[R] Issue with Rmath standalone (dwilcox, pwilcox, qwilcox, rwilcox) with unif_rand()

Mark Clements m@rk@c|ement@ @end|ng |rom k|@@e
Tue Feb 14 14:37:36 CET 2023


I may have found a bug in the Rmath standalone library when providing a user-defined random number generator (for context, I was investigating using Mersenne Twister with Rmath). The following C code does not compile using the Makefile:

// test_fails.c
#define MATHLIB_STANDALONE
#include <Rmath.h>
#include <stdio.h>

double unif_rand(void) { return 0.5; }

int main() {
  printf("%f\n", runif(0, 1));
  printf("%f\n", pnorm(1.96, 0, 1, 1, 0));
  printf("%f\n", dwilcox(100, 10, 20, 0));
  return 0;
}

// Makefile
LIBRMATH_A = $(shell pkg-config --variable=libdir libRmath)/libRmath.a
test_fails:
      gcc -o test_fails test_fails.c ${LIBRMATH_A} -lm

The error is "multiple definition of `unif_rand'". The code compiles and runs correctly after commenting out either (a) the unif_rand function or (b) the call to dwilcox(). I found similar issues with pwilcox, qwilcox and rwilcox. Can anyone explain this behaviour, please?

Sincerely, Mark.


När du skickar e-post till Karolinska Institutet (KI) innebär detta att KI kommer att behandla dina personuppgifter. Här finns information om hur KI behandlar personuppgifter<https://ki.se/medarbetare/integritetsskyddspolicy>.


Sending email to Karolinska Institutet (KI) will result in KI processing your personal data. You can read more about KI’s processing of personal data here<https://ki.se/en/staff/data-protection-policy>.


More information about the R-help mailing list