R help - Apr 2018 - Extracting specified pages from a lattice ("trellis") object.

Suppose that (e.g.) xyplot() returns an object "xxx" with (say) 3
pages.
I would like to extract/plot (print) just one of these pages, e.g.
page 2.

Here's a toy example:

x   <- rep(seq(0,1,length=11),12)
set.seed(42)
y   <- rnorm(3*44)
a   <- rep(letters[1:12],each=11)
dta <- data.frame(x=x,y=y,a=a)
xxx <- xyplot(y~x|a,data=dta,layout=c(2,2))

I would to extract from xxx and print page 2 (the page corresponding to
levels e, f, g and h).

Is there any (simple) way that I can do this?

I've mucked around with update.trellis() and [.trellis, but I cannot 
make head nor tail of the documentation.  The [.trellis method seems to 
work in some situations, but not in others, and since I cannot 
understand what it actually does, I cannot figure out why.

E.g. in my toy example "xxx[5:8]" seems to give me what I want, but in
the context of my real application a similar construction does not work.

Thanks for any insight.

cheers,

Rolf Turner

-- 
Technical Editor ANZJS
Department of Statistics
University of Auckland
Phone: +64-9-373-7599 ext. 88276

Hi

I think the subsetting works by giving you the panels for the 
corresponding levels of the conditioning variable(s).  Note that, if 
there is more than one conditioning variable, you will need more than 
one subsetting index.

For example, taking this plot with two conditioning variables and 12 
panels in total ...

dotplot(variety ~ yield | year * site, data=barley)

... this produces three pages ...

dotplot(variety ~ yield | year * site, data=barley,
         layout=c(2,2))

... and this produces the second page (both panels for the first 
conditioning variable and the third and fourth panels for the second 
conditioning variable) ...

dotplot(variety ~ yield | year * site, data=barley,
         layout=c(2,2))[1:2, 3:4]

Hope that helps

Paul

On 14/04/18 12:25, Rolf Turner wrote:
> 
> Suppose that (e.g.) xyplot() returns an object "xxx" with (say) 3
pages.
> I would like to extract/plot (print) just one of these pages, e.g.
> page 2.
> 
> Here's a toy example:
> 
> x?? <- rep(seq(0,1,length=11),12)
> set.seed(42)
> y?? <- rnorm(3*44)
> a?? <- rep(letters[1:12],each=11)
> dta <- data.frame(x=x,y=y,a=a)
> xxx <- xyplot(y~x|a,data=dta,layout=c(2,2))
> 
> I would to extract from xxx and print page 2 (the page corresponding to
> levels e, f, g and h).
> 
> Is there any (simple) way that I can do this?
> 
> I've mucked around with update.trellis() and [.trellis, but I cannot 
> make head nor tail of the documentation.? The [.trellis method seems to 
> work in some situations, but not in others, and since I cannot 
> understand what it actually does, I cannot figure out why.
> 
> E.g. in my toy example "xxx[5:8]" seems to give me what I want,
but in
> the context of my real application a similar construction does not work.
> 
> Thanks for any insight.
> 
> cheers,
> 
> Rolf Turner
> 
-- 
Dr Paul Murrell
Department of Statistics
The University of Auckland
Private Bag 92019
Auckland
New Zealand
64 9 3737599 x85392
paul at stat.auckland.ac.nz
http://www.stat.auckland.ac.nz/~paul/

On 24/04/18 15:17, Paul Murrell wrote:
> Hi
> 
> I think the subsetting works by giving you the panels for the 
> corresponding levels of the conditioning variable(s).? Note that, if 
> there is more than one conditioning variable, you will need more than 
> one subsetting index.
> 
> For example, taking this plot with two conditioning variables and 12 
> panels in total ...
> 
> dotplot(variety ~ yield | year * site, data=barley)
> 
> ... this produces three pages ...
> 
> dotplot(variety ~ yield | year * site, data=barley,
>  ??????? layout=c(2,2))
> 
> ... and this produces the second page (both panels for the first 
> conditioning variable and the third and fourth panels for the second 
> conditioning variable) ...
> 
> dotplot(variety ~ yield | year * site, data=barley,
>  ??????? layout=c(2,2))[1:2, 3:4]
> 
> Hope that helps.
Hmm.  Thanks Paul.  I may be able to work with that.  But what I really 
wanted was to take

     bar <- dotplot(variety ~ yield | year * site, data=barley)

and then do (something like)

     foo <- bar[<something>]

so that foo contains only the second page of bar, and then do print(foo)
to get a plot of (just) the second page.  Without re-issuing a 
(modified) plot command.

Is that not at all possible?

cheers,

Rolf.

-- 
Technical Editor ANZJS
Department of Statistics
University of Auckland
Phone: +64-9-373-7599 ext. 88276