March 2013 - mkgmap-dev - The mkgmap lists

Routing bugs in version 2464?
by Francisco Moraes 17 Mar '13

17 Mar '13

Hi, I've noticed some strange routing on certain junctions, where it either gives me a routing error on MapSource/Basecamp or the GPS does something crazy, like a zigzag or a direct line that doesn't follow a road. I can show a picture of basecamp showing the problem. I've generated the map with the following command and version 2464: java -Xmx1792M -ea -Dlog.config=logging.properties -jar %MKGMAP%/mkgmap.jar --index --add-pois-to-areas --adjust-turn-headings --latin1 --name-tag-list=name:en,int_name,name:zh_py,name:engels,name --remove-short-arcs --make-opposite-cycleways --route --description="NC Routable" --series-name="NC Routable" --series-name="NC Routable" --country-name="UNITED STATES" --country-abbr="US" --product-id=4 --family-id=2053 --tdbfile --nsis --overview-mapname=63240000 --family-name="NC Routable" --style-file=styles-mine --style=fran --location-autofill=bounds --gmapsupp --process-exits 6324*.osm.pbf M0000805.TYP The file NC file was first split with splitter version 282: java -Xmx1792m -ea -jar %SPLITTER%/splitter.jar --no-trim --max-areas=150 --cache=cache --mapid=63240001 --max-nodes=1200000 --write-kml=world.kml --geonames-file=cities15000.zip --output=pbf north-carolina.osm.pbf Francisco

3 4

uncommitted patches for mkgmap
by GerdP 17 Mar '13

17 Mar '13

Hi, I got no feedback on these posts regarding patches: http://gis.19327.n5.nabble.com/Why-do-we-have-the-Area-too-small-to-split-a… http://gis.19327.n5.nabble.com/PATCH-v1-Optimize-filters-for-shapes-tp57441… I used both of them and I think they are ok. Is it okay to commit them? Gerd -- View this message in context: http://gis.19327.n5.nabble.com/uncommitted-patches-for-mkgmap-tp5749600.html Sent from the Mkgmap Development mailing list archive at Nabble.com.

3 6

Commit: r2532: Suppress the "Area too small to split at " messages by using an alternative split algorithm when
by svn commit 17 Mar '13

17 Mar '13

Version 2532 was committed by gerd on Sun, 17 Mar 2013 Suppress the "Area too small to split at " messages by using an alternative split algorithm when two or more large ways or polygons have the same center point.

1 0

Commit: r2531: deadendcheck_v2.patch : move the code for dead end checks into the StyledConverter. This allows e.g. to suppress warnings for oneways leading to or from
by svn commit 17 Mar '13

17 Mar '13

Version 2531 was committed by gerd on Sun, 17 Mar 2013 deadendcheck_v2.patch : move the code for dead end checks into the StyledConverter. This allows e.g. to suppress warnings for oneways leading to or from underground parking by adding a fixme tag to the node in the points style file. If the first or last point of a oneway road has a fixme tag, the dead-end-check is skipped for this road.

