| 00:00.31 | elephantum | pb_: ! look ! I created something that gpe could use as logo |
| 00:00.39 | elephantum | http://dezcom.mephi.ru/~el/gpe.png |
| 00:00.57 | pb_ | elephantum: ah, neat. |
| 00:01.04 | elephantum | you like it? |
| 00:01.05 | elephantum | =) |
| 00:01.24 | elephantum | I`m so proud of myself =) |
| 00:02.03 | pb_ | yeah, it looks kind of cool. |
| 00:02.15 | moray | I'd be kind of worried it might upset the GNOME folk, though. |
| 00:02.39 | elephantum | moray: why? dont we have much in common with gnome? |
| 00:03.27 | moray | elephantum: We're not officially part of the GNOME project, though, and I guess they'd want to keep some control over their logo. |
| 00:04.11 | elephantum | then if most of gpe team will like it, we could simply ask gnome team about it |
| 00:04.39 | pb_ | ibot, botmail for florian: I uploaded a new gpe-bluetooth which I hope will work a bit better. Looks like there was some glade-related screwup in the previous version. |
| 00:05.18 | pb_ | hm, I suppose florian_kc would have been a better target for that. oh well. |
| 00:07.03 | moray | elephantum: I *think* it's a legal trademark of theirs, so I doubt they'd want us using a derived version unless we were offically GNOME - even then, I suspect they'd rather we used the normal version rather than confuse people. |
| 00:08.10 | moray | elephantum: http://mail.gnome.org/archives/foundation-list/2003-November/msg00098.html |
| 00:13.28 | elephantum | then how about smth like this http://dezcom.mephi.ru/~el/gpe-nofoot.png |
| 00:13.28 | elephantum | ? |
| 00:13.29 | elephantum | =) |
| 00:18.43 | elephantum | ok... I see everyone is sleeping. so I should go to bed too. |
| 00:18.53 | elephantum | night everyone. |
| 00:18.56 | mallum_ | gnight |
| 00:19.20 | elephantum | mallum_: hello and goodbye =) |
| 00:19.31 | *** part/#gpe elephantum (~andy@dezcom.mephi.ru) |
| 00:57.48 | *** join/#gpe abhi (~guest@128.97.251.50) |
| 01:53.25 | *** join/#gpe Debolaz (~debolaz@rgbqi71.cm.chello.no) |
| 03:40.04 | *** join/#gpe wonid (~destroyer@200.b.001lg.syd.iprimus.net.au) |
| 05:33.13 | *** join/#gpe litage (~litage@Kingston-HSE-ppp3560733.sympatico.ca) |
| 05:33.14 | litage | is this wifi + 128mb CF card supported by Familiar?: http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=2980334480&category=43465 |
| 06:37.26 | *** join/#gpe omi007__ (~omi@ztex11.ztl.ch) |
| 07:21.00 | *** join/#gpe Toi|work (~pleemans@oia-gw1.be.jnj.com) |
| 09:56.35 | *** join/#gpe florian_kc (~fuchs@kc.celox-online.de) |
| 09:56.52 | pigeon | yo florian_kc! |
| 09:58.05 | *** join/#gpe fluxx (c99fln@peppar.cs.umu.se) |
| 09:58.15 | florian_kc | hi pigeon |
| 09:58.17 | florian_kc | hi all |
| 10:10.23 | paul | hi florian |
| 10:12.02 | florian_kc | paul: hi |
| 10:12.19 | *** join/#gpe wonid (~destroyer@077.g.001lg.syd.iprimus.net.au) |
| 10:20.42 | *** join/#gpe pb_ (~pb@dsl-62-3-66-204.zen.co.uk) |
| 10:23.25 | florian_kc | pb_: good morning! |
| 10:23.34 | pb_ | florian_kc: good morning! |
| 10:24.21 | pb_ | florian_kc: I was just trying gpe-package on my ipaq, but it won't start up. It appears that some pixmaps are missing. |
| 10:25.00 | pb_ | <PROTECTED> |
| 10:25.06 | pb_ | and |
| 10:25.12 | pb_ | <PROTECTED> |
| 10:25.34 | florian_kc | pb_: oh well... this seems to happen frequently in my apps... |
| 10:25.43 | florian_kc | let me look... |
| 10:28.42 | pb_ | florian_kc: thanks. did you see my botmail about gpe-bluetooth, by the way? |
| 10:30.08 | florian_kc | pb_: nope, no message... |
| 10:30.15 | florian_kc | ibot: botmail read |
| 10:30.30 | florian_kc | "<ibot> You have no botmail." |
| 10:30.42 | florian_kc | ah, wait.... stupid me! |
| 10:31.03 | pb_ | florian: ah, yes, sorry. I stupidly sent it to florian, not florian_kc. |
| 10:31.12 | florian | ibot: botmail read |
| 10:32.02 | florian | pb_: np, i could have been faster too :-) ok, great... i'll try later :-) |
| 11:00.20 | paul | pb_: morn pb |
| 11:00.32 | paul | pb_: i think the g_io assertion was due to encoding. |
| 11:01.11 | paul | pb_: at least, since i've set encoding to UTF-8 the assert has gone away. (the devhelp docs arent terribly clear on requirements) |
| 11:02.23 | paul | pb_: also, i'm getting the following message from the d-bus init script when trying to start it (after doing an ipkg upgrade earlier this morning): |
| 11:02.26 | paul | Starting system message bus: Failed to start message bus: Attribute "send" is invalid on <allow> element in this context |
| 11:02.39 | paul | i'm not sure which conf file thats due to. |
| 11:03.44 | paul | (system.conf presumably) |
| 11:22.02 | *** join/#gpe {HyGy} (~hygy@caracas-1005.adsl.interware.hu) |
| 11:22.03 | {HyGy} | hi |
| 11:22.43 | {HyGy} | re |
| 11:26.22 | florian_kc | hi {HyGy} |
| 11:26.43 | {HyGy} | florian_kc :) |
| 11:28.33 | *** join/#gpe mallum (~mallum@82-35-113-164.cable.ubr07.dals.blueyonder.co.uk) |
| 11:33.24 | florian_kc | hi mallum |
| 11:33.38 | mallum | hey florian_kc |
| 11:33.51 | *** join/#gpe Hallski (~micke@h34n1c1o1035.bredband.skanova.com) |
| 11:47.35 | *** join/#gpe PaxAnima (~petter@b134b.studby.ntnu.no) |
| 11:57.57 | paul | pb_: committed gpe-irc changes, want to kick out a new package to unstable? (or tell me/point me to docs on how to do it myself) |
| 11:59.05 | *** join/#gpe jakmap (~jakmap@hibernia.jakma.org) |
| 11:59.17 | pb_ | paul: ah, great. |
| 11:59.46 | pb_ | paul: check in /etc/dbus-1/system.d to see which file is to blame. |
| 12:00.37 | paul | pb_: hehe, how do i tell? |
| 12:01.40 | pb_ | paul: look for the one that contains a "send" attribute. |
| 12:02.07 | pb_ | "send_destination", "send_interface" and "send_path" are all ok; plain "send" is not. |
| 12:02.56 | paul | pb_: is there any good glib list around? i'm really curious about g_io_channel, specifically how one is supposed to close a channel down. |
| 12:03.01 | paul | p# grep send= * |
| 12:03.01 | paul | bluetooth.conf: <allow send="org.handhelds.gpe.bluez.pin-request"/> |
| 12:03.10 | paul | pb_: what should that be? |
| 12:03.21 | paul | path presumably. |
| 12:03.39 | pb_ | oh, that file should just be deleted. |
| 12:03.45 | pb_ | I wonder where it came from. |
| 12:03.54 | paul | pb_: i upgrade all the time. |
| 12:04.30 | pb_ | does ipkg think that it belongs to any package? |
| 12:06.40 | jakmap | pb_: no it doesnt. |
| 12:07.07 | jakmap | pb_: is hcid supposed to run as root or user? |
| 12:09.44 | jakmap | i get "GPE-ERROR: Inquiry failed:" from gpe-bluetooth when i try look for devices. |
| 12:09.58 | pb_ | jakmap: as root |
