[Bug-gsl] bug in gsl_linalg_QRPT_decomp2 ?

[Mario Pernici]

gsl version: 1.3
Red Hat 7.3
gcc version 2.96

Dear Sirs,
 from the documentation I understand that
 the norms of each column of R are stored in the vector norm.
 However from the following example it seems that in the vector norm
 are stored the absolute values of the diagonal elements of R.
 Best Regards
  Mario Pernici
 ===========================
#include <stdio.h>
#include <gsl/gsl_linalg.h>
int main() {
  double a_data[] = {1,2,3, 1,5,6, 1,8,9, 1,11,12};
  gsl_matrix_view m = gsl_matrix_view_array(a_data, 4, 3);
  gsl_matrix *q, *r;
  gsl_vector *tau, *norm;
  int signum, i,j;
  gsl_permutation * p = gsl_permutation_alloc (3);
  q = gsl_matrix_alloc(4,4);
  r = gsl_matrix_alloc(4,3);
  tau = gsl_vector_alloc(3);
  norm = gsl_vector_alloc(3);
 gsl_linalg_QRPT_decomp2(&m.matrix, q, r, tau, p, &signum, norm);

  for(i=0; i < 3; i++) {
    printf("%lf\t", gsl_vector_get(norm, i));
  }
  puts("\n");
 for(i=0; i < 4; i++) {
   for(j=0; j <3; j++) {
     printf("%lf \t", gsl_matrix_get(r,i,j));
   }
   printf("\n");
 }
}
================================
Output:
16.431677       0.816497        0.000000

-16.431677      -1.825742       -14.605935
0.000000        -0.816497       0.816497
0.000000        0.000000        0.000000
0.000000        0.000000        0.000000