemangordonit is madness to allocate memory just to make free and strcmp happy
10:46.14freemangordonjust do if(p) free(p)
10:46.38Paliyou can call free(NULL)
10:46.40freemangordonand there is g_strcmp (or similar) function dealing with NULL input parameters
10:46.45freemangordonthat one too
10:46.48Palifree(NULL) it doing nothing
10:47.03Palistrcmp must be non NULL
10:47.12Pali(no idea if g_strcmp can be)
10:47.29freemangordoniirc strncmp is dealing with NULL, but could be wrong
10:48.09Palithere is no info what happen if strdup get NULL
10:48.23tadzikCompares str1 and str2 like strcmp(). Handles NULL gracefully by sorting it before non-NULL strings. Comparing two NULL pointers returns 0.
10:48.27tadzikthat's g_strcmp0
10:48.33freemangordonyep
10:48.33tadzik(http://developer.gnome.org/glib/2.32/glib-String-Utility-Functions.html#g-strcmp0)
10:48.47keriofree(NULL) is specified, it MUST be the same as a no-op
10:49.00kerioin any compliant libc
10:49.24PaliC functions strcmp and strncmp cannot be called with NULL
10:49.24freemangordon:nod:
10:49.25tadzikthere's always g_free, which does exactly the same
10:49.41freemangordontadzik: but you should use g_strdup
10:49.50tadzikuse where?
10:50.00freemangordonoh, wait
10:50.31freemangordonit IS used
10:50.38freemangordong_free that is
10:50.50tadziksounds reasonable
10:50.50freemangordonhttps://gitorious.org/community-ssu/operator-name-cbs-widget/blobs/master/sbpluginsrc/operator-name-cbs-home-item.c#line361
10:51.44tadzikyeah, looking at the file it should be ok for it to be NULL, as long as it switches to g_strfoo stuff
10:51.56freemangordonyep
10:52.10Palijonwil, for clock/time here is full lists of bugs: https://bugs.maemo.org/buglist.cgi?query_format=advanced&component=Clock&product=Utilities&classification=Maemo%20Official%20Applications
10:52.48jonwilso yeah my plan right now is to produce 1:1 clone of clockd and libtime
10:53.01jonwilif anyone wants more than that, they can do it themselves :)
10:53.15jonwilsame if anyone wants to make cbs-widget better :P
10:53.29*** join/#maemo-ssu M4rtinK (~M4rtinK@mail.melf.eu)
10:53.36freemangordonjonwil: what? why you take it as offence?
10:54.03freemangordoncome on. :(
10:54.10jonwilI dont take it as offense, I just dont have the skills to do really good code
10:54.15jonwilmy code is always messy
10:54.35jonwilso if someone wants to clean it up, feel free but dont expect me to do it :)
10:54.58freemangordonjonwil: and it will always be like that if you don't try :P
10:55.14freemangordonhowever
11:13.04*** join/#maemo-ssu kolp (~quassel@212.255.17.14)
11:32.54*** join/#maemo-ssu dhbiker (~dhbiker@95.87.145.172)
11:59.51*** join/#maemo-ssu LaoLang_cool (~LaoLang_c@221.226.175.140)
12:07.53jonwilI prefer writing useful code than writing clean code or well-commented code :P
12:13.50*** join/#maemo-ssu LaoLang_cool (~LaoLang_c@221.226.175.140)
12:31.21keriocode that's not clean or well-commented isn't really useful for long
12:37.42*** join/#maemo-ssu xmlich02 (~imlich@2001:67c:1220:80c:d1:e86a:13b3:b849)
12:37.47*** join/#maemo-ssu NIN101 (~NIN@p5DD28CDE.dip0.t-ipconnect.de)
12:39.13*** join/#maemo-ssu anthony__ (~anthony@AOrleans-652-1-93-169.w92-140.abo.wanadoo.fr)
13:01.53*** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali)
13:03.38Paliif somebody has scratchbox loaded: need to build new version of packages from git: profiled, dsme, upstart
13:12.32*** join/#maemo-ssu kolp (~quassel@212.255.17.14)
13:22.09keriooh, zlib is in cssu-devel already, neat
13:22.50jonwilkerio: you were having problens with the cbs widget before, can you test if the newest version gives you any problems?
13:23.03keriojonwil: i've been testing the newest version since this morning
13:23.07kerioso far, no problems
13:24.00jonwilso thats version 3.2.2?
13:24.11kerioyep
13:24.17jonwilok, great
13:27.46keriomy problem has been fixed for a while though
13:27.50kerioand it was *weird*
13:28.17jonwilthe point was to make sure that the new code changes I made didn't case people who had problems before to have problems again
13:30.32lufkerio: also new libpng ...
13:30.51kerioluf: do those two have to match? i still have the thumby libpng
13:32.39lufkerio. No it's not necessarry. But with the newest version freemangordon also recompile browser to use system libpng and zlib instead of it's own.
