| 00:05.35 | cr2_ | tsdogs: i think it depends on your goal. |
| 00:06.20 | tsdogs | cr2_: goal is make it speedy and use not that much memory |
| 00:06.22 | tsdogs | :) |
| 00:06.37 | cr2_ | or course mmap'ing the ozf2 file + caching tiles in a offscreen pixmap is the fastest solution. |
| 00:07.08 | cr2_ | but using RasterIO is portable and will work everywhere. |
| 00:07.20 | tsdogs | is creating/reading ozf2 files dificult? ok. |
| 00:07.37 | cr2_ | for reading - you have the code. |
| 00:08.07 | cr2_ | whic uses mmap. just don't write out the tiff and remove the respective code. |
| 00:08.10 | tsdogs | :/ does qgis read ozf2 ? |
| 00:08.24 | cr2_ | you may even overload QImage.load() |
| 00:08.26 | cr2_ | no. |
| 00:09.16 | *** join/#htc-linux donn (n=donn@p5085D015.dip.t-dialin.net) |
| 00:09.20 | cr2_ | but then you need to convert 8bit palette -> 16bit pixmap, and handle scaling yourself. |
| 00:09.21 | donn | hi :) |
| 00:09.25 | donn | hi cr2 |
| 00:09.26 | donn | :) |
| 00:09.31 | cr2_ | hi donn |
| 00:09.33 | donn | i have the bits of the sd card |
| 00:09.38 | donn | thanks to kevin2 |
| 00:09.43 | cr2_ | ok. |
| 00:10.07 | donn | He wrote: Okay. So GPIO 34 and GPIO 126 are your SD card detect and SD power gpios. (Not sure which is which, but cr2 can probably figure that out.) |
| 00:10.17 | cr2_ | guys, it's really late now. |
| 00:10.24 | donn | @cr2: no problem |
| 00:10.26 | tsdogs | cr2_: ok, it's late, maybe tomorrow I'll test the RasterIO. good night |
| 00:10.30 | cr2_ | which appears first when you insert hte card ? |
| 00:10.32 | donn | yes :9 |
| 00:10.47 | donn | at first it was inserted |
| 00:11.07 | donn | we can do it tomorrow |
| 00:11.09 | cr2_ | which gpio ? |
| 00:11.12 | cr2_ | ok. |
| 00:11.14 | donn | or on another day.. |
| 00:11.15 | cr2_ | good night. |
| 00:11.16 | donn | 34 and 126 |
| 00:11.41 | donn | good night :) |
| 00:11.42 | donn | cu |
| 00:11.48 | *** join/#htc-linux donn220 (n=donn@p5085D015.dip.t-dialin.net) |
| 00:11.57 | cr2_ | tsdogs: you should coordinate it with kiozen |
| 00:12.11 | tsdogs | cr2_: ok |
| 00:12.30 | tsdogs | cr2_: http://nopaste.php-q.net/300690 for insert |
| 00:12.34 | donn | okay, cu later |
| 00:12.48 | tsdogs | http://nopaste.php-q.net/300691 for remove (from donn) |
| 00:12.58 | donn | ah, thanks tsdogs :) |
| 00:13.12 | donn | thats great |
| 00:13.17 | tsdogs | night |
| 00:13.19 | donn | what would I do without you all |
| 00:13.21 | donn | night :) |
| 00:13.29 | donn | set in favorites for tomorrow |
| 00:14.25 | donn | cu |
| 00:18.48 | *** join/#htc-linux walshieau (n=walshiea@ppp92-1.lns1.mel3.internode.on.net) |
| 04:15.49 | *** join/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk) |
| 05:09.48 | *** join/#htc-linux sadeness_ (n=LamersIn@nat1.rlan.ru) |
| 05:57.40 | *** join/#htc-linux lpotter (n=ljp@c210-49-127-23.rochd1.qld.optusnet.com.au) |
| 06:08.38 | *** join/#htc-linux rob_w|mis (n=rob_w@p549B9DC7.dip0.t-ipconnect.de) |
| 06:17.59 | *** join/#htc-linux dion (n=dion@inhex.net) |
| 06:21.58 | *** join/#htc-linux _law_ (n=law@mail.stiftadmont.at) |
| 06:50.28 | *** join/#htc-linux sadeness_ (n=LamersIn@nat2.rlan.ru) |
| 07:27.34 | *** join/#htc-linux pleemans (n=peter@gast039a.ugent.be) |
| 07:48.10 | *** join/#htc-linux donn (n=donn@p5085FC01.dip.t-dialin.net) |
| 07:48.12 | donn | ** |
| 07:48.56 | donn | Hi ladys :) |
| 07:57.27 | *** join/#htc-linux rob_w|mis (n=rob_w@p549B9DC7.dip0.t-ipconnect.de) |
| 08:31.59 | *** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com) |
| 08:36.42 | *** join/#htc-linux jeanseb (n=jeanseb@ns.popipo.fr) |
| 08:36.46 | jeanseb | hi |
| 08:55.46 | goxboxlive | jeanseb, hi |
| 08:55.53 | goxboxlive | cr2_, are you awake? |
| 09:04.47 | jeanseb | ls |
| 09:04.53 | jeanseb | oups |
| 09:28.49 | *** join/#htc-linux sadeness_ (n=LamersIn@nat2.rlan.ru) |
| 09:56.33 | *** join/#htc-linux kiozen (n=oeichler@p54922CA9.dip0.t-ipconnect.de) |
| 10:09.40 | *** join/#htc-linux psokolovsky_ (n=psokolov@82.193.98.14) |
| 10:09.51 | *** join/#htc-linux tsdogs (n=twostupi@84.18.144.39) |
| 10:11.15 | *** join/#htc-linux bd2 (n=cbou@85.21.88.2) |
| 10:14.47 | cr2_ | goxboxlive: yes. |
| 10:38.36 | *** join/#htc-linux rob_w|mis (n=rob_w@p549B9DC7.dip0.t-ipconnect.de) |
| 11:17.28 | *** join/#htc-linux Ranpha1 (n=Ranpha@hlm-n-73ed.adsl.wanadoo.nl) |
| 11:17.58 | *** join/#htc-linux gundam (n=gundam@slackware.it/staff/gundam) |
| 11:21.01 | *** join/#htc-linux gundam (n=gundam@slackware.it/staff/gundam) |
| 11:38.50 | *** join/#htc-linux dion (n=dion@inhex.net) |
| 11:41.57 | *** join/#htc-linux BabelO (n=Miranda@faim1.itesoft.com) |
| 11:42.01 | BabelO | hi |
| 12:04.56 | *** join/#htc-linux wireddd (n=wired@97-81-78-105.dhcp.athn.ga.charter.com) |
| 12:27.22 | kiozen | tsdogs ? |
| 12:30.27 | BabelO | hi kiozen |
| 12:30.32 | kiozen | hi |
| 12:47.30 | *** join/#htc-linux __law__ (n=law@mail.stiftadmont.at) |
| 12:55.22 | tsdogs | kiozen: hi |
| 12:55.42 | kiozen | hi, cr2 is right we should sync our efforts |
| 12:56.02 | kiozen | I have got 2 goals |
| 12:56.20 | kiozen | the first is to make QLandkarte switch to a rastermap |
| 12:56.26 | tsdogs | yep. |
| 12:56.31 | kiozen | on close zoom levels |
| 12:56.35 | tsdogs | ok. |
| 12:57.03 | kiozen | the 2nd one is a leightweight canvas for vector and raster maps |
| 12:57.20 | kiozen | for the futur after my Garmin GPSr :) |
| 12:57.38 | tsdogs | :) ok. |
| 12:57.48 | kiozen | your turn |
| 12:58.02 | tsdogs | <PROTECTED> |
| 12:58.17 | kiozen | same as me :) |
| 12:58.37 | tsdogs | then will try to migrate roadmap to use RasterIO functions for map drawing. |
| 12:58.42 | kiozen | just started 5 min ago |
| 12:59.08 | tsdogs | roadmap is already using a canvas object, maybe you should give it a look |
| 12:59.19 | kiozen | doesn't roadmap work on vector maps? |
| 12:59.20 | tsdogs | stc/qt4/qt_canvas.cc |
| 12:59.27 | tsdogs | s/stc/src/ |
| 13:00.19 | cr2_ | hi guys. roadmap uses the fraction 3/2 (2/3) for zooming with the base value of 20. imho it will lead to a pseudorandom walk between 2^1 and 2^16 in the zoom factor, but we can easily switch to 2 (1/2) and 16(or 32) |
| 13:00.45 | kiozen | brainoverload, booom |
| 13:00.51 | tsdogs | :) |
| 13:00.53 | kiozen | what are you talking abaout |
| 13:01.41 | cr2_ | i'm trying to understand how roadmap data/render relates to garmin's. |
| 13:03.00 | cr2_ | the coordinate system is different too. (int) (lat*10^6) |
| 13:03.18 | cr2_ | (int) (long*10^6) |
| 13:03.45 | kiozen | it's decimal? |
| 13:03.55 | cr2_ | tsdogs: qt4 compilation failed on linuxppc. |
| 13:04.04 | cr2_ | kiozen: yes. |
| 13:04.23 | tsdogs | cr2_: qt4 or roadmap_qt4 ? |
| 13:04.24 | cr2_ | tsdogs: with some obscure objdump error (!) |
| 13:04.31 | cr2_ | tsdogs: qt4 |
| 13:04.47 | tsdogs | trolltech trouble I think. |
| 13:04.52 | cr2_ | but i've managed to force it to continue manually. |
| 13:04.58 | tsdogs | :) |
| 13:05.16 | cr2_ | qtcore is compiled. doing qtgui now. |
| 13:05.44 | cr2_ | kiozen: does qlandkarte work on big-endian ? :) |
| 13:05.49 | kiozen | no |
| 13:06.11 | cr2_ | ok. |
| 13:06.37 | cr2_ | roadmap mmaps the maps, so it needs byteswapped maps ;) |
| 13:07.42 | Kevin2 | cr2_: Can you run "loadfunc coredll _wcsicmp" on your jornada? -- psokolovsky_ wanted to add it to haret on startup, and there is some question if it will break 2.10 devices. |
| 13:07.47 | kiozen | using qt's QDataStream would solve the problem |
| 13:07.58 | kiozen | but it's slow |
| 13:08.14 | cr2_ | Kevin2: i'll check later. |
| 13:08.26 | psokolovsky_ | Hi! |
| 13:08.34 | psokolovsky_ | cr2_, Kevin2: _wcsnicmp to be exact |
| 13:09.14 | cr2_ | psokolovsky_: run 'nm' on coredll.lib from evc3 |
| 13:10.01 | psokolovsky_ | cr2_: damn, I'm switching boxes, and some time ago killed evc install ;-( |
| 13:10.38 | cr2_ | i'll look. |
| 13:10.42 | psokolovsky_ | thanks |
| 13:11.15 | cr2_ | i#ve copied the libs and headers to linux, so i can grep/nm them. |
| 13:11.49 | Kevin2 | I'll be back later. |
| 13:12.10 | cr2_ | 0000000000000000 T _wcsicmp |
