Hi Jan

It wasn't so much the type number as the size of a poi bitmap.

In the 'olden' days Mapsource would allow pois with lengths and width equalling zero. This would render them transparent, useful for names of places.

However, Basecamp would not accept such pois and I think has a problem displaying some pois with length and width  set to 1.

I'm glad it wasn't the typ file.

r

Nick

On 07/08/2021 14:59, jan meisters wrote:
Hi Gerd,

yes, 4805 renders correctly. Thanks!

@Nick: I tried without typ, but that made no difference. Do you remember the 
faulty POI?

Jan

Am 07.08.2021 um 11:24 schrieb Gerd Petermann <[email protected]>:

Hi Jan,

I think r4805 fixes the problem. I've not analysed why the default style 
doesn't show the problem. The routing data is very different.
It seems that the OFM style produces a lot more restrictions.

Gerd

________________________________________
Von: Gerd Petermann <[email protected]>
Gesendet: Samstag, 7. August 2021 09:47
An: Development list for mkgmap
Betreff: AW: [mkgmap-dev] Bad tile around East London / Greenwich

Hi Jan,

I can reproduce the problem with the OFM full style. With the just released 
r4804 I see this:
SCHWERWIEGEND (global): Bad file format: e:\Jan\10700105.osm.pbf
uk.me.parabola.imgfmt.FormatException: too many restrictions
        at uk.me.parabola.imgfmt.app.net.TableC.getOffsetSize(TableC.java:84)
        at 
uk.me.parabola.imgfmt.app.net.TableC.propagateSizeBytes(TableC.java:88)
        at 
uk.me.parabola.imgfmt.app.net.RouteCenter.updateOffsets(RouteCenter.java:87)
        at uk.me.parabola.imgfmt.app.net.RouteCenter.write(RouteCenter.java:98)
        at uk.me.parabola.imgfmt.app.net.NODFile.writeNodes(NODFile.java:101)
        at uk.me.parabola.imgfmt.app.net.NODFile.write(NODFile.java:71)
        at uk.me.parabola.mkgmap.build.MapBuilder.makeMap(MapBuilder.java:352)
        at uk.me.parabola.mkgmap.main.MapMaker.makeMap(MapMaker.java:114)
        at uk.me.parabola.mkgmap.main.MapMaker.makeMap(MapMaker.java:70)
        at uk.me.parabola.mkgmap.main.Main.lambda$1(Main.java:290)
        at java.util.concurrent.FutureTask.run(FutureTask.java:266)
        at 
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
        at 
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
        at java.lang.Thread.run(Thread.java:748)

I think this problem is caused by an error in mkgmap, it seems to put too many 
restrictions into one so called RouteCenter.
Have to check the details...

Gerd

________________________________________
Von: mkgmap-dev <[email protected]> im Auftrag von jan meisters 
<[email protected]>
Gesendet: Freitag, 6. August 2021 22:41
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] Bad tile around East London / Greenwich

Hi Gerd,

yes, but only on my own style:
SCHWERWIEGEND (global): Bad file format: 10700105.osm.pbf

File uploaded here<https://files.mkgmap.org.uk/download/517/10700105.osm.pbf>. 
It is splitter-generated from a geofabrik great-britain-latest with max-node=1400000.
Splitter 615 and mkgmap 4802.

For my own and and the default I have the same splitter options:
java -Xmx26000m -jar splitter.jar --output=pbf 
--output-dir=/great_britain-splitter --max-nodes=1400000 --mapid=10750001 
--geonames-file=PRECOMP/cities15000.txt --polygon-file=ARGS/great_britain.poly 
DATA/great_britain-latest.osm.pbf

And mkgmap then:
java -Xmx26000m -jar mkgmap.jar --output-dir=great_britain-output -c 
ARGS/great_britain-HGT.args -c great_britain-splitter/template.args 
ARGS/10000.txt

