--- Log opened Thu Jul 21 00:00:24 2011
--- Day changed Thu Jul 21 2011
00:00 -!- Netsplit over, joins: aspix
00:01 -!- KaZeR [~kazer@130.43.195.77.rev.sfr.net] has joined #navit
00:01 -!- mode/#navit [+o KaZeR] by ChanServ
00:12 -!- antenna1 [~ion@dslb-188-099-086-211.pools.arcor-ip.net] has quit [Quit: Leaving.]
00:22 -!- aspix [~Adium@186.65.213.129] has quit [Read error: Connection reset by peer]
00:22 -!- aspix1 [~Adium@186.65.213.129] has joined #navit
00:46 < pini> lindi-: just back from real life, and going to bed; will look at 634522 and 634918 tomorrow
00:47 < pini> well s/tomorrow/this evening/ :)
00:47 < lindi-> pini: hi, I didn't know you were on IRC :)
00:50 < lindi-> pini: patch for 634918 sent
00:56 < pini> lindi-: thanks
01:37 < lindi-> the crash looks like  http://iki.fi/lindi/navit/navit-crash1.png
02:08 -!- sera [~sera@174-32.62-81.cust.bluewin.ch] has joined #navit
02:11 -!- sera_ [~sera@32-169.62-81.cust.bluewin.ch] has quit [Ping timeout: 250 seconds]
03:04 -!- noradtux_ [~quassel@f054019143.adsl.alicedsl.de] has joined #navit
03:04 -!- noradtux [~quassel@g224050107.adsl.alicedsl.de] has quit [Ping timeout: 260 seconds]
03:28 -!- cp15` [hxybgn@p57B1C72B.dip0.t-ipconnect.de] has joined #navit
03:28 -!- cp15 [abkeei@p57B1CCDB.dip0.t-ipconnect.de] has quit [Disconnected by services]
03:28 -!- cp15` is now known as cp15
03:28 -!- mode/#navit [+o cp15] by ChanServ
03:32 -!- aspix1 [~Adium@186.65.213.129] has quit [Quit: Leaving.]
04:40 -!- laroche [~laroche@dslb-178-007-202-217.pools.arcor-ip.net] has quit [Ping timeout: 252 seconds]
04:54 -!- laroche [~laroche@dslb-092-074-007-241.pools.arcor-ip.net] has joined #navit
06:07 <@Horwitz> morgen
06:18 -!- korrosa [~netbook@81.253.14.11] has joined #navit
06:18 < korrosa> lindi-, thanks for updating the dbus wiki.
06:18 < korrosa> Can you also check and update the more detailed description of set_destination on that page: http://wiki.navit-project.org/index.php/Dbus#set_destination
06:19 < korrosa> (You'll notice there that it does mention a string:"comment" is needed, but doesn't show it in the examples.
06:19 < korrosa> And close)
06:19 < korrosa> In other news:
06:20 < korrosa> [TRAC] New comment "ME TOO" http://trac.navit-project.org/ticket/166 "Show the scale of the map on the screen"
06:20 < korrosa> [TRAC] Ticket closed http://trac.navit-project.org/ticket/902 "French translation enhancement"
06:23 < korrosa> [TRAC] New comment "Don't blame me" http://trac.navit-project.org/ticket/901#comment:2 "Tunnel extrapolation [PATCH]"
06:23 < korrosa> :)
06:23 -!- korrosa [~netbook@81.253.14.11] has quit [Quit: Leaving]
07:05 -!- chollya [~chollya@2a02:7200:0:7:92fb:a6ff:fe36:2aa4] has joined #navit
07:09 < chollya> Hi everyone
07:37 -!- tparcina [~tomo@cisco15.fesb.hr] has joined #navit
08:13 -!- z30 [~Z30@77.242.201.52] has quit [Ping timeout: 276 seconds]
08:14 -!- Zaba [~zaba@about/goats/billygoat/zaba] has quit [Ping timeout: 276 seconds]
08:19 -!- Zaba [~zaba@about/goats/billygoat/zaba] has joined #navit
08:30 -!- Horwitz [~mich1@baldrian.franken.de] has quit [Ping timeout: 240 seconds]
08:40 -!- antenna1 [~ion@dslb-178-007-000-148.pools.arcor-ip.net] has joined #navit
08:51 -!- goldfndr_ [~goldfndr@h-68-165-77-43.sttnwaho.dynamic.covad.net] has quit [Ping timeout: 246 seconds]
08:51 -!- goldfndr__ [~goldfndr@h-72-244-204-190.sttnwaho.dynamic.covad.net] has joined #navit
09:05 -!- woglinde [~heinold@g225145048.adsl.alicedsl.de] has joined #navit
09:15 < Number6_> Morning
09:16 < woglinde> hi number6
09:16 < woglinde> hm lets see if wince builded
09:17 < Number6_> It did
09:17 -!- Number6_ is now known as Number6
09:17 < woglinde> good
09:17 < woglinde> http://download.navit-project.org/logs/navit/wince/svn/navit-svn-4654.ok
09:17 < woglinde> args the wince package still includes .svn dirs
09:18 < woglinde> cp15 ping
09:18 < kenguest> Number6: but did the build wince?
09:18 < woglinde> kenguest follow the link
09:20 < Number6> kenguest: How comical of you
09:21 < woglinde> hm win32 still fails
09:21 < woglinde> http://download.navit-project.org/logs/navit/win32/svn/navit-svn-4654.failed
09:22 < Number6> woglinde: Win32 fails due to maptool
09:22 < woglinde> yes
09:22 < woglinde> http://bytes.com/topic/c/answers/884920-sbrk-function-windows
09:23 < kenguest> Number6: it is a rather cringeworthy platform name though.
