| 00:15.57 | *** join/#htc-linux unicat_ (i=foobar@dslb-088-073-018-016.pools.arcor-ip.net) |
| 05:00.44 | *** join/#htc-linux sadeness_ (n=LamersIn@nat1.rlan.ru) |
| 05:17.02 | *** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com) |
| 05:20.12 | *** join/#htc-linux slapin (n=slapin@iap-pxy-mow1.siemens.ru) |
| 05:28.06 | *** part/#htc-linux emiliano75 (n=emiliano@adsl-ull-116-216.42-151.net24.it) |
| 05:34.19 | *** join/#htc-linux _law_ (n=law@mail.stiftadmont.at) |
| 06:24.03 | *** join/#htc-linux pleemans (n=peter@leemanpe.ugent.be) |
| 06:55.49 | *** join/#htc-linux pH5 (n=ph5@p5485FE7B.dip.t-dialin.net) |
| 07:51.53 | *** join/#htc-linux HexM (n=HotCoffe@distancing-wreath.volia.net) |
| 08:47.20 | *** join/#htc-linux psokolovsky__ (n=psokolov@82.193.98.14) |
| 08:51.15 | goxboxlive | <PROTECTED> |
| 08:52.32 | goxboxlive | pH5, it's on Magician |
| 08:54.51 | pH5 | goxboxlive: nice! will you upload those images? |
| 09:19.47 | goxboxlive | pH5, of course i can. I'll tell when i have done it. Have to go to the store first. |
| 09:20.46 | pH5 | goxboxlive: great, no hurry. |
| 09:24.37 | *** join/#htc-linux dion (n=dion@inhex.net) |
| 09:38.42 | goxboxlive | pH5, i just want to show you another screenie here, now we only need to make that driver: http://gnulinux.biz/files/Universal/tmp/crap/sc_Fri_Jun_22_09.37.28_2007.png |
| 09:40.45 | *** join/#htc-linux rob_w|mis (n=rob_w@213.218.30.172) |
| 09:41.30 | pH5 | goxboxlive: heh, or find out who made that driver work on a780 and ask how they did it. |
| 09:43.30 | goxboxlive | I thinko cr2 is aware of who made it, or i know stefan_smith knows |
| 09:47.40 | *** join/#htc-linux kermit78 (n=copal@gw-out2.colpal.com) |
| 09:47.46 | kermit78 | hi |
| 09:47.54 | kermit78 | i have a question |
| 09:48.21 | kermit78 | do u have an irc client software for pda phone??? |
| 09:50.10 | Foxdie | o.o |
| 09:50.13 | pH5 | goxboxlive: ah, found it |
| 09:50.15 | pH5 | http://www.mail-archive.com/openezx-devel@lists.openezx.org/msg00270.html |
| 09:51.16 | pH5 | Foxdie: I think that's a new record :) |
| 09:52.24 | *** join/#htc-linux rob_w|mis (n=rob_w@213.218.30.172) |
| 09:53.24 | Foxdie | for the most impatient person ever? |
| 10:11.24 | pH5 | Foxdie: yup |
| 11:10.55 | goxboxlive | good |
| 11:25.33 | *** join/#htc-linux Ranpha1 (n=Ranpha@hlm-n-73ed.adsl.wanadoo.nl) |
| 12:17.12 | *** join/#htc-linux cyrill62 (n=user@ble59-5-82-233-205-36.fbx.proxad.net) |
| 12:17.57 | goxboxlive | pH5, opieII is upped. I have just upped the opieII content. Just unzip it to / ans use the provided opieII start file to start opieII. You will need to have libstdc++6 to your console image. |
| 12:18.03 | goxboxlive | Here is the link: http://www.linuxtogo.org/~htcpxa/htcmagician/images/OpieII/no_rootfs/opieII-4.2.2.tar.bz2 |
| 12:23.16 | goxboxlive | cr2, this opieII will also work with Universal (maybe others too) . Just remember to edit the screen size in the opieII startup file. |
| 12:27.32 | goxboxlive | Actually the 'export QWS_SIZE=240x320' is not required as long as you have 'export QWS_DISPLAY="VNC:Transformed:LinuxFb:mmWidth44:mmHeight58"' this set correct. This line is for magician. It shows the physical size of the lcd screen. |
| 12:28.18 | goxboxlive | AFAIR universal use 55 75 or something. |
| 12:39.07 | cr2 | goxboxlive: nice screenshots :) i've changed 'xplatform' to htcphone, so it can be compiled with -eabi from oe. |
| 12:39.30 | cr2 | now i need to get rid of /usr/local/arm links, and do it in ~/3rdparty |
| 12:40.47 | *** join/#htc-linux BabelO (n=Miranda@faim1.itesoft.com) |
| 12:51.34 | *** join/#htc-linux dion (n=dion@inhex.net) |
| 13:00.00 | *** join/#htc-linux rob_w|mis (n=rob_w@213.218.30.172) |
| 13:03.21 | BabelO | hi |
| 13:03.38 | BabelO | goxboxlive : your screenshot are with opie theme ? |
| 13:06.48 | *** join/#htc-linux aljen (n=aljen@gsa26.internetdsl.tpnet.pl) |
| 13:06.51 | aljen | hey :) |
| 13:13.14 | goxboxlive | . |
| 13:13.32 | goxboxlive | BabelO, no, the theme is smart with Qtopia style |
| 13:13.49 | goxboxlive | cr2, i have a bb file if you want it. But i havent testet it my self |
| 13:17.07 | cr2 | goxboxlive: with dbus and alsa ? |
| 13:25.43 | cr2 | i don't understand why it insists on -m32, but it can be resolved later. |
| 13:43.31 | cr2 | hmm. it's hardcoded. probably some buggy code needs it. |
| 13:46.32 | cr2 | no-bluetooth no-dbus and sound-system oss |
| 14:05.31 | goxboxlive | cr2, yes it with dbus, alsa, helix and bluetooth. |
| 14:18.36 | goxboxlive | http://www.mystifistisk.net/2007/01/30/it-came-from-outer-space |
| 14:28.35 | *** join/#htc-linux cyrill62 (n=user@ble59-5-82-233-205-36.fbx.proxad.net) |
| 14:29.19 | *** join/#htc-linux GameOn (n=savirc@resnet24.nat.lancs.ac.uk) |
| 14:29.30 | GameOn | Hi all |
| 14:35.51 | cr2 | goxboxlive: it builds the 32bit qt4 version. and then wants to link them with the arm libs. at least the arm libs are in the search path. |
| 14:36.08 | cr2 | it seems i must compile it on a 32bit machine ;-) |
| 14:36.26 | cr2 | where is your .bb ? |
| 14:55.14 | *** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com) |
| 14:57.27 | *** join/#htc-linux rmoravcik (n=rmoravci@r2i34.net.upc.cz) |
| 14:58.30 | *** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com) |
| 15:21.21 | *** join/#htc-linux scorpio16v (n=scorpio1@ACB15D81.ipt.aol.com) |
| 15:40.33 | *** join/#htc-linux paulproteus (i=paulprot@29.145.221.202.bf.2iij.net) |
| 16:34.47 | *** join/#htc-linux Ranpha1 (n=Ranpha@hlm-n-73ed.adsl.wanadoo.nl) |
| 16:49.02 | *** join/#htc-linux tsdogs (n=twostupi@84.18.144.39) |
| 16:49.37 | goxboxlive | hi tsdogs |
| 16:49.46 | tsdogs | hi goxboxlive |
| 16:49.57 | tsdogs | was it the libpng? |
| 16:50.15 | goxboxlive | tsdogs, no there where missing files in the source. |
| 16:50.23 | tsdogs | goxboxlive: ok |