The mkgmap ARGS for my style are:
family-id: 10700
product-id: 1
draw-priority: 30
series-name: Great_Britain-HGT
family-name: Great_Britain-HGT
description: Great_Britain-HGT
copyright-message: openstreetmap.org<http://openstreetmap.org> | 
earthexplorer.usgs.gov<http://earthexplorer.usgs.gov> | 
viewfinderpanoramas.org<http://viewfinderpanoramas.org> | 
openfietsmap.nl<http://openfietsmap.nl>
area-name: Great_Britain
country-name: Great_Britain
country-abbr: Great_Britain
style-file: Styles
overview-mapname: 10700000
precomp-sea: PRECOMP/sea-latest.zip
generate-sea: land-tag=natural=background
bounds: PRECOMP/bounds-latest.zip
location-autofill: is_in,nearest
housenumbers
split-name-index
tdbfile
latin1
name-tag-list: name,name:de,int_name,place_name
code-page: 1252
dem: HGT_NASADEM_Europe
dem-dists: 3314,4971,6628,13256,29826
overview-dem-dist=89478
show-profiles: 1
merge-lines
road-name-config=examples/roadNameConfig.txt
add-pois-to-areas
pois-to-areas-placement
add-pois-to-lines
link-pois-to-ways
poi-excl-index=0x3200-0x331f
nearby-poi-rules=*/named:5,*/unnamed:10,0x2b07:50,0x2f00-0x2f03:50,0x2f04-0x2f09:25,0x2f0a-0x2f0b:25,0x2f0c:50,0x2f0d-0x2f0f:25,0x2f10-0x2f16:50,0x2f17-0x2f18:25,0x2f19:50,0x2f1a-0x2f1f:50,0x3200-0x331f:50,0x13700-0x13702:50,0x13703-0x13707:25,0x13708-0x13709:50,0x1370a-0x1370f:50
min-size-polygon: 4
polygon-size-limits=16:2, 14:0
preserve-element-order
keep-going
route
index

And mkgmap ARGS for default:
family-id: 10750
product-id: 1
draw-priority: 30
series-name: Great_Britain-default
family-name: Great_Britain-default
description: Great_Britain-default
area-name: Great_Britain
country-name: Great_Britain
country-abbr: Great_Britain
style-file: examples/styles/default
overview-mapname: 10750000
precomp-sea: PRECOMP/sea-latest.zip
generate-sea: land-tag=natural=background
bounds: PRECOMP/bounds-latest.zip
# location-autofill: is_in,nearest
# housenumbers
# split-name-index
# tdbfile
# latin1
# name-tag-list: name,name:de,int_name,place_name
code-page: 1252
# dem: HGT_NASADEM_Europe
# dem-dists: 3314,4971,6628,13256,29826
# overview-dem-dist=89478
# show-profiles: 1
# merge-lines
# road-name-config=examples/roadNameConfig.txt
add-pois-to-areas
# pois-to-areas-placement
add-pois-to-lines
link-pois-to-ways
# min-size-polygon: 4
# polygon-size-limits=16:2, 14:0
# preserve-element-order
keep-going
route
index



Am 06.08.2021 um 19:58 schrieb Gerd Petermann 
<[email protected]<mailto:[email protected]>>:

Hi Jan,

Any error message from mkgmap?
I can try to reproduce if you upload the input osm file to 
https://files.mkgmap.org.uk/
(or post another download link)
and the options that you use.


Gerd

________________________________________
Von: mkgmap-dev 
<[email protected]<mailto:[email protected]>> im 
Auftrag von jan meisters <[email protected]<mailto:[email protected]>>
Gesendet: Freitag, 6. August 2021 19:35
An: Development list for mkgmap
Betreff: [mkgmap-dev] Bad tile around East London / Greenwich

Hi all,

since some time my map has a bad tile around East London (Greenwich).
I´ve checked earlier revisions of splitter and mkgmap, and decreased splitter´s 
max-nodes - without success.

Problem seems to be in my style, which is based on OpenFietsMap. With the 
original OFM-style the error is rendered as well, but not with the mkgmap 
default style.
But I´ve got no clue where to look for the mistake -  anyone has got a hint for 
me?

https://files.mkgmap.org.uk/download/516/bad_tile-east_london.jpg

Jan

_______________________________________________
mkgmap-dev mailing list
[email protected]
https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[email protected]
https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

_______________________________________________
mkgmap-dev mailing list
[email protected]
https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[email protected]
https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[email protected]
https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

Reply via email to