May 2020 - mkgmap-dev - The mkgmap lists

java.lang.StackOverflowError in length() function
by Joris Bo 28 May '20

28 May '20

Hello, I got errors in the length() function When I build a europe map the error is written 2000 lines I think (I'm not 100% sure) it start happening after adding the length calculation of waterways in the relations file to decide if rivers should be visible at lower zoomlevels For a smaller country(Belgium) there are no errors type = waterway & length() >= 0 { apply role = main_stream { set jbm_waterway_length = '${mkgmap:cache_length}'; set jbm_waterway_distance = '${distance}'; set jbm_waterway_from_relation = true; } } It starts at line 5 of the logging output, the first 4 lines are warnings about drive-left/drive-right tiles The map builds oke after that (besides a couple of tiles in the alps which maybe correlated to this length issue, I can prove that later) java.lang.StackOverflowError (Next line is written 2000 lines the first java:49 then java:59) at uk.me.parabola.mkgmap.osmstyle.function.LengthFunction.calcLength(LengthFunction.java:49) at uk.me.parabola.mkgmap.osmstyle.function.LengthFunction.calcLength(LengthFunction.java:59) at uk.me.parabola.mkgmap.osmstyle.function.LengthFunction.calcLength(LengthFunction.java:59) at uk.me.parabola.mkgmap.osmstyle.function.LengthFunction.calcLength(LengthFunction.java:59) Kind regards, Joris Bo

2 2

Commit r4504: Add robustness for length style function. Nested relations may form loops like A has member B and B has member A.
by svn commit 27 May '20

27 May '20

Version mkgmap-r4504 was committed by gerd on Wed, 27 May 2020 Add robustness for length style function. Nested relations may form loops like A has member B and B has member A. http://www.mkgmap.org.uk/websvn/revision.php?repname=mkgmap&rev=4504

1 0

Thanks
by Joris Bo 27 May '20

27 May '20

Hello, I like to say thank you, to Gerd, Ticker and others who put a lot of effort in the mkgmap development. With the new features for IS_IN and 'Nearby poi' it's possible to make the maps much nicer. The online tools for creating routes maybe better then what basecamp can do, but as soon as you want your gpx or route on your garmin device and experience the same nice layout, then have a look at my style which creates a "99,9%" identical copy of the https://www.openstreetmap.org/ layout. https://github.com/Jorisbo/Mkgmap-Mapnik-Style-Garmin/blob/master/README.md Kind regards, Joris Bo

5 4

Re: [mkgmap-dev] mkgmap-dev Digest, Vol 142, Issue 47
by ET Commands 27 May '20

27 May '20

> Date: Tue, 26 May 2020 12:46:35 -0300 (UYT) > From: muralito(a)montevideo.com.uy > To: mkgmap-dev <mkgmap-dev(a)lists.mkgmap.org.uk>, jorisbo(a)hotmail.com > Subject: Re: [mkgmap-dev] Thanks > > Yes. Thanks to all who contribute to this tool. > > Your work that mimics the web layout is impressive. I will try to build a map of my country with that style an test it. Anyway, the OSM web style is a generic one, and for driving, or cycling, it's probably better to have a specialized map for that activity. > > Regards, > M. > > > De: "Joris Bo" <jorisbo(a)hotmail.com> > Para: "mkgmap-dev" <mkgmap-dev(a)lists.mkgmap.org.uk> > Enviados: Martes, 26 de Mayo 2020 8:53:48 > Asunto: [mkgmap-dev] Thanks > > > > Hello, > > > > I like to say thank you, to Gerd, Ticker and others who put a lot of effort in the mkgmap development. > While we're thanking the developers, add my name to the list of people thanking them. For several years I've relied soley on my custom maps that I'm able to create with Mkgmap. It's a great tool! Mark B.

1 0

nearby-POI: named / unnamed
by Gerd Petermann 25 May '20

25 May '20

Hi all, in (1) Ticker suggested "I'd like to see a method that deletes all unnamed POI within {distance} of a named one. I guess this is also close to the meaning of "declutter" (I don't know how that works in Garmin devices, my Oregon doesn't have such an option) I don't fully understand the idea. I assume it is about overlapping icons? @Ticker: Should this be done for a given list of POI types or for all named POI? The default style often adds names to POI, e.g. for a barrier=bollard. Is this the name of the POI? An alternative would be to use the name=* tag of the corresponding node. Or we could introduce a new tag mkgmap:treat-as-unnamed=true which could be used in the style to mark generated names. Should it be done before or after the existing rules were applied? Remember that we have the "delete-name" action. Gerd (1) http://gis.19327.n8.nabble.com/nearby-POIs-tp5964757p5964995.html

