I don't understand at all what the problem is, but a workaround
is simply not to specify the "pdf" argument in "name=value" form
but rather in positional form. I.e. just make it the first argument
to .Fortran(). As in:
tmp <- .Fortran("pdf", <other parameters here>)
Works for me! :-)
cheers,
Rolf
On 11/09/12 13:21, Peter Dunn wrote:
Hi all
I've been getting some emails from the R package maintainers that I need
to update some code in a CRAN packge that uses FORTRAN, to comply with
(not so recent) changes. I've been a little busy...
I'm having trouble adjusting my code. I hope someone can help.
The package was working fine, and a few R functions in my package had
lines like this:
tmp <- .Fortran( name="pdf", <other parameters here>)
and similar. The "pdf" is the name of the FORTRAN subroutine being
called, of course.
When I replace that line with
tmp <- .Fortran( .NAME="pdf", <other parameters here>)
...the package then does not pass the checks. The library package builds
OK with no errors or warnings, and then during "R CMD check" the examples
give this error:
out <- tweedie.profile( y~1, p.vec=seq(1.1, 1.9, length=9),
do.plot=TRUE)
1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9
.Error in .Fortran(.NAME = "pdfun", as.double(power),
as.double(phi[i]/(y[i]^(2 - :
supplied argument name '.NAME' does not match 'name'
Calls: tweedie.profile -> dtweedie.inversion -> .Fortran
Execution halted
I've had a search--including a read of the "Writing R Extensions" manual
on CRAN (I did read it, but interestingly a search of that document did
not reveal any instance of .NAME)--and I cannot see the problem. It's
probably obvious.
Any help appreciated, even pointers to more info.
<SNIP>
______________________________________________
R-help@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.