09:23 < Number6> kenguest: Aye, aye. Sure tis all Windows Mobile, anyway
09:27 < Number6> wget http://dl.dropbox.com/u/2705524/rtuxdoc2.png -O - > /dev/dsp      // Look at the image before running the command
09:29 < woglinde> no
09:29 < woglinde> no viruses
09:31 < Number6> It's a PNG, being piped to your soundcard... I doubt it's even possible to get a virus doing that
09:34 < woglinde> o.O
09:42 < Number6> I take it your mind is blown!
09:50 -!- woglinde_ [~heinold@g225146126.adsl.alicedsl.de] has joined #navit
09:53 -!- woglinde [~heinold@g225145048.adsl.alicedsl.de] has quit [Ping timeout: 250 seconds]
10:04 < CIA-69> martin-s * r4655 /trunk/navit/navit/maptool/geom.c: Fix:maptool:Avoid integer overflow with point in polygon test
10:06 -!- Horwitz [~mich1@baldrian.franken.de] has joined #navit
10:06 -!- mode/#navit [+o Horwitz] by ChanServ
10:07 <@Horwitz> morgen
10:07 < Number6> Morgen, Horwitz
10:07 < woglinde_> hi horwitz
10:07 < woglinde_> longtime not seen
10:08 -!- KaZeR_W [~Z30@77.242.201.53] has joined #navit
10:11 <@Horwitz> hmm. I just had to turn off the computer for a short while.
10:11 < woglinde_> uhm?
10:12 <@Horwitz> I'm pretty much always there
10:16 -!- drlizau [~liz@billiau.net] has joined #navit
10:19 < CIA-69> martin-s * r4656 /trunk/navit/ (4 files in 2 dirs): Fix:maptool:Better check for sbrk
10:28 -!- Amorphous [jan@unaffiliated/amorphous] has quit [Read error: Connection reset by peer]
10:46 -!- Amorphous [jan@unaffiliated/amorphous] has joined #navit
12:02 -!- Prz_^ [~ju@231.109.103.84.rev.sfr.net] has joined #navit
12:26 -!- simon___ [c06dbe58@gateway/web/freenode/ip.192.109.190.88] has joined #navit
12:28 -!- berghmans [c06dbe58@gateway/web/freenode/ip.192.109.190.88] has joined #navit
12:29 < lindi-> any suggestions why navit crashes in http://iki.fi/lindi/navit/navit-crash1.png ?
12:30 < berghmans> Hello , wie ist der Dataflow innerhalb navit und die Auftragsteilung zwischen CPU und GPU für opengl ?
12:30 < lindi-> the crash does not occur if I build with -O0
12:31 -!- niekie [~niek@CAcert/Assurer/niekie] has joined #navit
12:32 -!- niekie_ [~niek@CAcert/Assurer/niekie] has quit [Read error: Connection reset by peer]
12:34 < woglinde_> lindi png's are bad
12:34 < woglinde_> berghmans english please
12:35 < woglinde_> berghmanns and navit is using librarys for opengl
12:39 < lindi-> woglinde_: bad?
12:39 < simon___> Hello,
12:39 < woglinde_> lindi and why did you open bookmarks.c
12:39 < woglinde_> lindi yes better use pastebin
12:39 < lindi-> woglinde_: the crash happened there
12:40 < woglinde_> for pasting the gdb backtrace
12:40 < lindi-> woglinde_: ah sure
12:40 < lindi-> woglinde_: I just wanted to show everything I see :)
12:40 < woglinde_> yes and I dont see everything
12:40 < lindi-> woglinde_: there's no easy way to pastebin all those parts in any sensible way
12:40 < woglinde_> because the windows overlaps
12:40 < lindi-> woglinde_: I can pastebin the whole backtrace of course
12:42 < lindi-> woglinde_: http://paste.debian.net/123646/
12:43 < woglinde_> hm at its only on arm right?
12:43 < lindi-> woglinde_: yes
12:43 < woglinde_> s/at/and
12:43 < lindi-> woglinde_: happens on two different ARM systems
12:43 < lindi-> woglinde_: does not happen on amd64
12:43 < lindi-> fully reproducible
12:43 < lindi-> woglinde_: needs -O2, -O0 does not make it occur
12:43 < simon___> Hello, we (berghmans and me) would like to use and adopt the opengl functionality of navit.. We achieved to let it run with "internal" gui and rendering with opengl on Ubuntu.. This runs, but fonts are rendered unreadable.. Calling cmake forbids some optioons, becuse libfreetype would not be found.. It is installed, but the findFreeType is maybe not so well configured..
12:43 < lindi-> woglinde_: is that memset quite right?
12:44 < lindi-> woglinde_: it's writing sizeof(int*) bytes to a memory area that contains integers, why?
12:44 < woglinde_> simon use autotools
12:44 < woglinde_> lindi mom
12:45 < woglinde_> lindi did you try to remove the destinations file?
12:45 < woglinde_> and try it again
12:45 < simon___> Well, besides this font rendering problem we are currently trying to understand the flow of the graphics (With a currently limited opengl/3d knowledge :( )..
12:45 < lindi-> woglinde_: yes it does not occur then I think
12:45 < woglinde_> simon___ why you need the opengl port?
12:46 < simon___> Is there a better alternative to having accelerated drawing?
12:46 < woglinde_> why you need accel drawing?
12:46 < simon___> We target embedded devices..
12:46 < berghmans> For rendering the maps to the surface by opengl do I use the graphics_opengl source. Which parts of the sources are executed by the CPU and which parts are executed by the GPU ?
12:46 < simon___> Our goal is to use anvit as a sort of reference navigation
12:47 < woglinde_> simon which class of embedded devices?
12:47 < woglinde_> even on openmoko and ce devices
12:47 < woglinde_> the rendering is fast
12:47 < simon___> Are there currently any issues with the current opengl implementation? is opengl es more mature?