| 12:10.50 | paul | thats what i have it running as (from the hcid init script on startup). running gpe-bluetooth as root gives same error, so its probably not permissions. |
| 12:11.21 | paul | hmm.. gpe-irc doesnt v6, must fix that. |
| 12:11.38 | pb_ | try "hcitool scan", see what happens. |
| 12:11.52 | pb_ | ah, yes, that'd be good. I think gpe-irc is one of our few remaining v4 only applications. |
| 12:12.10 | jakmap | ah, no, mightnt be so, i specified irc.freenode.net. |
| 12:12.18 | jakmap | pb_: hcitool scan works. |
| 12:12.51 | pb_ | hrm |
| 12:13.12 | jakmap | hcitool inq paulj gives: <address> clock offset: 0x66b0 class: 0x520204 |
| 12:13.25 | pb_ | can you run gpe-bluetooth under strace and find out what's up? |
| 12:13.33 | jakmap | hcitool info OTOH gives 'Can't create connection: Input/output error' |
| 12:13.41 | jakmap | pb_: yes, will try. |
| 12:15.05 | jakmap | sdptool browse paulj -> connection timed out (is that normal?) |
| 12:15.53 | pb_ | Depends. Do you think you were paired with paulj? |
| 12:16.10 | paul | dont think so. |
| 12:16.18 | pb_ | In that case, yeah, probably normal. |
| 12:16.21 | jakmap | ah |
| 12:17.20 | pb_ | I think it varies a bit with the device, but iirc my 6310 won't respond to SDP unless you're paired. |
| 12:17.25 | jakmap | pb_: and a pairing is created with hcitool cc right? |
| 12:18.09 | pb_ | jakmap: generally it will prompt you to pair if you just try talking to the device. SDP should be enough to trigger that, actually. |
| 12:18.43 | pb_ | did you restart hcid since repairing dbus? It won't do PIN authentication properly if dbus wasn't running when it started up. |
| 12:19.10 | jakmap | should sdpd be running? |
| 12:19.26 | pb_ | yes, ideally. |
| 12:20.23 | jakmap | hmm.. after restart of sdpd/hcid gpe-bluetooth now sees it. |
| 12:20.37 | jakmap | pb_: what can i do with it now?here |
| 12:22.21 | jakmap | i dont at any stage get prompted for a pin. |
| 12:26.10 | pb_ | hm. |
| 12:42.54 | jakmap | pb_: what should hcid.conf say for pin_helper? |
| 12:45.08 | pb_ | nothing. it should have "dbus_pin_helper", though. |
| 12:45.23 | pb_ | If you have a pin_helper line, it needs to be deleted. |
| 12:45.33 | paul | pb_: it says dbus_pin_helper. |
| 12:45.34 | paul | hmm.. |
| 12:47.34 | pb_ | mm. well, that should be OK then. |
| 12:47.47 | pb_ | Maybe there's some problem with the dbus config files. |
| 12:48.46 | jakmap | possibly. |
| 12:49.14 | jakmap | i have only a bluez.conf |
| 12:49.54 | jakmap | which isnt supposed to be there either is it? |
| 12:51.45 | pb_ | yeah, you want that one. |
| 12:51.54 | pb_ | It might not contain the right stuff, though. |
| 12:52.13 | jakmap | allow own="org.handhelds.gpe.bluez" |
| 12:52.18 | jakmap | allow send_destination="org.handhelds.gpe.bluez" |
| 12:52.18 | pb_ | You could try running the dbus-daemon-1 with DBUS_VERBOSE=1 set in the environment, and watch for messages that suggest it's blocking things. |
| 12:52.23 | jakmap | <allow receive_sender="org.handhelds.gpe.bluez" |
| 12:52.26 | jakmap | allow send_path="/org/handhelds/gpe/bluez/PinRequest" |
| 12:52.35 | jakmap | is the policy |
| 12:52.55 | pb_ | hm, that doesn't sound quite right. I think the path should be .../bluez/PinAgent/PinRequest, or something like that. |
| 12:53.32 | paul | what ipkg should provide that file? |
| 12:53.38 | pb_ | gpe-bluetooth, I think |
| 12:54.53 | jakmap | well, lets remove it, remove bluez.conf and reinstall gpe-bluetooth. |
| 12:55.26 | pb_ | It might well be wrong in the package. |
| 12:56.00 | paul | if what i pasted is wrong, then that must be the case. |
| 12:56.04 | paul | cause file is same. |
| 13:07.01 | paul | gpe-bluetooth/dbus.c has: #define REQUEST_NAME "org.handhelds.gpe.bluez.pin-request" |
| 14:01.24 | *** join/#gpe mallum (~mallum@82-35-113-164.cable.ubr07.dals.blueyonder.co.uk) |
| 14:12.54 | *** join/#gpe Disconnect (~dis@3ffe:bc0:91:0:203:6dff:fe1a:c5b6) |
| 14:15.36 | *** join/#gpe {HyGy} (~hygy@caracas-1005.adsl.interware.hu) [NETSPLIT VICTIM] |
| 14:15.37 | *** join/#gpe herzi (~herzi@kiwi.mediascape.de) [NETSPLIT VICTIM] |
| 14:15.37 | *** join/#gpe sabrewulf (~mason@fennec.stairclub.com) [NETSPLIT VICTIM] |
| 14:15.58 | *** join/#gpe Hallski (~micke@h34n1c1o1035.bredband.skanova.com) [NETSPLIT VICTIM] |
| 14:18.20 | *** join/#gpe PaxAnima (~petter@b134b.studby.ntnu.no) [NETSPLIT VICTIM] |
| 14:18.21 | *** join/#gpe fluxx (c99fln@peppar.cs.umu.se) [NETSPLIT VICTIM] |
| 14:18.21 | *** join/#gpe Toi|work (~pleemans@oia-gw1.be.jnj.com) [NETSPLIT VICTIM] |
| 14:18.21 | *** join/#gpe pigeon (~pigeon@adsl-60-232.swiftdsl.com.au) [NETSPLIT VICTIM] |
| 14:18.21 | *** join/#gpe tigert (~tigert@mopo.tigert.net) [NETSPLIT VICTIM] |
| 14:18.21 | *** join/#gpe paul (~paul@hibernia.jakma.org) |
| 14:19.13 | *** join/#gpe jakmap (~jakmap@hibernia.jakma.org) [NETSPLIT VICTIM] |
| 14:19.13 | *** join/#gpe pb_ (~pb@dsl-62-3-66-204.zen.co.uk) [NETSPLIT VICTIM] |
| 14:19.14 | *** join/#gpe gdsx (xsdg@xsdg.staff.freenode) |
| 14:26.05 | *** join/#gpe gdsx (xsdg@xsdg.staff.freenode) [NETSPLIT VICTIM] |
| 14:26.06 | *** join/#gpe pb_ (~pb@dsl-62-3-66-204.zen.co.uk) [NETSPLIT VICTIM] |
| 14:26.06 | *** join/#gpe jakmap (~jakmap@hibernia.jakma.org) [NETSPLIT VICTIM] |
| 14:26.06 | *** join/#gpe paul (~paul@hibernia.jakma.org) |
| 14:26.06 | *** join/#gpe tigert (~tigert@mopo.tigert.net) [NETSPLIT VICTIM] |
| 14:26.06 | *** join/#gpe pigeon (~pigeon@adsl-60-232.swiftdsl.com.au) [NETSPLIT VICTIM] |
| 14:26.06 | *** join/#gpe Toi|work (~pleemans@oia-gw1.be.jnj.com) [NETSPLIT VICTIM] |
| 14:26.06 | *** join/#gpe fluxx (c99fln@peppar.cs.umu.se) [NETSPLIT VICTIM] |
| 14:26.06 | *** join/#gpe PaxAnima (~petter@b134b.studby.ntnu.no) [NETSPLIT VICTIM] |
| 14:26.06 | *** join/#gpe Hallski (~micke@h34n1c1o1035.bredband.skanova.com) [NETSPLIT VICTIM] |
| 14:26.06 | *** join/#gpe cmarqu_ (colin@buug.mind.de) |
| 14:26.07 | *** join/#gpe florian_kc (~fuchs@kc.celox-online.de) [NETSPLIT VICTIM] |
| 14:26.07 | *** join/#gpe Disconnect (~dis@3ffe:bc0:91:0:203:6dff:fe1a:c5b6) [NETSPLIT VICTIM] |
| 14:26.07 | *** join/#gpe {HyGy} (~hygy@caracas-1005.adsl.interware.hu) [NETSPLIT VICTIM] |
| 14:26.07 | *** join/#gpe herzi (~herzi@kiwi.mediascape.de) [NETSPLIT VICTIM] |
| 14:26.07 | *** join/#gpe sabrewulf (~mason@fennec.stairclub.com) [NETSPLIT VICTIM] |
| 14:33.14 | pb_ | paul: hm, that doesn't sound quite right either. |
| 14:33.19 | pb_ | I'll take a look later this afternoon. |
| 15:36.57 | *** join/#gpe wonid (~destroyer@052.a.002lg.syd.iprimus.net.au) |
| 15:45.17 | *** part/#gpe litage (~litage@Kingston-HSE-ppp3560733.sympatico.ca) |
| 15:47.38 | *** join/#gpe PaxAnima (~petter@b134b.studby.ntnu.no) |
