mirror/guile
mirror/guile
1
Fork 0
mirror of https://git.savannah.gnu.org/git/guile.git synced 2025-06-11 22:31:12 +02:00
Code Activity

(scm_i_divide): For big/big wanting inexact, use mpq_get_d

Browse source 
rather than converting to doubles, to avoid inf or nan when the inputs
are too big for a double but the quotient does fit.  This affects
conversions exact->inexact of big fractions.
This commit is contained in:
Kevin Ryde 2006-05-09 00:25:11 +00:00
parent 101bd61cf5
commit 65581bc84d
1 changed files with 27 additions and 22 deletions
Download patch file Download diff file Expand all files Collapse all files

49
libguile/numbers.c
View file

@ -4779,28 +4779,33 @@ scm_i_divide (SCM x, SCM y, int inexact)
else
{
/* big_x / big_y */
int divisible_p = mpz_divisible_p (SCM_I_BIG_MPZ (x),
SCM_I_BIG_MPZ (y));
if (divisible_p)
{
SCM result = scm_i_mkbig ();
mpz_divexact (SCM_I_BIG_MPZ (result),
SCM_I_BIG_MPZ (x),
SCM_I_BIG_MPZ (y));
scm_remember_upto_here_2 (x, y);
return scm_i_normbig (result);
}
else
{
if (inexact)
{
double dbx = mpz_get_d (SCM_I_BIG_MPZ (x));
double dby = mpz_get_d (SCM_I_BIG_MPZ (y));
scm_remember_upto_here_2 (x, y);
return scm_from_double (dbx / dby);
}
else return scm_i_make_ratio (x, y);
}
if (inexact)
{
/* It's easily possible for the ratio x/y to fit a double
but one or both x and y be too big to fit a double,
hence the use of mpq_get_d rather than converting and
dividing. */
mpq_t q;
*mpq_numref(q) = *SCM_I_BIG_MPZ (x);
*mpq_denref(q) = *SCM_I_BIG_MPZ (y);
return scm_from_double (mpq_get_d (q));
}
else
{
int divisible_p = mpz_divisible_p (SCM_I_BIG_MPZ (x),
SCM_I_BIG_MPZ (y));
if (divisible_p)
{
SCM result = scm_i_mkbig ();
mpz_divexact (SCM_I_BIG_MPZ (result),
SCM_I_BIG_MPZ (x),
SCM_I_BIG_MPZ (y));
scm_remember_upto_here_2 (x, y);
return scm_i_normbig (result);
}
else
return scm_i_make_ratio (x, y);
}
}
}
else if (SCM_REALP (y))