| 16:50.52 | goxboxlive | I didnt have libpng installed on my image, and i have been using that image for opie2-4.2.1 |
| 16:51.25 | goxboxlive | So the source wich is in the svn now builds, installs and runs. :-) |
| 16:51.38 | tsdogs | :) ok, so no need for libpng |
| 16:52.11 | goxboxlive | But it seem a little more unstable than the previous version though. I have segfault some times |
| 16:52.19 | goxboxlive | no, libpng shouldnt be needed |
| 16:52.46 | goxboxlive | like this: |
| 16:52.48 | goxboxlive | QTimeZone::data Can't create a valid data object for 'America/Fortaleza' |
| 16:52.48 | goxboxlive | Segmentation fault |
| 16:52.48 | goxboxlive | Cannot get semaphore /tmp/qtopia-0/sharedCache 't' |
| 16:52.48 | goxboxlive | Error 2 No such file or directory |
| 16:52.48 | goxboxlive | Did not receive a connection event from the qws server |
| 16:52.51 | tsdogs | I'm in the process of building it |
| 16:52.57 | goxboxlive | i c |
| 16:53.16 | tsdogs | hmm. did you copy the TimeZone data ? |
| 16:56.05 | goxboxlive | tsdogs, i dont know how to do that |
| 16:56.15 | goxboxlive | pH5, what is all this spamming? http://pastebin.no/2163 |
| 16:56.29 | tsdogs | goxboxlive: don't think this is the cause of the segfault though |
| 16:56.51 | goxboxlive | no me neither, but it is anoying haveing it there |
| 16:57.28 | pH5 | goxboxlive: brightness control. |
| 16:57.30 | tsdogs | goxboxlive: _ljp told us how to fix it time ago, but don't remeber exactly how |
| 16:57.44 | pH5 | magician_cpld checked for same values before writing, htc-egpio does not. |
| 16:58.13 | pH5 | Kevin2: should htc-egpio check the cached values and only write to the registers if something actually changed? |
| 17:01.32 | *** join/#htc-linux psokolovsky_ (n=psokolov@82.193.98.14) |
| 17:02.42 | cr2 | tsdogs: i've submitted the qt4 patch again mainline, and afaik it is applied. |
| 17:02.58 | tsdogs | cr2: I saw it :) |
| 17:03.26 | cr2 | tsdogs: i've also tried to copy the qt4 code to the current roadmap_editor code |
| 17:03.36 | tsdogs | cr2: I cannot still subscribe to the mailing list though :( |
| 17:03.52 | tsdogs | cr2: does it work? |
| 17:03.54 | cr2 | with minor patches it works. |
| 17:03.59 | tsdogs | cool :) |
| 17:03.59 | cr2 | basic things. |
| 17:04.05 | cr2 | yes :) |
| 17:04.37 | tsdogs | ok, we used yours as we did not have any map available. but now we got better maps :) |
| 17:04.48 | cr2 | now i need to generate the map. |
| 17:05.07 | cr2 | it works only with the map of israel now. |
| 17:05.17 | tsdogs | ok, did you apply all the changes ? |
| 17:05.23 | cr2 | no. |
| 17:05.30 | cr2 | just copied the qt/ over |
| 17:05.39 | cr2 | and fixed the compilation problems. |
| 17:05.42 | tsdogs | ho ok. |
| 17:05.45 | cr2 | all new stuff does not work. |
| 17:05.54 | cr2 | because it's not there. |
| 17:06.16 | cr2 | but i can see the map. at least parts of it. |
| 17:06.43 | *** join/#htc-linux RoEn (n=roen@p54A64F3E.dip.t-dialin.net) |
| 17:06.49 | tsdogs | ok |
| 17:06.51 | cr2 | i've also learned how to simplify the polygons. |
| 17:07.00 | cr2 | if they were cut in pieces. |
| 17:07.18 | tsdogs | that's what I told you to do, splt them :) |
| 17:07.19 | cr2 | it's unbelievably slow, i need to understand how to do it in grass. |
| 17:07.23 | goxboxlive | tsdogs, yes i know, but i seemed like it wasent correct. I tried it but it didnt work. |
| 17:07.37 | tsdogs | goxboxlive: worked for me :) |
| 17:07.59 | goxboxlive | tsdogs, ok, the i obvioues did something wrong :-) |
| 17:08.08 | tsdogs | :) |
| 17:08.40 | cr2 | tsdogs: i'd also like to generate such DB: http://mpa.itc.it/corso_dit2004/grass04_4_vector_network_neteler.pdf |
| 17:09.02 | cr2 | so it will be possible to check the map independently. |
| 17:10.19 | cr2 | and we will be able to find unconnected pieces in the map. |
| 17:11.00 | tsdogs | I have no idea on how this works, but yes :) |
| 17:12.48 | cr2 | ok. |
| 17:12.59 | cr2 | it uses the same dglib library. |
| 17:13.21 | tsdogs | ha ok. so we only need to extend it's use? |
| 17:14.10 | tsdogs | goxboxlive: you have to copy the etc/zoneinfo folder from the source tree to /opt/Qtopia/etc/. |
| 17:14.11 | cr2 | i need to import the data in grass. |
| 17:14.28 | tsdogs | cr2: do we have all that routing data? |
| 17:14.43 | cr2 | it depends on the map. |
| 17:14.53 | cr2 | for the map that you have - no. |
| 17:15.39 | goxboxlive | tsdogs, ok thx |
| 17:15.41 | cr2 | there are maps on mapcenter, that have good routing data, even with turn restrictions. |
| 17:15.59 | tsdogs | cr2: ok |
| 17:17.05 | *** join/#htc-linux Babel (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 17:18.13 | cr2 | tsdogs: i'm driving to perugia on next weekend. |
| 17:18.42 | tsdogs | :) what's your shcedule? |
| 17:20.08 | cr2 | i think i must be there on 2th, then 1 stay for a week and then drive back. |
| 17:21.17 | cr2 | goxboxlive: where is your .bb file ? |
| 17:21.40 | goxboxlive | cr2, it's not mine, but i can mail it to you. |
| 17:21.51 | cr2 | ok. thanks. |
| 17:22.32 | goxboxlive | cr2, sent |
| 17:22.53 | goxboxlive | What is this: QPainter::setCompositionMode: Composition modes not supported on device |
| 17:23.14 | tsdogs | goxboxlive: a feature missing from QT4 |
| 17:23.38 | goxboxlive | Ok, it seems like it's causing the segfault |
| 17:23.56 | tsdogs | http://doc.trolltech.com/qtopia4.2/qpainter.html#setCompositionMode |
| 17:23.57 | goxboxlive | Is there a way i can remove that from the sourve |
| 17:24.27 | tsdogs | goxboxlive: where is it called from? |
| 17:24.37 | goxboxlive | Ni idea |
| 17:24.40 | goxboxlive | no |
| 17:24.46 | tsdogs | ho ok |
| 17:26.17 | tsdogs | goxboxlive: when does it give it to you? |
| 17:28.48 | cr2 | goxboxlive: got it. |
| 17:30.19 | goxboxlive | tsdogs, well here and there. Its randomly |
| 17:31.05 | tsdogs | goxboxlive: ok, but then it's hard to fetch, do you have debug enabled? |