13:33.01lufkerio: as usual I forgot the package name :D
13:40.40keriomicrob-engine probably
13:40.52keriohm, nope
13:41.06kerioat least, it's not in cssu-devel yet
13:42.06freemangordonyep, it is still not in -devel
13:42.26freemangordonkerio: BTW i'll compile thumby version of all this
13:42.28luffreemangordon: where is it? :)
13:42.31freemangordon*versions
13:42.37freemangordonluf: on my PC :P
13:43.16freemangordonremember, zlib was stuck in incoming last night ;)
13:46.04luffreemangordon: URL, IP + username + pwd? :D
13:52.22*** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali)
14:12.35jonwilonce the wiki is working again someone should go through http://wiki.maemo.org/Fremantle_closed_packages and annotate things that are worth cloning vs those that aren't (for example, the benefit to cloning clockd is much bigger than that of cloning, say, foreca-weather-applet)
14:14.26jonwilthat wiki page also will need updates for things that have been recently cloned like libcal
14:23.13*** join/#maemo-ssu Skry (~skry@77.109.215.222)
14:23.57jonwiland there are some third party items who's authors would rather we didn't clone them (I for one wouldn't feel comfortable being connected to a reverse-engineering of, say, telepathy-spirit :)
14:25.45freemangordonjonwil:  what's that responsoble of?
14:25.50freemangordontelephony?
14:27.00jonwilThats the skype plugin :)
14:27.24jonwiland I know the Skype people are very active in going after those who try to reverse engineer their protocols :)
14:32.27kerioi wouldn't feel comfortable using telepathy-spirit
14:32.34keriothat's why i uninstalled it
14:33.21keriosadly, i still have to keep the skype-ui-* shit
14:33.46jonwilI use Skype on my N900 (text only, no voice/video) to talk to a few people I know who are easiest to reach via Skype
14:34.28jonwilI figure that the Skype I already have on my N900 is a lot less likely to install all kinds of crap I dont want than the Windows version of Skype that I dont have on my desktop PC :)
14:36.59keriosrsly though, why does rtcom-notification-ui depend on skype-ui-l10n-mr0? :(
14:38.21jonwilbecause rtcom-notification-ui handles skype status display
14:38.39keriowhat's so fucking special about skype?
14:39.09jonwilit happens to be a very popular VoIP solution
14:39.11*** join/#maemo-ssu xes (~xes@host187-60-dynamic.2-87-r.retail.telecomitalia.it)
14:39.27kerioyeah but the actual skype plugin is *a plugin*
14:42.13freemangordonjonwil: skype people? last known about them is that M$ bought Skype. So you will RE M$ protocol :P
14:52.44*** join/#maemo-ssu kolp (~quassel@212.255.17.14)
15:24.36*** join/#maemo-ssu arcean (~Arcean@aadc19.neoplus.adsl.tpnet.pl)
15:54.35*** join/#maemo-ssu ekze (~nyan@bakaekze.ru)
17:23.00*** join/#maemo-ssu M4rtinK (~M4rtinK@mail.melf.eu)
17:58.39*** join/#maemo-ssu Woody14619 (~Woody@2620:4:4000:11:a8b6:b7:58ae:a08b)
17:58.39*** join/#maemo-ssu Woody14619 (~Woody@Maemo/community/contributor/Woody14619)
18:04.54*** join/#maemo-ssu toxaris (~toxaris@s83-180-246-172.cust.tele2.se)
18:21.23*** join/#maemo-ssu NIN101 (~NIN@p5DD29126.dip0.t-ipconnect.de)
18:21.56*** join/#maemo-ssu luf (~luf@ip-89-102-208-114.net.upcbroadband.cz)
18:27.34*** join/#maemo-ssu Martix (~martix@ip-62-245-106-78.net.upcbroadband.cz)
18:28.11*** join/#maemo-ssu wumpwoast (~jcassidy@63.251.248.156)
18:52.07*** join/#maemo-ssu luf (~luf@ip-89-102-208-114.net.upcbroadband.cz)
19:01.19*** join/#maemo-ssu M13 (~M13@170.133-224-87.telenet.ru)
19:20.59*** join/#maemo-ssu Sc0rpius (~naikel@190.79.197.57)
19:37.11*** join/#maemo-ssu nox- (noident@freebsd/developer/nox)
20:11.18luffreemangordon: I got it :) Now I'm able to read device properties ...
20:24.26*** join/#maemo-ssu Milhouse (~Milhouse@Maemo/community/contributor/Milhouse)
20:33.59*** join/#maemo-ssu Martix (~martix@ip-62-245-106-78.net.upcbroadband.cz)
20:35.04*** join/#maemo-ssu Martix_ (~martix@ip-62-245-106-78.net.upcbroadband.cz)
20:39.04*** join/#maemo-ssu zogg_ (~zoggrules@bzq-79-176-169-31.red.bezeqint.net)
20:42.34*** join/#maemo-ssu _rd (~rd@p57B491F1.dip0.t-ipconnect.de)
20:59.36*** join/#maemo-ssu _rd (~rd@p57B491F1.dip0.t-ipconnect.de)
21:15.32*** join/#maemo-ssu Martix (~martix@ip-62-245-106-78.net.upcbroadband.cz)
22:16.54lufCan anybody help me with C? I really have no idea what's wrong.
