[GRASS-user] Network problem and v.net.visibility approach

Moritz Lennert mlennert at club.worldonline.be
Fri Feb 28 07:26:06 PST 2014


On 28/02/14 15:58, Margherita Di Leo wrote:
> Dear All,
>
> Can this problem [1] posed on stackexchange be solved using
> v.net.visibility?

For those to lazy to go and look:

"I'm dealing with a quite big line shapefile containing road tracks. The 
roads are connected to each other, creating distinct networks. These 
networks are independent and they are not connected to each other.

I'm trying to find a way to identify these networks, count them and 
eventually measure their total length."


No. v.net.visibility connects features that are visible from each other.

> Or I am completely wrong? I'm trying it on a dataset but
> it takes forever, so I can't actually tell how the output looks like. Any
> other options that come on your mind?

You might want to look into v.spanningtree or some of Daniel Bundula's 
network connectivity modules, notably v.net.components, although I'm not 
sure how both of these handle seperated parts of the network.

Or you could work with v.net.allpairs to identify all nodes connected to 
each other (and thus, by inverse logic all those that are not connected).

Moritz


More information about the grass-user mailing list