[Rd] Possible bug in R 2.6.1 (PR#10565)

[ripley at stats.ox.ac.uk]

On Fri, 11 Jan 2008, galway at rand.org wrote:

> Colleagues,
>
> =20
>
> In using the paste command I have to spell out the collapse option:
>
> =20
>
>> paste(1:3,coll=3D"a")
>
> [1] "1 a" "2 a" "3 a"
>
>> paste(1:3,collapse=3D"a")
>
> [1] "1a2a3"
>
> =20
>
> My understanding is that the abbreviation coll should be adequate.
> Actually, even collaps isn't enough:

Your understanding is wrong.  The help page says

Usage:

      paste(..., sep = " ", collapse = NULL)

Argument names after ... cannot be abbreviated ('S Programming' p.40, 
amongst other places).

I do often wonder why people are 'sure you know for certain' (to quote the 
FAQ) that something as elementary as this would be a bug for so many 
years. It indicates a lack of respect for the R developers.


>
> =20
>
> paste(1:3,collaps=3D"a")
>
> [1] "1 a" "2 a" "3 a"
>
> =20
>
> LG
>
> =20
>
> Lionel Galway, Ph.D.
>
> Senior Statistician
>
> The RAND Corporation     voice: (+1) 310-393-0411 x7957
>
> 1776 Main St.                  fax:    (+1) 310-393-4818
>
> Santa Monica, CA 90407 USA
>
> email:  Lionel_Galway at rand.org <mailto:Lionel_Galway at rand.org>=20

-- 
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