| 13:12.55 | cr2_ | 0000000000000000 T _wcsnicmp |
| 13:14.22 | cr2_ | strings 820.us | grep _wcsnicmp |
| 13:14.32 | cr2_ | _wcsnicmp |
| 13:14.49 | cr2_ | hmm. the chances are very good that it works. |
| 13:18.45 | psokolovsky_ | cr2_: nice, thanks |
| 13:20.28 | kiozen | cr2: if I get the dilema right roadmap has an open format and routing but no good / many maps |
| 13:20.41 | kiozen | and Garmin we do not know routing but there are maps |
| 13:20.44 | kiozen | right? |
| 13:24.53 | cr2_ | right. but it's possible to convert mp to roadmap. |
| 13:26.14 | kiozen | who is MP once again? |
| 13:26.26 | cr2_ | roadmap does not have POI renderer and a murky code for POIs. and the polygon storage/rendering needs some changes. should be peanuts for you :) |
| 13:26.38 | cr2_ | format used by cgpsmapper |
| 13:27.19 | cr2_ | text format from http://mapcenter.cgpsmapper.com |
| 13:27.20 | kiozen | thus you want to change the format completely and make a new one? ;) |
| 13:27.45 | cr2_ | no. |
| 13:28.02 | kiozen | it sounds like that |
| 13:28.04 | cr2_ | i'd like to 'extend and embrace' it |
| 13:28.19 | cr2_ | which will make it binary incompatible. |
| 13:28.42 | cr2_ | the road network (graph) in roadmap is in a separate file |
| 13:28.53 | cr2_ | which can be created with other tools. |
| 13:28.57 | cr2_ | or dumped. |
| 13:28.57 | kiozen | sorry phone |
| 13:29.53 | tsdogs | cr2_: for what I understood you wanted to drop the rdm, and use geotiff for map drawing. then use routing data for navigation |
| 13:30.21 | cr2_ | tsdogs: you get it wrong. |
| 13:31.15 | cr2_ | i'd like to extend rdm to match img in capabilities. having the raster overlay as a background pixmap is just a (very) nice extension. |
| 13:32.08 | tsdogs | ok, but then drawing the road would overlay the map. |
| 13:32.08 | cr2_ | so if you have a raster map for an area, and no vector/road graph you can just use it. as in oziCE. |
| 13:32.21 | cr2_ | checkbox() |
| 13:32.38 | cr2_ | or 'schema' entry. |
| 13:32.47 | tsdogs | ho ok, |
| 13:32.49 | cr2_ | to disable rendering of layers. |
| 13:33.07 | cr2_ | you don't like/want/depends on the zoom/etc. |
| 13:33.10 | tsdogs | so we use 3 "layers" raster map, rdm, routing |
| 13:33.33 | cr2_ | yes, even a bit more. |
| 13:33.58 | cr2_ | rdm subdivides into point,line,polygon layers. |
| 13:34.13 | cr2_ | points are cities, POIs, etc. |
| 13:34.33 | cr2_ | lines are roads, rivers, powerlines, etc. |
| 13:34.45 | tsdogs | yes ok, rdm is layered also. |
| 13:34.50 | cr2_ | polygons are lakes,seas, forests, etc. |
| 13:35.00 | cr2_ | yes. |
| 13:35.12 | cr2_ | so you can attach a style to a layer. |
| 13:35.26 | cr2_ | font,type,size,brush, etc. |
| 13:36.01 | tsdogs | ok. |
| 13:36.08 | cr2_ | you can use postgis as a layered database, it's just not a solution for a pda :) |
| 13:36.14 | cr2_ | we need something compact. |
| 13:36.24 | cr2_ | and fast. |
| 13:38.29 | tsdogs | so rdm is a good solution (maybe with some zip extension too) |
| 13:39.09 | cr2_ | yes, but you know its current limitations. |
| 13:39.45 | tsdogs | ok, but you know it better :) it is extendable right? |
| 13:39.51 | cr2_ | yes. |
| 13:41.18 | cr2_ | bbl. |
| 13:41.31 | tsdogs | ok |
| 13:42.01 | jeanseb | cr2_: is there any fdisk that could be run on the pda to part the sd ram ? |
| 13:42.06 | jeanseb | sd card |
| 13:42.10 | jeanseb | sorry |
| 13:47.59 | cr2_ | jeanseb: afaik no. |
| 13:48.26 | *** join/#htc-linux _law_ (n=law@mail.stiftadmont.at) |
| 13:49.58 | cr2_ | tsdogs: you have already extended the roadmap "gui" api by some functions and parameters. can we document the changes and extend it further to support font names, sizes, colors, line styles and brushes ? |
| 13:50.53 | tsdogs | cr2_: yes, but I'd need to know where to call them ;) |
| 13:51.34 | cr2_ | right now we have colors and widths for lines. |
| 13:51.38 | cr2_ | right ? |
| 13:52.36 | tsdogs | yes. |
| 13:53.07 | tsdogs | Actually we only have Pen/Brush color/width |
| 13:53.29 | cr2_ | ok, so the linestyle is an obvious extension. |
| 13:53.35 | cr2_ | brush ? |
| 13:53.44 | cr2_ | ah, brush color. |
| 13:53.52 | tsdogs | We will need to have some more complex structure that describes the layer prop, inthead of only QPen object |
| 13:53.57 | tsdogs | :) |
| 13:54.10 | cr2_ | pixmap brush should be nice to have. |
| 13:54.14 | tsdogs | Well, main object it's Pen anyway |
| 13:54.24 | tsdogs | qt4 handles it |
| 13:54.28 | cr2_ | yes, but it should be done at the roadmap_canvas.h level |
| 13:54.48 | *** join/#htc-linux rob_w|mib (n=rob_w@p549B9DC7.dip0.t-ipconnect.de) |
| 13:54.51 | tsdogs | also qt4 has more fill options (gradient and so on) |
| 13:55.20 | tsdogs | we would have to drop qt3 compat though, or simply implement empty functions for it |
| 13:55.28 | cr2_ | well, but to call it from the main code in a portable way it should be in roadmap_canvas.h |
| 13:55.33 | cr2_ | empty functions. |
| 13:55.44 | tsdogs | :) |
| 13:55.49 | cr2_ | raster tiles too, btw. |
| 13:56.07 | tsdogs | ok. |
| 13:56.16 | cr2_ | it's then up to the window system how to implement them. |
| 13:56.21 | tsdogs | raster might be a little more dificult |
| 13:56.26 | tsdogs | ok |
| 13:56.57 | tsdogs | but I think it can be done. |
| 13:57.23 | cr2_ | i thinks so. |
| 13:57.39 | cr2_ | afair gdal can be compiled for wince :) |
| 13:58.11 | tsdogs | ok, I'll experiment a bit with RasterIO, and then will start including (yes, qgis uses it) code and implement more functions. |
| 13:58.25 | jeanseb | thxs |
| 13:58.27 | tsdogs | I'll try not to break the code :) |
| 13:58.33 | cr2_ | ok. |
| 14:42.31 | kiozen | cr2_ back from phone |
| 14:42.57 | kiozen | thus if I understand you right you want to keep the navigation stuff |
| 14:43.12 | kiozen | of roadmap / rmp |
| 14:43.52 | kiozen | but you want to extend the render capabilities of |
| 14:44.03 | kiozen | ploylines / polygon / pois |
| 14:44.57 | kiozen | s/rmp/rdm |
| 14:45.53 | kiozen | it would need a conversion tool from rdm,mp,... -> new rdm |
| 14:50.32 | cr2_ | kiozen: rdm-> new rdm is not needed. the rest is ok. |
| 14:52.02 | kiozen | if you are just interested in the routing info of rdm, you are right |
| 14:52.25 | kiozen | but if you want to use the graphical information, too |
| 14:52.56 | kiozen | you need two different file parser or a converter, don't you? |
| 14:55.40 | *** join/#htc-linux donn (n=donn@p5085FC01.dip.t-dialin.net) |
| 14:55.49 | donn | hi@all :9 |
| 14:56.33 | cr2_ | kiozen: the converter is already there, it's just the format has some deficiency. |
| 14:56.38 | cr2_ | hi donn |
| 14:57.08 | cr2_ | kiozen: and some things (like polygon labels) are stored in the format, but never rendered. |
| 14:58.17 | kiozen | ok, so you want to adapt the conversion tool |
| 14:58.54 | cr2_ | i'll do the conversion tool myself. |
| 14:59.52 | cr2_ | i need to remove the 65536 polygon line segments limit per file. and add rendering of the polygon labels. |
| 15:00.35 | kiozen | ok |
| 15:00.38 | cr2_ | find out what's wrong with the POI stored table, and add rendering POI labels (and icons). |
| 15:01.08 | kiozen | ok |
| 15:01.14 | cr2_ | then it will be on par with the qlandkarte. |
| 15:01.58 | kiozen | ok - so what's my part you need me for :) |
| 15:02.24 | cr2_ | and img will be irrelevant, and we can use qlandkarte to use rdm. |
| 15:02.36 | cr2_ | let's think about it. |
| 15:03.02 | cr2_ | grab the program from xanadux CVS and compile it first. |
| 15:03.10 | cr2_ | roadmap_editor |
| 15:03.34 | donn | @cr2: so, i am ready to take off :-p |
| 15:03.36 | kiozen | ok not the one from the original place - it's hell |
| 15:03.55 | cr2_ | make DESKTOP=QT4 SHAPEFILES=NO QTDIR=/where/is/your/qt4 MODE=DEBUG |
| 15:03.59 | tsdogs | kiozen: it's still hell :) |
| 15:04.01 | cr2_ | SSD=YES |
| 15:04.40 | cr2_ | works for me on x86, amd64, alpha and ppc. |
| 15:04.56 | cr2_ | don't know about arm yet. but should work too. |
| 15:05.12 | tsdogs | cr2_: last time I builded it worked :) |
| 15:05.22 | cr2_ | ok. |
