April 2012 - mkgmap-dev - The mkgmap lists

Resolution range for POIs ?
by Chris66 30 Apr '12

30 Apr '12

Hi, when I changed my point file to: place=city [0x0400 resolution 22-16] place=town [0x0800 resolution 22-18] the map was corrupt (blank tiles in BaseCamp and other effects). So, is it not allowed to have a Resolution Range for POIs or is this a known bug ? Chris

3 4

java.nio.Buffer.position(Unknown Source)
by Felix Hartmann 29 Apr '12

29 Apr '12

I get this error when compiling South America with boundaries... Anyone knows why this error happens? Exception in thread "main" java.lang.IllegalArgumentException at java.nio.Buffer.position(Unknown Source) at uk.me.parabola.imgfmt.Utils.bytesToString(Utils.java:80) at uk.me.parabola.imgfmt.sys.ImgHeader.setHeader(ImgHeader.java:25 at uk.me.parabola.imgfmt.sys.ImgFS.readInitFS(ImgFS.java:313) at uk.me.parabola.imgfmt.sys.ImgFS.openFs(ImgFS.java:131) at uk.me.parabola.imgfmt.sys.ImgFS.openFs(ImgFS.java:124) at uk.me.parabola.mkgmap.combiners.FileInfo.imgInfo(FileInfo.java: at uk.me.parabola.mkgmap.combiners.FileInfo.getFileInfo(FileInfo.j at uk.me.parabola.mkgmap.main.Main.endOptions(Main.java:429) at uk.me.parabola.mkgmap.CommandArgsReader.readArgs(CommandArgsRea at uk.me.parabola.mkgmap.main.Main.main(Main.java:114)

2 1

Commit: r2274: Add help text to note that if the index is created as an extra step then
by svn commit 29 Apr '12

29 Apr '12

Version 2274 was commited by steve on 2012-04-29 18:35:09 +0100 (Sun, 29 Apr 2012) Add help text to note that if the index is created as an extra step then you must use the same code-page arguments when creating the index as were used when creating the tiles.

1 0

flooded tile without any coastline
by aighes 29 Apr '12

29 Apr '12

Hi, I have a tile [1] in western part of China. There is no coastline inside, but this tile gets flooded. I use --generate-sea=extend-sea-sectors,close-gaps=6000. Any guesses why this happens? Henning [1] http://www.aighes.de/data/20000029.osm.pbf ~4mb, mapdata (c) Openstreetmap contributors, cc-by-sa

4 7

Subdivision width is 36627 at 3230916/1236133
by Marko Mäkelä 28 Apr '12

28 Apr '12

This is an old error that mkgmap has been spitting for a while. I spent some time today splitting my map tiles in Northern Finland, and this is what I got now: 2012/04/20 20:55:34 WARNING (Subdivision): 63240012.osm.pbf: Subdivision width is 36627 at 3230916/1236133 The tile boundaries are: 63240012: 3171072,1169984 to 3266944,1368576 # : 68.043823,25.105133 to 70.101013,29.366455 The point ought to be around here, in a sparsely mapped area: http://www.openstreetmap.org/?mlat=69.327936&mlon=26.524535&zoom=12 The primary suspects for this error are these ways: http://www.openstreetmap.org/browse/way/25684606 highway = primary ref = 92 source = gps [I added a few attributes when splitting this road] and an administrative border with a very long line segment (69.1476278, 25.7386032) to (69.3421159, 26.6885093): http://www.openstreetmap.org/browse/way/32060780 I split the highway a little (adding bridges), but this failed to address the issue. I do not think that it is proper to add points to the OSM data of the administrative border. What could be done to fix the error? Could mkgmap insert some intermediate points to the border line segment? If the error comes from the highway, I guess I could split it a bit further. Best regards, Marko

4 9

Every address and every item in category search turns up twice (r2272)
by Steve Hosgood 27 Apr '12

27 Apr '12

I use the following to generate a map from the four tiles extracted from http://download.geofabrik.de/osm/europe/great_britain/wales.osm.pbf java -Xmx2000M -jar mkgmap.jar --tdbfile --route --generate-sea=multipolygon --drive-on-left --check-roundabouts --remove-short-arcs --add-pois-to-areas --nsis --index --adjust-turn-headings --coastlinefile=../coastlines_europe-111004.osm.pbf --gmapsupp -c template.args 6324000*.osm.pbf I do have a directory called "bounds" in the working directory. Anyway - I get a navigable map with mostly-working address search. But every item in the address-search turns up duplicated as the following item in the list. Both items "work", but it isn't correct behaviour! Steve

3 3

Dubious interpretation of "Sports and Fitness Centre"
by Steve Hosgood 26 Apr '12

26 Apr '12

I see another issue. It seems that every closed polygon (and maybe POI) marked as sport=multi is being added to the list of what I see as "Sports&Fitness Centres" on my Streetpilot. Couldn't "Sports&Fitness Centres" be better considered as the catch-all for things marked "leisure=sports_centre" (which I think it does already). But don't lump "sport=multi" in there. Steve

1 0

Dubious interpretation of "Arena/Track"
by Steve Hosgood 26 Apr '12

26 Apr '12

The address-searching and associated searching for things based on their category is a very welcome recent innovation. Thank you to all who managed to decipher the file format sufficiently to allow that! I see an issue. It seems that every closed polygon (and maybe POI) marked as leisure=pitch is being added to the list of what I see as "Arena/Track" on my Streetpilot. Trouble is, this means that every occasional tennis court or football field appears seperately under this category - effectively rendering the category useless! There are hundreds of unattached pitches flagged in my area (Swansea, South Wales). Couldn't "Arena/Track" be better considered as the catch-all for things marked "leisure=track" and "amenity=stadium" (or whatever is typical for stadiums). Steve

1 0

Car routing issue with --make-all-cycleways at tile boundary with cycle way lane
by Michael Günnewig 25 Apr '12

25 Apr '12

Hi all. I have some interesting routing issue with my self-generated OSM map based on the All-In-One styles. I could track the issue down in the meantime to the --make-all-cycleways option being given to mkgmap. The issue also occurs with old mkgmap or old splitter versions, even though I'm quite sure that I didn't encounter it in the past. I first guessed that some additional data to the tiles leads to some overflow, but even reducing the number of nodes for splitting has not resolved the issue. The routing problem happens on this map excerpt: http://www.openstreetmap.org/?lat=51.50112&lon=7.34123&zoom=17&layers=M Coming from east on way 71933528 (Martener Straße) towards west to turn left onto 55188142 (Westricher Straße) crossing the segments 55262354 and then 30889150 doesn't work on my Oregon 450, when using car routing with --make-all-cycleways. It usually directs me from 71933528 towards east to do a great extra way avoiding streets with cycleways. Bicycle and foot routing that route works correctly nevertheless. Also if I just use the --make-opposite-cycleways option the car routing works also fine with the same tiles. I have also already tried to explicitly forbid bicycle traffic (bicycle=no instead of current designated) on the streets, as well as renaming the street to just its first letter, which did not help either. When splitting the germany pbf files from GeoFabrik the tile border is positioned approximately between way segments 55262354 and 30889150 (using 700k or 600k nodes). Any idea what the route cause could be here? Regards, Michael

3 5

why is mkgmap missing this road?
by michael lohr 23 Apr '12

23 Apr '12

http://www.openstreetmap.org/browse/way/158074029 i tried this: highway=tertiary [0x01 road_class=0 road_speed=0 resolution 14] and this: highway=* [0x01 road_class=0 road_speed=0 resolution 14] nothing else in the style. strangely, the default style *is* rendering this road.

1 0