12:48 -!- berghmans [c06dbe58@gateway/web/freenode/ip.192.109.190.88] has quit [Quit: Page closed]
12:48 < woglinde_> simon there is ongoing support for gles
12:48 < woglinde_> which is much more usefull for embedded devices
12:49 -!- gberghma [c06dbe58@gateway/web/freenode/ip.192.109.190.88] has joined #navit
12:49 < woglinde_> and the opengl port is a bit behing in codecleanes and fucntionality as gtk and qt
12:49 < lindi-> woglinde_: next thing I'd do is start reading the disassembly of that function and figuring out if the compiler did something wrong...
12:49 < woglinde_> lindi it should happen with internal gui too
12:50 < woglinde_> if you set a destination
12:50 < woglinde_> could you try it?
12:50 < woglinde_> hm I could try it on my ac100 too
12:51 < simon___> ok, thank you..
12:51 < simon___> so, we should start looking into the opengl es part? Is this meant with gles?
12:51 < woglinde_> okay
12:51 < woglinde_> phineas and ferb
12:51 < woglinde_> till later
12:52 < woglinde_> yes
12:52 < lindi-> woglinde_: I'm using internal gui
12:52 < woglinde_> opengl es == gles
12:52 < woglinde_> lindi the dbus call?
12:52 < lindi-> woglinde_: and I hit the problem first when I tried to set a destination
12:52 < woglinde_> okay till later now
12:52 < lindi-> woglinde_: I just used dbus-send to make it easier to reproduce
12:52 < woglinde_> ah okay
12:52 < simon___> more this opengl es 2 or earlier versions? they seem to be very different in implementing the rendering
12:52 < lindi-> woglinde_: I think I've seen this bug many times this year but it's always been difficult to reproduce
12:53 < lindi-> woglinde_: and I'm always somewhere in a hurry when it occurs :)
12:54 < lindi-> woglinde_: should that memset be memset(offsets, 0, sizeof(*offsets));?
12:55 <@cp15> simon, the opengl driver should work with normal opengl, opengles1 and opengles2
12:55 <@cp15> But with opengles there are currently some restrictions (like no filled polygons)
12:57 < simon___> So you use opengl rendering for all elements? (worst case, which i thin is not the case: a 2d Image is generated on the cpu, then passed to open gl, just to make the view transformation)
12:58 < simon___> what is currently the drawback of not being able to draw filled polygons? (sorry, maybe my questions show I'm not so deep into the problem yet..)
12:58 < simon___> Are they rendered by the CPU or not rendered at all? What elements are affected by this limitation?
13:01 <@cp15> Yes, we use opengl rendering for all elements
13:02 <@cp15> The problem is that glu is missing with opengles, requiring us to do the tesselation of polygons (which is not yet implemented)
13:03 <@cp15> Affected are roads which are wider than 1 pixel (which are nearly all), building outlines, wood, water and so on. This items will be displayed as wireframe currently
13:06 < simon___> ah, ok.. Do you know, if there is a current development for an opengl es 2.x adaption of glu?
13:09 <@cp15> No, I don't know... But the tesselation code of glu is available in source from sgi
13:13 < lindi-> woglinde_: my bug could be http://trac.navit-project.org/ticket/845
13:14 < woglinde_> simon___ there was some efforts to bring glu or glew to gles
13:14 < woglinde_> but they all didnt succeed
13:18 < lindi-> woglinde_: cp15: added all the information to http://trac.navit-project.org/ticket/845#comment:4
13:19 < woglinde_> thanks linid
13:20 < woglinde_> okay so its something in optimazations flags
13:20 < lindi-> either optimization bug or some other bug that only becomes visible when compiler optimizes
13:20 < lindi-> I'd still like ot hear what that memset is doing :)
13:27 -!- aspix [~Adium@186.65.199.7] has joined #navit
13:28 < CIA-69> martin-s * r4657 /trunk/navit/navit/maptool/osm.c: Fix:maptool:Better format for some strings
13:35 < simon___> I googled a little and found, that mesa has some eopengles2 support: http://www.mesa3d.org/opengles.html did you think about using this?
13:36 < simon___> or is this not suitable for embedded implementations?
13:39 <@cp15> That is just an opengles implementation, navit can use it, but it won't make a difference to any other opengles implementation
13:40 < woglinde_> simon mesa gles/gl is software based
13:40 < woglinde_> for most embedded devices
13:44 -!- skamster [~quassel@77-56-73-44.dclient.hispeed.ch] has joined #navit
13:46 -!- aspix [~Adium@186.65.199.7] has quit [Quit: Leaving.]
13:47 -!- drlizau [~liz@billiau.net] has quit [Remote host closed the connection]
13:48 < simon___> ok, so it would not reduce effort, but the framerate :( ..
13:49 < woglinde_> again
13:49 < woglinde_> navit is fast enough with softrendering even on slow devices
13:50 < gberghma> we want to have hardware rendering
13:51 < woglinde_> *sigh*
13:59 <@cp15> gberghma, simon___, may I ask what you are planning to do?
14:10 < gberghma> .
14:20 -!- woglinde [~heinold@g225073008.adsl.alicedsl.de] has joined #navit
14:20 -!- woglinde_ [~heinold@g225146126.adsl.alicedsl.de] has quit [Read error: Connection reset by peer]
14:22 -!- playya__ [~playya@unaffiliated/playya] has joined #navit
14:23 -!- tparcina [~tomo@cisco15.fesb.hr] has left #navit []
14:24 -!- AngusRF [~Angus@p5B083C29.dip.t-dialin.net] has joined #navit
14:26 -!- playya_ [~playya@unaffiliated/playya] has quit [Ping timeout: 240 seconds]
14:26 -!- gberghma [c06dbe58@gateway/web/freenode/ip.192.109.190.88] has quit [Quit: Page closed]
14:29 -!- __ge__ [c06dbe58@gateway/web/freenode/ip.192.109.190.88] has joined #navit
14:30 < simon___> Well,
14:31 < simon___> we target currently to implement navit on embedded linux systems. GPU is not fixed yet (Nvidia or OMAP/PowerVR) and do performance tests there
14:31 < woglinde> *sigh*
14:31 < simon___> Navit should not run standalone but with multitasking, so performance is very important, especcially CPU load
14:31 <@cp15> simon___, are you from germany?