| 15:05.28 | donn | so, I have looked for the gpios bits |
| 15:05.42 | donn | is there a tutorial what to do now with the linux files? |
| 15:05.55 | donn | well, i have still running a functionable linux distribution |
| 15:05.56 | cr2_ | and you need a map. unless i'll fix the sql escapes, so mp2rdm will work. |
| 15:06.13 | donn | so i think i have something to configure in the linux.ha file (default.txt) right? |
| 15:06.24 | cr2_ | donn: you need to hack at much lower level than a distribution. |
| 15:06.45 | donn | eh.means? |
| 15:07.09 | cr2_ | this is complete a mess now. |
| 15:07.17 | donn | hmm..k |
| 15:07.24 | cr2_ | all devices pretend that they are "tornado". |
| 15:07.24 | donn | but how can I have access to the sd card? |
| 15:07.47 | cr2_ | so you should get a compilable tornado kernel. |
| 15:08.01 | cr2_ | and change its SD gpio settings. |
| 15:08.14 | cr2_ | recompile, and use this kernel instead of the old one. |
| 15:08.23 | donn | hmm.and where do i get it? |
| 15:08.44 | cr2_ | read linwizard and typhoon sites. |
| 15:08.57 | cr2_ | i'm not quite sure myself. |
| 15:08.57 | donn | eh.. :-x kk |
| 15:09.00 | donn | ok : |
| 15:09.01 | donn | :) |
| 15:09.10 | *** join/#htc-linux \bash (n=sps@240.pool85-53-13.dynamic.orange.es) |
| 15:09.51 | cr2_ | well, it's the least developed PDA port. |
| 15:12.49 | \bash | Hi guys, I have been doing some beta test with the hermes these days. I have put the contents of the readable /sys files in http://213.27.196.132/cat.txt I have found interesting things when enabling usbhost ohci |
| 15:12.51 | donn | hmm... i have a special "haret.exe" and a tar.gz file with the sources |
| 15:13.04 | donn | do i have to compile the tar.gz? |
| 15:13.10 | donn | the content of it? |
| 15:14.49 | donn | i have devC++ |
| 15:15.01 | donn | its very difficult :-P |
| 15:15.36 | donn | the question is, what is the difference between the tar.gz and the exe file |
| 15:15.39 | cr2_ | \bash: lsusb ? |
| 15:15.58 | \bash | havent tried yet |
| 15:16.04 | \bash | i run on a minimal initrd |
| 15:16.05 | cr2_ | donn: you don't need to do anything with .exe |
| 15:16.10 | donn | k |
| 15:16.14 | donn | i have to recompile it? |
| 15:16.23 | cr2_ | \bash: can you extend inird ? |
| 15:16.27 | \bash | the file cat.txt is now ready |
| 15:16.46 | cr2_ | donn: recompile .tar.gz with some changes. |
| 15:16.55 | \bash | I'm trying to mount an nfs share |
| 15:16.55 | donn | the question is, where i have to enter the gpio numbers |
| 15:17.01 | donn | hm.k |
| 15:17.12 | \bash | anyway |
| 15:17.19 | donn | the result of the recompile should be a "haret.exe" file, or not? |
| 15:17.24 | cr2_ | donn: i'm not quite sure. |
| 15:17.33 | \bash | File: ./devices/platform/s3c2410-ohci/usb1/1-1/product |
| 15:17.33 | donn | do you know, who could know it? |
| 15:17.33 | \bash | Qualcomm CDMA Technologies MSM |
| 15:17.33 | \bash | File: ./devices/platform/s3c2410-ohci/usb1/1-1/manufacturer |
| 15:17.33 | \bash | Qualcomm, Incorporated |
| 15:17.33 | \bash | File: ./devices/platform/s3c2410-ohci/usb1/1-1/quirks |
| 15:17.42 | cr2_ | \bash: is the usbnet still dog-slow as before. |
| 15:17.58 | \bash | for me is slow |
| 15:18.15 | \bash | I have been able to work opening two telnets |
| 15:18.22 | cr2_ | \bash: that looks good. exactly like on the universal. |
| 15:18.31 | cr2_ | ok. |
| 15:18.45 | \bash | i ping my host from one telent |
| 15:18.48 | \bash | telnet |
| 15:18.56 | \bash | and write bash commands in the other |
| 15:19.00 | cr2_ | ok, so it's still buggy. |
| 15:19.21 | \bash | buggy but at least doesn't crash :D |
| 15:19.43 | \bash | I have read all the /sys files |
| 15:20.03 | \bash | their contents are in the http address I wrote |
| 15:20.33 | \bash | File: ./block/mtdblock0/slaves |
| 15:20.34 | \bash | File: ./block/mtdblock0/holders |
| 15:20.34 | \bash | File: ./block/mtdblock0/subsystem |
| 15:20.34 | \bash | File: ./block/mtdblock0/stat |
| 15:20.34 | \bash | <PROTECTED> |
| 15:20.34 | \bash | <PROTECTED> |
| 15:20.36 | \bash | File: ./block/mtdblock0/size |
| 15:20.40 | \bash | 262144 |
| 15:20.42 | \bash | File: ./block/mtdblock0/removable |
| 15:20.44 | \bash | 0 |
| 15:20.46 | \bash | File: ./block/mtdblock0/range |
| 15:20.48 | \bash | 1 |
| 15:20.50 | \bash | File: ./block/mtdblock0/dev |
| 15:20.52 | \bash | 31:0 |
| 15:20.54 | \bash | File: ./block/mtdblock0/uevent |
| 15:21.15 | cr2_ | \bash: what are you going to do next ? |
| 15:21.30 | \bash | I don't know, I would like to help |
| 15:21.41 | \bash | but don't know where to start |
| 15:21.46 | cr2_ | do you know the NAND partitioning ? |
| 15:21.58 | \bash | nope |
| 15:21.58 | cr2_ | can you 'dd' from NAND now ? |
| 15:22.38 | *** join/#htc-linux jeanseb (n=jeanseb@gazypan.dyndns.org) |
| 15:22.39 | cr2_ | hmm. the current partitioning is not good. |
| 15:23.01 | \bash | I have made mknod flash b 31 0; cat /flash |
| 15:23.07 | \bash | but I can't read |
| 15:23.24 | \bash | I made cvs up this morning |
| 15:23.29 | cr2_ | ok. |
| 15:23.48 | cr2_ | afair we have mapped only 1MB at the top. |
| 15:24.07 | cr2_ | you can try to change the map in htchermes.c |
| 15:24.28 | cr2_ | to convert 128MB in partitions. |
| 15:24.49 | kiozen | cr2: is dglib supposed to be part of roadmap or grass |
| 15:25.09 | \bash | im taking a look |
| 15:25.27 | cr2_ | \bash: here at the bootom are some hints http://wiki.xda-developers.com/index.php?pagename=HermesMemoryMap |
| 15:25.45 | donn | @ all: did someone edit the sources of Linwizard and add his GPIOS bits to it? |
| 15:26.05 | cr2_ | kiozen: it's taken from grass and included into roadmap_editor. cd dglib; make -f Makefile.alone |
| 15:26.16 | kiozen | thx |
| 15:26.23 | \bash | <PROTECTED> |
| 15:26.34 | cr2_ | \bash: at the bottom. |
| 15:26.51 | cr2_ | NAND Flash partitioning |
| 15:27.12 | \bash | tnks! |
| 15:27.17 | cr2_ | try to remove 0x50000000 from the "fake" numbers. |
| 15:28.16 | \bash | sorry, To many things |
| 15:28.52 | \bash | static struct s3c2410_nand_set htchermes_nand_sets[] = { |
| 15:28.52 | \bash | <PROTECTED> |
| 15:28.52 | \bash | <PROTECTED> |
| 15:28.53 | \bash | <PROTECTED> |
| 15:28.53 | \bash | <PROTECTED> |
| 15:28.53 | \bash | <PROTECTED> |
| 15:28.55 | \bash | <PROTECTED> |
| 15:28.57 | \bash | }; |
| 15:29.12 | \bash | You mean I should touch this struct? |
| 15:29.31 | cr2_ | i think no. |
| 15:29.39 | cr2_ | only htchermes_nand_part |
| 15:30.06 | donn | @cr2: do you someone who knows what to do in my case? |
| 15:30.59 | \bash | <PROTECTED> |
| 15:30.59 | \bash | <PROTECTED> |
| 15:30.59 | \bash | <PROTECTED> |
| 15:30.59 | \bash | <PROTECTED> |
| 15:31.07 | \bash | ok, I thunk I get it |
| 15:31.10 | \bash | think |
| 15:31.33 | cr2_ | donn: linwizard is silent. but you can try to contact the typhoon people. http://vivien.chappelier.free.fr/typhoon/index.html |
| 15:31.54 | donn | ah, thank you very much :) |
| 15:32.11 | cr2_ | donn: they have the same cpu and kernel. |
| 15:32.32 | cr2_ | \bash: modify .offset and .size |
| 15:32.35 | donn | you mean, the typhoon "distribution" will work on my pda? |
| 15:32.47 | cr2_ | \bash: for each partition. |
| 15:32.52 | \bash | i'm on it |
| 15:33.02 | \bash | I will copy paste as I finish |
| 15:33.03 | cr2_ | donn: no, but it's the best starting point. |
| 15:33.10 | donn | hm :-/ |
| 15:33.19 | cr2_ | \bash: use pastbin. too much spam otherwise. |
| 15:33.23 | donn | okay, seems that all work was for the cat :-P |
| 15:33.43 | \bash | :D |
| 15:33.47 | \bash | pastbin? |
| 15:33.54 | \bash | what is it? |
| 15:34.12 | cr2_ | donn: blueangel kernel was developed 2 years, before it reached the current state. |
| 15:34.22 | cr2_ | \bash: http://pastebin.ca |
| 15:35.00 | donn | the problem is, nobody can say how much work it is to run a stupid sd card on the linux pda |
| 15:36.23 | psokolovsky_ | donn: it would take ~1mon to do reasonable kernel port for a new pda, if done right. htcsable proven that. the keyword is "done right" though... |
| 15:38.47 | cr2_ | psokolovsky_: are you planning to merge omap patches ? |
