[R] "non-plot" plotting
Marc Schwartz
marc_schwartz at comcast.net
Fri Jan 18 01:03:20 CET 2008
Johannes Graumann wrote:
> I really do not know ho to else title this ... I want to draw something like
> the attached png with R and would like to poll you on how to start ... make
> an empty plot first and then start positioning the characterstring
> by 'text' and then drawing the lines ...
>
> Joh
Johannes,
Try this (PDF of the output attached):
# Open the initial plot window and set
# up a coordinate system based upon
# placement of the letters centered
# on integer 'x' values 1:12
plot(1, xlim = c(0, 13), ylim = c(0, 4),
type = "n",
ann = FALSE, axes = FALSE)
# Create the vector of letters
Vec <- c("T", "V", "F", "S", "Q", "A", "Q",
"L", "C", "A", "L", "K")
# Plot them
text(1:12, 2, Vec, cex = 2, font = 2)
# Get the height of the letters for spacing
# See ?strheight
height <- strheight("T", cex = 2) * .8
# Set a default width for the "boxes"
# around the letter
width <- 0.5
# Set the values for 'Y's
Y <- 10:2
X <- 3:11
# Loop over the Y's and using plotmath
# plot the values and subscripts in bold
# See ?plotmath and ?bquote
# While looping, do the colored segments
for (i in 1:9) {
text(X[i], 2 + (height * 1.6), bquote(bold(Y[.(Y[i])])),
col = "red", font = 2)
x <- c(X[i] - width, X[i] - width, X[i] + 0.3)
y <- c(2, 2 + height, 2 + height)
lines(x, y, col = "red", lwd = 2)
}
# Same here now for the 'b's
b <- c(2:6, 8, 9, 11)
X <- b
for (i in 1:8) {
text(X[i], 2 - (height * 1.6), bquote(bold(b[.(b[i])])),
col = "blue", font = 2)
x <- c(X[i] - 0.3, X[i] + width, X[i] + width)
y <- c(2 - height, 2 - height, 2)
lines(x, y, col = "blue", lwd = 2)
}
The above should provide the basic approach. You can then adjust/fine
tune spacing, etc. as you require.
HTH,
Marc Schwartz
-------------- next part --------------
A non-text attachment was scrubbed...
Name: plot.pdf
Type: application/pdf
Size: 5063 bytes
Desc: not available
Url : https://stat.ethz.ch/pipermail/r-help/attachments/20080117/473d224f/attachment-0002.pdf
More information about the R-help
mailing list