[R] wrapping lattice xyplot

[hadley wickham]

On 6/8/07, Zack Weinberg <zackw at panix.com> wrote:
> This is an expanded version of the question I tried to ask last night
> - I thought I had it this morning, but it's still not working and I
> just do not understand what is going wrong.
>
> What I am trying to do is write a wrapper for lattice xyplot() that
> passes a whole bunch of its secondary arguments, so that I can produce
> similarly formatted graphs for several different data sets.  This is
> what I've got:
>
> graph <- function (x, data, groups, xlab) {
>   g <- eval(substitute(groups), data, parent.frame())
>
>   pg <- function(x, y, group.number, ...) {
>     panel.xyplot(x, y, ..., group.number=group.number)
>     panel.text(2, unique(y[x==2]),
>                levels(g)[group.number],
>                pos=4, cex=0.5)
>   }
>
>   xyplot(x, data=data, groups=substitute(g),
>               type='l',
>               ylab=list(cex=1.1, label='Mean RT (ms)'),
>               xlab=list(cex=1.1, label=xlab),
>               scales=list(
>                 x=list(alternating=c(1,1), tck=c(1,0)),
>                 y=list(alternating=c(1,0))
>                 ),
>               panel=panel.superpose,
>               panel.groups=pg
>               )
> }
>
> "pg" is supposed to pick "g" up from the lexical enclosure. I have no
> idea whether that actually works, because it never gets that far.  A
> typical call to this function looks like so:
>
> > graph(est ~ pro | hemi, sm, obs, "Probe type")
>
> (where 'sm' is a data frame that really does contain all four columns
> 'est', 'pro', 'hemi', and 'obs', pinky swear) and, as it stands above,
> invariably gives me this error:
>
> Error in eval(expr, envir, enclos) : object "est" not found
>
> I tried substitute(x) (as that seems to have cured a similar problem
> with "g") but then x is not a formula and method dispatch fails.
>
> Help?
> zw

It's not lattice, but ggplot2, http://had.co.nz/ggplot2, is designed
to make this easy because you don't have to specify the data set when
creating the plot. e.g.

install.packages("ggplot2", dep=T)
library(ggplot2)

# This is an abstract definition of a plot - it doesn't have any data yet
p <- ggplot(mapping = aes(x=cyl, y=mpg)) + geom_point() +
geom_smooth(method="lm")

mt2 <- mtcars * 2
mt3 <- as.data.frame(mtcars ^ 2)

# Add datasets
p %+% mtcars
p %+% mt2
p %+% mt3
# (the syntax isn't great, but you get the idea)

# Or even changing the default mapping from data to visual properties
p %+% mt3 + aes(x = mpg, y=wt)

Obviously, you can do even more within a function, and the aes call is
relatively easy to create programmatically (although not well
documented currently, so please ask me for more details if you are
interested).

Hadley