"know" inside a function if current plot is base-graphics

Question

Is there a way to know programmatically and non-destructively whether the current graphics device is base graphics?

I need this to be non-destructive, in that it doesn't open a new device, close the current device, or otherwise modify it (e.g., adding points/lines, etc).

For instance,

isbasegx <- function() {
  length(dev.list()) > 0 && # dev.interactive() && 
    !identical(par("usr"), c(0, 1, 0, 1))
}

(Edited to remove dev.interactive(), as this is intended to be useful in report-generation as well.)

That might be a start, but anything that relies on par() is going to fail since grid-based graphics do not update them (this is not a surprise):

plot(1, 1)
par1 <- par()
ggplot(mtcars, aes(mpg, disp)) # same graphics device
par2 <- par()
all(mapply(identical, par1, par2))
# [1] TRUE

What other "hints" at graphics type are there?

Answer 1

The problem is that in R, the graphics device does not have two mutually exclusive "modes" that we can classify as base R graphics or grid graphics. These are just the (?only) two different graphics APIs that we can use to draw pixels onto a plotting device in R. In fact, the following code example demonstrates that it is perfectly possible to mix and match:

library(grid)
  
plot(1:10)
par(new = TRUE)
grid.rect(width = 0.5, height = 0.5, gp = gpar(lwd = 3), name = "gr")
lines(2:6, c(6, 4, 5, 6, 1), col = 'red')

Here, we have drawn a scatterplot in base R graphics, then drawn a rectangle using grid, then drawn a red line over the top in base R graphics again.

When we have drawn the above image, we cannot say that the current graphics device 'is' base R or 'is' grid. It contains graphics primitives that were drawn using calls to both.

---

EDIT

If you are trying to work out whether the last plot was a ggplot or a base R graphics, you could try interrogating the contents of recordPlot, which keeps a record of the calls that drew the graphics primitives. Therefore, you can do something like:

is_it_a_ggplot <- function() {
  'ggplot2' %in% unlist(lapply(recordPlot()[1][[1]], function(x) {
    x <- x[[2]][[1]]
    if('name' %in% names(x)) x$name else as.character(x)
  }))
}

is_it_base <- function() {
  'C_plot_new' %in% unlist(lapply(recordPlot()[1][[1]], function(x) {
    x <- x[[2]][[1]]
    if('name' %in% names(x)) x$name else as.character(x)
  }))
}

library(ggplot2)

ggplot(iris, aes(Sepal.Width, Petal.Length, col = Species)) +
  geom_point()

is_it_a_ggplot()
#> [1] TRUE
is_it_base()
#> [1] FALSE

plot(1:10)

is_it_a_ggplot()
#> [1] FALSE
is_it_base()
#> [1] TRUE