[R] adding 3D arrows to 3D plots

[Peter Ehlers]

Cool, Barry.
I set n=30, col="red" and stuck it into my daiquiri!

  -Peter Ehlers

Barry Rowlingson wrote:
> Have a go with this:
> 
> arrow3d <- function(p0=c(0,1,0),p1=c(1,1,1),s=0.1,theta=pi/4,n=3,...){
>  ##    p0: start point
>  ##    p1: end point
>  ##     s: length of barb as fraction of line length
>  ## theta: opening angle of barbs
>  ##     n: number of barbs
>  ##   ...: args passed to lines3d for line styling
> 
>  require(geometry)
>  require(rgl)
> 
>  ## rotational angles of barbs
>  phi=seq(0,2*pi,len=n+1)[-1]
> 
>  ## length of line
>  lp = sqrt(sum((p1-p0)^2))
> 
>  ## point down the line where the barb ends line up
>  cpt=(1-(s*lp*cos(theta)))*(p1-p0)
> 
>  ## draw the main line
>  line = lines3d(c(p0[1],p1[1]),c(p0[2],p1[2]),c(p0[3],p1[3]),...)
> 
>  ## need to find a right-angle to the line. So create a random point:
>  rpt = jitter(c(
>    runif(1,min(p0[1],p1[1]),max(p0[1],p1[1])),
>    runif(1,min(p0[2],p1[2]),max(p0[2],p1[2])),
>    runif(1,min(p0[3],p1[3]),max(p0[3],p1[3]))
>    ))
> 
>  ## and if it's NOT on the line the cross-product gives us a vector
> at right angles:
>  r = extprod3d(p1-p0,rpt)
>  ## normalise it:
>  r = r / sqrt(sum(r^2))
> 
>  ## now compute the barb end points and draw:
>  pts = list()
>  for(i in 1:length(phi)){
>    ptb=rotate3d(r,phi[i],(p1-p0)[1],(p1-p0)[2],(p1-p0)[3])
>    lines3d(
>            c(p1[1],cpt[1]+p0[1]+lp*s*sin(theta)*ptb[1]),
>            c(p1[2],cpt[2]+p0[2]+lp*s*sin(theta)*ptb[2]),
>            c(p1[3],cpt[3]+p0[3]+lp*s*sin(theta)*ptb[3]),
>            ...
>            )
>  }
>  return(line)
> }
> 
> This creates a line with 'n' arrow barbs at one end, equally spaced
> when you look at the line end-on. The barb length is 's' times the
> length of the line, and the opening angle is theta. Just do arrow3d()
> to get something.
> 
> Might be useful, plus I wanted to brush up on my vector geometry anyway...
> 
> Barry
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 
> 

-- 
Peter Ehlers
University of Calgary
403.202.3921