14:31 < simon___> we need CPU for other tasks as well.
14:32 < simon___> yes
14:32 < simon___> Munich..
14:32 <@cp15> I guessed that from your domain name... Actually I am not that far from munich, so maybe we can even meet and discuss that in person
14:34 -!- dnjaramba [~rht@41.72.193.86] has joined #navit
14:36 < woglinde> .o(hm I could measure both on tegra2 and omap/power right now)
14:37 < simon___> well, this could be very helpful for us.
14:37 <@cp15> simon___, can you write me an email? cp15 (at) navit-project.org?
14:37 < simon___> Sorry, i cannot PM you, I am here with the webchat interface..
14:37 <@cp15> Yes, I guessed that already
14:38 < woglinde> hm webchat dont works with /msg?
14:38 < woglinde> there were always a way
14:50 < simon___> @woglinde: how do you run your tests? Do you measurec CPU/GPU load while scrolling? does there already exist some documentation how to do this?
15:11 < woglinde> simon no
15:12 < woglinde> I only said I could mesuare it because I have both hw
15:14 -!- dnjaramba [~rht@41.72.193.86] has quit [Read error: Connection reset by peer]
15:16 < simon___> So,
15:17 < simon___> is there a way to show the fps? How would you measure this?
15:17 < simon___> @cp15: got my mail..
15:27 <@cp15> simon___, yes
15:27 <@cp15> I will reply soon
15:31 < simon___> thank you..
16:06 -!- woglinde_ [~heinold@g225146141.adsl.alicedsl.de] has joined #navit
16:09 -!- woglinde [~heinold@g225073008.adsl.alicedsl.de] has quit [Ping timeout: 240 seconds]
16:28 -!- __ge__ [c06dbe58@gateway/web/freenode/ip.192.109.190.88] has quit [Ping timeout: 252 seconds]
16:28 -!- __ge__ [c06dbe58@gateway/web/freenode/ip.192.109.190.88] has joined #navit
16:34 -!- z30 [~Z30@77.242.201.52] has joined #navit
16:38 -!- KaZeR_W [~Z30@77.242.201.53] has quit [Ping timeout: 260 seconds]
16:42 -!- antenna1 [~ion@dslb-178-007-000-148.pools.arcor-ip.net] has quit [Quit: Leaving.]
17:11 -!- noradtux_tp [~wircer@f054019143.adsl.alicedsl.de] has joined #navit
17:17 < AngusRF> Is it a known bug that navit changes its language as soon as it connects to a gpsd?
17:21 -!- khetzal [~quetzal@2001:470:1f13:6e7:6ef0:49ff:fee6:8e93] has joined #navit
17:24 <@cp15> AngusRF, no
17:24 <@cp15> From which language to which one? What OS?
17:25 < woglinde_> re
17:30 < AngusRF> it's on arch linux and changes from German to English.
17:30 < woglinde_> arch linux again
17:30 < AngusRF> I think this started with the svn commit that enabled gpsd 2.96
17:31 < AngusRF> and I just updated to the latest revision and it still persists
17:32 < AngusRF> I also tried the stock navit.xml, but same behaviour
17:48 -!- simon___ [c06dbe58@gateway/web/freenode/ip.192.109.190.88] has quit [Quit: Page closed]
17:52 < woglinde_> AngusRF hm which behaviour again?
17:53 < AngusRF> When I start up navit with gpsd NOT running, it's German, as I want it to be. As soon as I start gpsd or gpsfake and Navit connects to it, it changes to English.
17:54 -!- skamster [~quassel@77-56-73-44.dclient.hispeed.ch] has quit [Remote host closed the connection]
17:54 < AngusRF> When I start Navit with gpsd (or gpsfake) running, it's English right away
17:54 < AngusRF> I'm just trying to downgrade to GPSd 2.95 and see if it's the same
17:55 <@cp15> AngusRF, I assume there is some code in libgps that sets LANG to en_US, since languages which use "," as decimal separator caused problems in the past
17:56 <@cp15> Have you compiled navit by yourself?
17:56 < AngusRF> yes and no. I'm using Arch Linux's build system
17:57 <@cp15> Are you capable of changing the code and doing a rebuild?
17:57 < AngusRF> I've never tried, but I suppose I am
17:58 <@cp15> Ok, then look for vehicle_gpsd.c
17:59 <@cp15> Search for dbg(0,"Trying to connect...")
17:59 < woglinde_> hm ah that looks the same behavoiur pini discovered
17:59 <@cp15> and add a line dbg(0,"LANG=%s\n",getenv("LANG")); after it
17:59 < AngusRF> ok, just one note: I now recompiled navit linking against gpsd 2.95 and the bug is gone
17:59 < AngusRF> so it's only there with 2.96
18:00 <@cp15> Ok, I will look what changed between them
18:00 < AngusRF> I'll try to find out if your suggestion helps with 2.96
18:01 <@cp15> No, it is just to find out where the problem is
18:01 < AngusRF> ot, what it says :)
18:01 < woglinde_> hm debian unstable is still at 2.95
18:03 <@cp15> Found the problem
18:03 < woglinde_> cp15 hm what is it?
18:04 < AngusRF> great
18:04 <@cp15> libgps is changing the locale but it seems it is doing it wrong and it can't restore its previous state
18:04 < woglinde_> ieehks
18:04  * woglinde_ curses raymond
18:06 -!- Benja1972 [~sergei@227.213.10.109.rev.sfr.net] has joined #navit
18:17 <@cp15> AngusRF, wanna report the bug to gpsd?
