Hi,

Just wondering: what is the most efficient way to convert a Math::MPFR object into a Math::GMPq. Ideally, without loosing too much precision.
So far, I tried converting Math::MPFR to Math::GMPf, then Math::GMPf to Math::GMPq, which works correctly, but it's not the best solution in terms of efficiency.

Here is the code for that:

use strict;
use warnings;

use Math::MPFR qw(:mpfr);
use Math::GMPf qw(:mpf);
use Math::GMPq qw(:mpq);

my $PREC = 128;

# Compute log(3)
my $fr = Rmpfr_init2($PREC);
Rmpfr_set_ui($fr, 3, MPFR_RNDN);
Rmpfr_log($fr, $fr, MPFR_RNDN);

print "$fr\n";    # print as float

# MPFR -> floating-point
my $f = Rmpf_init2($PREC);
Rmpfr_get_f($f, $fr, MPFR_RNDN);

# floating-point -> rational
my $q = Rmpq_init();
Rmpq_set_f($q, $f);

print "$q\n";     # print as rational
[download]

I'm pretty sure this can be done better, but just can't figure it how. I would be very happy if someone will point out an easier way for doing this.
Thank you very much and happy holidays!