Add Float::INFINITY and Float::NAN.

* numeric.c (Init_Numeric): Add Float::INFINITY and Float::NAN.
  [ruby-dev:1657] [ruby-dev:4760] [ruby-list:7023] [ruby-list:46690]
  [ruby-core:26632] [ruby-talk:41352] [ruby-talk:203333]

* include/ruby/defines.h (INFINITY): defined.

* include/ruby/defines.h (NAN): defined.

* include/ruby/util.h (ruby_div0): removed.

* numeric.c (fix_pow): use INFINITY and NAN instead of ruby_div0(1.0).

* marshal.c (r_object0): ditto.

* bignum.c (big_fdiv): ditto.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@26197 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

1 files changed, 1 insertions, 1 deletions

diff --git a/bignum.c b/bignum.c
index f6fca9f546..7489ba5231 100644
--- a/bignum.c
+++ b/bignum.c
@@ -2485,7 +2485,7 @@ big_fdiv(VALUE x, VALUE y)
 #if SIZEOF_LONG > SIZEOF_INT
 	{
 	    /* Visual C++ can't be here */
-	    if (l > INT_MAX) return DBL2NUM(ruby_div0(1.0));
+	    if (l > INT_MAX) return DBL2NUM(INFINITY);
 	    if (l < INT_MIN) return DBL2NUM(0.0);
 	}
 #endif