Routable Garmin maps: http://emexes.powweb.com/osm/

You can use the ‘merge nodes’ function in JOSM, after doing a rectangular selection around these nodes.

Split the roundabout into pieces at the crossings. Having ‘roundabout’ as topological object is a bad idea anyway.

I’m using some quick and dirty awk scripts to extract, download and edit certain OSM topological errors, but maybe we can extend
osm2mp to work as an osm2osm preprocessor. Then it will be possible to re-upload preprocessed data directly to the server.
It opens a big can or worms with concurrency, if two people will try to reprocess the same or overlapping areas simultaneously.
Fortunately, there are not that much people editing OSM late at night :slight_smile:

You can’t avoid ‘mapping-for-garmin’ here, so think out some soultion that will be commonly acceptable.

Error R009: Limit of 60 segments for a single road exceed.

osm2mp should deal with it, since it’s a garmin-specific problem.

Distance is below 5.4, NODID1: 9513, NODID2: 8571 RgnID: 14

It does not show up in gpsmapedit, but i will check again.

Internal error during split of RGN80 objects - if this is supported version of program - contact with gps_mapper

I doubt we can do something about cgpsmapper internal bugs without having the source code.
One more reason to finish the NOD research, add it to mkgmap and put this cgpsmapper guy out of the business :wink:

this fatal error is caused while trying to recover this error:

the way has nodes with different ID but with same coords. just redraw that way

I’m doing the full n(n-1)/2 distance qsort() on the original routable nodes table, so it just can’t happen.

edit:
sorry, the bridleways and similar weird ways were removed from the table. maybe it explains
what’s going on here.

Garmin has a specific type for roundabout (0xc, polyline). Can we have all ways tagged as junction=roundabout use this type. Modifying poly.cfg doesn’t seem to work. You might also want to set all roundabouts as oneway=yes if it was not tagged as such.

Thanks for such a great tool

Hi Guys,

I got a program written in python that corrects all the situations that do not fulfil Garmin requirements. But it makes the corrections while reading data. I have never planed to update OSM database itself. This is because I think, that these requirements are Garmin specific and OSM is not. What if there is someone preparing routable maps for X devices and they require roundabouts from one piece and nodes laying no closer that 10 meters? Should he correct the data in OSM? I’m sure not. Correcting data in fly should be no problem.

I quickly (=without lot of work) generated a map of Petersburg with my tool. You can download it here: [link removed]. I made a mistake declaring an area and the city is a bit off centre, but this shouldn’t be an issue. Everything that I have done manually was to declare an area of the map, it’s ID, name and some paths to planet and osm files. The whole process is automatic. I don’t correct any data in OSM. It took about 1,5 hour of computing time, but almost all the time was needed for Osmosis to extract the data from planet file. And obviously I am not yet ready to support cyrillic languages :wink:

I checked the map quick on my eTrex and it seems to be routable. Here is the screenshot:

Admin edit: Link not working anymore

this osm2mp version uses latest tag to determine object type
so if it sees “junction=roundabout” after “highway=” it sets Type=0x0c, or highway type if before
some priority mechanism must be imbplemented to have it always use 0x0c
i’ll think about it :slight_smile:

add line

      $polydir     = 1                  if ("$1=$2" eq "junction=roundabout");

before

      $polydir     = $yesno{$2}         if ($1 eq "oneway");

I think that certain changes should be put back into the database (all that are non-garmin specific).

It is much faster (and easier) to use OSMXAPI DB for this purpose.

I think it’s garmin to blame here for not using the utf8 fonts.

That should be easy to solve, if koi8r is the russian charset.

iconv -f utf8 -t koi8r planet-extract.osm >russian.osm

Might not parse in all OSM tools (osmosis) after that but it should convert from utf8 to koi8r.

garmin uses cp1251, but it is not an issue here. St. Petersburg area
borders with Finnland (iso8859-1) and Estonia (iso8859-4), so
it’s a pity to loose all national characters, which do not fit into
one codepage.

I agree. But I received several “Internal Server Error” responses while unsuccessfully trying to fetch London area so I switched to planet file which is a pain, but works. I am converting huge amounts of data (several GB) and I currently temporary have rather poor connection.

yes… I’m thinking when we get more routable maps people will see the real benefit of OSM and that can only boost the number of users contributing to the “map”
good job.

Hi,
i created a wikipage about routable garmin maps. Please add some information: http://wiki.openstreetmap.org/index.php/Routable_garmin_maps

Jonas

Do we know how these maps were created ?
I didn’t find any published source code, so it has zero value for me,
compared to the excellent work of liosha.

I think Radomir may release code once the code has reached some point of full completion , but this is what I heard…

This is what actually have kept me away from OSM for some time, couldn’t make use of the maps, but with the recent tools being available I’ve started contributing to the project. A couple of thousand nodes in the last week :slight_smile:
I believe this is the key to getting succesful, along with improving the map making software (I’m currently using latest beta of Meerkartor, which is ok but not perfect, maybe time to convince the guy making GPSMapEdit to build in support for OSM??)

I have read his comments too, let’s hope it will happen soon :slight_smile:

release early and often… :slight_smile:

Radomir, could you please edit the start post of this thread and update it with latest info (like download location). That way new readers don’t have to go through the entire topic when they just want to download the maps. Thanks!

I just received a report from a user of your maps that he can’t search for place/city names. I assume this is a bug somewhere?