| 15:39.24 | psokolovsky_ | cr2_: well, I don't have omap patches ;-). if you'd like to do that, that might be good idea ;-) |
| 15:39.47 | cr2_ | and the keyword is PDA here. |
| 15:40.09 | cr2_ | psokolovsky_: i think it's a 3.1MB compressed patch for 2.6.21 |
| 15:40.36 | psokolovsky_ | cr2_: hm, well, that's too much ;-) |
| 15:41.04 | psokolovsky_ | cr2_: what's wrong with selecting only what's needed? and what's wrong with omap support already in mainline? |
| 15:41.29 | cr2_ | device support. i'll look for a link. |
| 15:41.40 | \bash | http://pastebin.ca/564342 |
| 15:41.42 | \bash | done |
| 15:42.06 | psokolovsky_ | cr2_: well, I know there's a separate tree for linux-omap. the q, if we need their whole stuff... |
| 15:42.55 | psokolovsky_ | cr2_: I personally have palmte2 or something, which I'm not yet started to hack on yet, so I'd be interested in omap too with very background priority... |
| 15:43.36 | \bash | kernel crash :) |
| 15:43.38 | cr2_ | https://www.muru.com/linux/omap |
| 15:43.57 | cr2_ | the patches are at the bottom. |
| 15:44.38 | cr2_ | \bash: which one ? |
| 15:45.02 | \bash | I did the changes to the struct |
| 15:45.17 | cr2_ | ok. |
| 15:45.20 | \bash | http://pastebin.ca/564342 |
| 15:45.52 | \bash | I'm going to remove the unnecesary (risky) parts (ipl and so on) |
| 15:45.55 | cr2_ | \bash: i've told you to subtract 0x50000000 |
| 15:46.05 | cr2_ | from the addresses. |
| 15:46.05 | \bash | To all of them? |
| 15:46.13 | cr2_ | of course it will crash . |
| 15:46.20 | cr2_ | .offset = 0x50000000, |
| 15:46.23 | cr2_ | should be |
| 15:46.27 | cr2_ | .offset = 0x0 |
| 15:46.38 | \bash | ok |
| 15:46.39 | \bash | done |
| 15:47.04 | cr2_ | from all. |
| 15:47.14 | psokolovsky_ | cr2_: I'd guess thsi stuff would be even more "official": http://source.mvista.com/git/gitweb.cgi?p=linux-omap-2.6.git;a=summary |
| 15:48.28 | cr2_ | psokolovsky_: may be. but htc tornado/typhoon uses the patch above. so it's known to be working. |
| 15:51.06 | donn | @cr2: its not possible to set the results from the bits into the defaul.txt, or? |
| 15:52.14 | \bash | I have changed it but still crashes on boot |
| 15:52.24 | \bash | see pastebin |
| 15:54.43 | cr2_ | donn: no. |
| 15:55.11 | cr2_ | \bash: the link ? |
| 15:55.44 | \bash | http://pastebin.ca/564373 |
| 15:56.39 | \bash | I have leaved only extrom |
| 15:56.45 | \bash | doesn't crash by now |
| 15:56.57 | \bash | going to telenet |
| 15:57.01 | cr2_ | \bash: ok. maybe the size is wrong ? |
| 15:57.27 | \bash | 64 * 1024 * 1024 |
| 15:57.28 | \bash | 67108864 |
| 15:57.58 | cr2_ | the page says ~64 |
| 15:58.30 | cr2_ | still strange. |
| 15:58.55 | cr2_ | tsdogs: 17:52:8 s## roadmap_dbread.c, line 108 illegal offset 469762048 in database usdir |
| 15:59.48 | cr2_ | hehe. |
| 15:59.51 | cr2_ | 17:57:41 s## roadmap_locator.c, line 141 cannot open directory database (usdir) |
| 15:59.56 | tsdogs | cr2_: it seems something wrong with the usdir.rdm :) |
| 15:59.57 | cr2_ | if it is removed. |
| 16:00.02 | cr2_ | linuxppc |
| 16:00.05 | tsdogs | lol |
| 16:00.18 | cr2_ | qt4 version. |
| 16:01.14 | cr2_ | i think building the global index file (usdir.rdm) should be built-in |
| 16:01.57 | kiozen | :) found out 5 min ago |
| 16:02.05 | kiozen | how do I see my maps |
| 16:02.22 | kiozen | I put them into /usr/local/share/roadmap/maps |
| 16:02.24 | donn | does somebody knows a Pocket PC where a linuxkernel runs without any problems? |
| 16:02.38 | kiozen | but nothing to see |
| 16:02.40 | donn | i want to resolve my problem, but its an informal question |
| 16:04.14 | \bash | (for kiozen) Sharp Zaurus?? |
| 16:04.19 | cr2_ | kiozen: which maps ? i use ~/.roadmap/maps |
| 16:04.31 | kiozen | Europ/Asia |
| 16:04.46 | cr2_ | kiozen: where did you get it from ? |
| 16:04.53 | kiozen | \bash No PC |
| 16:05.05 | \bash | PC? |
| 16:05.07 | kiozen | cr2 offial site |
| 16:05.15 | cr2_ | kiozen: forget it. |
| 16:05.22 | kiozen | link? |
| 16:05.24 | \bash | don't get it |
| 16:05.49 | \bash | no luck cr2: |
| 16:05.51 | \bash | end_request: I/O error, dev mtdblock0, sector 0 |
| 16:05.51 | \bash | Buffer I/O error on device mtdblock0, logical block 0 |
| 16:05.51 | \bash | end_request: I/O error, dev mtdblock0, sector 16 |
| 16:05.51 | \bash | Buffer I/O error on device mtdblock0, logical block 2 |
| 16:05.51 | \bash | end_request: I/O error, dev mtdblock0, sector 0 |
| 16:05.52 | \bash | Buffer I/O error on device mtdblock0, logical block 0 |
| 16:06.17 | cr2_ | kiozen: it's binary incompatible. you can get the map of Israel, or i can send you the map i've sent tsdogs. |
| 16:06.58 | kiozen | what's yours covering? |
| 16:07.00 | cr2_ | \bash: ok. we need to think about it. |
| 16:07.27 | cr2_ | kiozen: verona & up north. |
| 16:07.53 | kiozen | ok, I have got a better orientation in this region |
| 16:08.07 | kiozen | use the gmx on |
| 16:08.10 | cr2_ | i guess so. |
| 16:08.12 | cr2_ | ok. |
| 16:08.41 | psokolovsky_ | cr2_: ok, just to have it discussed a bit more: 3.1Mb *un*compressed size. and large part of it consists of specific omap machines support. In the end, we could do it, if there would be maintenance and development of that stuff, it wouldn't be too much worse comparing to s3. |
| 16:10.15 | \bash | Ok, I will be working on the ~64mb issue |
| 16:12.25 | cr2_ | kiozen: sent |
| 16:12.29 | kiozen | got it |
| 16:13.42 | cr2_ | psokolovsky_: then apply it. |
| 16:14.19 | \bash | solved crash |
| 16:14.28 | cr2_ | ok. |
| 16:14.57 | cr2_ | kiozen: the position is 10,45 |
| 16:15.17 | psokolovsky_ | cr2_: well, it doesn't work liek that ;-). some who is really interested to do that, should carefully pickup patches to apply, test, and submit them ;-) |
| 16:15.19 | cr2_ | unlike garmin the bbox is not shown on lowzoom ;) |
| 16:16.37 | kiozen | still no map, but now it says sook |
| 16:16.37 | *** join/#htc-linux scorpio16v (n=scorpio1@ACB66F10.ipt.aol.com) |
| 16:16.38 | cr2_ | psokolovsky_: i don't have an omap device. hehe. garmin |
| 16:16.52 | cr2_ | kiozen: set the position to 10,46 |
| 16:16.55 | cr2_ | then zoom. |
| 16:17.25 | psokolovsky_ | cr2_: then let's not haste with that, just keep as an option |
| 16:17.36 | kiozen | cr nothing to see |
| 16:18.00 | kiozen | s/cr/cr2 |
| 16:18.50 | kiozen | delete maps list no files (?!) |
| 16:19.15 | cr2_ | psokolovsky_: ok. |
| 16:19.24 | cr2_ | kiozen: where did you put the maps ? |
| 16:19.37 | kiozen | ~/.roadmap/maps |
| 16:19.49 | kiozen | del the other ones |
| 16:20.19 | cr2_ | tsdogs: created buildus for ppc, run in the empty dir. now the program runs. can't generate the map, because there is no postgres on that machine. |
| 16:20.36 | cr2_ | kiozen: what else do you have in ~/.roadmap ? |
| 16:21.21 | kiozen | history |
| 16:21.22 | kiozen | maps |
| 16:21.27 | kiozen | postmortem |
| 16:21.30 | kiozen | schema |
| 16:21.33 | kiozen | session |
| 16:21.34 | kiozen | trips |
| 16:21.37 | kiozen | . |
| 16:21.59 | cr2_ | does your schema have ^M at the end ? |
| 16:22.02 | tsdogs | kiozen: you tryed moving the map left, zooming out? |
| 16:22.35 | kiozen | no ^M |
| 16:22.49 | \bash | Well, I have been able to read 8K from extrom at aaa000, but I still get a lot of errors from all the devices. I think the problem may be the offsets, or the mtd driver itself |
| 16:22.51 | tsdogs | oterwise simply change the values in session |
| 16:22.54 | tsdogs | Hold.Position: 10965158,46039825 |
| 16:23.02 | tsdogs | Selection.Position: 10952180,46047220 |
| 16:23.11 | tsdogs | General.Zoom: 126 |
| 16:23.38 | tsdogs | that should do the trick. cr2_: should defenatly implement the map management :) |
| 16:24.25 | cr2_ | tsdogs: the map indexing and the .tdb kiozen known what i'm talking about :) |
| 16:24.40 | kiozen | second :) |
| 16:25.22 | kiozen | still no map :/ |
| 16:25.38 | cr2_ | 'buildus' should be a menu option, so it searches the maps and plots the bboxes on canvas. with names :) |
| 16:25.55 | cr2_ | buildidx is a better name. |
| 16:26.15 | cr2_ | kiozen: strace ? |
