[Talk-ee] Administrative division

[Jaak Laineste]

> Okay ... as usual I gave it a try and hacked some proof of concept ...
> http://andre.grueneberg.de/osm/remove_spikes.pl ... at least it's
> removing 3543 "suspicious nodes" ...
> But maybe I'm totally off the track ...

This spike removal could be actually good idea. I found out now that MapInfo
had discovered automatically maybe 30% of them, perhaps only when it created
topological error (self-intersecting area). In many cases "spike" means that
there is node sequence 1-2-3(spike)-2'-4-5-6..., where there is very small
difference between 2 and 2' (0.4 cm in one sample), but it is still
technically valid polygon and therefore not in the errors file. Here nodes 3
and 2' should and could be removed, unless they are also nodes of some other
neighbor polygon (happens sometimes, but rarely). Usually 3 is not, but 2'
or 2 could be. It might be easier to find it after merging of nodes.

I have removed all tags from my file, except minimum clc_code and clc_id
(clc_key is just temporary, remove it). Maybe right place to generate Corine
tagging could be modified polyshp2osm script, based on clc_code? It does
some tagging now also, but not all of it.

/Jaak