1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
|
\name{thigmophobe.labels}
\title{Place labels away from the nearest point}
\usage{
thigmophobe.labels(x,y,labels=NULL,text.pos=NULL,...)
}
\alias{thigmophobe.labels}
\arguments{
\item{x,y}{Numeric data vectors or a list with two components.
Typically the x/y coordinates of plotted points.}
\item{labels}{A vector of strings that will be placed adjacent to
each point. Defaults to the indices of the coordinates.}
\item{text.pos}{An optional vector of text positions (see
\link{text}).}
\item{...}{additional arguments are passed to \samp{text}}.
}
\description{
\samp{thigmophobe.labels} places labels adjacent to each point,
offsetting each label in the direction returned by
\samp{thigmophobe}.
}
\details{
Typically used to automatically place labels on a scatterplot or
similar to avoid overlapping labels. \samp{thigmophobe.labels}
will sometimes place a label off the plot or fail to separate
labels in clusters of points. The user can manually adjust the
errant labels by running \samp{thigmophobe} first and saving
the returned vector. Then modify the position values to place
the labels properly and pass the edited vector to
\samp{thigmophobe.labels} as the \samp{text.pos} argument. This
takes precedence over the positions calculated by \samp{thigmophobe}.
\samp{thigmophobe} will fail with only two labels, as it can't figure out
the nearest neighbors. If you really want to use this with two labels, just
eyeball the plot and work out in which direction the labels will go. Then
pass the directions to \samp{thigmophobe.labels} as the \samp{text.pos}
argument.
Both \samp{pointLabel} in the \pkg{maptools} package and \samp{spread.labs}
in the \pkg{TeachingDemos} package use more sophisticated algorithms to
place the labels and are worth a try if \samp{thigmophobe} just won't get it
right.
}
\value{
A vector of directions away from the point nearest to each point.
}
\keyword{misc}
\author{Jim Lemon (thanks to Stephen Milborrow for finding the single point bug
and Erik Aronesty for finding the two point problem.)}
\seealso{\link{thigmophobe}, \link{text}}
\examples{
x<-rnorm(20)
y<-rnorm(20)
xlim<-range(x)
xspace<-(xlim[2]-xlim[1])/20
xlim<-c(xlim[1]-xspace,xlim[2]+xspace)
ylim<-range(y)
yspace<-(ylim[2]-ylim[1])/20
ylim<-c(ylim[1]-yspace,ylim[2]+yspace)
plotlabels<-
c("one","two","three","four","five","six","seven","eight","nine","ten",
"eleven","twelve","thirteen","fourteen","fifteen","sixteen","seventeen",
"eighteen","nineteen","twenty")
plot(x=x,y=y,xlim=xlim,ylim=ylim,main="Test thigmophobe.labels")
# skip the almost invisible yellow label, make them bold
thigmophobe.labels(x,y,plotlabels,col=c(2:6,8:12),font=2)
}
|