| 17:32.34 | goxboxlive | tsdogs, no, i can enable it. But first i will try to fix the zoneinfo stuff. I uploaded the whole folder to gnulinix.biz (that is allowad right? I am not breaking any copyright laws?) |
| 17:33.15 | tsdogs | goxboxlive: it's opieII right? so it's all GPL stuff |
| 17:33.18 | goxboxlive | Why doesnt it install the whole zoneinfo folder when it is installed to the SD card? |
| 17:33.21 | goxboxlive | good |
| 17:33.43 | cr2 | OPIE2/opie/trunk/qtopiacore/qt/src/gui/embedded/qsoundqss_qws.cpp:50: error: declaration of 'int* __errno_location()' throws different exceptions |
| 17:34.18 | cr2 | <PROTECTED> |
| 17:36.14 | cr2 | hehe. qtopia redefines errno ;) |
| 17:36.19 | tsdogs | :) |
| 17:36.57 | cr2 | imho it's a bug. |
| 17:39.39 | cr2 | tsdogs: i have the canvas icons with the latest roadmap_editor qt4 code. |
| 17:41.02 | *** join/#htc-linux kris (n=kris@host81-129-113-105.range81-129.btcentralplus.com) |
| 17:41.22 | kris | hello |
| 17:41.50 | kris | I was wondering if anyone has had trouble with GPE on the universal? |
| 17:42.52 | goxboxlive | kris, what do you mean ? |
| 17:43.29 | kris | i have an htc universal, when i go to the program folder to boot it off the sd card (pre prepared) it gives me a message saying it can't find the device... |
| 17:45.55 | *** join/#htc-linux pleemans (n=peter@d51A5E76A.access.telenet.be) |
| 17:46.40 | kris | i have got a feeling that it might even be the partition that is causing the problem, maybe i have not made it correctly... |
| 17:46.57 | *** part/#htc-linux scorpio16v (n=scorpio1@ACB15D81.ipt.aol.com) |
| 17:48.03 | tsdogs | cr2: bmp or png? |
| 17:49.56 | *** join/#htc-linux FossiFoo (n=Fossi@e176126238.adsl.alicedsl.de) |
| 17:51.02 | cr2 | tsdogs: i think they must be png |
| 17:51.03 | goxboxlive | tsdogs, are you able to build libstdc++6 in OE anymore? |
| 17:51.16 | cr2 | afaik windows port also uses png now. |
| 17:51.21 | tsdogs | goxboxlive: no idea. have to check |
| 17:51.35 | cr2 | heh. |
| 17:51.37 | tsdogs | cr2: I saw bmps there before, I converted them to png |
| 17:51.39 | cr2 | /* On Win32 memset and memcpy are already declared in windows.h */ |
| 17:51.39 | cr2 | #else |
| 17:51.39 | cr2 | #if __STD_C |
| 17:51.39 | cr2 | void* memset(void*, int, size_t); |
| 17:51.39 | cr2 | void* memcpy(void*, const void*, size_t); |
| 17:52.11 | goxboxlive | tsdogs, thx |
| 17:52.38 | cr2 | too many windows monkeys worked on qtopia :) |
| 17:52.50 | tsdogs | :) |
| 17:53.35 | cr2 | defining errno, memset and memcpy it too much. |
| 17:54.18 | cr2 | as if it's only 'already declared in windows.h' |
| 17:54.47 | tsdogs | where aren't they defined ? dos ? |
| 17:56.07 | cr2 | they could do #include <stdlib.h> ;-) |
| 17:57.25 | cr2 | it's usually windows headers that produce a lot of warnings and are not very ansi c friendly. |
| 17:59.02 | cr2 | oh. they even have crypto :) i've seen sha512.c |
| 17:59.34 | cr2 | it's openssl |
| 18:03.09 | *** join/#htc-linux FossiFoo_ (n=Fossi@e176123060.adsl.alicedsl.de) |
| 18:04.50 | cr2 | tsdogs: should we hack on 0.8.* or move to the current _editor CVS ? |
| 18:05.50 | goxboxlive | tsdogs, it happend agin, i just hit document and uit began scanning: |
| 18:05.52 | goxboxlive | QTimeZone::data Can't create a valid data object for 'Europe/Oslo' |
| 18:05.52 | goxboxlive | Cannot get semaphore /tmp/qtopia-0/sharedCache 't' |
| 18:05.52 | goxboxlive | Error 2 No such file or directory |
| 18:05.52 | goxboxlive | Segmentation fault |
| 18:05.52 | goxboxlive | QAbstractSocket::waitForReadyRead() is not allowed in UnconnectedState |
| 18:05.53 | goxboxlive | ApplicationLayer: Application layer server unexpectedly terminated. |
| 18:05.57 | goxboxlive | QAbstractSocket::waitForReadyRead() is not allowed in UnconnectedState |
| 18:05.59 | goxboxlive | ApplicationLayer: Application layer server unexpectedly terminated. |
| 18:11.50 | *** join/#htc-linux rob_w|usa (n=bob@AC9EA3A2.ipt.aol.com) |
| 18:14.17 | *** join/#htc-linux Ro_En (n=roen@p54A66F2C.dip.t-dialin.net) |
| 18:14.19 | *** part/#htc-linux paulproteus (i=paulprot@29.145.221.202.bf.2iij.net) |
| 18:17.07 | *** join/#htc-linux BabelOued (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 18:17.18 | tsdogs | goxboxlive: we need to investigate. |
| 18:18.05 | tsdogs | cr2: are the changes relevant or not? will we keep updated to it? |
| 18:18.10 | goxboxlive | tsdogs, another one: http://pastebin.no/2164 |
| 18:18.45 | tsdogs | goxboxlive: it obviusly the should go in /usr/share :) |
| 18:19.08 | goxboxlive | tsdogs, ?? zoneinfo? |
| 18:19.22 | goxboxlive | of course i am blind |
| 18:19.23 | tsdogs | yep. the QPainter thing is pretty strange. |
| 18:20.39 | tsdogs | QPainter::begin: Paint device returned engine == 0, type: 3 this should be the problem, but what it is, no idea... |
| 18:21.08 | goxboxlive | tsdogs, yes now the zoneinfo stuff is gone |
| 18:21.17 | tsdogs | :) |
| 18:23.08 | tsdogs | goxboxlive: could it be a libpthread issue ? |
| 18:23.22 | tsdogs | goxboxlive: is libpthread liked? |
| 18:23.25 | goxboxlive | I dont know, i ghave asked ljp about it now |
| 18:23.28 | tsdogs | s/liked/linked/ |
| 18:24.17 | goxboxlive | tsdogs, is libpthread a part of Qtopia or a part of rootfs |
| 18:24.26 | tsdogs | rootfs |
| 18:24.42 | tsdogs | I think you need ldd libQtCore or something |
| 18:27.02 | goxboxlive | tsdogs, it seems like it failes when this is happening: Cannot get semaphore /tmp/qtopia-0/sharedCache 't' |
| 18:28.15 | tsdogs | that's why I'm thinking about threads ... |
| 18:28.47 | tsdogs | also this one: QTextBrowser: Cannot open 'image/help/dpad' for reading |
| 18:30.21 | goxboxlive | ok |
| 18:32.51 | tsdogs | goxboxlive: where is the image on linuxtogo? |
| 18:33.50 | goxboxlive | tsdogs, you need to aply your own console image. This is just /opt folder and a startup script |
