[GRASS5] [bug #3704] (grass) d.rast.arrow: magnitude raster map sets arrow length

[Harmish Bowman via RT]

> Someday I would like to see a general raster vector display tool in
> GRASS. Something like grdvector in GMT. 

At some point it is better to let GMT do the job of cartography versus 
duplicating the effort within GRASS. Each to our own strengths.
(As long as we make it easy to im|export GRASS maps to GMT, Matlab, et al.)

> What's the MATLAB equivalent pswiggle or something like that?

quiver.  I'm not familiar with anything like "pswiggle".


> Anyway, something that can handle direction and magnitude from either 
> U and V components, or a single grid with bearing/angle.
[..]
> As for vector plots, wouldn't it (also) make sense to take two rasters
> U and V where the arrow direction:
>  dir =3D atan2(u, v)
>  mag =3D sqrt(u^2 + v^2)

I prefer magnitude,direction maps to U,V maps as GRASS already works with 
direction maps. Converting U,V to theta,r maps is only a r.mapcalc away.


> It should also probably handle both Cartesian and Nautical conventions 
> for representing angles 

[done]

> and both Atmospheric and Oceanographic conventions for plotting the 
> arrows (currents coming from or currents going to).

Again, I prefer to keep the module simple and let the user use r.mapcalc if
they want the other convention. This makes us all think about it a bit more?
d.rast.arrow expects aspect maps to point downslope, ie "going to".


Hamish


-------------------------------------------- Managed by Request Tracker