1 0

Re: [mkgmap-dev] Problem with bounds_*.zip
by Bernd Weigelt 16 Mar '13

16 Mar '13

I habe used the bounds.ZIP since six days without any problem my download is IHMO correct Bernd Hier sollte eigentlich eine Signatur stehen. -----Original Message----- From: "Carlos Dávila" <cdavilam(a)orangecorreo.es> To: Development list for mkgmap <mkgmap-dev(a)lists.mkgmap.org.uk> Sent: Sa., 16 Mär 2013 21:03 Subject: Re: [mkgmap-dev] Problem with bounds_*.zip El 16/03/13 18:51, Bernd Weigelt escribió: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hello > > I've got a mail from an user in Sri Lanka, who use a map on a Nüvi > > > | The last OSM Sri Lanka only map that you sent me has a Region > | Error.Garmin nuvi device shows the country > | name of the map as *Sh??rii Lnkaa.* It spells Sri Lanka deadly wrong. > | > | You can see this map error in a nuvi device just touch *Where To? > | *then *Junction. *There you can see it shows *Sri Lanka* as *Sh??rii > | Lnkaa.* > | Can you re correct this error and show me the up dated map file? > > I can see this on my Oregon too. > > I use latin1 in my conf, but i had to remove > '--bounds=bounds_20130803.zip' > from the mkgmap-options to show *Sri Lanka* > > Any hints? I downloaded bounds_20130308.zip when WanMil announced it in the list but yesterday I noticed mkgmap wasn't able to use it. I re-downloaded it and now it works fine. Maybe you have the same problem. _______________________________________________ mkgmap-dev mailing list mkgmap-dev(a)lists.mkgmap.org.uk http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

1 0

(no subject)
by Michael Breeger 16 Mar '13

16 Mar '13

1 0

Routing Problems in Mapsource
by Geoff Sherlock 16 Mar '13

16 Mar '13

Hi, I generate routable maps for walking in the UK, but have come across a couple of problems trying to route down certain ways in Mapsource. These ways do not cross any tile boundaries in Mapsource. For some reason the following unclassified highway cannot be routed along. This may be as it contains a gate at one end, but I do not show gates in my rules files for points of interest. Even though I do not show gates does this act as a barrier for routing purposes? http://www.openstreetmap.org/browse/way/166723108/history lines style file contains the following lines for unclassified highway: highway=unclassified & access!=private [0x06 road_class=1 road_speed=1 level 1] . . . . . . # No routing for private road highway=unclassified [0x06 level 1] The second one concerns a bridleway with no special attributes: http://www.openstreetmap.org/browse/way/39771110/history I thought at first it was because the bridleway I was trying to route from contained foot=unknown, so I forced this in the lines style with: highway=bridleway {set bicycle=yes; set foot=yes} highway=cycleway {set horse=no; set foot=yes} highway=footway {set horse=no; set bicycle=no} And it is drawn with: highway=bridleway [0x11 road_class=0 road_speed=1 level 1] I have attached a couple of jpegs showing how Mapsource is avoiding these ways for routing purposes. Any ideas? Cheers, Geoff.

2 4

Commit: r2530: check_connected_v2.patch: if a road is tagged with mkgmap:check_connected=true and that road shares no node with another road, add it as a line, not as a road.
by svn commit 16 Mar '13

16 Mar '13

Version 2530 was committed by gerd on Sat, 16 Mar 2013 check_connected_v2.patch: if a road is tagged with mkgmap:check_connected=true and that road shares no node with another road, add it as a line, not as a road.

1 0

Patch v1: check if road is connected
by GerdP 16 Mar '13

16 Mar '13

Hi all, attached is a patch that implements the check discussed here: http://gis.19327.n5.nabble.com/Style-function-is-connected-tp5750777.html check_connected_v1.patch <http://gis.19327.n5.nabble.com/file/n5751822/check_connected_v1.patch> If a way is a road and has has the mkgmap:check_connected=true tag mkgmap does a check if this road is connected to any other road. If not, the way is not added to the road network. Sample change in style file lines: highway=service {add mkgmap:check_connected = true}[0x07 road_class=0 road_speed=2 resolution 22] I am not sure if this gives the wanted result, so please try and let me know. Ciao, Gerd -- View this message in context: http://gis.19327.n5.nabble.com/Patch-v1-check-if-road-is-connected-tp575182… Sent from the Mkgmap Development mailing list archive at Nabble.com.

6 28

Flooding in Biograd
by RheinSkipper 15 Mar '13

15 Mar '13

After deleting the remove-short-arcs option from my settings, I can see the correct coastline in Biograd (Croatia) now. But even when I precompile sea from recent shape files or use the latest from Wanmil, I have some flooding in the harbor. I have seen other OSM maps which do not have this problem. Which option does the trick?

4 11