| 16:27.16 | kiozen | loads of stuff :) but I think it's not reading my local session file |
| 16:27.30 | kiozen | kwrite did not report a change during sessions |
| 16:28.12 | cr2_ | strace -ff qtroadmap >& ERR |
| 16:28.17 | cr2_ | grep open ERR |
| 16:29.22 | *** join/#htc-linux pH5 (n=ph5@e178252039.adsl.alicedsl.de) |
| 16:30.22 | kiozen | hm last is from ~/.roadmap/session |
| 16:31.16 | *** join/#htc-linux rob_w (n=bob@X1d26.x.pppool.de) |
| 16:32.39 | tsdogs | kiozen: try this: Location.Position: 11042977,45889720 |
| 16:33.12 | *** join/#htc-linux pH5_ (n=ph5@e178252039.adsl.alicedsl.de) |
| 16:34.09 | kiozen | does it need GPS source? |
| 16:34.19 | cr2_ | no. |
| 16:35.38 | kiozen | hm, I will try from scratch at home. |
| 16:35.47 | kiozen | have to go now, bye |
| 16:38.31 | cr2_ | tsdogs: i've enabled 2^ based zoom, with 2^5=32 initial zoom. |
| 16:39.07 | cr2_ | now i have 15 zoom levels. |
| 16:39.08 | tsdogs | perfect. |
| 16:39.30 | cr2_ | they can be visualized as a bar or something. |
| 16:40.15 | cr2_ | maybe to plot a m/km scalebar on the canvas a la garmin ? |
| 16:41.01 | cr2_ | i have another radical idea :) |
| 16:41.27 | tsdogs | I thought about it, but on small devices would take too much display space |
| 16:41.39 | cr2_ | to extend the number of layers to match garmin names. |
| 16:41.51 | cr2_ | tsdogs: garmin is 176x220 |
| 16:42.19 | cr2_ | but extending the layers will require map rebuild. |
| 16:42.54 | tsdogs | what is the bm2rdm status? |
| 16:43.11 | tsdogs | s/bm/mb |
| 16:43.14 | cr2_ | the mainline CVS has a variable number of layers (<256). |
| 16:43.17 | cr2_ | mp2rdm ? |
| 16:43.23 | tsdogs | yes :) |
| 16:43.58 | cr2_ | i have it in postgis, so it's only a new 'buildmap+buildus' run. 5minuts. |
| 16:44.11 | tsdogs | ok |
| 16:44.35 | cr2_ | mp2rdm needs escaping the ' in commands. |
| 16:44.51 | cr2_ | pita. |
| 16:45.39 | tsdogs | ok, substitute ' with \' |
| 16:46.09 | cr2_ | but after i'll extend it, we need some changes in the renderer. |
| 16:46.22 | cr2_ | ok. |
| 16:46.36 | tsdogs | ok. what will it need to be done? |
| 16:50.09 | *** join/#htc-linux goxboxlive (n=goxboxli@176.84-48-210.nextgentel.com) |
| 16:51.24 | cr2_ | brushes/fonts and polygon names. |
| 16:52.01 | tsdogs | :) |
| 16:52.21 | cr2_ | hmm. RoadMapContext is static, can get the zoom level in other places. |
| 16:52.30 | tsdogs | for polygons names you mean poi ? |
| 16:55.39 | cr2_ | no. |
| 16:55.51 | cr2_ | poi is more difficult. |
| 16:56.43 | cr2_ | roadmap_math_zoom_getlevel() |
| 16:56.47 | tsdogs | ok, so we have to extend basic structures to support them |
| 16:58.27 | cr2_ | yes, there is a table 'places' for pois, but something is broken there. |
| 16:58.48 | cr2_ | the polygon names are available but are not rendered on the canvas. |
| 16:59.16 | cr2_ | places should be read from a shapefile. text is not supported. |
| 16:59.57 | cr2_ | but shapefile read/sort is broken on 64bit machines. and you need the shapefil.h and -libshp. |
| 17:00.04 | cr2_ | weird. |
| 17:00.18 | cr2_ | need to add it to postgres. |
| 17:02.26 | tsdogs | ok. polygons have names, so we simply need to paint the name. brushes are somehow supported. fonts are not, but shold not be that dificult. |
| 17:04.23 | cr2_ | polygon rendering is broken in general. |
| 17:05.55 | tsdogs | cr2_: what do you mean? |
| 17:07.24 | cr2_ | sometimes they are filled in stripes or triangles. |
| 17:07.46 | cr2_ | something for kiozen :) |
| 17:08.52 | *** join/#htc-linux kiozen (n=oeichler@rgnb-d9bde00e.pool.mediaWays.net) |
| 17:09.12 | kiozen | does roadmap use the polygon drawing from qt |
| 17:09.24 | tsdogs | kiozen: yes |
| 17:09.34 | tsdogs | :) layers could be Categories? |
| 17:09.41 | kiozen | some cheap graphic adapter have problems |
| 17:10.00 | kiozen | especially the stripes are known to me |
| 17:10.23 | kiozen | are the horizontal |
| 17:10.27 | kiozen | ? |
| 17:11.03 | kiozen | do they disappear if you move the map? |
| 17:11.11 | cr2_ | yes. |
| 17:11.43 | kiozen | hm, seems to be some bug in QT/X |
| 17:11.59 | tsdogs | kiozen: I have nvidia, but they have problems too. |
| 17:12.00 | cr2_ | ok. |
| 17:12.10 | cr2_ | ati here. |
| 17:12.19 | kiozen | hm nvidia should be ok |
| 17:12.19 | cr2_ | nonbinary of course. |
| 17:12.26 | tsdogs | also ati on desktop :) |
| 17:12.42 | kiozen | I experienced this with Centrino chipset |
| 17:13.04 | cr2_ | hmm. zoom=20 is hadcoded somewhere. |
| 17:13.10 | tsdogs | kiozen: amd turion 64x2 |
| 17:13.19 | kiozen | I will look at this later, if roadmap wants to install on my laptop |
| 17:13.44 | kiozen | cu later |
| 17:13.48 | cr2_ | heh. the zoom value is opposite to garmin. lower means hi-res. |
| 17:13.53 | tsdogs | I thought it was some bug in the rendering of the zoom/thikness of objects |
| 17:14.08 | tsdogs | :) |
| 17:14.11 | cr2_ | tsdogs: i think there is some roadmap bug there. |
| 17:14.53 | tsdogs | I think it's some editor, as I didn't experience it with roadmap afair |
| 17:15.43 | cr2_ | ok. |
| 17:18.03 | cr2_ | found :) |
| 17:18.06 | cr2_ | ("preferences", &RoadMapConfigGeneralDefaultZoom, "20"); |
| 17:19.05 | tsdogs | :) |
| 17:21.00 | cr2_ | tsdogs: the scrollwheel is not used :) |
| 17:32.31 | *** part/#htc-linux scorpio16v (n=scorpio1@ACB66F10.ipt.aol.com) |
| 17:45.47 | goxboxlive | pH5, i am trying out your latest changes in gsmd in OE, but still i get the same bug. |
| 17:53.19 | goxboxlive | BabelO, is it posible to build and run xgnokii in a x11 image, or do i have to change the source? |
| 17:53.58 | pH5 | goxboxlive: does http://en.pastebin.ca/564634 help? |
| 17:54.21 | goxboxlive | i'll try |
| 18:01.03 | *** part/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk) |
| 18:03.24 | goxboxlive | pH5, it failes compiling pastebin.no/2047 ( i am building on rev 2234) |
| 18:08.22 | pH5 | goxboxlive: yes, let me fix that.. |
| 18:08.34 | goxboxlive | thx |
| 18:09.42 | *** join/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk) |
| 18:16.08 | pH5 | http://en.pastebin.ca/564713 |
| 18:16.21 | *** join/#htc-linux SmallR2002 (n=SmallR20@80-44-105-242.dynamic.dsl.as9105.com) |
| 18:17.09 | goxboxlive | :-) thx i'll try it out |
| 18:17.52 | *** join/#htc-linux pleemans (n=peter@d51A5E76A.access.telenet.be) |
| 18:25.55 | goxboxlive | pH5, nope still the same : http://en.pastebin.ca/564742 |
| 18:27.19 | goxboxlive | i'm running strace now |
| 18:28.46 | *** join/#htc-linux punkgeek (n=jeanseb@gazypan.dyndns.org) |
| 18:30.33 | pH5 | goxboxlive: this is confusing. can you http://en.pastebin.ca/564752 |
| 18:31.52 | *** join/#htc-linux kiozen (n=oeichler@rgnb-d9bde00e.pool.mediaWays.net) |
| 18:31.55 | goxboxlive | pH5, i added -w 5 at the end of the gsmd -P/dev blabla and now something happends, but still it's dead. I'll pastebin the output |
| 18:32.47 | goxboxlive | Now it just hangs there ate modem alive |
| 18:33.26 | goxboxlive | Why is it sending at+cfun=0. I have allways used at+cfun=1 when doing trying it with 'cu'. |
| 18:33.54 | pH5 | argh, -w should just set interpreter_ready=0, same as your vendor_qc patch already does. why does it make a difference? |
| 18:34.27 | goxboxlive | no idea. i'll add your printf. |
| 18:34.29 | pH5 | goxboxlive: harald changed gsmd to start with the phone disabled, the moko dialer now calls +cfun=1 in the beginning. (I think this is a bad idea) |
| 18:34.56 | goxboxlive | i agree. |
| 18:35.07 | pH5 | and that apart from the fact that it crashes gsmd here when at+cops=0 is sent before the answer to at+cfun=1 |
| 18:35.55 | goxboxlive | now , something is happenin, this happend after modem alive |
| 18:35.57 | goxboxlive | Tue Jun 12 20:09:09 2007 <3> gsmd.c:80:alive_tmr_cb() modem alive! |
| 18:35.57 | goxboxlive | Tue Jun 12 20:13:39 2007 <1> gsmd.c:119:alive_interval_tmr_cb() interval expired, starting next alive inquiry |
| 18:35.57 | goxboxlive | Tue Jun 12 20:13:39 2007 <1> atcmd.c:448:atcmd_submit() submitting command `ATE0' |
| 18:35.57 | goxboxlive | Tue Jun 12 20:13:39 2007 <1> atcmd.c:180:ml_parse() buf=`[WCDMA] ASSERT CD'(17) |