| 16:15.23 | *** join/#gpe pattieja (~pattieja@63-252-5-1.ip.mcleodusa.net) |
| 16:32.26 | *** join/#gpe pb_ (~pb@cpc2-cmbg4-3-0-cust239.cmbg.cable.ntl.com) |
| 16:33.23 | pb_ | re |
| 16:33.51 | mallum | pb_: hi |
| 16:34.40 | pb_ | hi mallum |
| 16:46.09 | *** join/#gpe cmarqu (colin@buug.mind.de) |
| 17:08.29 | pb_ | gpe-bluetooth: relocation error: gpe-bluetooth: undefined symbol: sdp_uuid16_cmp |
| 17:08.30 | pb_ | Argh. |
| 17:09.42 | florian_kc | eh :-/ |
| 17:09.50 | pb_ | How did this happen? |
| 17:15.45 | paul | pb_: lp1 on fire? |
| 17:15.58 | pb_ | paul: heh |
| 17:16.15 | {HyGy} | bye |
| 17:19.21 | mallum | pb_: hi, so name I have a matchbox that has all the font rendering abstracted though libmb so in theory you only need to change the libmb to switch between core/xft/pango fonts |
| 17:19.51 | mallum | pb_: I should be able to provide 2 libmb packages which provide:libmb one just xft and one for pango right ? |
| 17:20.34 | pb_ | libmb1, but yes. |
| 17:22.09 | mallum | pb_: you do get a slight performance degration with Pango so Id only recommend if people want stuff like rtl rendering, clever font substitution etc |
| 17:22.34 | mallum | pb_: for all ltr languages xft should be fine ( all text is treated as utf8 now ) |
| 17:23.08 | pb_ | ok, sounds cool. Does the xft version do font selection for glyph coverage, or do we need Pango for that? |
| 17:24.54 | mallum | pb_: no you need pango for that or a big ass unicode font |
| 17:25.05 | pb_ | righto |
| 17:27.28 | pb_ | florian_kc: uh, looks like bluez-sdp is out of date in familiar. |
| 17:28.32 | florian_kc | Sounds like i'll have some fun testing the new stuff this evening :-) |
| 17:28.42 | pb_ | hehe, yes |
| 17:29.34 | paul | pb_: that have anything to do with my problems earlier? |
| 17:29.54 | pb_ | paul: yah, could be. |
| 17:30.11 | pb_ | paul: the symptom on my machine seems to be that gpe-bluetooth gets jammed at the "retrieving service information" dialog. |
| 17:30.13 | abhishek | is there a list of packages in the bootgpe images? |
| 17:30.31 | pb_ | abhishek: no, but you can refer to the logs or chase dependencies from gpe2 downwards. |
| 17:31.46 | florian_kc | I should work on gpe-package today... |
| 17:31.59 | pb_ | florian_kc: that'd be very cool |
| 17:32.17 | abhishek | pb_: ah, cool. thanks |
| 17:32.44 | paul | pb_: thats what i was seeing. |
| 17:32.57 | florian_kc | herzi: alive? |
| 17:34.31 | pb_ | paul: ah right |
| 17:34.36 | paul | pb_: if i look in logs there's stuff about hcid "pin_code_request: ..." and "hotplug: no runnable /etc/hotplug/bluetooth.agent is installed" a |
| 17:35.04 | pb_ | paul: I think that's all good. |
| 17:35.31 | pb_ | oh, except you said that you weren't actually seeing any PIN dialogs? that's not so good then. |
| 17:36.18 | paul | pb_: no i wasnt. |
| 17:36.43 | paul | pb_: tried setting that env var and running dbus, but got nothing, be it to stdout or syslog. |
| 17:37.02 | pb_ | strange, you should get all kinds of stuff on stdout. |
| 17:37.12 | pb_ | or stderr maybe, but the terminal in any case. |
| 17:38.11 | paul | pb_: what kind of stuff can gpe-bluetooth do if system is configured correctly? |
| 17:39.10 | pb_ | oh, looks like our dbus package is built with verbose mode disabled. whoops. |
| 17:39.36 | pb_ | paul: well, it will let you make network connections to access points. Jamey also wrote some code to do something with headsets, but I'm not sure what. |
| 17:39.49 | pb_ | it would be nice to have it support dialup: someone just needs to write a dialer for that to work. |
| 17:41.10 | abhishek | ah, i kept meening to ask about that |
| 17:42.35 | pb_ | dialup is basically the same as LAP style networking, except that you need a chat script to do the dial/login bits. |
| 17:44.20 | paul | pb_: ah headsets would be neat. |
| 17:44.47 | pb_ | yeah, I might try to buy a headset for myself to play with. |
| 17:44.51 | paul | pb_: colleague of mine did that actually, i have his mail somewhere. |
| 17:45.05 | paul | (former colleague even) |
| 17:45.14 | abhishek | pb_: how hard would it be to make the ipaq emulate a headset? |
| 17:45.15 | pb_ | paul: where are you working now, by the way? |
| 17:45.23 | pb_ | abhishek: I'm not sure. Probably not all that hard. |
| 17:45.56 | pb_ | abhishek: it depends a bit on the bluetooth chipset though. I think one of the ipaq models is not capable of that, but I forget which one. |
| 17:45.58 | paul | pb_: mostly from home. for a certain unix vendor whose name consists of 3 letters and starts with S. |
| 17:46.10 | pb_ | SCO! |
| 17:46.17 | paul | :) |
| 17:46.20 | paul | no no no :) |
| 17:46.22 | paul | Sun. |
| 17:46.22 | abhishek | pb_: lemme guess, the 5400:( |
| 17:46.23 | pb_ | heh |
| 17:46.53 | pb_ | abhishek: actually, I think the 5400 is ok and the 3900 is deficient. But I'm not certain. |
| 17:47.02 | abhishek | oh cool |
| 17:47.32 | abhishek | i was just going with the trend of the chipsets on the 5400 being lacking |
| 17:48.28 | pb_ | heh, right |
| 17:49.12 | abhishek | btw when are you leaving for ny? |
| 17:49.27 | pb_ | in about ten hour |
| 17:49.28 | pb_ | s |
| 17:50.07 | abhishek | cool |
| 17:50.23 | pb_ | I think my flight lands about 2pm tomorrow afternoon, local time. |
| 17:52.44 | abhishek | my flifht leaves 5 hours after that |
| 17:53.10 | pb_ | ah, heh. what time do you get in? |
| 17:53.43 | abhishek | around midnight |
| 17:53.50 | florian_kc | bbl |
| 17:54.07 | abhishek | miss setup day because of some stupid finals |
| 17:54.30 | pb_ | right, that's too bad. |
| 17:57.02 | abhishek | yeah, too bad they don't fly concordes coast to coast |
| 17:57.15 | pb_ | heh, yah. well, they don't fly concordes at all any more. |
| 17:57.15 | mallum | tada! http://handhelds.org/scap/port.20839.png |
| 17:57.29 | moray | aha |
| 17:57.33 | abhishek | i thought the last flight was sometime this year |
| 17:57.54 | moray | mallum: are the window decorations there a normal theme, or something special because this is the panel? |
| 17:58.01 | moray | (I'm hoping the former) |
| 17:58.35 | moray | abhishek: they had the last flights |
| 17:59.15 | abhishek | moray: oh, too bad |
| 17:59.42 | mallum | moray: its a theme Im working on that I've not released as yet |
| 18:00.53 | moray | mallum: ok - looks nice |
| 18:01.59 | mallum | moray: thanks, mb now supports putting panels in titlebar, thats why there is a clock in the corner |
| 18:02.36 | moray | mallum: all seems good |
| 18:03.08 | mallum | moray: gpe panel code need to support multiple trays I dont think it does, easy to add though |
| 18:04.02 | moray | right |
| 18:04.45 | mallum | this wm is scary : http://www.eyetap.org/research/medr/rwm.html |
