- mkgmap-dev - The mkgmap lists

Documenting the various sorts of file
by ael 10 Jun '20

10 Jun '20

The current help doesn't explain what sorts of files can be processed. What documentation there is seems to be scattered, and difficult to discover. Having skimmed the source, it looks as if FileKind.java is the place where the various sorts are handled. I am not quite sure where the test-map:all-elements and test-map:test-points fit in. Is there any chance that someone who knows the source properly could enhance the help messsage? Or document it somewhere else. ael

2 1

Re: [mkgmap-dev] Performance of POI search on the Device
by Franco Bez 10 Jun '20

10 Jun '20

Hi Gerd, I tried out my old Oregon 550 and my ZUMO 396 The Oregon 550 behaves almost identical to the Oregon 750t. The boundary layer slows down the POI search by factor 10. Round about 1 sec without and 9 sec with the boundary layer. If I put all transparent layers on the SD Card it's 50 sec. I assume the biggest impact comes fron my contourlines layer with DEM Data for all of Europe (3,7GB size). The Zumo seems to be faster. The boundary layer is no big problem (maybe 1 sec penalty). But when I put my contourlines layer on the Zumo it slows down extremely. The GARMIN Firmware bug seems to affect several devices, if not all. Ciao, Franco

2 3

Maps with contours shows two maps in some devices (i.e. Zumo XT)
by Enrico Liboni 09 Jun '20

09 Jun '20

Hi there. I create my maps with contours using the following command java -jar ./mkgmap/mkgmap.jar --description="*$family_name - OSM*" --output-dir="$dest"/ \ --family-id=10030 --product-id=1 --family-name="*$family_name - Map*" --route --remove-short-arcs --bounds=bounds.zip \ --precomp-sea=sea.zip --location-autofill=is_in,nearest --draw-priority=20 --gmapsupp --index --housenumbers "$work"/"$suffix"*osm.pbf ./typ/gspeed.typ \ --family-id=10031 --product-id=1 --family-name="*$family_name - Contours*" --draw-priority=10000 --transparent $contours ./typ/gspeedContours.typ In blue the part that refers to contours. Mapids for the two sets are different. On my Garmin Montana I see just one map with the name passed in --description option (*$family_name - OSM*). On an old Garmin Nuvi I see two maps, the real map with streets (named *$family_name - Map*) and the one with contours (named *$family_name - Contours*), which I can enable/disable individually. On the new Garmin Zumo XT I see two maps, both named *$family_name - OSM. *I can't disable just one of the two, they seems to be linked and it is weirs they show up with very same name in the list. Question: is there a way I can merge the two maps so I get the same behaviour anywhere? Thanks! Enrico

4 5

Split gmapsupp.img files over 4GB
by Felix Hartmann 09 Jun '20

09 Jun '20

Could mkgmap automatically split gmapsupp.img over 4GB (or have an option to do so - but besides Oruxsmaps/locus and some other not often used tools - I don't know any others that can use larger gmapsupp.img files duet to FAT32). It would be nice if those gmapsupp.img files were somehow rastered - but even just according to mapid would be fine. -- Felix Hartman - Openmtbmap.org & VeloMap.org

2 4

is_in() function for point on line
by Joris Bo 09 Jun '20

09 Jun '20

Hi Does the is_in() function should also work for a point on a line in stead of in a polygon? highway = turning_circle & is_in(highway, track, in_or_on) = true [0x641a resolution 24] It returns false for this situation where the turningcircle is attached to a highway = track. (mkgmap:cache_is_in_POINT_highway_track_in_or_on=false) Would there be another way to know if the point is attached to a highway = track rather then a for example highway = residential? I use add-pois-to-lines, and link-pois-to-ways but as far as I can see, that doesn't give me more information. https://www.openstreetmap.org/#map=18/49.65888/5.45845 [cid:image002.jpg@01D636A9.5643B8A0] Kind regards Joris

3 25

add-pois-to-lines documentation
by Mike Baggaley 09 Jun '20

09 Jun '20

Hi Gerd, please find attached a proposed patch to the documentation for the add-pois-to-lines option as requested. Cheers, Mike >I've implemented the suggested improvements with r4507. Please suggest improvements for the documentation changes.

