[Rd] Getting the name of a SEXP type in package code?

Prof Brian Ripley ripley at stats.ox.ac.uk
Wed Nov 8 08:05:06 CET 2006


type2char is itself a fairly recent addition (there is also type2str, 
which is older) that I found useful for improving messages in R itself.

I have not encountered such a need in package code, but the function is 
exported.

This is not a request to move type2char from Defn.h to Rinternals.h, and I 
don't understand why not: what else is needed?


On Tue, 7 Nov 2006, Martin Maechler wrote:

>>>>>> "Seth" == Seth Falcon <sfalcon at fhcrc.org>
>>>>>>     on Tue, 07 Nov 2006 06:55:06 -0800 writes:
>
>    Seth> Hi,
>    Seth> Is there an alternative to type2char that is available for use in
>    Seth> package code?  It would be useful for producing more informative error
>    Seth> messages.
>
> I haven't looked closely,  but I vaguely remember that I had
> thought in the past,
> (a version of)  type2char() should be made part of R's API.
> Exactly because of the reason you give.
>
> A more sensible name for the exported function would probably be
> R_type2char    or
> Rtype_to_char  or
> Rtype_as_char

It is exported as Rf_type2char.

> If you are going to provide a patch (against R-devel),
> you'd ideally also include the corresponding entry in
> doc/manual/R-exts.texi ... (:-)
>
> Martin
>
> PS: I will be mostly absent from e-mail for the rest of the week.
>
> ______________________________________________
> 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



More information about the R-devel mailing list