18:19 <@cp15> http://git.berlios.de/cgi-bin/gitweb.cgi?p=gpsd;a=blob;f=json.c;h=c1ed524ce8e070bdd9e61773ddd9a00e40361849;hb=HEAD#l604
18:19 <@cp15> In this function setlocale(LC_ALL, NULL) is called and the result is saved
18:20 <@cp15> then setlocale(LC_ALL,"C") is called to set the locale to "C"
18:20 -!- Prz_^ [~ju@231.109.103.84.rev.sfr.net] has quit [Remote host closed the connection]
18:20 <@cp15> later setlocale(LC_ALL, savedlocale) is called to restore the original locale
18:20 < AngusRF> I don't see why it doesn't work the way they do it
18:21 <@cp15> But the string returned from setlocale is only valid until the next call to setlocale, so its content is nonsense after setlocale(LC_ALL,"C") and the final setlocale doesn't work
18:21 < AngusRF> oh, so they'd have to copy the result string
18:24 <@cp15> Yes, but generally this is bad code. The locale will be changed for all threads, so if the user has multiple threads, other threads will have their locale changed for a short time
18:26 < woglinde_> ieehks
18:26 < AngusRF> okay I see. I also see they're very concerned about getting only useful bug reports. they offer a 13-step instruction on how to report bugs
18:27 < AngusRF> So it might be better if you wrote the bug report, if you have time? After all, I can more or less just copy what you have told me
18:27 <@cp15> And the setlocale call is expensive... It does several opens and mmaps
18:31 -!- __ge__ [c06dbe58@gateway/web/freenode/ip.192.109.190.88] has quit [Quit: Page closed]
18:32 <@cp15> Ok, I have to take back the last comment. When just changing to the "C" locale, the last locale will stay cached and can be re-activated cheaply
18:33 < pini> bzed: if you're around there, please have a look above about gpsd-2.96 :)
18:33 <@cp15> pini, known problem?
18:34 < pini> cp15: no, not for me ; but I know that locale handling is part of the changelog of gpsd-2.96 ; bzed is a gpsd dev; that's why I ping him
18:36 <@cp15> Ah, ok :-) Thanks for the info. Then we probably can save us the work to report a bug
18:36 < pini> will depends on bzed being there...
18:36 <@cp15> I am starting to wonder once again whether we should code gpsd access into navit
18:37 <@cp15> The incompatibilities gpsd introduces are only topped by qt
18:38 < lindi-> cp15: have you talked with esr?
18:38 <@cp15> A few times... He thinks it is ok the way it is
18:39 < lindi-> cp15: would be nice to see this discussion somewhere publicly, for example on foss-gps list
18:39 < lindi-> foxtrotgps also had some trouble initially
18:42 <@cp15> Well, I guess I could discuss to the end of my life there. gpsd is horrible
18:42  * woglinde_ agrees
18:42 < lindi-> cp15: yes but we can change that only through discussion :)
18:43 < woglinde_> lindi only change is to make your own gpsd
18:43 < lindi-> woglinde_: possibly
18:44 < lindi-> but the problems should be discussed
18:44 < woglinde_> lindi- as cp15 sayed it's waste of time
18:44 < woglinde_> with esr
18:44 <@cp15> Well, best choice for us would be to reimplement the gpsd access in navit
18:45 <@cp15> This would have several advantages:
18:45 < woglinde_> lindi- atleast he comes with his guns
18:45 < woglinde_> and forces you to believe hom
18:45 < woglinde_> him
18:45 < lindi-> woglinde_: yes yes but this needs to be public so that people understand the problem
18:45 < lindi-> woglinde_: ogpsd people wrote their own but did not tell others why they did not use gpsd
18:45 <@cp15> - API stability: The basics of gpsd access change quite rare, so we would have one gpsd module which would fit all gpsd versions
18:45 < lindi-> cp15: would using for example geoclue help?
18:46 < pini> cp15, woglinde_: I reported bugs to gpsd a couple of times and always found esr responsive
18:46 < woglinde_> pini he changes the api a7bi at will
18:46 < woglinde_> ups abi
18:46 < lindi-> woglinde_: but I do agree that gpsd is not very nice, especially on an embedded system with limited battery life
18:47 <@cp15> - Portability: Would work on android, wince, iphone... Just in case someone needs it
18:48 < lindi-> cp15: well those don't run gpsd anyway?
18:49 <@cp15> Not on itself, but you can access a gpsd over the internet and show the position
18:49 < lindi-> cp15: ok sure
18:50 <@cp15> And it would also help noradtux_ who has a webos tablet without gps to access the gps of his phone
18:50 < lindi-> cp15: libgps does not work on those architectures?
18:51 <@cp15> libgps even doesn't work well on linux, so I don't expect much from it on other platforms
18:52 < lindi-> cp15: well whatever you do, please don't hardcode stuff inside navit :) there are other gps applications that would like to use it too
18:53 <@cp15> Well, I guess we could add a libgps_tiny to the support directory so other projects can use it too
18:53 -!- Benja1972 [~sergei@227.213.10.109.rev.sfr.net] has quit [Remote host closed the connection]
18:56 < Number6> Evening
18:56 <@cp15> Good evening
18:56 < woglinde_> hi numnber6
18:56 < Number6> I've the Mac with me, still going through that damn iOS book!
18:56 < woglinde_> lol
18:58 < Number6> Memory management in Objective-C / iOS is interesting. Classes has a reference counter for "owners" of an object. If there are no owners, the class is destroyed
18:58 < woglinde_> yes
18:59 < woglinde_> pretty common gc concept
19:00 < woglinde_> hm good that native speakers making the same mistakes as me
19:00 < woglinde_> Classes has/have
19:00 < woglinde_> hm hm
19:00 < Number6> I'm Irish, we love to make mistakes. If only to annoy Korossa
19:01 < woglinde_> lol
19:04 -!- antenna1 [~ion@dslb-178-007-000-148.pools.arcor-ip.net] has joined #navit
19:06 -!- antenna1 [~ion@dslb-178-007-000-148.pools.arcor-ip.net] has left #navit []
19:30 -!- bremner [~bremner@yantan.tethera.net] has joined #navit
19:31 < bremner> lindi-: hi!
