An Example of “Linking” Graphical Objects in
grid
Paul Murrell
June 23, 2022
Suppose that I am drawing two graphs on a page, which are the results from
two subjects in an experiment. I want the graphs to have the same axes to aid
in comparison of the subjects.
First of all, I will split the page up into two bits for the two graphs.
> pushViewport(viewport(layout = grid.layout(1, 2, respect = TRUE)))
Now I generate some data and draw the first plot.
> x <- 1:10
> y1 <- rnorm(10)
> vp1a <- viewport(layout.pos.col = 1)
> vp1b <- viewport(width = 0.6, height = 0.6,
+ xscale = c(0, 11), yscale = c(-4, 4))
> pushViewport(vp1a, vp1b)
> grid.xaxis(name = "xaxis")
> grid.yaxis(name = "yaxis")
> grid.points(x, y1)
> popViewport(2)
Notice that I have named the graphical objects for the axes. When I draw the
second plot I will use the same names for the axis objects.
> y2 <- rnorm(10)
> vp2a <- viewport(layout.pos.col = 2)
> vp2b <- viewport(width = 0.6, height = 0.6,
+ xscale = c(0, 11), yscale = c(-4, 4))
> pushViewport(vp2a, vp2b)
> grid.xaxis
> grid.xaxis(name = "xaxis")
> grid.yaxis(name = "yaxis")
> grid.points(x, y2)
> popViewport(2)
1
The output looks like the figure below.
4 4
2 2
0 0
−2 −2
−4 −4
0 2 4 6 8 10 0 2 4 6 8 10
Because I have used the same name for the axis objects in both plots, I can edit
the axes for both plots simultaneously rather than having to edit each one in
turn. For example ...
> grid.edit("xaxis", at = c(1, 5, 9), global = TRUE)
>
The output now looks like the figure below.
4 4
2 2
0 0
−2 −2
−4 −4
1 5 9 1 5 9
This might seem a very small gain in this example, but it is potentially of great
use in, for example, the implementation of a scatterplot matrix.