22:17.13lufuuid_filter[len] = '\0'; g_warning("get_dict_property: uuid_filter[len] = %c", uuid_filter[len]);
22:17.38lufI see in log: "get_dict_property: uuid_filter[len] = *"
22:30.38DocScrutinizer05what can I say, '\0' ?
22:30.53DocScrutinizer05you're sure this works as intended?
22:31.33lufDocScrutinizer05: yes. It usually works ;) I need to make the string shorter ;)
22:31.35DocScrutinizer05also printing a char(0) is errr
22:32.52DocScrutinizer05nfc how g_warning is defined, but i'd guess passing 0x00 to a %c parameter is no good idea
22:32.52lufIt also works in e.g. bluez: devname[15] = '\0';
22:33.58lufDocScrutinizer05: please don't elaborate on g_warning. The problem is that uuid_filter[len] = '\0'; does nothing.
22:34.44DocScrutinizer05what do you mean by "does nothing"?
22:35.09lufIt doesn't assign end of "string" to that position.
22:35.20DocScrutinizer05how do you know?
22:36.00DocScrutinizer05you're not printing a string, you're printing a %c har
22:36.06lufbecause g_warning I posted above and also next g_warning:
22:36.11lufg_warning("get_dict_property: UUIDs filter = %s, len = %d, char = %c, full = %d", uuid_filter, len, uuid_filter[len], full);
22:36.20lufget_dict_property: UUIDs filter = 00001106-*, len = 9, char = *, full = 0
22:37.07DocScrutinizer05I just can say printf("%c", 0) is a bad idea
22:37.37lufDocScrutinizer05: I don't want to have the first g_warning. I added it because the code doesn't work ...
22:37.59DocScrutinizer05then add something better than that
22:38.25lufWhat? I don't understand.
22:38.41DocScrutinizer05e.g g_warning("get_dict_property: uuid_filter[len] = %s", uuid_filter)
22:38.56DocScrutinizer05or g_warning("get_dict_property: uuid_filter[len] = %x", uuid_filter[len]);
22:41.46lufDocScrutinizer05: can you skip the %c? It doesn't work even without all g_warnings ... It doesn't work with your suggested g_warning("get_dict_property: uuid_filter[len] = %s", uuid_filter).
22:42.48DocScrutinizer05you pasted some code and asked what's wrong with it, I answered you shouldn't print char(0)
22:43.32lufThe problem is that it doesn't assign \0 to the uuid_filter at position len.
22:43.36DocScrutinizer05dunno what else I could say about that line of code
22:45.24lufIs there some "mode" when the char *uuid_filter = "00001106-*" can't be set using the way uuid_filter[0] = 'l' ?
22:45.59merlin1991luf: is uuid_filer filled by something or by a static string?
22:46.13lufstatic string
22:46.22merlin1991because char* whatever = "something"; whatever[5] = 'x'; does not work
22:46.31merlin1991because something is allocated in a read only block
22:46.41DocScrutinizer05yep
22:46.49DocScrutinizer05points to const area
22:46.59merlin1991what I meant :D
22:47.01lufif (!get_device_properties (conn, (const char*) remote_devname, devprops, "00001106-*")) {
22:47.20lufSo this is const initialization?
22:47.45lufstatic gboolean get_device_properties (Connection *conn, const char *dev, DeviceProperties *devprops, char *uuid_filter)
22:49.47merlin1991uuid_filter inside the function is a pointer to const memory if you cal lthe function like you do
22:49.51DocScrutinizer05anyway I'd thing you should get compile time warning
22:49.56DocScrutinizer05think*
22:52.17lufDocScrutinizer05: no warning :)
22:52.30lufThanks you (both of you).
22:52.38DocScrutinizer05yeah, compiler flags can do amazing things ;-D
22:53.41DocScrutinizer05-Wall ?
22:53.58lufyes
22:54.00DocScrutinizer05I really dunno shit about those flags
22:56.13merlin1991luf: did you fix it?
22:56.36lufI'm rewriting it a little bit. But I'm now sure what's the problem.
22:57.04merlin1991you could make the function axcept a const char* and just copy the data inside
22:57.41lufI can't change const char* (do you mean strdup?)
22:59.37lufI recognized if it's prefix or whole string (based on ending *). So I rewrite it that I add parameter like gboolean is_prefix.
23:01.30merlin1991yep strdup
23:06.21*** join/#maemo-ssu dhbiker (~dhbiker@193.2.218.150)
23:37.35lufIt's working now :)
23:53.01*** join/#maemo-ssu LaoLang_cool (~LaoLang_c@221.226.175.140)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.