| 18:33.52 | goxboxlive | http://www.linuxtogo.org/~htcpxa/htcmagician/images/OpieII/no_rootfs/opieII-4.2.2.tar.bz2 |
| 18:34.15 | *** join/#htc-linux SmallR2002 (n=SmallR20@88-107-35-245.dynamic.dsl.as9105.com) |
| 18:34.22 | goxboxlive | tsdogs, do you have a 3ggp video file? |
| 18:34.35 | goxboxlive | i need one, but i cant see to find any on google |
| 18:34.53 | tsdogs | 3ggp? don't think so what is it? |
| 18:37.18 | goxboxlive | it handhelds movie format |
| 18:38.29 | tsdogs | goxboxlive: http://nds1.nokia.com/phones/files/software/Nokia_6600_Multimediaplayer_1_1_en.exe |
| 18:38.35 | tsdogs | maybe it works with wine |
| 18:38.59 | tsdogs | goxboxlive: also http://www.gromkov.com/faq/faq2004-0072.html |
| 18:40.43 | goxboxlive | tsdogs, no i want to test it in helix player on qtopia. 3gp is just a mime-type, instaed of mpeg |
| 18:41.08 | tsdogs | goxboxlive: u can convert an avi to 3gp with this player |
| 18:41.38 | goxboxlive | ahh, i c |
| 18:41.40 | tsdogs | look at the second link. |
| 18:41.48 | goxboxlive | i got one file now |
| 18:41.57 | tsdogs | ok my OE is fskup :( |
| 18:42.03 | goxboxlive | how? |
| 18:42.14 | tsdogs | ERROR: Could not inherit file classes/rootfs_${IMAGE_PKGTYPE}.bbclass |
| 18:42.42 | tsdogs | I haven't updated it for some time, and now this. |
| 18:43.40 | *** join/#htc-linux Foxdie` (i=Foxdie@77-98-48-193.cable.ubr05.harb.blueyonder.co.uk) |
| 18:48.47 | tsdogs | going home now. bbl cya |
| 18:48.58 | goxboxlive | tsdogs, ok, you dont have to delete your work folder to use a new database. You can just checkout a new one and build on your existing work folder |
| 19:18.28 | *** join/#htc-linux tsdogs (n=tsdogs@62.123.180.130) |
| 19:20.20 | *** join/#htc-linux tsdogs (n=tsdogs@62.123.180.130) |
| 19:24.35 | cr2 | tsdogs: i suggest to update to 0.8.2 first |
| 19:25.39 | tsdogs | cr2: ok. |
| 19:28.03 | *** join/#htc-linux GSV (i=GSV@c-71-202-155-188.hsd1.ca.comcast.net) |
| 19:28.54 | *** join/#htc-linux Ranpha1 (n=Ranpha@hlm-n-73ed.adsl.wanadoo.nl) |
| 19:37.21 | cr2 | ok, i've compiled qtopia4. but still it's too much manual help in the progress. |
| 19:37.57 | cr2 | and i don't have alsa, bt and dbus |
| 20:09.12 | goxboxlive | cr2, on universal: http://gnulinux.biz/files//Universal/tmp/crap/menu_on_universal.png |
| 20:11.37 | cr2 | goxboxlive: looks good :) |
| 20:14.52 | goxboxlive | I dont have any segfault with universal (no so far) |
| 20:15.31 | goxboxlive | ljp told me that i might have to configure with -depth 8,16 for magician and 8,16,32 for Universal |
| 20:16.00 | goxboxlive | So i have to do that i have to have two different htcphone folders. |
| 20:16.14 | goxboxlive | BabelO, are you here? |
| 20:21.13 | BabelOued | yes |
| 20:22.19 | goxboxlive | BabelOued, come send me the phone source. I dont care about sms, i am to ols anyway to send sms. I'm only using the phone. |
| 20:23.02 | BabelOued | ok |
| 20:23.23 | goxboxlive | gracias |
| 20:43.16 | *** join/#htc-linux tsdogs (n=twostupi@62.123.180.130) |
| 20:43.35 | *** join/#htc-linux Hymie (n=Hymie@l8r.net) |
| 20:44.53 | cr2 | tsdogs: found one more bug/limitation for the polygons. |
| 20:45.06 | tsdogs | cr2: :) |
| 20:45.27 | tsdogs | cr2: it seems to me that 0_8_2 is older than the source we are using (?) |
| 20:45.35 | Hymie | hmm.. my hermes has that lame, classic screen bug, where I have to keep re-aligning it.. so, I'm going to send it in for repairs. However, that's likely to take a few weeks, so.. I'm going to buy a replacement and then sell the refurb hermes when it returns. Anyone have any suggestions on a device thar has 1) keyboard 2) bluetooth 3) UMTS, supporting N/A frequencies |
| 20:45.38 | cr2 | really ? |
| 20:46.08 | tsdogs | I think so, buildmap_line_speed.x are missing |
| 20:46.47 | cr2 | tsdogs: strange. |
| 20:47.14 | tsdogs | also dglib/cr_from_a.c is missing |
| 20:48.42 | tsdogs | hmm |
| 20:48.46 | cr2 | i've added cr_from_a.c myself. |
| 20:48.53 | tsdogs | cr2: :) |
| 20:48.58 | Hymie | http://www.engadget.com/2006/12/21/htc-athena-x7500-a-umtshsdpa-quad-band-UMPC/ <-- heh, wish that was slightly smaller, it seems cool |
| 20:49.14 | cr2 | Hymie: it's not a phone. |
| 20:49.29 | Hymie | cr2: I know, but soon a phone will have those specs ;) |
| 20:49.32 | Hymie | maybe 1 year |
| 20:49.34 | Hymie | 2 outside |
| 20:50.12 | Hymie | cr2: you've used 4 billion different cell phones, no ideas on a hermes like device, that isn't a hermes? |
| 20:50.28 | cr2 | Hymie: it's the physical LCD size that it nice. pushing VGA on 2.8" just does not make sense. |
| 20:50.43 | GSV | what do you guys think about kaiser? |
| 20:51.02 | cr2 | Hymie: hermes is the best in its class. |
| 20:51.06 | Hymie | cr2: I can hardly wait for those flexible displays to make mainstream, then we could have a display that rolls out when needed |
| 20:51.35 | Hymie | cr2: outside of a hack, I wouldn't mind WM6 though.. |
| 20:52.01 | cr2 | wm6 is the same insecure "os" as wince5. |
| 20:52.04 | Hymie | cr2: I have two options, spend $150 on a temp cell phone, and wait for my hermes to come back, or buy an even cooler phone, and roll that $150 into it |
| 20:52.29 | Hymie | cr2: I had heard that the bluetooth support was v2 in vm6, a bit better for car adapters, etc |
| 20:52.38 | Hymie | cr2: but, I could be wrong |
| 20:52.45 | Hymie | cr2: I know it's still just WM though ;) |
| 20:53.02 | Hymie | hmm, it's been a few months, any luck on the hermes port? |
| 20:53.17 | Hymie | I know you guys have been working on 1000 devices though, so I imagine it may not have moved far |
| 20:53.19 | cr2 | Hymie: i have athena, it's great for the gps navigation in the car :) |
| 20:53.41 | Hymie | cr2: heh, I bet it is... |
| 20:53.43 | cr2 | the main problem is the ATI SD chip. |
| 20:53.50 | GSV | do you think kaiser will be a good phone? |
