[GRASS5] [bug #3704] (grass) d.rast.arrow does not save a line in the monitor history

[Hamish]

> This is great. Thanks much. Following up on the rest of the thread
> (not reproduced here), this is moving toward a very useful new
> visualization tool, especially if arrow magnitude (width or length,
> but maybe width is better because of length limits imposed by grid
> sizes) can be read from another map.

Changing the arrow size based on magnitude is the common convention.
Scaling both length & width is an option (be it more work) after the
simple line case is working. There is no limit to fitting the arrow
within the cell, that's just how it has been programmed so far.

The new function would place the center of the arrow's tail in the
center of the source cell, then rotate and scale the arrow size based on
the magnitude raster. A scale factor option would be given on the
command line, but the the default would be to make the longest arrow 10x
the grid size or so.

I have started on this, but with no great haste.

 
> Currently, the arrows fill grid cells, set by the region resolution.
> However, to be useful in a wider variety of uses (e.g., wind
> directions, cost surface/travel visualization, animal
> movements/migration modeling, flow intensities across channels, etc.)
> it would be useful to be able to drape a map of arrows at a coarse
> resolution over a relief map (or nviz topography) at a much higher
> resolution. I can't think of any way to do this, but maybe someone out
> there can.

g.region rast=source_map
d.rast source_map
g.region res=  (res * 16)
d.rast.arrow ...

perhaps that leads to aliasing problems?


Hamish