[Rd] Difference between .C and .Fortran (on Windows)

Prof Brian Ripley ripley at stats.ox.ac.uk
Thu Apr 24 16:42:34 MEST 2003


On Thu, 17 Apr 2003, Duncan Murdoch wrote:

> In updating my web page on using non-standard compilers in Windows,
> I've been trying to figure out exactly what the differences are
> between .C() and .Fortran(), but the code is pretty convoluted. 
> 
> Can someone let me know if the following is correct?
> 
> There are two differences between .C("foo", ...) and .Fortran("foo").
> 
>  1.  .C looks for the symbol "foo" in the external library, whereas
> .Fortran("foo", ...) looks for the symbol "foo_" (which is how g77
> would export the subroutine "foo").

There are symbol.C() and symbol.For() to show you the mappings used.
Not all systems appends an underscore for Fortran, and some
case-convert the Fortran names (to lower, usually but not always).
R does not currently cope with the latter, so there is an assumption
that Fortran symbol names are written in lower case.  (That precludes
some Fortran compilers.)

I don't think do_symbol is convoluted, BTW!

>  2.  .C passes character mode vectors as a pointer to an array of
> pointers to the strings, whereas .Fortran just passes a pointer to a
> 255 character buffer containing the first string.  In both cases the
> strings are null-terminated.
> 
> Are there other differences?  Are these platform-dependent?

As Thomas noted, arbitrary language objects can be passed through .C.
Also, .Fortran does analogous things on returned strings (which may not be 
null-terminated, BTW).

-- 
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