| 20:54.15 | cr2 | GSV: it seems to be a omap850 line replacement |
| 20:54.24 | cr2 | lol |
| 20:54.39 | Hymie | cr2: well, they just lie all the time about linux support and linux drivers |
| 20:54.45 | Hymie | cr2: it pisses me off... |
| 20:55.01 | GSV | what does that mean? omap850 |
| 20:55.07 | Hymie | cr2: say what you will about nvidia, they supported Linux _while_ saying "we won't ever release the code".. honest, blunt, and support all at once |
| 20:55.28 | cr2 | they should better provide the datasheet. it's not that much what we need to support non-video part. |
| 20:55.41 | Hymie | http://www.pdadb.net/index.php?m=pdalist&list=phone3g |
| 20:55.42 | Hymie | hmm |
| 20:55.49 | Hymie | interesting, that seems to be a good list of pdas |
| 20:56.11 | cr2 | Hymie: i never use binary drivers. |
| 20:56.21 | Hymie | cr2: I want my good 3d |
| 20:56.23 | cr2 | so nvidia can go to hell. |
| 20:56.28 | Hymie | cr2: hehe |
| 20:56.46 | cr2 | i have hw opengl supported on an ati card in 2000. |
| 20:56.54 | Hymie | cr2: hey, I understand your attitude, but regardless, nvidia has been honest, and has supported linux in their own way. that's a lot more than many companies have done |
| 20:57.12 | cr2 | i ran tuxracer once, and then forgot about 3d :) |
| 20:57.33 | Hymie | cr2: 3d is over for me, I have a playstation 2 to play a few old games on, that's all I wanted it for anyhow |
| 20:57.40 | cr2 | Hymie: on nvidia i use 'nv' driver. it's ok for 2d. |
| 20:57.46 | Hymie | yeah |
| 20:57.48 | Hymie | I've heard that |
| 20:58.22 | GSV | guys i need your advice. i got a 8525 2 weeks ago and last week i found out that kaiser will come in august. I can return my 8525 and wait for kaiser. so what do you suggest? |
| 20:59.04 | cr2 | GSV: only if you urgently need a gps on your phone. |
| 20:59.20 | GSV | i dont need gps |
| 20:59.31 | cr2 | it's certainly a cheaper phone than hermes. |
| 20:59.39 | GSV | it's rom is 256 and it's ram is 128 |
| 20:59.55 | cr2 | because if i get it right, it's a 1 cpu device. |
| 21:00.34 | cr2 | so it's more of a classical phone. |
| 21:00.53 | GSV | so it's better that i keep my hermes and wait for better phone |
| 21:00.55 | Hymie | man alive, htc seems to have a lot of phones coming out in the next few months |
| 21:01.01 | tsdogs | cr2: now I confirm, our version looks more recent than the 0_8_2 in roadmap_canvas.c we have some changes which are not in 0_8_2 |
| 21:01.14 | Hymie | http://www.pdadb.net/index.php?m=pdalist&list=phone3g |
| 21:01.46 | cr2 | tsdogs: it seems i've used some intermediate CVS version. |
| 21:02.07 | tsdogs | I think they started tagging after that one :/ |
| 21:02.10 | Hymie | hmm, cavalier seems interesting |
| 21:02.23 | tsdogs | cr2: ok, so we go for the MAIN? |
| 21:03.08 | GSV | is samsung s3c2442 400 Mhz better or Qualcomm MSM7200 384 Mhz |
| 21:03.48 | cr2 | samsung means samsung+qualcomm. it's not a phone chipset. |
| 21:04.05 | cr2 | tsdogs: you'll be left without a map for some time. |
| 21:04.14 | Hymie | cr2: have you seen the cavalier yet? |
| 21:04.26 | tsdogs | :) are your changes completly uncompatible? |
| 21:04.37 | GSV | oh so that means hermes has 2 cpu?! |
| 21:04.51 | tsdogs | cr2: I want to merge your map changes too :) |
| 21:05.05 | cr2 | Hymie: i evaluate only the hardware side :) |
| 21:05.25 | Hymie | cr2: so, have you seen the hardware? ;) |
| 21:05.46 | cr2 | GSV: all not-cheapskate (i.e. non-omap) htc phones have 2 cpus |
| 21:05.52 | tsdogs | cr2: then you should probably go for a laptop :) |
| 21:05.55 | cr2 | Hymie: no. |
| 21:06.12 | cr2 | tsdogs: there are no laptops with pxa270 cpu. |
| 21:06.19 | tsdogs | lol |
| 21:06.24 | cr2 | athena is the closest :) |
| 21:06.46 | GSV | cr2: so hermes is better than kaiser in cpu.right? |
| 21:06.55 | cr2 | tsdogs: i have a laptop with a risc cpu: linuppc :) |
| 21:07.00 | tsdogs | :) |
| 21:07.33 | goxboxlive | pH5 / cr2 do you know if Magican supprts iwmmxt? |
| 21:07.35 | cr2 | GSV: certainly. htc just cuts the production costs. adding +64MB and gps. |
| 21:07.50 | cr2 | goxboxlive: yes, it#s a pxa27x feature. |
| 21:07.56 | goxboxlive | thx |
| 21:08.28 | cr2 | tsdogs: we can drop the 'roadmap' tree from xanadux now. it's all in the mainline CVS |
| 21:08.29 | GSV | cr2:thanks for the information. you are great |
| 21:08.45 | tsdogs | cr2: how do I do that? |
| 21:09.24 | Hymie | cr2: tell me whether or not to buy the cavalier. Then, if it is good, I can take credit for being wise enough to ask you. If it is bad, then I can blame you for the mistake. ;Þ :Þ |
| 21:09.33 | cr2 | tsdogs: neved did it on such a big scale :) |
| 21:09.53 | tsdogs | cr2: ok I'll look on how to do it. |
| 21:10.22 | cr2 | Hymie: i have a personal dislike for the s3c24xx cpus. but they are the future. |
| 21:12.44 | cr2 | tsdogs: we need to find some simple way to support the 'point' layers. that's the biggest problem for me now. |
| 21:13.27 | tsdogs | can we implement it like a new layer type? we have Road/Area/Feature |
| 21:14.29 | cr2 | tsdogs: yes. but we need to pick some supporting code from 'places' |
| 21:15.04 | tsdogs | cr2: sure as places are broken we need to finish them up. |
| 21:15.40 | cr2 | let's assume it will work. |
| 21:15.50 | tsdogs | :) |
| 21:16.10 | cr2 | if you'll install the shapelib on an i386 machine, then it will work. i can do it. |
| 21:16.30 | cr2 | but then you need some data in the format it wants. |
| 21:16.54 | tsdogs | so it is only a matter of x64 incompatilibity? |
| 21:16.54 | cr2 | the code was written by some guy from Canada, so he used it for some DMTI data afair. |
| 21:17.32 | cr2 | the sort routine is certainly broken there. |
| 21:17.48 | cr2 | i think the code does the following: |
| 21:18.11 | cr2 | it adds a new table to the file, so you can use a search. |
| 21:18.14 | cr2 | nothing more. |