2 1

New style function is_drive_on_left()
by Dave Jackman 09 Jun '20

09 Jun '20

Hi Gerd, Have done a quick test and the new function seems to work fine, I will compile a map covering both Zambia and DRC to fully test it. But I have received an error message: Error in style Error: (lines:395) : Error: No function with name 'length()' So it seems the length() function may no longer work. Dave >

4 4

Re: [mkgmap-dev] Nearby POIs
by Mike Baggaley 09 Jun '20

09 Jun '20

Hi Gerd, The solution gets rid of the unexpected error message, however, it does not delete POIs that were originally coincident. Perhaps a better solution would be not use the set of locations for testing uniqueness in removeSimpleDuplicates, instead using a list and trawl the list using high precision comparison to determine whether the coordinate matches another. It could even compare both in high precision and displayed precision so that POIs which have been moved to be coincident are also removed. As the list will only contain a few coordinates it shouldn't be expensive. To help minimise enumerations, the list could be ordered by longitude or latitude if necessary. Another option would be to derive a class with different iterator comparator which might be useful elsewhere. Regards, Mike >I added options --route --link-pois-to-ways and finally could reproduce the problem with the small file. >Should be fixed with r4523. The high-prec-stuff makes many things better but it is also is a constant source for problems like these

2 1

Commit r4525: nearby-poi: compare high prec locations in removeSimpleDuplicates() to handle duplicate POI which were moved to different places by WrongAngleFixer.
by svn commit 09 Jun '20

09 Jun '20

Version mkgmap-r4525 was committed by gerd on Tue, 09 Jun 2020 nearby-poi: compare high prec locations in removeSimpleDuplicates() to handle duplicate POI which were moved to different places by WrongAngleFixer. http://www.mkgmap.org.uk/websvn/revision.php?repname=mkgmap&rev=4525

1 0

default style points enhancements
by Ticker Berkin 09 Jun '20

09 Jun '20

Hi Gerd and followers of the default style Here are some enhancements to default style "points" - patch attached. I've been building up a list of changes to the default style for a while, and, with the current discussion about multiple POI for a single OSM object and the benefit of benches (I've just added this), it seemed like a good time to put forward these changes. Some of these derive from: http://www.mkgmap.org.uk/pipermail/mkgmap-dev/2019q4/030097.html Some from other comments in this forum and the rest are from when I've noticed something wrong or that could be improved. Summary of changes: - for amenity={atm/bar/cafe} allow additional POI from other tags, eg shop=bakery or any others: leisure/tourism/sport/...=... - create POI 0x6605 for bench (Geographic>Land>Bench) - don't create POI for private parking - create parking POI for parking_entrance - if parking unnamed, use access tag if exists, giving names like "Customer parking", "Permissive parking"... - explicit shelter_type=basic_hut rather that trying to exclude bus shelters etc for Find>Lodgings - marina code from 0x4300 to 0x2f09; searchable as "Others" - POI for nature_reserve, even if not named - don't create POI for pitch if no name or sport - don't create POI for private swimming pools - don't create POI for unnamed cliff - add POI 0x6503 for named bay (Geographic>Water>Bay) - for platforms, just 1 POI when --add-pois-to-lines - name unnamed shops as category, regardless of category being matched - add, commented out, extra shop categories supported by some devices - add POI for shop=outdoor - for {camp/caravan}_site have BOTH 0x2b03 AND 0x2b05; 03 not shown on modern devices, 05 not shown on old devices - have both Food & Lodging POI for same point if appropriate - landuse=basin code from 0x650f (Geographic>Water>Reservoir) to 0x6603 (Geographic>Land>Basin) - water=reservoir as well as landuse=reservoir - add POI 0x650d for named water=lake/pond (Geograhic>Water>Lake) I'm happy to explain any of these change in more detail and, if anyone considers a particular change incorrect, provide a revised patch. Ticker

3 6