| 18:08.12 | pb_ | paul: right, I uploaded the new bluez-sdp. This seems to fix the sdp_uuid16_cmp thing, but gpe-bluetooth is still hanging somewhere. |
| 18:09.01 | paul | will give it a try in a bit |
| 18:17.44 | mallum | pb_: whats the command to get xst to list all of its props ? |
| 18:18.18 | pb_ | xst read, I think |
| 18:18.58 | mallum | hmm not giving me anything |
| 18:19.18 | pb_ | That might just mean nothing is set. |
| 18:21.37 | mallum | pb_: yep your right |
| 18:45.24 | mallum | owww, I hate it when everything runs fine on my desktop, I then build for ipaq and it all falls to pieces :(( |
| 19:12.22 | *** join/#gpe DimiCh (~dimich@195.50.1.120) |
| 19:12.41 | DimiCh | re all |
| 19:12.56 | DimiCh | help plz |
| 19:13.30 | DimiCh | i h problemm with usbf |
| 19:14.37 | DimiCh | ifconfig usbf up returned no such device |
| 19:17.09 | *** join/#gpe florian (~fuchs@pD9E5F74D.dip.t-dialin.net) |
| 19:17.17 | florian | re |
| 19:18.08 | mallum | hey florian |
| 19:18.10 | pb_ | florian: re |
| 19:18.19 | mallum | florian: http://handhelds.org/scap/port.20839.png |
| 19:18.34 | mallum | florian: I'll upload it as soon as I fix annoying problems |
| 19:18.40 | mallum | florian: with mb |
| 19:21.57 | florian | mallum: wow, this looks good in two ways... which theme do you use there? |
| 19:25.35 | florian | pb_: I just updated to unstable... ipkg installed several gpe-todo language packages which is definitely not correct. |
| 19:27.38 | pb_ | florian: oh dear. |
| 19:30.20 | paul | where are the logs kept? |
| 19:30.48 | pb_ | which logs? |
| 19:31.00 | paul | (dont see any sign of them in /mnt/ramfs - but gpe-conf logread seems to be able to access them) |
| 19:31.04 | paul | pb_: syslog |
| 19:31.12 | florian | paul: In a memory buffer - use logread to display. |
| 19:31.30 | florian | ok - or gpe-conf :-) |
| 19:31.42 | florian | s/gpe-conf/gpe-info :-) |
| 19:31.44 | paul | cute :) |
| 19:33.24 | pb_ | florian: by the way, I uploaded a version of nsqld that can run from sshd. |
| 19:36.05 | pb_ | and also yet another gpe-bluetooth, which might fix the earlier problem I was having. |
| 19:36.23 | paul | pb_: ooh. |
| 19:37.14 | pb_ | and hadess made a new dbus config file for it too. |
| 19:37.32 | paul | yay |
| 19:37.50 | paul | how do you get ipkg to tell you what package are installed? |
| 19:38.00 | pb_ | ipkg status | grep ^Package |
| 19:43.37 | paul | is the second bank supported yet? |
| 19:44.05 | pb_ | as a separate filesystem, yes, but not combined with the first bank. |
| 19:45.15 | pb_ | or more accurately, linux can merge the banks but bootldr won't work if you do. |
| 19:45.58 | paul | ah |
| 19:48.26 | paul | hcid appears to be dying on me for some reason |
| 19:48.35 | pb_ | oh dear |
| 19:49.08 | florian | pb_: oh, this is nice... then we have a start for syncing now :-) |
| 19:49.29 | pb_ | florian: yep :-) |
| 19:49.56 | pb_ | florian: I will start work on a multisync plugin in the next few days. |
| 19:52.45 | florian | ah btw. did anyone ask the gnome people about their opinion concerning the gpe logo proposal? |
| 19:53.25 | pb_ | Not that I've heard of. That might be a good idea, though. I know Moray was concerned about it. |
| 19:53.34 | florian | brb, lunch |
| 19:54.03 | florian | maybe herzi will awake later... ;-) |
| 19:54.06 | pb_ | maybe :-) |
| 19:54.27 | paul | hcitool scan gives "Device not available..." hmm. |
| 20:01.05 | *** join/#gpe nilsf (~nils@kc.celox-online.de) |
| 20:01.15 | nilsf | Wup! |
| 20:01.27 | mallum | florian: its the Indutriasl gtk theme tweaked abit. will release soon |
| 20:02.17 | nilsf | pb_: I still seem to have the same problem with gpe-bluetooth-0.30, i.e. it hangs while "Scanning for services"... |
| 20:02.26 | pb_ | nilsf: Please try 0.31 which I just uploaded. |
| 20:02.33 | nilsf | pb_: :)) |
| 20:03.10 | nilsf | Oh, and I found a problem with ipkg ... seems it does not like large /usr/lib/ipkg/info/ directories. At some point it simply stops working and only emits SIGSEGV :( |
| 20:03.19 | pb_ | Oh dear, that's not good. |
| 20:03.28 | nilsf | After clenaing up the dir it works again but then info files are missing :(( |
| 20:03.46 | pb_ | I think there might be a debugging version of ipkg in the feed. Maybe you could try running this under gdb to locate the segv. |
| 20:04.49 | nilsf | Ouch :( |
| 20:13.00 | *** join/#gpe mallum_ (~mallum@82-35-113-164.cable.ubr07.dals.blueyonder.co.uk) |
| 20:13.00 | *** join/#gpe nilsf (~nils@kc.celox-online.de) |
| 20:13.00 | *** join/#gpe cmarqu (colin@buug.mind.de) |
| 20:13.00 | *** join/#gpe PaxAnima (~petter@b134b.studby.ntnu.no) |
| 20:13.00 | *** join/#gpe wonid (~destroyer@052.a.002lg.syd.iprimus.net.au) [NETSPLIT VICTIM] |
| 20:13.01 | *** join/#gpe sabrewulf (~mason@fennec.stairclub.com) [NETSPLIT VICTIM] |
| 20:13.01 | *** join/#gpe herzi (~herzi@kiwi.mediascape.de) [NETSPLIT VICTIM] |
| 20:13.01 | *** join/#gpe Disconnect (~dis@3ffe:bc0:91:0:203:6dff:fe1a:c5b6) |
| 20:13.01 | *** join/#gpe Hallski|away (~micke@h34n1c1o1035.bredband.skanova.com) [NETSPLIT VICTIM] |
| 20:13.01 | *** join/#gpe fluxx (c99fln@peppar.cs.umu.se) |
| 20:13.01 | *** join/#gpe Toi|work (~pleemans@oia-gw1.be.jnj.com) |
| 20:13.01 | *** join/#gpe pigeon (~pigeon@adsl-60-232.swiftdsl.com.au) |
| 20:13.01 | *** join/#gpe tigert (~tigert@mopo.tigert.net) [NETSPLIT VICTIM] |
| 20:13.02 | *** join/#gpe paul (~paul@hibernia.jakma.org) |
| 20:13.02 | *** join/#gpe gdsx (xsdg@xsdg.staff.freenode) [NETSPLIT VICTIM] |
| 20:14.46 | *** join/#gpe mallum_ (~mallum@82-35-113-164.cable.ubr07.dals.blueyonder.co.uk) [NETSPLIT VICTIM] |
| 20:14.46 | *** join/#gpe nilsf (~nils@kc.celox-online.de) [NETSPLIT VICTIM] |
| 20:14.46 | *** join/#gpe cmarqu (colin@buug.mind.de) |
| 20:14.46 | *** join/#gpe PaxAnima (~petter@b134b.studby.ntnu.no) [NETSPLIT VICTIM] |
| 20:14.46 | *** join/#gpe wonid (~destroyer@052.a.002lg.syd.iprimus.net.au) [NETSPLIT VICTIM] |
| 20:14.46 | *** join/#gpe gdsx (xsdg@xsdg.staff.freenode) [NETSPLIT VICTIM] |
| 20:14.46 | *** join/#gpe paul (~paul@hibernia.jakma.org) |
| 20:14.47 | *** join/#gpe tigert (~tigert@mopo.tigert.net) [NETSPLIT VICTIM] |
| 20:14.47 | *** join/#gpe pigeon (~pigeon@adsl-60-232.swiftdsl.com.au) [NETSPLIT VICTIM] |
| 20:14.47 | *** join/#gpe Toi|work (~pleemans@oia-gw1.be.jnj.com) [NETSPLIT VICTIM] |
| 20:14.47 | *** join/#gpe fluxx (c99fln@peppar.cs.umu.se) [NETSPLIT VICTIM] |
| 20:14.47 | *** join/#gpe Hallski|away (~micke@h34n1c1o1035.bredband.skanova.com) [NETSPLIT VICTIM] |
| 20:14.47 | *** join/#gpe Disconnect (~dis@3ffe:bc0:91:0:203:6dff:fe1a:c5b6) [NETSPLIT VICTIM] |