4 9

Excluding words from index using x-mdr7-excl
by Joris Bo 25 May '20

25 May '20

Hi, Searching my Oregon almost becomes impassable due to my usage of big maps and detail. So I like to test what I can do with the search index. Maybe somebody can give me some quick directions or confirmations before I spent to much time on a dead end? I personally don't like the idea of the road-name-config=filename because it also changes the names on the map itself So I like to focus on the option --x-mdr7-excl="weg,street,road" * Am I correct that poi-index and address search index are two different indexes with two independent/different approaches for keeping them smaller? * Does this x-mdr7-excl option only affects address search and not the poi search? * Can I use a config file for it (I'm now trying to use -read-config=c:\x-mdr7-excl.config and within this file I use x-mdr7-excl="weg,street,road" * but this feels like pretty soon I'll bounce on some length limits? * As far as I tried now it does not accept a path name * Will this approach really helps me to make the index smaller by cleaning all the 'rue,de,la,le,van,von,road,weg,square,platz,lane' words? * Does it somehow conflict with the split-name-index? * Can I open the index osmmap_mdr.img? in excel or other tool to analyse the effects of my work (now I'm testing trough my Oregon which is extremely time consuming and a source of noise) * If lets say the word 'Road' is removed from the index and I search for 'Main' to find the 'Main Road' will the search-result still display the original full name 'Main Road' or just the part 'Main' which was indexed? * Indexing highway refs seems less important to me in this situation, any idea how I could filter them out. * It looks like if I have two maps overlapping the same area on my Oregon, it searches both indexes, displaying two results, even if I only have one of the maps enabled? * One of my (disabled) overlapping maps is a occasionally used transparent contourlines map. Regarding the previous finding: are contourlines and their number/names indexed ? * Will basecamp mapinstall rebuild the indexes when transferring the map? I think I see some hints about this passing by when transferring large maps. Probably throwing all this effort away? Probably easily to be solved by placing a gmapsupp.img manually. Hopefully not to many questions at once, Thanx in advance, Kind regards, Joris Bo

2 1

Commit r4503: merge the nearby-POI branch which implements and documents new options --nearby-poi-rules and --nearby-poi-rules-config.
by svn commit 25 May '20

25 May '20

Version mkgmap-r4503 was committed by gerd on Mon, 25 May 2020 merge the nearby-POI branch which implements and documents new options --nearby-poi-rules and --nearby-poi-rules-config. http://www.mkgmap.org.uk/websvn/revision.php?repname=mkgmap&rev=4503

1 0

Commit r4502: add one more method to GpxCreator which helps debugging
by svn commit 25 May '20

25 May '20

Version mkgmap-r4502 was committed by gerd on Mon, 25 May 2020 add one more method to GpxCreator which helps debugging http://www.mkgmap.org.uk/websvn/revision.php?repname=mkgmap&rev=4502

1 0

Compilation fails building France map
by Steph 24 May '20

24 May '20

Hi, Since a couple of days the compilation for my France map fails. Perhaps the output log could help you to indicate me the direction I should investigate… - mkgmap-r4486 - java -version: openjdk version "13.0.2" 2020-01-14 OpenJDK Runtime Environment (build 13.0.2+8) OpenJDK 64-Bit Server VM (build 13.0.2+8, mixed mode) - France.log: Time started: Sat May 23 10:15:31 EAT 2020 The following styles are available: default 1.27: "Default style" modifié. Style détaillé pour un usage pédestre mais fonctionnel aussi en usage cycliste ou motorisé. Found one style in /home/steph/osm/styles/default/ finished check-styles Number of MapFailedExceptions: 0 Exception in thread "main" java.lang.AssertionError: 164041 at uk.me.parabola.imgfmt.app.BufferedImgFileWriter.put2u(BufferedImgFileWriter.java:164) at uk.me.parabola.imgfmt.app.trergn.Zoom.write(Zoom.java:81) at uk.me.parabola.imgfmt.app.trergn.TREFile.writeMapLevels(TREFile.java:215) at uk.me.parabola.imgfmt.app.trergn.TREFile.writeBody(TREFile.java:123) at uk.me.parabola.imgfmt.app.trergn.TREFile.write(TREFile.java:342) at uk.me.parabola.mkgmap.build.MapBuilder.makeMap(MapBuilder.java:305) at uk.me.parabola.mkgmap.combiners.OverviewBuilder.writeOverviewMap(OverviewBuilder.java:191) at uk.me.parabola.mkgmap.combiners.OverviewBuilder.onFinish(OverviewBuilder.java:104) at uk.me.parabola.mkgmap.main.Main.endOptions(Main.java:674) at uk.me.parabola.mkgmap.CommandArgsReader.readArgs(CommandArgsReader.java:125) at uk.me.parabola.mkgmap.main.Main.mainStart(Main.java:143) at uk.me.parabola.mkgmap.main.Main.main(Main.java:114) Two little overlays (for ski resorts and particular osm tags I want to verify) are correctly produce. The main map compilation fails! Regards. Steph

4 11

Explanation of the is_in function
by Joris Bo 23 May '20

23 May '20

Hi When testing the IS_IN function I get very nice results ! Thank you very much for this improvement. Could somebody help me to explain why a certain pedestrian is not catched by the rules? Also not a big deal, I'm already happy wit the current results, it is just for understanding. I am using pedestrian streets without an outline if crossing pedestrian areas. 95% works perfectly but some seem to fool me. Before (The red lines are nicely replaced in the after situation, but the blue one is not) highway = pedestrian & jbmsquare != true & is_in(highway,pedestrian, any) = true [0x10d0c resolution 24 continue] area: München Viktualienmarkt: https://www.openstreetmap.org/#map=19/48.13504/11.57637 pedestrian : https://www.openstreetmap.org/way/240014256 square: https://www.openstreetmap.org/relation/7431621#map=19/48.13512/11.57670 [cid:image002.jpg@01D62CF3.C553B200] After [cid:image005.jpg@01D62CF3.C553B200] Met vriendelijke groet, Joris Bo jorisbo(a)hotmail.com<mailto:jorisbo@hotmail.com>

3 16