[Rd] order() fails on a chr object of class "AsIs" with "\265" in it

[Prof Brian Ripley]

On Fri, 15 Jan 2010, Don MacQueen wrote:

> Here's an example (session info at the end).
>
>>  tmpv <- c('\265g/L','Bq/L')
>>  order(tmpv)
> [1] 2 1
>>  tmpv <- I(tmpv)
>>  order(tmpv)
> Error in if (xi > xj) 1L else -1L : missing value where TRUE/FALSE needed
>>  foov <- gsub('\265','',tmpv)
>>  order(foov)
> [1] 2 1
>>  str(tmpv)
> Class 'AsIs'  chr [1:2] "\265g/L" "Bq/L"
>>  str(foov)
> Class 'AsIs'  chr [1:2] "g/L" "Bq/L"
>
> I can easily work around this in my scripts, but shouldn't order() succeed 
> with such an object?

Not in the C locale.  There is no pre-defined ordering for non-ASCII 
characters in that locale and the string is invalid in a strict C 
locale.

> (I suppose this could be Mac-specific, but I'm assuming it's not...)

No, but the handling of invalid strings in C is OS-specific.

> For context:
> The character "\265" causes the Greek letter mu to be displayed in various 
> output devices. For example, the character vector eventually gets written to 
> an html file, which when displayed in Firefox (Mac) is displayed as Greek mu. 
> Also in Excel 2004 (Mac).
>
> I first wrote these scripts 6 years ago, when "\265" was a way I could find 
> to display the Greek mu in output text files of various kinds. They worked as 
> recently as 3 months ago. Maybe there's a better way now to display a mu in 
> text-based contexts?

Use UTF-8 and Unicode \u03BC 
(http://www.alanwood.net/unicode/greek.html).

The issue is that you need a xtfrm method for 'AsIs': it falls back to 
comparisons via .gt and those (correctly) fail.

xtfrm.AsIs <- function(x) xtfrm(unclass(x))

would keep get you going until you fix the scripts.

>
>>  sessionInfo()
> R version 2.10.1 (2009-12-14)
> i386-apple-darwin8.11.1
>
> locale:
> [1] C
>
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
>
> Thanks
> -Don
> -- 
> --------------------------------------
> Don MacQueen
> Environmental Protection Department
> Lawrence Livermore National Laboratory
> Livermore, CA, USA
> 925-423-1062
>
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595