| 18:35.58 | goxboxlive | Tue Jun 12 20:13:39 2007 <1> atcmd.c:334:ml_parse() Calling cmd->cb() |
| 18:35.59 | goxboxlive | Tue Jun 12 20:13:39 2007 <1> gsmd.c:151:gsmd_test_atcb() `AT+CFUN=0' returned `[WCDMA] ASSERT CD' |
| 18:36.01 | goxboxlive | Tue Jun 12 20:13:39 2007 <1> atcmd.c:180:ml_parse() buf=`OK'(2) |
| 18:36.03 | goxboxlive | Tue Jun 12 20:14:09 2007 <1> gsmd.c:74:alive_tmr_cb() gsmd_alive timer expired |
| 18:36.05 | goxboxlive | Tue Jun 12 20:14:09 2007 <8> gsmd.c:77:alive_tmr_cb() modem dead! |
| 18:36.37 | pH5 | is this '[WCDMA] ASSERT CD' unsolicited? |
| 18:36.56 | pH5 | (can this just pop up any time when you talk to the modem with cu?) |
| 18:37.58 | goxboxlive | yes i guess until cops? returns something |
| 18:38.56 | pH5 | is there any way to turn this message off? |
| 18:39.24 | goxboxlive | i have no idea |
| 18:40.17 | kiozen | cr2 ? |
| 18:41.05 | pH5 | goxboxlive: ok, for now I'll just assume that it can't. I'll try to make vendor module handle this. |
| 18:41.14 | *** join/#htc-linux Babel (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 18:41.23 | goxboxlive | pH5, thanks a lot |
| 18:41.44 | goxboxlive | prinf didnt gave me any extra output |
| 18:42.53 | pH5 | there has to be a line "V0: 0" or "V0: 1" before the first "ATE0V1" |
| 18:43.25 | pH5 | if there isn't you should make sure that you indeed use the patched gsmd. |
| 18:44.31 | goxboxlive | hold on i'll pastebin after it dies out. |
| 18:44.39 | goxboxlive | to rebuild i do: |
| 18:45.21 | goxboxlive | deleting the gsmd.tar.gz file in my source directory. I bitbake -cclean libgsmd && bitbake libgsmd && bitbake -crebuild package-index |
| 18:46.02 | *** join/#htc-linux jeanseb|out (n=jeanseb@gazypan.dyndns.org) |
| 18:46.17 | goxboxlive | and then i ipkg remove gsmd -recursive (also gsmd-plugins) and ipkg update and ipkg install gsmd (also gsmd-plugins) |
| 18:46.24 | goxboxlive | thats my procedyre |
| 18:47.36 | goxboxlive | pH5, here is the output form gsmd : http://en.pastebin.ca/564789 |
| 18:49.29 | cr2_ | kiozen: i'm back. |
| 18:49.42 | goxboxlive | hi ChanServ |
| 18:49.46 | goxboxlive | hi cr2_ :-) |
| 18:49.53 | kiozen | hi, now I get icons, there seems to be one map |
| 18:49.58 | kiozen | but still nothing to see |
| 18:51.41 | cr2_ | nothing to see ? |
| 18:51.51 | cr2_ | there should be a road network there. |
| 18:52.03 | kiozen | if I start roadmap a yellow box says no map available |
| 18:52.57 | kiozen | if I look at map disk space it says 1 / 4132K |
| 18:53.36 | kiozen | any preference to set? |
| 18:53.57 | kiozen | I had to setup the map path to ~/.roadmap/maps |
| 18:54.56 | cr2_ | disk space is not very helpful . |
| 18:55.23 | kiozen | well one hour ago it was zero :) |
| 18:55.33 | kiozen | and no icons showed up |
| 18:55.54 | cr2_ | ok. |
| 18:56.10 | kiozen | I "made install" and start qtroadmap from ./src/qt4 |
| 18:56.10 | cr2_ | does it open the *.rdm map ? |
| 18:56.23 | kiozen | how can I see strace? |
| 18:56.34 | cr2_ | strace qtroadmap >& ERR |
| 18:56.42 | cr2_ | grep open ERR |
| 18:57.06 | pH5 | goxboxlive: that's strange. I upload the patches to oe.dev, could you please pull and rebuild in a minute? |
| 18:57.14 | kiozen | open("/home/oeichler/.roadmap/maps/usdir.rdm", O_RDONLY) = 11 |
| 18:57.27 | goxboxlive | pH5, of course |
| 18:57.59 | kiozen | nothing else |
| 18:58.43 | pH5 | cr2_: do you know what the universal modem means when it sends "[WCDMA] ASSERT CD" or "[WCDMA] DEASSERT CD" over the AT link? |
| 19:00.34 | pH5 | goxboxlive: hold on, I made another error |
| 19:00.41 | goxboxlive | :.) |
| 19:01.02 | cr2_ | kiozen: then you are not in the map area. do a long mouse press. it will show you your mouse coordinates |
| 19:01.34 | cr2_ | pH5: it may correlate with the "DCD" gpio going up. |
| 19:01.42 | kiozen | 1,-312 |
| 19:02.30 | kiozen | now 46,10 |
| 19:02.31 | cr2_ | i don't know how fast is the 'open(/dev/ttyS0)'. maybe it really makes sense to define the DCD function in pxa_serial.- |
| 19:02.35 | *** join/#htc-linux jeanseb|out (n=jeanseb@gazypan.dyndns.org) |
| 19:02.55 | cr2_ | kiozen: it should be 10000000,46000000 |
| 19:03.24 | pH5 | goxboxlive: ok, missing file uploaded |
| 19:03.44 | goxboxlive | pH5, should i revert gsmd.c before building? |
| 19:03.49 | kiozen | no map availbale |
| 19:03.50 | cr2_ | kiozen: call Find->Position |
| 19:04.02 | kiozen | yes I do |
| 19:04.04 | cr2_ | and enter 10 |
| 19:04.06 | cr2_ | 46 |
| 19:04.13 | cr2_ | OK. |
| 19:04.29 | cr2_ | then do a long press. |
| 19:04.43 | pH5 | goxboxlive: all needed changes should be in the default build if you run gsmd with -w 1 |
| 19:04.44 | kiozen | 46000084, 10000943 |
| 19:04.49 | kiozen | no map |
| 19:05.01 | goxboxlive | ok, then i revert it |
| 19:06.04 | cr2_ | kiozen: zoom ? |
| 19:06.15 | kiozen | tried ++++++ |
| 19:06.18 | kiozen | no map |
| 19:07.01 | cr2_ | strange. i'll check myself. |
| 19:08.26 | kiozen | open("/home/oeichler/.roadmap/maps/usc07011.rdm", O_RDONLY) = 12 |
| 19:08.37 | kiozen | open("/home/oeichler/.roadmap/maps/edt07011.rdm", O_RDWR) = -1 ENOENT (No such file or directory) |
| 19:09.15 | cr2_ | ok. |
| 19:09.26 | cr2_ | go to ~/.roadmap/schema |
| 19:09.29 | cr2_ | .Declutter: 2147483647 |
| 19:09.42 | cr2_ | and set .Declutter for the layers |
| 19:10.09 | kiozen | that's already |
| 19:10.16 | cr2_ | street layers. others are not in the file. |
| 19:10.33 | cr2_ | Highways.Declutter: |
| 19:10.41 | cr2_ | Streets.Declutter: |
| 19:10.50 | cr2_ | Trails.Declutter: |
| 19:11.23 | psokolovsky_ | Btw, what did you do about MaemoMapper? |
| 19:12.34 | kiozen | nope |
| 19:15.04 | *** join/#htc-linux tsdogs (n=twostupi@62.123.180.130) |
| 19:15.57 | kiozen | hi tsdogs, still try to get this map visible |
| 19:16.00 | psokolovsky_ | http://www.flickr.com/photos/atmasphere/375733253/ |
| 19:17.37 | tsdogs | kiozen: schema http://www.pastebin.ca/564853 |
| 19:18.12 | tsdogs | session http://www.pastebin.ca/564854 |
| 19:19.33 | tsdogs | kiozen: do you get some unknown color or other messages? beside the gps one |
| 19:20.06 | kiozen | QColor::setNamedColor: Unknown color name 'DARKSEAGREEN4' |
| 19:20.06 | kiozen | 21:20:8 == qt_canvas.cc, line 491 (roadmap_screen_repaint): color invalid: DarkSeaGreen4 |
| 19:20.06 | kiozen | QColor::setNamedColor: Unknown color name 'DARKSEAGREEN4' |
| 19:20.06 | kiozen | 21:20:8 == qt_canvas.cc, line 491 (roadmap_screen_repaint): color invalid: DarkSeaGreen4 |
| 19:20.19 | tsdogs | ok |
| 19:20.35 | tsdogs | try with the 2 files I gave you. |
| 19:20.46 | kiozen | but it was worse a few startups ago :) |
| 19:21.38 | goxboxlive | pH5, are you still here |
| 19:21.39 | cr2_ | kiozen: i can commit my patches, then you'll see the zoom level, not only the coordinates. |
| 19:21.52 | kiozen | ack |
| 19:22.00 | kiozen | that would help a lot |
| 19:22.13 | pH5 | goxboxlive: yup |
| 19:22.23 | goxboxlive | pH5, i love you |
| 19:22.26 | goxboxlive | :-) |
| 19:22.28 | kiozen | what happened to the original author? |
| 19:22.32 | goxboxlive | it works now |
| 19:23.07 | kiozen | he didn't like is baby, did he? |
| 19:23.11 | tsdogs | kiozen try with files I gave you and do Find -> destination |
| 19:23.33 | kiozen | no map available |
| 19:23.50 | goxboxlive | pH5, Thanks a lot. Now when i start it manually it just waits for at+cfun=1. And when i start openmoko-dialer it runs weither |
| 19:23.54 | tsdogs | hmm. do you have an index.rdm in maps? |
| 19:23.57 | goxboxlive | further i mean. |
| 19:24.12 | kiozen | no |
| 19:24.19 | cr2_ | done. |
| 19:24.35 | tsdogs | cr2_: index.rdm is needed? I think so. |
| 19:24.38 | cr2_ | tsdogs: usdir.rdm |
| 19:24.45 | goxboxlive | pH5, I also noticed that i can hangup without changing ath0 to ath. |
| 19:24.48 | cr2_ | it is the index file |
| 19:24.58 | pH5 | heh, great. now I can rest peacefully :) |
| 19:25.05 | goxboxlive | pH5, but there is still one issue, i dont know if you havethe same problem |
| 19:25.24 | goxboxlive | but openmokodialer doesnt understand that the other part has quit the call |
