[GRASSLIST:4849] Re: transition probability

[Radim Blazek]

Michael Barton wrote:
> Dylan,
> 
> Thanks for summarizing this. I've tried to follow this and more or less see
> what you did, although I'd actually need to go through it to really
> understand it.
> 
> I've got a similar, but slightly different problem. I will be trying to
> figure out how to do it, but hope you or someone else out there can suggest
> a solution. It is a general problem that many people might run into.
> 
> I have a lot of CAD maps that are digitizations in MicroCAD (*.dgn) of
> topographic maps from an area in Spain. When I import them in v.in.ogr, I
> get the vector shapes I need. However, the attributes that need to be
> attached to the vectors are actually attached to points that mark where the
> CAD folks put the label for the vector shape. So, for example, for each
> topographic line, I have a nice line without an attribute and a point
> somewhere near the line with the elevation value. The same goes for roads,
> streams, etc. I can extract each kind of vector shape and each kind of point
> to separate maps. 
> 
> I've done this for areas, but these are lines. I suspect that I can
> associate points with their vector shapes  geographically by giving each
> point a tiny buffer so that it overlays the associated line. But where do I
> go from here? If I could get the relevant line cat value into the proper
> point attribute table, I could reconnect the lines to the attributes, but
> I'm not quite sure how to do this.

v.distance

Radim