[GRASS-dev] v.clean meters against feets

[Moritz Lennert]

On 27/08/14 10:01, Luca Delucchi wrote:
> Hi everybody,
>
> I'm testing v.clean with meters and feet projection system.
> I'm trying to run v.clean in the same map and with the same
> parameters, but I obtain a different result. For the testing I'm using
> zipcodes_wake map of nc_spm_08
>
> Here my procedure:
>
> - download nc_spf location from here [0]
> - reproject zipcodes_wake from nc_spm_08 to nc_spf
>
> - run v.clean on meters location with "v.clean in=zipcodes_wake
> out=zipcodes_wake_clean tool=rmarea thre=10000000" the result is
>
> Number of nodes: 106
> Number of primitives: 166
> Number of points: 0
> Number of lines: 0
> Number of boundaries: 135
> Number of centroids: 31
> Number of areas: 31
> Number of isles: 2
>
>
> -  run v.clean on feet location with "v.clean in=zipcodes_wake
> out=zipcodes_wake_clean tool=rmarea thre=107638674"
>
> Number of nodes: 84
> Number of primitives: 103
> Number of points: 0
> Number of lines: 0
> Number of boundaries: 93
> Number of centroids: 10
> Number of areas: 10
> Number of isles: 1
>
> The threshold in feet should be the same of the meters one because:
>
> 1 meters is 0.3048006096012192 feet (according with g.proj output)
>
> and
>
> 10000000 squared meters is 107638674 squared feet ( 10000000 / 0.3048^2 )
>
> I expected the same output map, but this not happen; I'm wrong or
> there is something wrong in the v.clean code?

Have you compared the topo information of the two files before v.clean ? 
Maybe reprojection changes some of the topology (i.e. some lines do not 
close, or something similar) ?

Moritz