| 20:14.47 | *** join/#gpe herzi (~herzi@kiwi.mediascape.de) [NETSPLIT VICTIM] |
| 20:14.47 | *** join/#gpe sabrewulf (~mason@fennec.stairclub.com) [NETSPLIT VICTIM] |
| 20:16.32 | florian | re |
| 20:16.32 | florian | hi nilsf |
| 20:21.17 | nilsf | florian: Wup... |
| 20:22.12 | *** join/#gpe elephantum (~andy@dezcom.mephi.ru) |
| 20:22.29 | elephantum | good evening |
| 20:22.45 | nilsf | pb_: I have been staring (cluelessly) at the H3800 ASIC battery code this weekend :( Do you have an idea how this is meant to work, i.e. the general idea? |
| 20:22.49 | florian | hi elephantum |
| 20:23.10 | elephantum | florian: hi =) |
| 20:24.28 | pb_ | nilsf: I think I understand most of it, yes. Was there any particular part you were confused about, or just the general operation of the whole thing? |
| 20:24.30 | elephantum | did anyone do a conclusion from mail-list discussion of gui standards? I mean does it exist in wiki ? |
| 20:24.34 | pb_ | elephantum: good evening |
| 20:24.42 | elephantum | pb_: good evening =) |
| 20:25.20 | florian | elephantum: No, not yet - but i'll try to create an example... |
| 20:25.32 | florian | after making other stuff _usable_. |
| 20:25.35 | nilsf | pb_: For example is there any point of code where the charging process, i.e. puuting the current to the battery, is toggled? I find lots of led on and off but no clear point where it is decided if the battery is to be charged or not ... :( |
| 20:25.56 | elephantum | florian: that would be great if it will look like GNOME HIG |
| 20:26.02 | pb_ | nilsf: I think the charger is enabled whenever AC is present. |
| 20:26.43 | pb_ | nilsf: If you search for calls to set_charger(), that should find the right spots. |
| 20:26.57 | elephantum | florian: I like standardisations, so you can count on me =) though my english is not so good to create it myself |
| 20:27.05 | florian | elephantum: yes, this should be the aim... but there is quite a long way to make it that good. |
| 20:27.32 | elephantum | florian: draft document in wiki will be good beginnig |
| 20:28.31 | nilsf | pb_: mompls... |
| 20:28.35 | florian | elephantum: hehe, i'm not sure abmout mine - but so far noone told me to stop using english language ;-) |
| 20:29.12 | elephantum | florian: everyone is very polite I see =) |
| 20:30.16 | elephantum | oh! may be we could ask mallum to put his conclusion into wiki and then develop it |
| 20:30.32 | florian | :-) |
| 20:31.34 | elephantum | btw florian did you see my gpe logo proposal, though moray sais it voilates GNOME team copyrights, I like it. http://dezcom.mephi.ru/~el/gpe.png |
| 20:32.12 | mallum_ | elephantum: it would be nicer to come up with something a little more original |
| 20:32.41 | elephantum | mallum_: may be. It`s only an idea |
| 20:33.06 | nilsf | pb_: Back again... phone call... |
| 20:33.29 | nilsf | pb_: The problem was that I could not really find the point where it is decided that the battery is full or not. |
| 20:33.37 | nilsf | pb_: Or does this dallas chip decide that? |
| 20:33.45 | elephantum | mallum_: could you place your conclusion on interface standardisation to wiki? |
| 20:33.49 | florian | elephantum: yes, and i like it - but it would really be necessary to ask the gnome people about an opinion. |
| 20:34.11 | pb_ | nilsf: The kernel never actually turns the charger off. It reports the battery as "full" when the charging current drops to near zero, but the only effect is to stop the LED flashing. |
| 20:34.34 | mallum_ | elephantum: wiki sucks for things like that |
| 20:35.02 | elephantum | mallum_: why? we need something easy editable. |
| 20:35.03 | nilsf | pb_: Err... really? No wonder I couldn't figure this out - it's not there :) |
| 20:35.35 | pb_ | nilsf: :-) |
| 20:35.42 | mallum_ | elephantum: I dunno then |
| 20:35.50 | florian | elephantum: For a first opinion we could try to wake up herzi... |
| 20:36.19 | pb_ | nilsf: There are two places where it checks the charging current, depending on whether CHARGING_RATE_VIA_ADC is set or not. If not, it uses this block of code in h3600_asic_battery_update(): |
| 20:36.19 | pb_ | if (active_cur_to_ma(ds->cur,ds->resistor) < 10) |
| 20:36.20 | pb_ | h3600_asic_battery_set_led (CHARGING_STATE_FULL); |
| 20:36.20 | pb_ | else |
| 20:36.21 | pb_ | h3600_asic_battery_set_led (CHARGING_STATE_ACTIVE); |
| 20:36.22 | elephantum | florian: herzi ? |
| 20:36.38 | nilsf | pb_: But what does set_charger() do then? The LED is toggled somewhere else... |
| 20:36.43 | elephantum | mallum_: I think we should try, or this would die eventually |
| 20:37.02 | florian | elephantum: He is a gnome guy and i see he is in this channell... |
| 20:37.04 | pb_ | nilsf: set_charger() is called when the AC supply is either connected or disconnected. It controls the GPIO bit which enables the charger. |
| 20:37.44 | nilsf | pb_: That code you posted is only executed if the ADC is used, which is currently disabled... |
| 20:38.03 | nilsf | pb_: Ah, so there is some GPIO controlling the charger! |
| 20:38.04 | elephantum | florian: aah... so des ka =) |
| 20:38.05 | mallum_ | elephantum: the ml thread has gone dead |
| 20:38.06 | pb_ | nilsf: no, that's for the non-ADC path. The ADC path uses code in h3600_asic_battery_probe_task_handler. |
| 20:38.26 | nilsf | pb_: Sorry, overlooked the "!"... |
| 20:38.27 | pb_ | nilsf: Yes. The implementation of set_charger() is in battery_h3x00.c. It's a very simple function. |
| 20:38.43 | elephantum | mallum_: because it`s difficult to discuss without viewable examples |
| 20:38.53 | nilsf | pb_: #define set_charger(_x) (g_battery_ops->set_charger(_x)) |
| 20:39.11 | mallum_ | elephantum: well mock some up then |
| 20:39.12 | nilsf | pb_: I love those macros expanding to macros expanding to some structs expnding to... |
| 20:39.13 | pb_ | nilsf: yup. that ends up at either h3800_set_charger() or h3900_set_charger(), both in battery_h3x00.c. |
| 20:39.30 | elephantum | mallum_: we should make a draft with examples of what we purpose, then some people, I think, would argue something about it |
| 20:39.43 | pb_ | (or h5400_set_charger(), which is in a separate file) |
| 20:39.52 | mallum_ | elephantum: someone will always argue about something. its impossible to keep everyone happy |
| 20:40.21 | elephantum | mallum_: I hear sadness in your voice |
| 20:40.23 | elephantum | ^_^ |
| 20:40.25 | mallum_ | elephantum: no one has moaned about florians summary - we should just go with that |
| 20:40.54 | elephantum | mallum_: ahh... sorry... =( I was sure it was your summary |
| 20:40.54 | nilsf | pb_: What does GPIO1_CH_TIMER mean? What kind of "timer" is this? |