| 19:25.35 | goxboxlive | it is still active |
| 19:25.52 | pH5 | I wouldn't know - here it crashes when that happens. |
| 19:26.08 | tsdogs | kiozen: do a find destination and then zoom in. |
| 19:26.11 | goxboxlive | ok |
| 19:26.29 | pH5 | I can start phone calls, but I have to beg the other end to hang up :) |
| 19:26.37 | cr2_ | pH5: lol |
| 19:26.40 | tsdogs | lol |
| 19:26.44 | kiozen | that puts me to -245,-168 |
| 19:27.03 | goxboxlive | pH5, i am not using openmoko , i am running dialer in Opie with Xqt2. I found out that it was more stable |
| 19:27.05 | cr2_ | kiozen: what cpu do you have ? |
| 19:27.13 | kiozen | intel dual core |
| 19:27.32 | goxboxlive | pH5, ok, for me it is oposite, i have to end the call |
| 19:27.45 | pH5 | I'm running it in my olde GPE image, but without the slow-ass openmoko theme |
| 19:27.54 | goxboxlive | ok |
| 19:27.58 | pH5 | that causes segfaults from time to time. |
| 19:28.02 | cr2_ | kiozen: got the patches ? |
| 19:28.22 | kiozen | I did cvs up, but had the feeling nothing update |
| 19:28.23 | goxboxlive | pH5, could you update the gsmd file in OE and add -w 1 to it? |
| 19:28.41 | kiozen | compiled, pressed buttons, no zoom level |
| 19:28.59 | kiozen | I am on roadmap_editor rep. |
| 19:29.21 | pH5 | oh, right. |
| 19:29.32 | goxboxlive | pH5, thanks a lot for your work. |
| 19:30.15 | tsdogs | kiozen: yes that's correct |
| 19:30.28 | tsdogs | can you paste bin the session? |
| 19:30.43 | kiozen | never did, I try :) |
| 19:31.41 | *** join/#htc-linux dion (n=dion@inhex.net) |
| 19:31.57 | kiozen | http://www.pastebin.ca/564877 |
| 19:33.05 | kiozen | cr2 now there a patches |
| 19:33.20 | tsdogs | kiozen: change Focus.Name=Destination |
| 19:33.21 | *** join/#htc-linux Foxdie` (i=Foxdie@77-98-48-193.cable.ubr05.harb.blueyonder.co.uk) |
| 19:33.43 | tsdogs | change also zoom to 65400 |
| 19:34.54 | kiozen | still in cvs up |
| 19:35.15 | tsdogs | kiozen: for me doing Find Position and then zooming out works (even with your schema) |
| 19:36.33 | kiozen | now, there are some black lines |
| 19:36.41 | tsdogs | ok now zoom in |
| 19:37.20 | kiozen | ok, let's hunt bugs, where are those polygons? |
| 19:38.17 | cr2_ | not in this map. |
| 19:38.36 | tsdogs | no idea, cr2_ knows better, but moving when strings are there makes some lines disappear. |
| 19:38.46 | tsdogs | no idea if it's roadmap or qt error though. |
| 19:39.25 | kiozen | it uses the same string coding like Garmin ~[0x1e] |
| 19:39.46 | tsdogs | dinner time. bbl :) the data comes from it I think. |
| 19:39.59 | kiozen | ahhhhh |
| 19:40.21 | kiozen | it did not take the data correctly from Garmin |
| 19:40.29 | kiozen | bon appetit |
| 19:42.13 | kiozen | cr2 can you send me that polygon map? |
| 19:42.44 | Babel | kiozen: it is french :) good lunch |
| 19:42.49 | goxboxlive | cr2_, How do i make the universal ring when it recives calls? Is there anything in the gsmd sources or is it the kernel source wich needs modifying? |
| 19:43.05 | kiozen | didn't new the italian version |
| 19:43.31 | kiozen | my wife usually takes over the italian part |
| 19:44.20 | Babel | :) |
| 19:44.39 | Babel | ~ping BabelO |
| 19:44.50 | apt | pong BabelO |
| 19:44.59 | Babel | grhh |
| 19:45.52 | cr2_ | kiozen: take the map of israel. there is a lake there. |
| 19:47.23 | kiozen | link? israel and roadmap is a bad pair to search |
| 19:47.26 | pH5 | goxboxlive: gsmd just emits an usock signal, I think the dialer or the moko gsm applet do the ringing. |
| 19:48.19 | goxboxlive | pH5, ok, do you have it working? |
| 19:48.56 | pH5 | no, not on the magician |
| 19:48.58 | cr2_ | kiozen: http://www.freemap.co.il/roadmap/maps |
| 19:48.59 | goxboxlive | I van not have the applet in opie, so i hope it is the dialer, i'll check the source |
| 19:49.06 | goxboxlive | s/van/can/ |
| 19:49.48 | cr2_ | goxboxlive: it's difficult, you need to do some black magic with the mixer. which is in a hacked state right now. |
| 19:50.28 | goxboxlive | cr2_, ok |
| 19:50.33 | kiozen | cr2: do I need the usdir.rdm, too? |
| 19:50.53 | cr2_ | goxboxlive: you need to mix the sound from the phone with the sound from stereo DAC. |
| 19:50.57 | cr2_ | kiozen: yes. |
| 19:51.17 | kiozen | thus I have to replace the italian map? |
| 19:51.21 | goxboxlive | cr2_, ahh of course, why did i not think of that :-) |
| 19:51.29 | cr2_ | kiozen: because the program can't build the index itself. even if you have 1 map. |
| 19:51.36 | goxboxlive | i found a alsa.c file |
| 19:51.39 | kiozen | ack |
| 19:52.21 | cr2_ | kiozen: but i think that it's wrong. the program should be caable of that. |
| 19:52.36 | kiozen | file -> import map :) |
| 19:52.46 | cr2_ | yes. |
| 19:53.28 | kiozen | where do I have to go for this lake? |
| 19:54.37 | cr2_ | hmm. 33,33 ? |
| 19:54.44 | cr2_ | 33,34 ? |
| 19:56.39 | kiozen | no map available, grr this thing makes me look stupid |
| 20:01.00 | kiozen | 22:0:55 s** editor/db/editor_db.c, line 843 editor_db_locator - can't find fips. |
| 20:01.00 | kiozen | <PROTECTED> |
| 20:01.00 | kiozen | <PROTECTED> |
| 20:01.12 | kiozen | cr2 does that ring a bell? |
| 20:03.21 | cr2_ | no. |
| 20:10.52 | kiozen | cr2 the files on the server have todays timestamp |
| 20:11.09 | kiozen | are they still subject to change? |
| 20:14.59 | cr2_ | hmm. good question. i#ll try myself |
| 20:15.19 | *** part/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk) |
| 20:17.54 | cr2_ | the map is in the right lower corner. |
| 20:17.58 | cr2_ | 31.19, 35.39 |
| 20:18.07 | cr2_ | is in the lake. zoom=64 |
| 20:18.31 | goxboxlive | cr2_, what is the universals dpi value? |
| 20:18.48 | cr2_ | 225 |
| 20:18.51 | cr2_ | i think |
| 20:20.26 | kiozen | no map available, this app stinks |
| 20:20.39 | cr2_ | :) |
| 20:20.51 | cr2_ | it's just not user-friendly. |
| 20:20.57 | kiozen | wouldn' it be better to make a new one? |
| 20:22.10 | kiozen | surely there is a core to use |
| 20:22.25 | cr2_ | sorry, revert lat and long |
| 20:22.54 | cr2_ | it's easier to fix the bad behaviour. |
| 20:23.18 | cr2_ | the program is very portable. |
| 20:24.31 | cr2_ | 35.40 31.20 |
| 20:24.41 | pH5 | sqrt(640²+480²)/3.7 is a bit over 216. |
| 20:25.08 | kiozen | hm, at least no missing map |
| 20:25.17 | kiozen | but no polygons either |
| 20:26.00 | cr2_ | zoom |
| 20:26.24 | kiozen | did it from 2 to 512 |
| 20:27.13 | cr2_ | then it's the schema |
| 20:27.46 | cr2_ | i suggest to hardcode the garmin layer numbers and colors. |
| 20:28.01 | cr2_ | do you have such style ? |
| 20:28.10 | kiozen | yes, and I always thought Garmin is wacky |
| 20:28.27 | kiozen | compared to that they are brilliant |
| 20:28.52 | kiozen | what style? |
| 20:31.12 | cr2_ | layer, color, brush, linewidth |
| 20:31.49 | kiozen | it's hardcoded in CGarminMap.cpp in a switch case structure. |
| 20:32.11 | kiozen | never had the guts to change it into a map |
| 20:32.19 | kiozen | and make it configurable |
| 20:32.52 | kiozen | Sofar nobody complained :) |
| 20:33.23 | cr2_ | ok. |
| 20:33.30 | kiozen | maybe because they fear to be threatened to make a setup dialog |
| 20:33.46 | cr2_ | :) |
| 20:33.56 | cr2_ | mapsource uses a different style ? |
| 20:34.08 | kiozen | yes, it's common in the US |
| 20:34.21 | kiozen | I like the German style more |
| 20:34.30 | cr2_ | ok. |
| 20:34.47 | kiozen | there is not much choice in the color scheme |
| 20:35.11 | kiozen | a good contrast scheme is hard to find |
| 20:35.32 | kiozen | once you got it you don't want to change it |
| 20:36.01 | kiozen | took me quite some time to make trails and altitudes visible in the woods |
| 20:36.45 | cr2_ | your style is good. |
| 20:37.05 | cr2_ | maybe the red in the cities is too intense. |
| 20:37.38 | kiozen | can't tell, red and green are not my favorite colors :) |
| 20:37.57 | cr2_ | i can tell you how i've remapped the layers now. |
| 20:38.03 | cr2_ | lol. |
| 20:38.14 | cr2_ | white and blue :) |
| 20:38.36 | kiozen | for cities? |
| 20:38.51 | cr2_ | the favorite |