| 21:18.37 | cr2 | btw, the shapefile can be created by ogr2ogr |
| 21:19.02 | cr2 | i use a similar technique for creating a map now. |
| 21:19.14 | tsdogs | well I have the shapefiles. |
| 21:19.36 | cr2 | generate a mapinfo file (ascii) with a script, and then convert with ogr2ogr. |
| 21:19.47 | tsdogs | cr2: are you able to add the point layer to the map? |
| 21:19.54 | cr2 | you need 2 files: |
| 21:20.02 | cr2 | designations.txt ? |
| 21:20.09 | cr2 | and the shapefile itself. |
| 21:20.55 | cr2 | actually i'd like to have 2 tools for the rdm. |
| 21:21.08 | cr2 | 'rdminfo' like 'ogrinfo' |
| 21:21.11 | GSV | what's the difference between hermes 160 and hermes 200? |
| 21:21.19 | cr2 | and 'rdm2pgsql' |
| 21:21.48 | cr2 | the current 'dumpmap' is cryptic, undocumented and imho useless. |
| 21:22.16 | cr2 | there is a better tool in the mainline code - 'rdmxchange' |
| 21:22.51 | tsdogs | what does that do? |
| 21:23.00 | cr2 | but it's also not much better, because it produces some cryptic ascii output, so you need very deep knowldege. |
| 21:23.08 | tsdogs | :) |
| 21:23.37 | tsdogs | cr2: I have no idea bout ogrinfo, but rdminfo and rdm2pgsql sould be done. |
| 21:23.38 | cr2 | makes an ascii 'rdm', so you can convert it on a bigendian machine, for example. |
| 21:23.49 | tsdogs | cr2: ho ok |
| 21:23.59 | cr2 | we don't need to add it to ogrinfo. |
| 21:23.59 | goxboxlive | cr2, did you say that i could use -xplatform linux-htcphone-g++ intstead of linux-arm-g++? |
| 21:24.21 | cr2 | goxboxlive: i've used it, because i've added such config :) |
| 21:24.26 | tsdogs | cr2: shouldn't we use some socket ntoi and so on for bigendian/littlendian compatibility? |
| 21:24.32 | goxboxlive | ok |
| 21:24.52 | cr2 | tsdogs: i'd like to get the 'ogrinfo'-style output from rdm. |
| 21:25.13 | cr2 | because it's a standard format, which can be reused by other GIS tools. |
| 21:25.38 | cr2 | you can simply use qgis to verify your maps. |
| 21:26.27 | cr2 | tsdogs: a visualization of the rdm table format can help enormously. |
| 21:26.42 | cr2 | i still a bit confused about certain things. |
| 21:26.48 | tsdogs | a tree you mean? |
| 21:26.54 | cr2 | it's unnecessary complicated now. |
| 21:27.14 | cr2 | yes, what structures are there, and how they are linked with each other. |
| 21:27.31 | cr2 | do you know what 'netcdf' is ? |
| 21:27.32 | tsdogs | cr2: ok I might have a look. |
| 21:28.38 | cr2 | not really necessary. |
| 21:28.40 | cr2 | http://www.unidata.ucar.edu/software/netcdf |
| 21:29.04 | cr2 | it's a very nice 'array database' format. |
| 21:29.05 | tsdogs | cr2: what's the difference between roadmap_db_point and roadmap_db_place ? |
| 21:29.25 | cr2 | i think point has only the coordinates. |
| 21:29.43 | cr2 | and place is a pointer to point *, and the name. |
| 21:30.01 | cr2 | and layer. |
| 21:30.06 | cr2 | all you need. |
| 21:30.10 | cr2 | typedef struct { |
| 21:30.10 | cr2 | <PROTECTED> |
| 21:30.10 | cr2 | <PROTECTED> |
| 21:30.11 | cr2 | <PROTECTED> |
| 21:30.11 | cr2 | <PROTECTED> |
| 21:30.11 | cr2 | } BuildMapPlace; |
| 21:30.34 | tsdogs | cr2: ok |
| 21:30.36 | cr2 | cfcc is a fancy name for 'layer' |
| 21:30.59 | tsdogs | :) yep. |
| 21:31.22 | cr2 | i guess 'landmarks' are actually a redundant things. |
| 21:31.31 | goxboxlive | ok a new one, this is nice colors: http://gnulinux.biz/files/Universal/tmp/crap/home-vga.png |
| 21:31.37 | tsdogs | cr2: NetCDF seems to be nice. |
| 21:31.38 | cr2 | because they are 'places' in this terminology. |
| 21:32.13 | cr2 | tsdogs: the greatest thing about netcdf is 'ncdump'. it shows you all metadata with comments. |
| 21:32.18 | cr2 | ncdump -h |
| 21:32.23 | tsdogs | goxboxlive: I had this 1 month ago, only it was green :) |
| 21:32.27 | cr2 | is wish we had the same for rdm. |
| 21:32.51 | goxboxlive | tsdogs, i know, but it looks so much better with blue color |
| 21:33.01 | tsdogs | cr2: you didn't want to use this for rdm ? |
| 21:33.11 | tsdogs | goxboxlive: true :) |
| 21:33.26 | cr2 | tsdogs: the header ? actually we can borrow the header design. |
| 21:33.27 | tsdogs | goxboxlive: I think we need to define the opie theme. |
| 21:33.46 | cr2 | tsdogs: we don't need the vectors themselves. |
| 21:33.56 | tsdogs | cr2: ok. |
| 21:34.02 | goxboxlive | tsdogs, i dont like the Opie theme, it doesnt look that good as Smart/Qtopia |
| 21:34.14 | cr2 | but looking how the rdm is constructed, we can actually simulate it with netcdf. |
| 21:34.31 | tsdogs | goxboxlive: what I meant is that we need a new nice opieII theme. |
| 21:34.42 | cr2 | it'll be a bit tricky with 'mmap', but netcdf api does not support it, but once i've did such hack :) |
| 21:34.50 | goxboxlive | tsdogs, ahh, of course, but after we have a working phone |
| 21:35.09 | cr2 | because i know how it's done internally. at least for the netcdf3 version. |
| 21:35.34 | cr2 | i use netcdf for all my scientific data, because it's extremely comvenient. |
| 21:35.49 | tsdogs | cr2: I wouldn't want to add to much overhead, will be running on handhelds. |
| 21:36.07 | tsdogs | s/to/too |
| 21:36.09 | cr2 | tsdogs: it's only a header. |
| 21:36.19 | tsdogs | cr2: ok. |
| 21:36.24 | goxboxlive | hmm BabelO said that he would send me the phone app source. What happend, did he fell a sleep ? |
| 21:36.32 | cr2 | you are not forced to use netcdf api if you don't want. use mmap. |
| 21:37.15 | goxboxlive | hmm it where in my inbox. hmm my google applet doesnt work. |
| 21:37.27 | cr2 | tsdogs: we can also pack 16bit QImage format for tiles :) nonstandard, but fast :) |
| 21:38.06 | tsdogs | cr2: :) need to implement that code too |
| 21:38.27 | cr2 | it's nice to have fast app+data, but we need a reliable way to verify it too. |
| 21:38.35 | cr2 | tsdogs: it's more or less done. |
| 21:38.54 | cr2 | since the png icons are drawn with *draw_image* |