| 20:41.05 | elephantum | florian: comenasai |
| 20:41.17 | nilsf | pb_: (OK, CH means charger... but why "timer"?) |
| 20:41.26 | pb_ | nilsf: As far as I know it's just a bit in a GPIO register. I'm not sure why it has "timer" in the name. |
| 20:41.54 | mallum_ | elephantum: if nobody posts anymore mails to florians summary, I'll post another mail saying right we go with this |
| 20:41.57 | elephantum | mallum_: sure we must do it with florian summary |
| 20:42.14 | pb_ | nilsf: that's certainly the situation on h3900, at any rate. |
| 20:42.14 | nilsf | pb_: OK, reason I ask is that I suspected that the charger might draw current when no AC is online and it is still enabled, current back flow... |
| 20:42.14 | elephantum | mallum_: yes. that`s good |
| 20:42.45 | mallum_ | elephantum: people have either said 'yeah I like this idea' or 'The current way sucks but so does your idea but I cant suggest anything better' |
| 20:42.58 | pb_ | nilsf: Indeed, yes. In theory at least, this should be taken care of by h3600_asic_battery_up, which is called from h3600_asic_ac_in_isr() when the AC status changes. |
| 20:43.37 | elephantum | mallum_: sorry to say, but it was just hard for me to read all thread. I think I almost support your idea. |
| 20:43.54 | nilsf | pb_: Yes, I know, but very obviously something goes wrong there. One indication for me is that the LED keeps flashing though the charging has long finished. Thus I suspect that the whole code path does nuts at times... |
| 20:44.40 | pb_ | nilsf: Yep, that's possible. |
| 20:44.58 | nilsf | pb_: :) At least one hint... |
| 20:45.05 | pb_ | It would be interesting to see the contents of /proc/asic/battery when it's in this state where the LED is flashing but it's apparently fully charged. |
| 20:45.20 | nilsf | pb_: But the whole ASIC code look very suspicious to me... it's like a bug bunch of spaghetti- |
| 20:45.27 | nilsf | pb_: :)) |
| 20:45.42 | nilsf | pb_: But that's like looking in the dark. |
| 20:45.55 | elephantum | pb_: how is your progress in multisync plugin? |
| 20:46.15 | nilsf | Did I hear multisync plugin!? |
| 20:46.16 | pb_ | elephantum: I haven't started on multisync itself yet, though I built a version of nsqld that can be used with ssh. |
| 20:46.20 | pb_ | nilsf: Yeah, it's a bit of a mess. |
| 20:46.35 | elephantum | pb_: hmm... explain pls |
| 20:46.51 | florian | pb_: Is this the dbus issue you talked about? |
| 20:46.54 | florian | pb_: Starting system message bus: Failed to start message bus: Attribute "send" is invalid on <allow> element in this context |
| 20:46.57 | pb_ | elephantum: well, this was just a way to avoid the SASL drama that I got bogged down on previously. |
| 20:47.00 | elephantum | pb_: using port forwarding? or using ssh authentication? |
| 20:47.08 | elephantum | pb_: yep |
| 20:47.11 | pb_ | florian: yes, you may need to rm /etc/dbus-1/system.d/bluetooth.conf, if you have one. |
| 20:47.20 | nilsf | florian: Just remove /etc/dbus/system.d/bluetooth.conf |
| 20:47.20 | elephantum | pb_: so what did you do? |
| 20:47.24 | nilsf | Ah :) |
| 20:47.30 | florian | okay :-) |
| 20:47.47 | elephantum | pb_: ah... I`ll better get sources and watch them myself |
| 20:48.00 | pb_ | elephantum: I just changed nsqld to use standard input and output, so you can do "ssh ipaq /usr/bin/nsqld" and then talk SQL at it. |
| 20:48.12 | florian | ahhh |
| 20:48.36 | pb_ | elephantum: obviously you need to have public key authentication set up, and your passphrase in ssh-agent, so it isn't ideal. But it will do for a start. |
| 20:48.43 | elephantum | pb_: oh. =) you`re paranoid =) |
| 20:49.43 | elephantum | pb_: I think it doesnt matter now, do we use auth or not. we should develop algorithms now |
| 20:49.43 | nilsf | pb_: Idea: Shouldn't it be possible to save the GPIO register to some address during the short "maintenance" wakeup in the kernel? Then we would have a chance to see the registers while "almost" sleeping... |
| 20:49.51 | pb_ | nilsf: Yep, that would be a good idea. |
| 20:50.37 | nilsf | Oh well, lot of things to try. |
| 20:50.56 | nilsf | Just a pity that this "old" H3870 iPAQ is out of date by now... |
| 20:51.22 | pb_ | Yes. But many people still have h3800 and h3900 series machines. |
| 20:51.58 | pb_ | florian: did your "ahhh" indicate that bluetooth has started to work? |
| 20:51.58 | nilsf | Surely, but they can only become less since about half a year :( |
| 20:52.09 | florian | weeee! gpe-bluetooth is working again! |
| 20:52.18 | pb_ | Aha! |
| 20:52.21 | nilsf | florian: What does work? |
| 20:52.48 | florian | nilsf: i started a pan connection... i logged in successfully :-) |
| 20:53.04 | elephantum | mallum_: btw, what about matchbox and composite? does it still use a lot of cpu? |
| 20:53.14 | nilsf | florian: Strange... it does not work for me :( Still hangs while querying services... |
| 20:53.47 | pb_ | nilsf: try dragging the dialog box around the screen and see if that un-hangs it. |
| 20:54.03 | florian | nilsf: 0.31? |
| 20:54.16 | nilsf | florian: Yep. |
| 20:54.48 | pb_ | nilsf: You also need the latest version of bluez-sdp. |
| 20:54.51 | mallum_ | elephantum: yes |
| 20:55.09 | nilsf | pb_: Ah, that might be a problem - wrong dependencies, hae? |
| 20:55.13 | nilsf | pb_: :) |
| 20:55.13 | mallum_ | elephantum: it will always use some cpu , comes with the territory |
| 20:55.26 | pb_ | nilsf: Yeah, sorry. I forgot to update the Depends:. |
| 20:55.35 | mallum_ | elephantum: until we get hw accel X servers on pda's |
| 20:56.00 | elephantum | mallum_: that`s bad. could you make matchbox ipk without composite? |
| 20:56.11 | mallum_ | elephantum: there is one in stable |
| 20:56.23 | elephantum | mallum_: ok... I`ll try it |
| 20:56.25 | pb_ | nilsf: It needs version 1.5-hh1. The older package (1.1-hhsomething, I think) is missing some symbols. |
| 20:56.58 | mallum_ | elephantum: the next matchbox ( or very soon one ) you'll be able to turn compositing off completely |
| 20:57.03 | florian | mallum_: Is it possible to influence this font size? http://handhelds.org/scap/port.751.png |
| 20:57.09 | mallum_ | elephantum: on/off on the fly |
| 20:57.25 | nilsf | pb_: Congrats! That was it! |
| 20:57.35 | pb_ | nilsf: Phew! :-) |
| 20:57.40 | elephantum | mallum_: and that will increase/reduce cpu usage on the fly? |
| 20:57.43 | nilsf | pb_: Strangely noone (i.e. no software) did complain... |
| 20:58.01 | mallum_ | florian: yeah should probably copy the gtk theme font ... you can put a new line in there btw |
| 20:58.12 | nilsf | pb_: PAN connect also worked! |
