Hi all,

validator would not find ways like these two:
https://www.openstreetmap.org/changeset/64081765

They are in the middle of South America. I found them only because they 
visually stuck out in JOSM. I would not find such an island when it is close to 
the real coastline
but on land.

Gerd

________________________________________
Von: mkgmap-dev <[email protected]> im Auftrag von Gerd 
Petermann <[email protected]>
Gesendet: Donnerstag, 1. November 2018 15:15
An: Thorsten Kukuk; Development list for mkgmap
Betreff: Re: [mkgmap-dev] splitter: option for maximum tile area?

Hi Thorsten,

thanks for the info. What is it that makes the correction so complicated?
My approach would be this:
1) download planet
2) filter ways with natural=coastline
3) split the file with splitter into e.g. 100 parts
4) for each part:
- open in JOSM
- run validator
- If an error is found, download that area to have current data and fix it

This should allow to find and correct unconnected coastline ways within a few 
hours.
Am I missing something?

Gerd


________________________________________
Von: mkgmap-dev <[email protected]> im Auftrag von 
Thorsten Kukuk <[email protected]>
Gesendet: Donnerstag, 1. November 2018 12:38
An: [email protected]
Betreff: Re: [mkgmap-dev] splitter: option for maximum tile area?

On Thu, Nov 01, Gerd Petermann wrote:

> Hi Bernhard,
>
> my understanding is that you should never see broken precompiled sea data, 
> because the data source is not directly OSM
> but http://openstreetmapdata.com/data/land-polygons where Jochen Topf invests 
> time to check that the data is OK before updating it.
> So, sea.zip might not be up to date but should always be OK.

Not really, as he writes on his web page, his tools tries to repair the
coastlines. And as I use them, too, I know that this automatic "repair"
can break the result even more. I have added some additional
post-precessing checks and patched osmcoastline to bail out on more
errors to make sure that the result is for me really useable.

Currently the coastlines are so heavy broken, that osmcoastline only
seg.faults. So there will be no update with the broken costlines.
I assume that why geofabrik does not report coastlines errors anymore,
too.

  Thorsten

--
Thorsten Kukuk, Distinguished Engineer, Senior Architect SLES & MicroOS
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nuernberg, Germany
GF: Felix Imendoerffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nuernberg)
_______________________________________________
mkgmap-dev mailing list
[email protected]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[email protected]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[email protected]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

Reply via email to