R help - Mar 2005 - Base and lattice graphics on the same graphics page

Although base graphics does not mix with lattice in the one graph,
I've found that print.trellis(position=..., ) and the use of 
par(fig=...)
to put regular and trellis graphics on the one graphics page works
like a treat, at least in version 2.0.1 of R.  [Base graphics functions
that are themselves inconsistent with par(fig=...) are obviously
disallowed.]

I am wondering whether there are caveats of which I and others
should be aware, or whether there is a risk that the ongoing
development of R's graphics abilities will render such a cohabitation
unworkably fractious.

Example:

gph <- bwplot(voice.part ~ height, data=singer)
print(gph, position=c(0, 0.5, 1, 1)) # x0, y0, x1, y1
par(fig=c(0, 1, 0,0.5), new=TRUE) # x0, x1, y0, y1
boxplot(height ~ voice.part, data=singer, horiz=TRUE)

John Maindonald             email: john.maindonald at anu.edu.au
phone : +61 2 (6125)3473    fax  : +61 2(6125)5549
Centre for Bioinformation Science, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200.

On Tuesday 29 March 2005 22:32, John Maindonald wrote:
> Although base graphics does not mix with lattice in the one graph,
> I've found that print.trellis(position=..., ) and the use of
> par(fig=...)
> to put regular and trellis graphics on the one graphics page works
> like a treat, at least in version 2.0.1 of R.  [Base graphics
> functions that are themselves inconsistent with par(fig=...) are
> obviously disallowed.]
>
> I am wondering whether there are caveats of which I and others
> should be aware, or whether there is a risk that the ongoing
> development of R's graphics abilities will render such a cohabitation
> unworkably fractious.
Paul would know better, but I think that's unlikely. In fact, the 
gridBase package allows you to do use grid (and hence lattice) 
functions to add to a base plot, as well as (I didn't realize this 
before) the other way round. The only caveat is that resizing the 
device may mess things up. 

You may have to be careful with new devices. Your example pasted on a 
fresh session (tested only on r-devel) starts a new page for the 
boxplot since it thinks that new=TRUE doesn't make sense (because 
there's no 'old' plot yet).

Deepayan