| 20:39.23 | kiozen | ah, and that looks good? |
| 20:39.42 | cr2_ | i'm joking. |
| 20:40.10 | kiozen | ah |
| 20:40.49 | kiozen | thus no woods and flooded cities |
| 20:41.01 | kiozen | like the world map bug :) |
| 20:41.16 | cr2_ | :) |
| 20:42.18 | kiozen | let's see urban areas are usual some purple pinkish red, aren't they? |
| 20:43.01 | cr2_ | yes, but your color is a bit intense for my taste. |
| 20:43.09 | cr2_ | but it really does not matter. |
| 20:43.29 | cr2_ | to have some standard set is better. |
| 20:43.42 | kiozen | it's simply red, I try something different |
| 20:47.53 | kiozen | "#ed122b" that looks quite nice |
| 21:03.31 | *** join/#htc-linux rob_w (n=bob@X1d26.x.pppool.de) |
| 21:16.24 | *** join/#htc-linux snakefreak (n=dennisco@84.130.66.232) |
| 21:16.29 | snakefreak | hi all |
| 21:16.46 | goxboxlive | snakefreak, hi |
| 21:20.22 | snakefreak | babel0 are you there? |
| 21:21.55 | Babel | snakefreak: yes |
| 21:22.57 | snakefreak | i have problems with the extraction. many errors wizh symlinks in dev dir |
| 21:23.16 | Babel | ?? |
| 21:23.54 | snakefreak | when i extract the image with winrar or under linux with bunzip2 and tar i become a lot of errors. |
| 21:24.08 | Babel | winrar :) bad tools |
| 21:24.24 | Babel | tar xvjf <Angstrom image> |
| 21:24.43 | snakefreak | oh ok^^ |
| 21:26.51 | snakefreak | i test it to night. |
| 21:27.02 | Babel | ok :) |
| 21:27.14 | Babel | i have to eat a little |
| 21:27.14 | Babel | bbl |
| 22:05.01 | Babel | yesss :) |
| 22:05.52 | snakefreak | ^^ |
| 22:06.33 | BabelO | my Office computer is now down ;) i hate windoze |
| 22:06.42 | snakefreak | ^^ |
| 22:07.08 | snakefreak | not only you |
| 22:28.29 | snakefreak | good night all |
| 22:29.19 | tsdogs | snakefreak: night |
| 22:29.29 | BabelO | snakefreak: good night |
| 22:33.25 | cr2_ | hmm. have some bug here. |
| 22:35.26 | *** part/#htc-linux snakefreak (n=dennisco@84.130.66.232) |
| 22:36.14 | tsdogs | cr2_: is this normal int*float = 0 (int=600 float=0.33) |
| 22:36.39 | tsdogs | assigned to an int |
| 22:37.02 | cr2_ | no, but depends on the casts. |
| 22:37.28 | cr2_ | if the float is casted to int first, then it's zero. |
| 22:37.55 | tsdogs | ok |
| 22:44.57 | tsdogs | cr2_: I can draw the map now. |
| 22:49.12 | cr2_ | tsdogs: raster ? |
| 22:49.25 | tsdogs | yes the raster. |
| 22:49.35 | cr2_ | wow. |
| 22:49.53 | tsdogs | not in roadmap yes though, still a lot to do. |
| 22:49.54 | cr2_ | i'm still looking for a bug in a very simple code :( |
| 22:50.00 | cr2_ | ok. |
| 22:50.32 | tsdogs | I suppose I need to have an index of the tiles. |
| 22:50.33 | cr2_ | roadmap uses QPixmap for the canvas ? |
| 22:50.57 | tsdogs | yes, it's faster for drawing polygons. |
| 22:51.15 | cr2_ | ok. |
| 22:51.23 | cr2_ | index of the tils ? |
| 22:51.35 | cr2_ | s/tils/tiles/ |
| 22:51.48 | tsdogs | I mean rasters. |
| 22:52.09 | cr2_ | ok. |
| 22:52.19 | cr2_ | gdaltindex :) |
| 22:52.25 | tsdogs | If I do a RasterIO outside the range, I get an error |
| 22:52.33 | tsdogs | ho ok. |
| 22:52.51 | cr2_ | shp can be read with ogr. |
| 22:52.54 | tsdogs | I should do more rasters with more tiles |
| 22:53.35 | cr2_ | you run gdaltindex, and get a shapefile with the bounding boxes. |
| 22:54.04 | tsdogs | and the associated file? |
| 22:54.08 | cr2_ | it#s probbaly better to reuse the source to gdaltindex. |
| 22:54.12 | cr2_ | it's very easy. |
| 22:54.33 | cr2_ | gdaltindex doq_index.shp doq/*.tif |
| 22:54.41 | cr2_ | so you can use opendir() |
| 22:55.06 | tsdogs | ok. |
| 22:55.33 | cr2_ | and then just dumps geotiffs to maps/raster subdir. |
| 22:55.45 | cr2_ | the program will index them. |
| 22:55.50 | tsdogs | what's the best size of a raster in pixels? |
| 22:56.04 | cr2_ | i think we should do the same with *.rdm |
| 22:56.20 | cr2_ | <4GB |
| 22:56.30 | tsdogs | :) |
| 22:56.49 | cr2_ | geotiff has some limits :) |
| 22:57.58 | tsdogs | what I meant is: suppose we create rasters for the whole world. what size in pixels is the best? or we should have as less rasters as we can? |
| 22:58.47 | cr2_ | one big raster saves you 100000000 directory entries. |
| 22:59.04 | cr2_ | but it's limited by 4GB |
| 23:00.15 | tsdogs | ok, so a possibility would be to have as few rasters as we can for each nation. |
| 23:01.14 | cr2_ | i think it#s not a big issue. |
| 23:01.25 | cr2_ | the problem are the zoom levels. |
| 23:01.40 | tsdogs | ok. I need to make more rasters with more tiles in it. |
| 23:02.12 | cr2_ | if you have 2 files for the same region, one in 1:20000 resolution and another one in 1:1000000 |
| 23:02.30 | cr2_ | you don't want to use the second one at 1:20000 scales. |
| 23:02.42 | cr2_ | the overviews from the second one. |
| 23:03.00 | cr2_ | but it's more theoretical now. |
| 23:03.07 | cr2_ | let's assume there is no overlap. |
| 23:03.26 | tsdogs | that should be acomplished by overviews I think. |
| 23:04.55 | tsdogs | btw, can gdal simply give me the rasters/tiles I need to draw a region? or I must choose them? |
| 23:06.39 | cr2_ | you build the index with gdaltindex, then by reading it you know the names. |
| 23:07.39 | tsdogs | ok, but rasterIO is done on the band/tile not at raster level. |
| 23:08.02 | tsdogs | I suppose tile = band in gdal |
| 23:10.04 | cr2_ | no. band is the 'color' channel. |
| 23:10.17 | cr2_ | paletted raster has 1 band. |
| 23:10.23 | cr2_ | rgb has 3. |
| 23:10.30 | tsdogs | ha ok. |
| 23:10.46 | cr2_ | it's possible to have more that 3 bands. |
| 23:10.57 | tsdogs | so If I open the index as dataset, then I have automagically all the rasters. |
| 23:11.08 | cr2_ | for multispectral satellites for example. |
| 23:11.08 | tsdogs | this makes more sense for me. |
| 23:11.32 | tsdogs | ok, I saw there were different color handlings than palette. |
| 23:11.36 | cr2_ | no, you have the bounding box for each raster. |
| 23:12.11 | cr2_ | you search which rasters you need to cover your required area. |
| 23:12.19 | tsdogs | so each dataset is 1 raster, but then how do I open the index? cannot fine the class to handle that |
| 23:12.32 | cr2_ | it's like double loop. |
| 23:12.36 | tsdogs | s/fine/find |
| 23:12.58 | cr2_ | shp is vector, not raster, so it's the OGR part of gdal. |
| 23:13.33 | cr2_ | http://www.gdal.org/ogr/ogr_apitut.html |
| 23:13.45 | tsdogs | ho ok. |
| 23:13.48 | tsdogs | thx |
| 23:14.29 | cr2_ | create the index with gdaltindex first |
| 23:14.38 | tsdogs | hm yes. |
| 23:14.50 | cr2_ | and then use 'ogrinfo' on the index. you'll see a lot of useful info |
| 23:15.29 | cr2_ | even if you have only 1 geotiff, create an index. |
| 23:15.50 | tsdogs | ok |
| 23:17.36 | tsdogs | btw, I thought zoom levels in roadmap where 27~ |
| 23:20.32 | cr2_ | 27 ? |
| 23:21.26 | tsdogs | I did some tests time ago, but actually don't remember the exact number. |
| 23:21.33 | cr2_ | there are 15 levels now. if you don't set weird values in ~/.roadmap/* |
| 23:21.34 | tsdogs | what's the issue with the zoom |
| 23:21.43 | tsdogs | :) ok |
| 23:22.09 | cr2_ | the zoom is limited 2 <= zoom <= 32768 |
| 23:22.25 | cr2_ | 2^1 - 2^15 |
| 23:22.42 | tsdogs | ok. ain't that enough? |
| 23:23.21 | cr2_ | if you'll set an initial zoom to not a power of 2 (=weird), then maybe less. |
| 23:23.40 | cr2_ | garmin uses 26 |
| 23:23.58 | cr2_ | for the maps with autorouting 24. |
| 23:24.00 | cr2_ | afair. |
| 23:24.08 | cr2_ | kiozen should know better. |
| 23:24.23 | cr2_ | how much google uses ? |
| 23:24.26 | cr2_ | 18 ? |
| 23:24.41 | cr2_ | hmm. google is a bit different. |
| 23:24.43 | tsdogs | 17 for my region I think |
| 23:24.47 | cr2_ | they use mercator. |
| 23:24.56 | *** join/#htc-linux Ranpha1 (n=Ranpha@hlm-n-73ed.adsl.wanadoo.nl) |
| 23:25.27 | tsdogs | well, zoom is only a factor, so can be changed. |
| 23:27.41 | *** join/#htc-linux cyrill62 (n=user@ble59-5-82-233-205-36.fbx.proxad.net) |
| 23:30.41 | cr2_ | it can be used to create a more optimized (=faster) data format. |
| 23:56.32 | tsdogs | cr2_: ok I can read some data from the index. will continue tomorrow. good night. |