19:34 < bremner> I'm playing with navit on xmonad (a tiling WM) with the gtk qui. I'm stuck entering a destination.  I have a country/city/street/number entered, but now none of "Map/Bookmark/Destination" seems to do anything. I'm using svn 4494 from Debian
19:38 < lindi-> bremner: hi
19:38 < noradtux_> hi
19:38 < lindi-> bremner: did you watch my screencast on how to use navit?
19:38 < woglinde_> hi noradtux
19:39 < bremner> lindi-: no, I guess I should huh?
19:39 < lindi-> bremner: might help :)
19:39 < lindi-> bremner: it's in debconf-discuss
19:39 < bremner> yeah, that's where I got the bih map info
19:39 < lindi-> bremner:   http://lists.debconf.org/lurker/message/20110718.142313.8fe393d3.en.html
19:40 < lindi-> bremner: the UI is not the most intuitive. anyways, I use the internal ui, not gtk
19:40 < noradtux_> I have a GPX track of the two tunnels I tried with tunnel-extrapolation
19:40 < noradtux_> which didn't work for me ..
19:40 < bremner> lindi-: watching...
19:43 < lindi-> bremner: thanks to that email heard about this other routing program that we have in debian :)
19:46 < bzed> pini, cp15: please write a bug report, I'm completely out of spare time
19:47 < bzed> pini, cp15: the main reason for that setlocale stuff is to avoid issues with converting numbers from various locale dependent formats into other formats
19:48 < bremner> lindi-: oh, the problem seems to be it doesn't know the street number.
19:48 < lindi-> bremner: that's possible, is it a long street?
19:49 < bremner> Sarajevo, Zelenih beretki
19:49 < bremner> not long
19:49 < lindi-> bremner: well then  you can just not enter street number
19:50 < bremner> sure, except that is what I was trying to figure out.
19:51 -!- antenna1 [~ion@dslb-178-007-000-148.pools.arcor-ip.net] has joined #navit
19:52 < lindi-> heh
19:53 < antenna1> hey everybody
19:55 < noradtux_> hi antenna1
19:55 < woglinde_> hi antenna1
19:56 < antenna1> noradtux_ did you log your last route? Or did TE work somehow?
19:57 < noradtux_> nope, didn't work. but I have a gpx track
20:01 < antenna1> noradtux_ If you add "pipe:/home/ion/navit/navit/script/gps_emu4 &lt; /PATH/TO/FILE.txt" to source in <vehicle> you can replay your log and see whether te works at least theoretically
20:02 < antenna1> korossa: show yourself!
20:06 < noradtux_> ah, ok
20:07 < antenna1> lindi-: aren't you running navit on debian?
20:07 < lindi-> antenna1: yes
20:09 < antenna1> but not on a netbook, I presume?
20:10 < lindi-> antenna1: I use on openmoko and aspire one laptop
20:10 < woglinde_> hm you could convert gpx track to nmea
20:10 < woglinde_> and than use gpsfake
20:10 < woglinde_> to replay it too
20:11 -!- rj58 [~aap@a83-161-209-240.adsl.xs4all.nl] has joined #navit
20:18 < noradtux_> antenna1: I guess navit needs libgps to read those?
20:19 < antenna1> noradtux_ at least I have libgps-dev installed
20:20 < noradtux_> ok, I thought so
20:46 < noradtux_> it looks like maptool got official support for the tunnel-flag now. when will the map from maps.navit-project.org have that data?
20:47 < noradtux_> oh, btw, it compares the value of the tunnel-flag with 1. the tunnels I tested have tunnel=yes in OSM. could that be a problem?
20:48 < woglinde_> noradtux mostly the day after
20:48 < noradtux_> woglinde_: okay, so maptool is rebuild regularly, yes?
20:54 < woglinde_> yes
20:54 < woglinde_> think so
20:54 < woglinde_> but cp15 knows better
20:55 < noradtux_> I just checked, the tunnel seems to be recognized by maptool
20:55 < noradtux_> i'm off, bye
20:55 < woglinde_> ciao noradtux
20:56 -!- OwenLA7QZ [47e34853@gateway/web/freenode/ip.71.227.72.83] has joined #navit
20:59 -!- anonymous [~anonymous@f048137083.adsl.alicedsl.de] has joined #navit
21:00 -!- anonymous is now known as Guest32747
21:01 -!- noradtux_tp [~wircer@f054019143.adsl.alicedsl.de] has quit [Ping timeout: 264 seconds]
21:02 -!- Guest32747 [~anonymous@f048137083.adsl.alicedsl.de] has quit [Client Quit]
21:02 < OwenLA7QZ> Hi. Quick question for everyone. I've now got my Viewpad 7. I installed Navit from the android market on it, but removed it immediately as it does not allow installing on SD. Is there a newer Andoid version that will install on SD?
21:03 < woglinde_> try the one from svn
21:03 < woglinde_> but I may have other problems
21:03 -!- noradtux_tp [~wircer@f054019143.adsl.alicedsl.de] has joined #navit
21:03 < OwenLA7QZ> How do I install the one from svn on my Android Device?
21:03 < woglinde_> http://download.navit-project.org/navit/android_armv5te/svn/
21:04 -!- OwenLA7QZ_V7 [47e34853@gateway/web/freenode/ip.71.227.72.83] has joined #navit
21:09 -!- bob_ [81a2012a@gateway/web/freenode/ip.129.162.1.42] has joined #navit
21:10 -!- latouche [~Latouche@reverse-169.fdn.fr] has quit [Ping timeout: 246 seconds]
21:10 -!- bob_ [81a2012a@gateway/web/freenode/ip.129.162.1.42] has quit [Client Quit]
21:20 < OwenLA7QZ> Hi again. That didn't help. I downloaded navit-current.apk direct to the Tablet and installed it. It installs to the internal memory and can not be moved to SD. :(
21:21 < woglinde_> then use maps.navit-project.org directly
21:21 -!- latouche [~Latouche@reverse-169.fdn.fr] has joined #navit
21:22 < woglinde_> and read http://wiki.navit-project.org/index.php/Navit_on_Android
21:22 < woglinde_> hi latouche
21:25 < OwenLA7QZ> woglinde Can't find anything there on installing on SD card. I can't see any justification for applications insisting on being installed on the internal memory. On the Viewpad 7 with only 512 Mb that is painful. Doesn't matter how big my SD is if Apps insist on being installed on internal memory.I'll try Navit again when this is fixed.