| 21:39.28 | tsdogs | cr2: I know, but I need to implement the index reading, and code to get the right tiles to be drawn. |
| 21:39.40 | tsdogs | also zoom, needs some work |
| 21:39.50 | cr2 | it should not be very difficult. |
| 21:40.03 | tsdogs | cr2: but need some time. |
| 21:40.07 | tsdogs | :) |
| 21:40.08 | cr2 | i'm more concerned about the zoom |
| 21:40.19 | cr2 | we need to sync the zooms. |
| 21:40.24 | tsdogs | yep zoom can be triky |
| 21:40.41 | tsdogs | I now (for test) have implemented a % zoom. |
| 21:40.53 | cr2 | you don't want to do rescaling on the universal without any hw support. |
| 21:40.58 | tsdogs | based on the pixel |
| 21:41.05 | tsdogs | :) |
| 21:41.14 | cr2 | 2x is easy. just drop every seconf pixel |
| 21:41.37 | tsdogs | actually It's libgdal that does that. |
| 21:41.46 | cr2 | yes. |
| 21:42.10 | cr2 | but we can optimize it. |
| 21:42.16 | tsdogs | :) |
| 21:42.38 | cr2 | gdal allows direct block access. |
| 21:43.36 | tsdogs | cr2: I read it, but then we'd need to implement all the scalings and rotating. |
| 21:43.43 | cr2 | i need to think about 16bit clor palette. |
| 21:43.49 | cr2 | forget the rotation. |
| 21:43.56 | tsdogs | cr2: ok |
| 21:44.00 | cr2 | it's unfeasible. |
| 21:44.11 | cr2 | that's why i'd like to get mercator. |
| 21:44.15 | tsdogs | cr2: why not use a 8bit palette, would be easyer |
| 21:44.24 | cr2 | your display is 16bit. |
| 21:44.58 | tsdogs | hmm ok |
| 21:45.05 | cr2 | then you'll be able to memcpy() to the fb directly. |
| 21:46.14 | cr2 | hmm. it's a good idea with a netcdf header. |
| 21:46.20 | tsdogs | I wouldn't know how to do it but that would be the fastest as it can do |
| 21:46.29 | cr2 | yes. |
| 21:47.04 | cr2 | and all known pda lcds are 16bit. |
| 21:47.30 | cr2 | well, if you don't count old 4096 color pdas. |
| 21:47.34 | cr2 | 12bit |
| 21:47.42 | tsdogs | well, we'd also want to support PCs. |
| 21:48.05 | cr2 | qtopia is a custom compilation anyway. |
| 21:48.19 | cr2 | and gdal provides a generic interface. |
| 21:48.47 | tsdogs | cr2: not really with QT4 not many changes need to be done. |
| 21:49.08 | tsdogs | but ok, who uses a PC for navigation :) |
| 21:49.25 | tsdogs | and then PCs are faster. |
| 21:49.40 | cr2 | in the car. |
| 21:50.25 | tsdogs | cr2: I saw that Qtopia can be built with openGL support :), when will you enable 3d on Universal ? :) |
| 21:51.00 | tsdogs | cr2: ok I need priorities. |
| 21:51.56 | tsdogs | If it's ok for you, I'd go first for rdminfo and rdm2pgsql |
| 21:52.33 | tsdogs | btw, are you able to do a small dump of your map database? |
| 21:52.57 | tsdogs | so I can create a small map, and then copy it back to test? |
| 21:53.48 | tsdogs | cr2: keyboard is not the right place to sleep on ;) |
| 21:54.41 | cr2 | ? |
| 21:54.53 | cr2 | small dump ? |
| 21:55.04 | cr2 | have i sent you the .mp file ? |
| 21:56.03 | tsdogs | cr2: no idea |
| 21:56.54 | cr2 | ok. it's not a problem. |
| 21:57.02 | cr2 | do you have cgpsmapper ? |
| 21:57.10 | cr2 | well, can put that too :) |
| 21:57.20 | cr2 | you have ogr2ogr. |
| 21:57.21 | tsdogs | cr2: yes |
| 21:57.25 | tsdogs | I have it |
| 21:57.38 | tsdogs | also the mp |
| 21:57.41 | cr2 | postgres and postgis ? |
| 21:57.46 | cr2 | ok. |
| 21:57.53 | cr2 | postgis-1.2.1 ? |
| 21:58.10 | cr2 | well, you don't need the reprojection stuff. |
| 21:58.18 | cr2 | so 1.1.3 will work too. |
| 21:58.36 | tsdogs | cr2: but don't think its yours |
| 21:58.51 | tsdogs | have all |
| 21:59.17 | cr2 | ok. |
| 21:59.18 | tsdogs | postgis-1.2.0-4 |
| 21:59.20 | cr2 | if [ $# -ne 3 ] |
| 21:59.20 | cr2 | then |
| 21:59.21 | cr2 | <PROTECTED> |
| 21:59.21 | cr2 | <PROTECTED> |
| 21:59.21 | cr2 | fi |
| 21:59.31 | cr2 | so i have the 'import.sh' script. |
| 22:00.03 | cr2 | you choose the DB name and fips_number. |
| 22:00.10 | tsdogs | ok |
| 22:00.17 | cr2 | the mp file should have routing enabled. |
| 22:00.30 | tsdogs | that's what I miss I think |
| 22:00.32 | cr2 | generate wit gpsmapedit if it's not. |
| 22:00.43 | cr2 | autogenerate. |
| 22:00.51 | cr2 | that's what i've done. |
| 22:01.03 | cr2 | because the netgen output is too bad. |
| 22:01.15 | cr2 | can be resolved later :) |
| 22:01.36 | tsdogs | ok |
| 22:01.38 | cr2 | i'm just describing the general procedure. |
| 22:01.42 | cr2 | ok. |
| 22:01.51 | cr2 | so import.sh calls |
| 22:02.02 | cr2 | ./mp2mifR < ${MP} |
| 22:02.11 | cr2 | mp2mifR is an awk script. |
| 22:02.43 | cr2 | creates a mapinfo file. i could have done sql directly, but it's historical ;) |
| 22:02.54 | cr2 | can be viewed with qgis too. |
| 22:03.30 | cr2 | so it creates the segments.mi[fd] files |
| 22:04.00 | cr2 | with the road network and attributes (routable network, i.e. 'Road' in roadmap') |
| 22:04.01 | tsdogs | ok |
| 22:04.18 | cr2 | converted to shp |
| 22:04.20 | cr2 | ogr2ogr segments.shp segments.mif |
| 22:04.33 | cr2 | shp can be also seen with qgis |
| 22:04.49 | cr2 | so you can check if the network makes sense. |
| 22:04.57 | cr2 | and overlay it over other maps. |
| 22:05.10 | cr2 | vector or raster in any projection. |
| 22:05.30 | cr2 | shp2pgsql -s 4326 segments.shp segments | psql ${MAP} |
| 22:05.44 | cr2 | imports segments into the DB. |
| 22:06.03 | cr2 | ./cgpsmapper0076Astatic shp ${MP} |
| 22:06.27 | cr2 | creates the point, polyline and polygon shapefiles from .mp |
| 22:06.45 | cr2 | ogr2ogr -select "type,label" -where "type=20 or type=21 or type=24 or type=25 or type>=28" borders.shp line.shp |
| 22:07.09 | cr2 | 'borders' is for all non-road polyline layers. |
| 22:07.36 | cr2 | 'water' is for the polygons |
| 22:07.39 | cr2 | ogr2ogr -select "type,label" -where "type!=75" water.shp polygon.shp |
| 22:08.11 | cr2 | type=75 is some garmin weirdness aka background polygon. |
| 22:08.36 | cr2 | the main problem here happens when the number of polygon line segments exceed 65535 |