| 20:59.05 | pb_ | nilsf: Yeah. The effect of the old library was to crash one of gpe-bluetooth's threads with an unresolved symbol error, but you wouldn't see the message unless you ran gpe-bluetooth from the terminal. And the main thread (which controls the windows) continued to run. |
| 20:59.10 | pb_ | nilsf: Ah, excellent. |
| 20:59.25 | mallum_ | elephantum: yeah hopefully |
| 20:59.47 | mallum_ | elephantum: keithp says hes not really tested turning the compositing extension on/off though ;-) |
| 20:59.52 | florian | mallum_: That's a great idea for an easy solution... this will make looking this all consistent. |
| 21:00.11 | elephantum | mallum_: oh. that would be nice. I hope it`ll work =) |
| 21:00.45 | mallum_ | florian: yeah thats the plan :-) mb can now parse gtk font descriptions so its just a case of plugging it into xsettings |
| 21:00.56 | nilsf | pb_: Perfect! Thanks for now, gotta go now. CU tomorrow maybe! |
| 21:01.00 | nilsf | Bye! |
| 21:01.03 | *** part/#gpe nilsf (~nils@kc.celox-online.de) |
| 21:01.42 | florian | Oh yes, Nils will need to get up early tomorrow... he needs to be at office at 10:00 ;-) |
| 21:02.55 | pb_ | florian: hehe |
| 21:06.21 | pb_ | Maybe I should make a new bootstrap with the updated bluetooth bits. |
| 21:10.10 | florian | pb_: Wait, i could update gpe-package later... maybe i make it to work better and we can do a kind of field test then :-) |
| 21:17.31 | pb_ | florian: Ah, excellent idea. |
| 21:29.05 | elephantum | great! my girl came and said that it`s better for us to break it =( |
| 21:30.16 | elephantum | no apologises are accepted. sorry that I wrote it here. |
| 21:37.29 | florian | elephantum: oops :-( |
| 21:45.56 | florian | interface.c:66: braced-group within expression allowed only inside a function |
| 21:45.56 | florian | interface.c:66: initializer element is not constant |
| 21:46.05 | florian | oops |
| 21:46.21 | florian | Is there an easy way to handle menu translations? |
| 21:47.33 | pb_ | florian: I guess you can mark them for translation like any other string. |
| 21:47.46 | pb_ | oh, I see, are you trying to use an ItemFactory? |
| 21:48.33 | pb_ | in which case, maybe http://www.geocrawler.com/archives/3/488/2003/2/0/10340170/ will help. |
| 21:55.19 | florian | pb_: let me look... |
| 21:57.50 | florian | pb_: Well... N_ seems to be different from _... whow is N_ defined? |
| 21:58.29 | florian | s/whow/how |
| 21:58.42 | pb_ | florian: #define N_(x) x |
| 21:59.09 | florian | pb_: oh... ok |
| 21:59.11 | pb_ | florian: the N_() macro only exists to mark the strings so that xgettext can find them. It doesn't have any actual function in the code. |
| 22:05.22 | pb_ | Hm, I should make gpe-announce support the h5400 buzzer. |
| 22:05.46 | florian | pb_: OK, greaat |
| 22:07.16 | florian | pb_: oops... works, well... i need to find out what i'm doing here ;-) |
| 22:07.26 | pb_ | :-) |
| 22:11.17 | pb_ | Hm, freaky. My h3900 was just lying on its back on my desk, suspended. And all of a sudden, it woke up, all on its own. |
| 22:11.44 | pb_ | I wonder if this is a common problem. |
| 22:12.04 | florian | oops... secret acceleration sensors? ;-) |
| 22:12.06 | *** part/#gpe elephantum (~andy@dezcom.mephi.ru) |
| 22:12.30 | pb_ | Well, now it's gone back to sleep. I guess ipaq-sleep's timeout expired. |
| 22:12.51 | pb_ | Still, if this is happening often, I suppose it might explain the poor battery life that people experience. |
| 22:19.44 | pb_ | florian: btw, I just uploaded a new build of hostap-cs (for ipaqpxa only, at the moment) to unstable. |
| 22:20.42 | pb_ | hehe |
| 22:20.48 | pb_ | florian: that doesn't sound too good. |
| 22:21.31 | florian | pb_: It isn't - this day doesn't like me... i just hpe tomorrow will be better. |
| 22:21.54 | pb_ | Let's hope. |
| 22:21.54 | florian | s/hpe/hope |
| 22:22.38 | florian | Well... seems to be the same for elephantum - poor guy :-( |
| 22:23.29 | pb_ | Yes. :-( |
| 22:25.46 | florian | One light at the horizon: I have a working menu, at the correct üposition and an accelerator :-) |
| 22:25.54 | pb_ | Aha. |
| 22:26.02 | florian | and can't type anymore :-( |
| 22:30.51 | moray | florian: A guy on another channel always finishes his lines like this by accidentß |
| 22:30.52 | moray | ;) |
| 22:31.10 | florian | mallum: :-) |
| 22:31.23 | pb_ | florian: that was moray |
| 22:31.34 | florian | hehe |
| 22:31.58 | pb_ | heh |
| 22:40.44 | florian | We could need a kind of help system in gpe... |
| 22:41.17 | pb_ | What kind of thing do you have in mind? |
| 22:42.49 | moray | If the GNOME/KDE people agree on something, we could use that ;) |
| 22:43.05 | florian | Well... something i'd call "system" if i'd try to sell it... maybe a call in libgpewidget that opens a help window with some formated text. |
| 22:43.17 | moray | The GNOME system seems reasonable, but is perhaps overkill for our current purposes. |
| 22:43.31 | florian | *gg* |
| 22:44.00 | moray | florian: Can't we just use a web browser for displaying it? |
| 22:44.08 | *** join/#gpe wonid (~destroyer@059.c.002lg.syd.iprimus.net.au) |
| 22:44.09 | florian | The idea is to populate an help icon or menu. |
| 22:44.16 | moray | florian: Then you just need a script that calls the web browser with the appropriate page. |
| 22:44.29 | moray | florian: Sure, but if we're having menus you can just have something like in GNOME2 |
| 22:44.36 | moray | (a help menu, with a 'Contents' entry) |
| 22:44.49 | florian | Sounds like a good idea... |
| 22:45.32 | pb_ | Yup, sounds fine to me. |
| 22:46.07 | florian | The only problem i see is that some systems maybe don't want to have a browser installed. |
| 22:47.23 | florian | We could handle this with a libgpewidget call and change things without changing all apps... |
| 22:47.29 | moray | florian: Hm, I'd tend to feel that's their problem - but at worst we can presumably do something like display a plain-text version of the page |
| 22:47.39 | moray | (just by removing tags and displaying) |
| 22:48.10 | florian | moray: hmm... right, it should even fit into 16mb systems... |
| 22:48.29 | moray | florian: we may as well have a call, but I think it should just call some external script to display the help (which can check for a preferred display method) |
| 22:49.11 | florian | moray: yes, sounds fine - then we'd be very flexible |
| 22:49.41 | moray | florian: It would be worth checking what exact format e.g. the GNOME people are using for help - it may be we can use it / a simpler subset of it, rather than invent another new thing |
| 22:49.43 | florian | imho we have only one app coming with something like "help" and this is gpe-wlancfg |