21:25 < woglinde_> http://wiki.navit-project.org/index.php/Navit_on_Android#Manual_map_download_.5Bless_recommended.5D
21:29 -!- drlizau [~liz@billiau.net] has joined #navit
21:32 -!- KaZeR_W [~Z30@77.242.201.53] has joined #navit
21:34 < OwenLA7QZ> wogline, was that for me? Can't see any info on installing the App to SD there.
21:35 < woglinde_> ah the app
21:35 < woglinde_> I thought you mean the map
21:35 < woglinde_> app is only few mp
21:35 < woglinde_> ups mb
21:35 < woglinde_> dont know why you want it on sd
21:35 -!- z30 [~Z30@77.242.201.52] has quit [Ping timeout: 276 seconds]
21:37 < OwenLA7QZ> woglinde: The Viewsonic Viewpad 7 has only 512Mb of internal memory. A small part of that is actually available for installing apps and I have had to remove several useful apps for this reason. I have an 8Gb SD card on the Tablet with loads of space.
21:38 < OwenLA7QZ> Reading the market reviews for other apps that I have had to remove indicates that I'm not the only one with this problem by far.
21:39 < woglinde_> hm the pad7 suckz anyway
21:39 < woglinde_> sorry
21:39 < woglinde_> to slow
21:39 < woglinde_> slow gpu
21:39 < OwenLA7QZ> Most apps do allow for installing on SD. (I have Copilot, but they don't have maps for the Caribbean where I will be in a couple of weeks.)
21:40 < woglinde_> hm I wonder if you will be get satisfied with navit
21:40 < OwenLA7QZ> No, I don't find it slow. Works fine. My only real complaint is that it should have had at least 1 gig internal. Can't find anything else around that price with GPS and 3G.
21:41 < woglinde_> how much did you paid?
21:41 < OwenLA7QZ> $ 300
21:41 < OwenLA7QZ> Didn't want a 10" tablet as they are too big for my use.
21:41 < woglinde_> ah okay
21:41 < OwenLA7QZ> Bought it on Amazon.
21:42 -!- OwenLA7QZ_V7 [47e34853@gateway/web/freenode/ip.71.227.72.83] has quit [Ping timeout: 252 seconds]
21:42 < woglinde_> hm sorry I dont think we get it going this fast to sdcard installation
21:42 < woglinde_> I should look up whats necessary
21:43 < OwenLA7QZ> It may be wrong, but I read somewhere that the apps that don't install to SD are that way because the developer prevented it...?
21:44 < woglinde_> we are opensource
21:44 < woglinde_> why should we prevent it
21:44 < pini> Could someone please point me to the irc logs for this channel?
21:44 < Number6> pini: irclogs.navit.ie
21:44 < pini> Number6: thanks!
21:44 < OwenLA7QZ> I don't know. I thought it was weird. Can't imagine it's too difficult to allow SD installation.
21:46 < Number6> pini: The logs are created by NavitLogger, and it generates http://irclogs.navit.ie/stats/
21:46 < woglinde_> ah found it
21:46 < OwenLA7QZ> Must be a forum for Android devs somewhere. I'm sure they'd let you know.
21:46 < woglinde_> OwenLA7QZ are you familiar with compiling stuff your self?
21:47 < woglinde_> http://developer.android.com/guide/appendix/install-location.html
21:48 < pini> Number6: doh... just seen the link given by the topic; sorry for the noise :/
21:48 < Number6> pini: Who reads topics!? :-)
21:48 < OwenLA7QZ> Aha! I suggest compiling wth preferExternal.
21:49 < OwenLA7QZ> :)
21:49 < OwenLA7QZ> That would make me very happy!
21:50 < OwenLA7QZ> The first thing I do after downloading an app is move it to SD. If it already installed there by default I'm happy. If it won't allow me to move it, I usually delete the app immediately unless it is something absolutely mission critical. (Like the Norwegian keyboard)
21:50 < woglinde_> OwenLA7QZ wait we need to support lower api too
21:51 < OwenLA7QZ> ... or TuneIn Radio
21:55 -!- khetzal [~quetzal@2001:470:1f13:6e7:6ef0:49ff:fee6:8e93] has quit [Remote host closed the connection]
22:04 < pini> lindi-: ping
22:04 < lindi-> pini: pong
22:05 < pini> lindi-: would you mind try the latest package from git? it has commit b4befb27 which fix a valgrind error related to bookmarks.
22:06 < lindi-> sure
22:06 < lindi-> but we are already debugging gcc miscompilation
22:07 -!- rj58 [~aap@a83-161-209-240.adsl.xs4all.nl] has left #navit []
22:07 < pini> lindi-: will you have enough time to try it this evening? I'll be offline for the w-e...
22:08 < woglinde_> OwenLA7QZ I will try to build you a version with android:installLocation auto
22:08 < CIA-69> martin-s * r4658 /trunk/navit/ (8 files in 2 dirs): Add:Core:Initial work for graphics_cocoa
22:09 < Number6> `o/
22:09 < Number6> \o/ , rather.
22:09 <@cp15> noradtux_, no, the tunnel=1 isn't a problem, since tunnel=yes will be converted to tunnel=1
22:12 < lindi-> pini: I doubt that, need to sleep before debconf
22:12 < pini> lindi-: lucky you :)
22:13 < pini> will upload by next week then... the fix for 634918 is in the git repo anyway.
