rgeos (version 0.2-13)

calc.labpt.strings: optimally compute label points

Description

optimally compute label points

Usage

calc.labpt.strings(pol, label, gridpoints = 50, cex = 1,...)
polygonsLabel(pol, labels, ...)

Arguments

pol

object of class or deriving from SpatialPolygons

label

character; vector of length one with label

labels

character; vector with labels

gridpoints

integer; number of grid points to try

cex

numeric; magnification factor for text labels

...

argument(s) passed on to calc.labpt.strings, and from there to strwidth and strheight

Value

• This version of the code, which uses Nelder-Mead to zero in on the 'optimal' label position.

  But as alluded to in an earlier post, it won't necessarily work well for more regular polygons, as it's only optimal in the sense that the label position is *a* position with greatest distance to the polygon boundary. If there are many positions, this might be far from the visually optimal position.

  In the example below, the 'Hi!' label will always be placed in centre of the square, but the position of the two other labels can differ much (rerun the code to see them change). For example, the 'Narrow' label can be positioned almost at the bottom or the top of the narrow rectangle (but always horizontally in the middle), since moving it up or down doesn't change the (shortest) distance to the polygon boundary.

  calc.lbpt.strings is for use on a SpatialPolygons object with a single (multi)polygon).

Examples

x=c(0,1.8,1.8,1,1,3,3,2.2,2.2,4,4,6,6,14,14,6,6,4,4,0,0)
y=c(0,0,-2,-2,-10,-10,-2,-2,0,0,1.8,1.8,1,1,3,3,2.2,2.2,4,4,0)
xy=data.frame(x,y)

xy.sp=SpatialPolygons(list(Polygons(list(Polygon(xy)), ID="test")))

labels=c("Hi!", "A very long text string", "N
a
r
r
o
w")
plot(xy.sp, col="khaki")
for( label in labels) {
  xy=calc.labpt.strings(xy.sp, label)
  text(xy[1], xy[2], label)
}
plot(xy.sp, col="khaki")

# example with multiple polygons & text labels
x1=c(0,4,4,0,0)
y1=c(0,0,4,4,0)
x2=c(1,1,3,3,1)
y2=c(-2,-10,-10,-2,-2)
x3=c(6,14,14,6,6)
y3=c(1,1,3,3,1)
xy.sp=SpatialPolygons(list(
	Polygons(list(Polygon(cbind(x1,y1))), ID="test1"), # box
	Polygons(list(Polygon(cbind(x3,y3))), ID="test3"), # wide
	Polygons(list(Polygon(cbind(x2,y2))), ID="test2")  # high
	))
plot(xy.sp, col=bpy.colors(3))
text(polygonsLabel(xy.sp, labels), labels, col = c('white', 'black', 'black'))