| 22:52.04 | florian | moray: gnome uses XML help files |
| 22:57.26 | pb_ | right. I guess xml would be fine for us too. |
| 22:57.57 | florian | ohoh, and how to make dillo display this? |
| 22:58.33 | florian | hmm... cool idea - what about plucker? |
| 22:58.43 | pb_ | hm, yes, that could work. |
| 22:59.48 | florian | This is a common format and intendes do be used on mobile devices... |
| 23:01.32 | pb_ | right. |
| 23:04.45 | pb_ | florian: by the way, I tweaked the gpe-todo toolbar as we discussed the other day. I think it looks better now, though I'm still not entirely keen on chrisime's separators. |
| 23:06.48 | florian | pb_: hehe, i just started it - these seperators look strange, but the bar at all is much better than before imho. |
| 23:07.19 | mallum_ | the seperators look really bad in some themes |
| 23:08.01 | pb_ | yeah. I don't really understand what the rules are for where you're meant to put them. |
| 23:08.28 | pb_ | I think I might quietly back out that change while nobody is watching. |
| 23:09.05 | florian | :-) |
| 23:09.47 | pb_ | crumbs |
| 23:12.59 | florian | This C++ compiler is a little bit stricter... |
| 23:13.22 | pb_ | btw, space is getting tight again on 16MB machines. The last round of bootstraps came out at 14680064 bytes for h3600 and 14942208 for h3900. |
| 23:13.41 | pb_ | I guess we need to roll in jg's changes to shrink libx11. |
| 23:13.55 | florian | Oh this would be good! |
| 23:14.59 | pb_ | Yeah. I'm not quite sure what's involved, but I'll look into it. |
| 23:20.43 | pb_ | hm, looks like libxml2 is the only remaining library that isn't prelinkable. |
| 23:21.33 | pb_ | oh, and sqlite. |
| 23:21.53 | florian | :-/ |
| 23:22.02 | florian | I see why... |
| 23:22.33 | florian | I think someone shoud update the compiler in the GTK2 crosstoolchain.... |
| 23:22.41 | pb_ | Yeah. |
| 23:24.44 | pb_ | I guess we should freshen the gtk2 toolchain based on jamey's latest gcc 3.3.2. |
| 23:25.10 | mallum_ | pb_: you just build xlib with --disable-xcms --disable-xkb --disable-xcms |
| 23:25.24 | mallum_ | --disable-xlocale on that last one rather |
| 23:25.34 | pb_ | ok, sounds straightforward enough |
| 23:26.33 | florian | pb_: yes, this sounds good... maybe i can nils to do this :-) |
| 23:27.27 | mallum_ | we should try and get rid of the libxml dep |
| 23:28.08 | florian | who uses this? |
| 23:28.20 | mallum_ | it always seems a bit extravagent to me having 3 xml parsers on such a tiny system |
| 23:28.25 | pb_ | florian: figment, gpe-contacts, gpe-nmf, gstreamer, libgnomevfs2 and linphone. |
| 23:29.30 | mallum_ | I think I convinced someone to remove libxml from figment |
| 23:29.38 | florian | hmm nasty... ok who wuses the two others? |
| 23:29.52 | florian | mallum_: mallum? ;-) |
| 23:29.56 | mallum_ | what does gpe-contacts use it for ? |
| 23:30.05 | mallum_ | mallum is upstairs |
| 23:30.07 | pb_ | just reading the structure file. |
| 23:30.24 | pb_ | possibly writing it as well, but I have a feeling it does that by hand. |
| 23:31.21 | pb_ | gpe-nmf likewise for playlists, I think. |
| 23:32.01 | mallum_ | pb_: yeah looking at the source for gpe-contacts it looks like it could be converted over relatively painlessly to glib parser |
| 23:32.41 | pb_ | right |
| 23:41.41 | florian | is there an easy way to make one toolbar button stay on the right edge? |
| 23:46.32 | moray | florian: the GNOME HIG discourages that kind of thing - are you sure that's what you want? |
| 23:46.55 | moray | florian: (while the problem might be less on a small screen, users tend to miss e.g. help menus that are set off to the right) |
| 23:47.44 | *** join/#gpe Flu-Shot (name@66-215-160-185.riv-mres.charterpipeline.net) |
| 23:48.22 | florian | moray: hehe, maybe i should read it completely... i'd only say it looks better if it was on the same position for every app. |
| 23:48.58 | moray | florian: having it at the right-hand end of the existing buttons has the same visual effect, I think |
| 23:50.25 | florian | moray: ok, then i'll keep the current layout for this - another problem solved :-) |
| 23:51.03 | pb_ | florian: for gpe-todo I made a separate toolbar for the exit button, and packed it into an hbox with gtk_box_pack_end(). |
| 23:51.26 | florian | pb_: okay... another solution :-) |
| 23:52.01 | moray | as far as the exit buttons go, I'm tending to think that it's a problem with our WM themes if they're needed |
| 23:52.28 | moray | I think it's better to have an extra-large close button in the theme if needed, rather than duplicate it on the toolbar |
| 23:52.38 | pb_ | I dunno. I think most people have come to expect an explicit "quit" option in programs, nowadays. |
| 23:53.03 | mallum_ | I agree with moray |
| 23:53.04 | moray | pb_: they'll (potentially now) have the menu if they really want it |
| 23:53.15 | moray | pb_: I think X users may expect one, but not normal computer users ;) |
| 23:53.30 | moray | pb_: as in, because some X applications do the wrong thing if you use the WM's close button |
| 23:53.55 | moray | GPE applications will, of course, be models of behaviour in this respect, so don't need that workaround |
| 23:54.03 | florian | Well, i'm not sure about this... this exit button makes the applications look uniform. |
| 23:54.20 | moray | florian: using the WM close button makes them look even more uniform |
| 23:54.53 | florian | Additionally this wm button is a pain to use if you are not sitting on a table. |
| 23:55.12 | moray | florian: how's that, that's not 'because the current matchbox themes are like that'? |
| 23:55.27 | moray | (I'm not sure what you mean to start with there, actually.) |
| 23:55.59 | pb_ | moray: I quite often fold the titlebar away to reclaim more vertical space. |
| 23:56.48 | moray | pb_: maybe we need the close button to stay visible at that point, then? |
| 23:57.02 | florian | moray: If you add a larger icon to the titlebar you waste space... this is like adding an additional bar only for the exit button. |
| 23:57.23 | pb_ | moray: maybe, but it's hard to see how you could do that without obscuring part of the application's own window. |
| 23:57.50 | mallum_ | florian: much of the gui is a pain if your not sitting at a table, but thankfully for most app and most of the time users should be fairly static, except for something like a gps application |
| 23:59.31 | florian | mallum_: Yes, but no reason to make it worse... i'd prefer to make it easier to make more of the applications work better if you are really mobile. |
| 23:59.33 | moray | mallum_/pb_: do you think it is often useful to display the window's title in the first place? |