22:17 < lindi-> ah it's so small patch
22:18 < lindi-> pini: I guess I can build and run to see if it crashes in the same way
22:19 < pini> lindi-: ok; I'll be around 'til midnight (paris time)
22:20 < lindi-> pini: git-buildpackage -S should create source package which I can feed to debuild?
22:20 < pini> lindi-: never tried this way :/
22:21 < lindi-> pini: which way do you prefer?
22:21 < pini> lindi-: I use : DIST=sid git-buildpackage --git-builder="git-pbuilder" on my box
22:21 < pini> with cowbuilder
22:22 < woglinde_> pini do you have some patches open?
22:22 < woglinde_> pini I fixed libshape
22:22 < woglinde_> but you must make a cvs version of libshape
22:22 < woglinde_> so navit can use it
22:22 -!- AngusRF [~Angus@p5B083C29.dip.t-dialin.net] has quit [Quit: Verlassend]
22:22 < woglinde_> args okay
22:22 < woglinde_> I fixed it only for autotools
22:23 < pini> woglinde_: have a look at the git repo for the package; It's some weeks since I last work on it :/
22:23 < lindi-> pini: git-buildpackage -S produced navit_0.5.0~svn4494+dfsg.1-1.dsc as expected
22:23 < OwenLA7QZ> HI woglinde. Sorry I went missing. Phone call from wife. Thanks for your help. Have to go now. you can e-mail me on la7qz <at> hotmail dot co dot uk :)
22:23 < pini> lindi-: \o/
22:23 < lindi-> pini: I like debuild since it creates build logs :)
22:24 < pini> lindi-: as does g-bp
22:24 < lindi-> pini: ok
22:24 < lindi-> pini: I also like it since I can give these source packages to people who don't know my version control system of choice
22:25 -!- OwenLA7QZ [47e34853@gateway/web/freenode/ip.71.227.72.83] has quit [Quit: Page closed]
22:29 < lindi-> pini: FTBFS /home/lindi/scm/navit-0.5.0~svn4494+dfsg.1/navit/graphics/qt_qpainter/graphics_qt_qpainter.cpp:46:21: fatal error: qglobal.h: No such file or directory
22:29 < woglinde_> lindi- hm
22:30 < woglinde_> looks like you are missing qt-dev stuff
22:30 < pini> lindi-: yep... it's 634522 reported by lucas
22:30 < pini> I'm on it
22:30 < pini> something has changed in qt4
22:31 < lindi-> blah, I'll workaround since I want to get routing work :P
22:32 < woglinde_> pini on unstable I can build it without probs
22:33 < pini> woglinde_: strange... as 634522 was encountered using unstable
22:33 < lindi-> pini: magically the patch seems to avoid the miscompilation
22:33 < lindi-> pini: we still think it is a gcc bug
22:34 < lindi-> pini: I'm still waiting for real test (we've only read the disassembly so far...)
22:34 < pini> lindi-: I expected it in some way since on of the last commits does some cleanups in debian/rules; but I haven't tried recently (updating my chroot)
22:37 < woglinde_> libqt4-dev: /usr/include/qt4/Qt/qglobal.h
22:37 < woglinde_> libqt4-dev: /usr/include/qt4/QtCore/qglobal.h
22:43 < pini> woglinde_: does it have both?
22:44 < woglinde_> sure
22:44 < woglinde_> backward compat
22:44 < lindi-> pini: would be nice to have the bitmaps in a different source package :-)
22:44 < lindi-> pini: rsvg-convert is sloooow
22:46 -!- drlizau [~liz@billiau.net] has quit [Remote host closed the connection]
22:46 < pini> lindi-: doesn't make much sense to me
22:47 < lindi-> or maybe I should build only the arch packages
22:49 < pini> lindi-: worth giving a try
22:50 -!- antenna1 [~ion@dslb-178-007-000-148.pools.arcor-ip.net] has quit [Quit: Leaving.]
22:51 < lindi-> pini: not going to interrupt anymore :)
22:51 < pini> :)
22:52 < lindi-> dh_pysupport: This program is deprecated, you should use dh_python2 instead. Migration guide: http://deb.li/dhs2p
22:52 < lindi-> so almost done
22:55 < lindi-> pini: dpkg-shlibdeps: warning: dependency on libdl.so.2 could be avoided if "debian/maptool/usr/bin/maptool" were not uselessly linked against it (they use none of its symbols).
22:55 -!- sera_ [~sera@204.69.63.81.cust.bluewin.ch] has joined #navit
22:56 < pini> lindi-: known
22:57 < lindi-> pini: yes, think of it as a progress message
22:58 < pini> lindi-: I prefer to keep pkg-config doing his stuff than digging manually into makefiles to track these glitches
22:58 -!- sera [~sera@174-32.62-81.cust.bluewin.ch] has quit [Ping timeout: 240 seconds]
23:01 < lindi-> pini: the patch does not fix the crash unfortunately
23:01 < lindi-> pini: so you can upload if you want but it won't affect this
23:02 < pini> lindi-: yes, seen that too; it's something related to cmake / find_package(Qt4)
23:10 < woglinde_> ah
23:10 < woglinde_> right
23:10 < woglinde_> iam using autotools
23:10 < woglinde_> but good nite
23:10 -!- woglinde_ [~heinold@g225146141.adsl.alicedsl.de] has quit [Quit: leaving]
23:16 < pini> lindi-: this patch seems to do the trick for the FTBFS: http://pastebin.com/DkiDRcKw
23:17 < pini> betime now :)
23:17 < pini> gn
23:17 < pini> s/betime/bedtime/
23:39 < Number6> Bed, night all
23:39 < Number6> /o/
23:58 -!- goldfndr__ [~goldfndr@h-72-244-204-190.sttnwaho.dynamic.covad.net] has quit [Ping timeout: 260 seconds]
23:58 -!- goldfndr__ [~goldfndr@h-72-244-204-190.sttnwaho.dynamic.covad.net] has joined #navit
--- Log closed Fri Jul 22 00:00:38 2011