[R] Variable scope R 2.6.1

[John Fox]

Dear Stefan,

I don't think that your question was answered.

If you invoke the formula method for splom(), then your function works; that
is, you can use

	splom(~as.data.frame(rbind(data, outliers)), . . . .

It looks to me as if this is a scoping problem produced by the way in which
splom.data.frame() re-invokes the generic splom():

> lattice:::splom.data.frame
function (x, data = NULL, ...) 
{
    ccall <- match.call()
    if (!is.null(ccall$data)) 
        warning("explicit 'data' specification ignored")
    ccall$data <- list(x = x)
    ccall$x <- ~x
    ccall[[1]] <- quote(lattice::splom)
    eval.parent(ccall)
}

(I'm copying this response to  Deepayan Sarkar, in case he wants to
comment.) 


Finally, the warnings reflect an incorrect call to trellis.par.set(); you
should use

   trellis.par.set(background=list('white'))

I hope this helps,
 John 

--------------------------------
John Fox, Professor
Department of Sociology
McMaster University
Hamilton, Ontario, Canada L8S 4M4
905-525-9140x23604
http://socserv.mcmaster.ca/jfox

> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of Stefan Eichenberger
> Sent: January-01-08 6:50 AM
> To: r-help at r-project.org
> Subject: [R] Variable scope R 2.6.1
> 
> 
> I have the following procedure which worked just fine for in R 2.2.0.
> Recently I upgraded to 2.6.1 and now get an error:
> 
>   > ScatterOutlier(pass_500_506[1:1000,6:12], marginal_500_506[,6:12])
>    Error in eval(expr, envir, enclos) : object "out" not found
> 
> Note that I use the same workspace (and hence data) as in 2.2.0.
> When I make sure that the object "out" exists at console level
> by executing
> 
>    out <- c(rep('norm', nrow(pass_500_506[1:1000,6:12])),
> rep('outliers',
>    nrow(marginal_500_506[,6:12])))
> 
> the procedure seemingly works but spits out a few warnings:
>    Warning messages:
>    1: In trellis.par.get("background")$col :
>      $ operator is invalid for atomic vectors, returning NULL
>    2: In trellis.par.get("background")$col :
>      $ operator is invalid for atomic vectors, returning NULL
> 
> Any ideas? Have rules for scoping changed?
> 
> Here is the full procedure:
> 
> ScatterOutlier <- function(data, outliers, ...) {
>    # splot of rbind(data, outliers) is perfomed; outliers are
> highlighted
>    require(lattice)
>    out <- c(rep('norm', nrow(data)), rep('outliers', nrow(outliers)))
>    trellis.par.set(background='white')
>    cols <- c('black', 'red')
>    syms <- c('*', '*')
>    splom(as.data.frame(rbind(data, outliers)), groups=out, col=cols,
>          pch=syms,
>          panel = panel.superpose,
>          key = list(title = paste(deparse(substitute(data)), '; ',
>                     deparse(substitute(outliers)), sep=''),
>                     columns = 2,
>                     points = list(pch = syms, col = cols),
>                     text = list(c("norm", "outlier"))))
> }
> 
> Thanks for your help!
> 
> Kind regards                            Stefan
> 
> --
> _________________________________________________________________
> Stefan Eichenberger
> Brammenfeld 13
> D-47533 Kleve / Germany
> 
> 
> _________________________________________________________________
> 
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.