| 22:08.51 | cr2 | but it's not the case for your map afair. |
| 22:09.24 | cr2 | then 'import.sql' is called |
| 22:09.27 | cr2 | psql -d ${MAP} -f import.sql |
| 22:09.50 | cr2 | it does |
| 22:09.53 | cr2 | COPY streets FROM '/var/lib/pgsql/test/streets.map' DELIMITERS '|' CSV; |
| 22:10.12 | cr2 | so you must be poostgis in the /test subdir ;) |
| 22:10.27 | tsdogs | :) |
| 22:10.38 | cr2 | and some other silly DB maintenance. |
| 22:10.49 | cr2 | like |
| 22:10.51 | cr2 | ALTER TABLE segments rename COLUMN from_car_a TO from_car_allowed; |
| 22:11.19 | cr2 | because the shapefile column names are limited by 14 chars. |
| 22:11.43 | cr2 | can be solved to generating sql directly, instead of mif->shp->sql |
| 22:12.03 | cr2 | and an important thing |
| 22:12.05 | cr2 | UPDATE segments SET from_cross_time=length(the_geom)*400000/from_max_speed; |
| 22:12.21 | cr2 | time=distance/speed |
| 22:12.48 | tsdogs | what's the 400000? |
| 22:12.49 | cr2 | ok. that's all |
| 22:13.22 | cr2 | meters in the equator. |
| 22:13.29 | tsdogs | ok |
| 22:13.45 | cr2 | 2*M_PI*6371/360*1000 and so on. |
| 22:13.53 | tsdogs | :) |
| 22:14.07 | cr2 | assumes the speed in km/h, so 360 disappears. |
| 22:14.22 | cr2 | ./buildmap -v -m ${RDMDIR} -f PG ${FIPS} dbname=${MAP} |
| 22:14.39 | cr2 | and the index table |
| 22:14.41 | cr2 | ./buildus -m ${RDMDIR} |
| 22:14.52 | tsdogs | ok remember this :) |
| 22:15.20 | cr2 | buildmap usually only fails if there are too many segments. not in your case :) |
| 22:15.40 | tsdogs | k |
| 22:15.45 | cr2 | then the DB needs some manual interventions. |
| 22:15.59 | cr2 | and the buildmap needs to be repeated. |
| 22:16.19 | cr2 | ok, i'll copy the programs to the usual location. |
| 22:16.25 | tsdogs | ok. |
| 22:16.35 | cr2 | this can be simplified actually, but it's how it is now. |
| 22:17.00 | cr2 | if the polygon problem is solved, then it's 100% automated. |
| 22:17.58 | tsdogs | :) ok, shp files should be removed though. |
| 22:18.18 | tsdogs | but let's leave it for the future. |
| 22:18.22 | cr2 | well, yes. |
| 22:18.40 | cr2 | the script does not clean much ;) |
| 22:18.53 | cr2 | but i needed the intermediate steps for debugging. |
| 22:18.59 | cr2 | done. |
| 22:19.01 | tsdogs | ok |
| 22:19.08 | cr2 | try to run it, and let me know. |
| 22:19.26 | cr2 | tested on i386. |
| 22:19.43 | cr2 | hmm. and you also need the buildmap and buildus binaries in /test |
| 22:21.03 | tsdogs | cr2: ok |
| 22:21.27 | tsdogs | i get access denied on cgsmapper, should I use this one or can I use the one I have? |
| 22:21.34 | cr2 | hmm. |
| 22:22.21 | cr2 | changed. |
| 22:22.43 | tsdogs | ok |
| 22:22.53 | cr2 | you can use yours, but then you should change the script. |
| 22:23.23 | tsdogs | :) |
| 22:24.02 | tsdogs | ok, now I need the mp with routing |
| 22:29.40 | cr2 | i can give you my copy |
| 22:29.49 | cr2 | done with gpsmapedit. |
| 22:30.00 | tsdogs | cr2: ok |
| 22:30.19 | cr2 | but it's nothing more than 'generate routing nodes' running in wine. |
| 22:31.00 | cr2 | done. |
| 22:31.45 | tsdogs | :) |
| 22:32.15 | tsdogs | cr2: but it's very detailed :) |
| 22:34.32 | cr2 | what ? |
| 22:34.38 | tsdogs | the map |
| 22:34.58 | cr2 | hmm. the *.mp format is not very complex. |
| 22:35.28 | cr2 | remove the elements you don't need and regenerate the routing data. |
| 22:35.56 | cr2 | once you've done it for this map, you'll know how to do it for others. |
| 22:36.06 | cr2 | at least i know that this map works. |
| 22:36.51 | tsdogs | ok |
| 22:52.12 | cr2 | now it seems that you've fallen asleep :) |
| 22:52.21 | tsdogs | :) |
| 22:53.04 | tsdogs | I'm in the process of building the map. but have some truobles with postgis (I updated postgres, but postgis was not :( |
| 22:53.59 | tsdogs | also some paths are different |
| 22:54.29 | Kevin2 | pH5: I'm okay with that change to htc-egpio. We can also remove the printk - it is just for debugging anyway. |
| 22:59.28 | psokolovsky_ | Kevin2: Hi! do you remember out of head the name of wince func to change objstore ratio? |
| 23:01.45 | cr2 | psokolovsky_: extern "C" DWORD SetSystemMemoryDivision( DWORD dwStorePages ); ? |
| 23:02.01 | cr2 | but it does not work on wince5. |
| 23:03.08 | psokolovsky_ | cr2: yeah, that thing! |
| 23:03.19 | psokolovsky_ | cr2: why it doesn't - no objstore? ;-) |
| 23:04.07 | cr2 | it's changed dynamically by the wince kernel. |
| 23:04.24 | cr2 | it was implemented in gnu-haret, and always failed afair ;) |
| 23:05.44 | psokolovsky_ | hm, weird |
| 23:05.46 | cr2 | you can do crazy things like it's done by GetKPhys(), but the chances that you'll crash the system are very high. |
| 23:06.21 | psokolovsky_ | cr2: I see the issues that booting ~23mb initrd on hx4700 after reboot always causes "not enough memory" dialog and fails eventually. |
| 23:06.32 | psokolovsky_ | it works form the second try though... |
| 23:06.46 | psokolovsky_ | so, I was thinking on moving the slider in code ;-) |
| 23:08.32 | cr2 | :) |
| 23:08.45 | cr2 | if you are on wince2003 then it should work. |
| 23:09.01 | psokolovsky_ | ok, thanks for info, will try to play with it |
| 23:10.48 | tsdogs | cr2: will finish the test tomorrow. I need to build postgis again. :( |
| 23:10.52 | tsdogs | good night |
| 23:10.57 | cr2 | good night |
| 23:30.37 | *** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com) |
| 23:33.38 | cr2 | goxboxlive: how do i crosscompile an app for qtopia ? |
| 23:34.48 | goxboxlive | cr2, cant you use qmake? |
| 23:35.27 | goxboxlive | If you check a Makefile in OPieII source you can use that one, just change the name of some of the paths and the *pro file. |
| 23:37.30 | cr2 | the program is portable, so .pro will not be used. |
| 23:40.06 | cr2 | goxboxlive: where is the crosscompiled qt libs ? and include ? |
| 23:40.38 | cr2 | ok. found. |
| 23:40.49 | goxboxlive | it's getting late. night |
