ggplot2: faceting on a function of column

Question

Couldn't find an answer to this -- in ggplot2, is it possible to facet on a function of a column, rather than on the value of the column directly?

Simple reproducible example:

Sample data:

df=data.frame(dat=c(1,2,5,5,7))

This works:

ggplot(df, aes(x=1:5, y=dat, colour=factor(dat > 3))) + 
       geom_point() + facet_grid(dat ~ .)

This does not:

ggplot(df, aes(x=1:5, y=dat, colour=factor(dat > 3))) + 
       geom_point() + facet_grid((dat > 3) ~ .)

One solution is to add a column just for the facet. This works:

df$facet=df$dat>3
ggplot(df, aes(x=1:5, y=dat, colour=factor(dat > 3))) + 
       geom_point() + facet_grid(facet ~ .)

But is there a way to do it without having to add a new column to the data.frame?

Answer 1

I've come up with a compromise solution, which is a trade-off between arguments of @Nathan and @Яaffael posted in the comment section.

FacetingFunction <- function(df) {df$dat > 3}
ArbitraryFacetingPlot <- function(df, FacetingFunction) {
  df$facet <- FacetingFunction(df)
  p <- ggplot(df, aes(x=1:5, y=dat, colour=factor(dat > 3))) + 
    geom_point() + 
    facet_grid(facet ~ .)
  df$facet <- NULL
  p
}

ArbitraryFacetingPlot(df, FacetingFunction)
ArbitraryFacetingPlot(df, function(df) {df$dat==5})